CPUC Workshop - 2021-03-23
No chapters available yet
Legend
Transcript markers
- text Sentence being played
- text Search result
- text Intervention cited by the link you opened
Click a sentence to play the video from that moment. Select text to quote it with the speaker, timestamp and link.
More options
California Public Utilities Commission
Follow its sittingsSpeaker 2
And this last question, sentence around PSPS or de energization events. So how will initiatives reduce the risk of emissions, wildfire spread and impact to the decision to use PSPS? This last question is embedded in the previous three questions and is not discussed as a separate item here. Before I hand it over to our moderator for the first portion of the workshop, I'll go over a few meeting logistics. So just switch to the next slide.
Thanks. So most participants are in listen only mode. That means you cannot speak but you are still welcome to interact and ask questions using the chat feature. Please do not use the Q and A feature. The chat and the Q and A tools are in the lower right hand corner of the WebEx screen.
Again, only use the chat feature to ask technical questions. The designated WebEx host is a technical resource and is not a member of the Wildfire Safety Division. Please only direct a chat message to the host to resolve the technical issue. If you wish to ask questions, please select all panellists in the chat function. Someone from the Wildfire Safety Division will be monitoring posts.
Again, please don't use the Q and A feature. We will not be monitoring that. We will route questions as appropriate to panellists. Finally, we have asked several stakeholder organisations to serve as panellists. They will be given an opportunity to ask questions directly of the utilities.
If you just flip to the next slide. As an example, in the Q and A during the risk assessment and mapping section, stakeholders designated as panellists may use the raise hand function in order to ask questions. The moderator will coordinate those questions. Everyone else, please use the chat function as described earlier. Depending on the number of questions received, we may not be able to cover all of them, but we'll do our best.
Right, and with that, I'll hand it over to Alan Wu to moderate our third portion of today's workshop. Alan?
Speaker 3
Hi, thank you for the introduction, Lucy. Hi, everyone. Good morning. My name is Zhao Wu and I'm a wildfire safety analyst with the mitigation branch of the Wildfire Safety Division. I'll be your moderator for the upcoming session.
Wildfires continue to be a big threat to the environmental, social and economical well-being for the State of California and wildfire risks or wildfire risk models will play an important part of the battle by being the threat that holds DWMP together. Risk models should show the utilities where their risks lie now, where their risks will be in the future, and also act as a guiding compass to pinpoint where resources should be distributed.
Speaker 3
will focus on risk assessment, mapping and resource allocation methodology. The three SMJ utility companies will describe the risk modeling efforts, but more specifically, how the models have progressed since the 2020 WMP and how the utility company incorporates risk models with decision making. We will present in the order of Pacificor, Liberty and Bear Valley. Members of the panel, please reserve all questions for the Q and A session. And members of the audience, feel free to submit questions via chat throughout the presentation and Coco Tomassian, our chat moderator will read them during the Q and A session.
All right, Pacifico, we'll start with you. Thank you.
Speaker 5
Good morning. This is Heidi Caswell. I don't know if you're going to bring up the slide deck. Can we bring up the slide deck?
Speaker 5
Pacific Core. It's the one with risk at the end of the title. Thank you. So good morning. My name is Heidi Caswell.
I'm the Director of T and D Asset Performance and Wildfire Mitigation at Pacific Corp. And I'd like to discuss how we have moved forward with our risk assessment process. I'll try to tie it to the specifics around the risk mapping activity as well as locational identification of risks and as well as then tying it to the specific initiative that reduce that risk. I have some kind of feedback going on with my That phone. Don't
Speaker 5
Thank you. So, next slide. So, PACIFIC CORE has made substantial progress in our risk assessment methodology. As pointed out in our 2020 feedback or the feedback received from a 2020 WMP, we were fairly focused on the use of the high fire threat district and the various tiers And we're inspired to really advance how we looked at locational risk. Like many peer utilities, especially those that discuss their processes through the large three investor owned utilities.
We've been using a lot of data science, machine learning and other kind of more advanced techniques to try and model the risk with data sets that we have, incorporate data sets that previously we haven't incorporated and then try to extend that into kind of the what if to be able to then serve against the as an input to selecting mitigation initiatives and prioritizing around those. Again, previously we were applying the designation of the High Fire Threat District as an indicator of, know, should this circuit or these customers served by the circuit be part of the elevated and more highly prioritized activities or not. And with this update, we've made some substantial strides to advance beyond that. We also previously looked at seasonal impact of ignition probability and had less focus on risk event drivers. In other words, many we looked at unplanned outages as very similar risks and with this update have tried to really align more specifically against specific types of risk events and understand the why, the when, the how, the where kind of aspect to working against them.
As one of the large elements incorporated into our risk assessment, we look at the vegetation canopy as well at a zonal level and we quantify what we call utility risk, which is the result of those unplanned outages, vegetation exposure, recorded history, and then a probabilistic ARC energy calculation. In the future, we anticipate looking at elements such as equipment types, some specific components that may be more prone to that catastrophic failure mode, as well as aspects around asset health and getting better precision and granularity about some of the other inputs like vegetation. Next slide. This graphic is trying to illustrate for you the concept that's behind our risk assessment model which we've called the localized risk assessment model, LRAM, in our WMT. And you can see I've tried to make sure that you can link the different graphics back to the WMP so that there's good reference.
Some of these are eye tests and I apologize for that. This graphic is showing you that what we take is the basic service territory map, so terrain and, you know, land bases behind it. We overlay on top of that our circuit topology. We consider the historic fire weather risk fire spread risk. So that would be as expressed in the past development of the HFTD, the integrated utility threat index.
We overlay our tree canopy risk, modeling results from ARC Energy calculations done through our sign modeling, utility ignition fault risk, so those are the unplanned outages, and utility fires and equipment that comes from the fire database. And then fire weather risk, which is a more contemporary view around fire weather. So, the last four years at a more tightly gridded value. So, that forms for us both a long term view and a recent view with the topology and with the characteristics and performance that have recently been experienced within each of those circuits segments. And a zone of protection, that core that we're considering, the modelable core is from an isolation point like a fuse or breaker, recloser, that kind of device to the next downstream.
So all of the conductor and the equipment elements that are within that protective zone are considered one ZOP for modeling purposes and that's how we are then quantifying and calculating various aspects up along that segment. Within the state we have, if my memory serves right, 2,700 or so zones of protection where we have 70 some circuits. So just to help you understand the segmentation that's going on. The next graphic actually depicts the model results as of our most recent model run. And what you it's this is for illustrative purposes.
This really is a model output. If you look at this graphic, the combined fire risk is expressed by the width and the intensity, the darkness of the blue. So if you see certain areas where there are a very lightly revealed blue and not very thick, that would be the lower fire, the lower combined fire risk area. As and the corollary to that is the areas that are thicker and darker are the ones that are the higher fire risk. And again, this is the combination of what we call the utility risk blended with the climate impact.
It has that consideration of vegetation, outage performance, ARC energy, and then the climatology in the area. The next graphic or next slide, if you wouldn't mind. I wanted to actually walk through the process that we went through so that that map maybe starts to make a little bit more sense or that model output makes a little more sense. And it really is a five step process. We look at our outage data and analyze it, categorize it to look at that risk event driver.
We then look at risk events by the probability. And we also then look at fire records against the risk event drivers. Then the next step is to look at the initiatives that target against specific risk event drivers and then prioritize those mitigation efforts using a variety of measures including, you know, customers impacted, combined risk, utility risk, and so it's kind of like a multidimensional view that happens at the very end. So, this first graphic shows you how we map our outage cause code data and some of the data mining that we have to do within those records in order to align them against the risk events and the drivers. So we've got contact from object, we've got equipment failure, we've got some unknown, some human errors, all of those.
And we basically repurposed what would be traditional categorization of our outage causes in order to align against the risk events that are relative to fire risks. On the next slide, we then look at on the number two shows you where we analyze it by cause category. And you can see that forty two percent of the risk events during the twenty fifteen through twenty twenty fire season time period are the result of equipment failure and twenty two percent of the total is associated with contact from object. It's important for me to note that sometimes in equipment failure it's actually a contact from an object that causes an equipment failure like a tree branch breaks something. And so we attempt to separate recognize there can be a little bit of fuzziness within that blue wedge, big blue wedge of the pie.
On in pie chart that's designated number three, we take the fire utility fire database and go backwards with that and see that 47% of the fires, and it is a sparse data set, but 40% of the fires that we've records of are associated with equipment failure, while 41% of them are associated with contact from objects. So those two continue to kind of stand out as the areas to be responsive to. And while not part of our WMP, we were asked to actually score the risk events. And so that's actually on the right side we identify during fire season and non fire season the high, medium high, medium, medium low and low values for each of those risk event drivers versus the FHIR risk expectation. On the next slide, we try to align again back against the risk event driver the different mitigations that are part of our thought process when we're looking at them.
And this is actually in the WMT. It shows up as three separate tables just to make it readable in the document. The first series are focused around the ARC energy conductor damage and then the equipment failure piece and then the asset hardening piece. Sorry, equipment failure and vegetation management. So what you can see is that hopefully if there's an X in the box, it indicates that that mitigation can be, that mitigation measure can be effective against that risk event driver.
So when you would look at as an example, the animal contact has several ways that animal contact can be mitigated against. It can be improved by coordinating protective equipment, by additional protective equipment, by a variety of hardening activities that are over on the right side of the table and such, but that's really to try and identify for individuals that we don't have a one size fits all solution for how you might mitigate and it really is very targeted against the specifics. So there's, as you can see here, animal guarding, spreading the construction, enhancing the insulation, covering conductor, all of those are possible for animal contact, which again is contact from object under the risk event driver. So, that's how we look at the initiative in order to understand what kind of solutions might resolve that specific driver. On the next slide, the focus here is on prioritization.
So, we understand that there is a risk and what the elements of risk are, so when we see the zone of protection, we can understand is it a vegetation contact issue, is it history around a specific type of outages, what is the probability of any kind of arc energy issue if a fault event were to occur, we then are given the ability to see the zones of protection within each of the circuits. And in that case, in the upper left chart, which is shown on page 75 in our WMP, the Figure four-five-fourteen. This actually shows you in kind of a histogram sort of view against the combined score, which is the expression of the combination of utility and climate risk. Each of the zones of protection performance for every circuit. So you can see that there are a couple of circuits that have kind of some outlier zones of protection, which then would mean that in order to improve fire risk as well as possible, you would attack those specific ones.
You can also see that there are certain circuits that have very similar characteristics for the zones of protection, which then would suggest that since they are at perhaps an elevated level, all of that particular circuit is worthy of quick reaction from the mitigation perspective. The next chart to the right looks at the fire weather score. So this is really the climatology and the combined score. So that helps us understand those areas where it's not the utility risk that maybe is driving the situation. It's actually the location.
So that helps us understand what about it is the where. And you can see that we had circuits that had projects that were prioritized and certain circuits that previously had not yet been prioritized for some of the mitigation measures which were revealed with this view of the data as opposed to the expression of the HFPD and our PSPS plan. So it gives us another way to validate and modify the prioritization of our work now that we see it in this through this lens. In the upper right, we can use that view to help us understand the combined score risk across the bottom against the impact to customers. And what I failed to note is that the size of the bubble is the circuit length or indicates the circuit length.
So a small bubble is a small circuit is a short circuit short and length circuit, whereas a larger bubble is one that has more circuit mileage to be dealt with. And so we can start to understand both the customer piece as well as then by the big or little circles how exhaustive our efforts would need to be versus the risk that we would be buying down essentially with the mitigation efforts. So that there are a variety of other ways that we also look at kind of the alignment of our mitigation measures against the sections of circuits within our service territory. And those are also expressed further in the section 4.5.1 in the WMP. But this was just to try and help you work help work you through that thought process.
And later on we'll talk a little bit of or other panelists will talk about how we apply this for the asset mitigation and vegetation management activities that are also important parts of delivering the WMP.
Speaker 3
And Heidi, this is Alan. We do have to be aware of the time, so two minute warning here.
Speaker 5
And I'm actually, if you go to the next slide, I can say thank you and give you back two minutes. Sorry, I just talked Thank too
Speaker 3
you so much, Heidi. And Joe, next we're going to have Liberty Utilities present. Thank you.
Speaker 7
All right. I'll do a short introduction here. My name is Greg Campbell, and I'm a senior analyst in our Regulatory Affairs division. And I'm here to talk about Liberty's risk assessment, mapping and resource allocation methodology.
Speaker 7
So our progress from the 2020 WMP toward 2021 WMP, Liberty has made some significant strides. And coming into 2021, Liberty was the only utility who had not filed a risk based decision making framework as part of its GRC filing. And this includes how Liberty evaluates welfare risk. So, our first generation wildfire risk models were completed 02/20/2021. So, and this is adopted the same approach as what the larger IOUs used in their MAVF in MARS risk assessment methodology.
And so we used this approach and this approach replaced our old approach, which mainly consisted of qualitative scoring and using heat maps. So these were some techniques that the commission expressed some preference to move away from and to incorporate more quantitative analysis and that's what Liberty has done. So the company, we've established risk mapping capabilities through our partnership with React's engineering. And what we've been able to do is combine that analysis that we worked with React's with tools like Hexagon and Power BI to complete Liberty's analysis. Okay.
Next slide. So prior to Liberty having this risk mapping capability, a lot of our decisions were made on whether something was cyclical or whether something was compliance based. And another feature of the strides that we've made is that if you look in our maturity survey from the 2020 WMP, a lot of our answers refer to our projections being able to look at risk at the regional level. And so what we've been able to do is the work that we've made between 2020 and 2021, we're now able to look at risk at the circuit level and in many cases also at the segment or the SPAM level. One of the other strides that Liberty has made is we've also upgraded our outage management system responder.
And what this is able to do now is to produce a specific report that's going to capture a lot of the data points that the commission has asked Liberty to collect with respect to wildfire risk. And again, even with some of the advancements that Liberty has made from 2020 to 2021, there still is a lot of work to be done. Again, this is just our first crank of our models, our first results of our models. And we do plan to dedicate more resources towards our risk based decision making framework initiative. Okay, next slide.
So as we can see on this slide, in 2020, in the six step risk based decision making framework process, we were at the beginning of the risk analysis stage. So this involved us collecting data, looking at certain data points, whether it was coming from our outage management system, whether it was coming from our inspection reports, we really hadn't organized anything to be set into models and to be looked at and really analyzed. So now in quarter twenty twenty one, we've completed steps three and four and we are now at the beginning of step five, which now allows Liberty to look at some of the models outputs in the form of the risk and efficiencies and start to incorporate this into discussions in terms of what capital investment or operating and maintenance program decisions are going to look like forward looking. Next slide. So as I mentioned earlier, as part of Liberty's risk mapping efforts, Liberty commissioned REAC's engineering and they conducted a detailed wildfire risk analysis over our entire service territory.
And as part of this analysis that REAC's conducted, they looked at fire spread, they looked at the number of structures that would be impacted in the form of houses or in the form of businesses. They looked at what the cost would be to suppress the fire and also what the number of people or the populations impacted for 33 distinct areas within the Liberty service territory. So looking at these factors is very important because as we see in some slides that we'll see coming up, this will feed in directly as the data points to model the consequences of the outcomes that Liberty's used in its risk based decision making framework bow ties. So we'll see a slide where we'll see the four outcomes and how those consequence modeling fits into that. Next slide.
So here we see a visual and this is an example of the areas for which REAC looked in our service territory and how they've segmented it off. Here, this is our South Lake Tahoe region. And you can see that there's four different colors in this graphic right here. So we have the red region and this is going to represent very high wildfire risk. We have the pink region, this is going to represent high wildfire risk.
We look at the yellow region, this is going to represent moderate wildfire risk. And then if we look at this light blue region, this is going to represent low wildfire risk. And these green lines here, these are our primary overhead distribution lines. And a lot of these lines are going to crisscross into different segmented areas. And these segmented areas all have different wildfire risk profiles.
So this will be important to keep in mind in the following slide when I talk about how we look at what the wildfire risk rating for individual circuits are. And lastly, it's important to note here that this dark red region, if you look off to the eastern section of this dark red, very high wildfire risk region, this is currently what our HFTD-three region is. So this very high wildfire risk rating that Re X has found for this section in our service territory aligns with our current HF TD-three region. Okay. Next slide.
So as the next step, Liberty, we combined our outage management system data along with our vegetation management inspections as well as our detailed system survey inspections and looked at all of this data in those segmented areas that REACTS had looked at in terms of what its wildfire risk scoring was. So what we were able to do is integrate all Liberty's internal data with the ratings that REACTS looked at in which they segmented our service territory. So why this is important is that, as I explained earlier, a lot of our circuits will be located across two, three, sometimes four different wildfire risk areas. So some of our circuits could crisscross into the very high, the high and the moderate. And what Liberty was able to do is take all of these different risks that you see on this slide, our vegetation risk, our performance risk and our asset condition risk, look at each of those segments on the circuit and then roll that up into an overall wildfire risk rating for that circuit, looking at each of the different segments on that particular circuit.
So on the WMP, when we had to score and rank each of those circuits, that's the methodology that Liberty has used. We looked at the span risk on a particular circuit and then rolled that up to the circuit level. Okay. Next slide. So as I mentioned earlier, Liberty adopted the same MAVF MARS approach that the larger IOUs have used in their risk based decision making framework.
And here is a bow tie and this should look pretty familiar to everybody where you have your wildfire risk drivers off the left hand side. You have the actual risk event. In this case, it's a utility caused wildfire. It's your you're not in the bow tie. And then off to the right, you have your outcomes.
And in this case, Liberty worked with REACTs to come up with four distinct outcomes. And each of these outcomes has three consequences attached to them. So those consequences were safety consequences, financial consequences and reliability consequences. And this should all be pretty familiar with anyone who's worked closely with the FMAP and ramp proceedings. So what we looked at was when we looked at those risk factors, when I mentioned the structures impacted, the numbers of population affected, all of this was modeled in these consequences of the four outcomes that you see on this slide right here.
So the technique was the same that we've seen with the larger IOUs and that's the methodology that Liberty adopted to calculate its RSEs. And Liberty's RSEs with some of its scores for its first generation wildfire models along with each of those individual circuit risk ratings can be found in Attachment C in its 2021 WMP. All right. And that should conclude our discussion. So thank you.
Speaker 3
Great. Thank you, Greg, for that presentation. Next up, we're going to have Fair Valley Utilities and we're going to have Paul present. Thank you.
Speaker 6
Brief introduction. I'm Paul Marconi. I'm the President of Bear Valley Electric Service Incorporated. And I'll actually be the analyst for all three presentations. Can you go to the next slide?
Before I get into the risk assessment, just a brief if you've had a chance to go through our plan, some of the key areas of our plan aims to tailor our actions to our community. It's a small area, 32 square miles. It's up in the mountains. As you can see from the pictures, lots of trees. The trees get snow laden, so they get weak.
And then when the winds pick up in September, there's potential a for branches to come off and make contact with lines. So and because we're in a generally a dry environment during that time of the year, we're in the High Fire Threat District. For those with eagle eyes that people on the chairlift, that's pre pandemic, okay. Substantially, mitigate the risk to wildfire. That's what our plan aims to do.
There's emergency response for restoration and wildfire events and so forth. One of the big things that I take away from, I think this is my third technical workshop after filing our plan is the incorporate industry best practices. We always learn quite a bit from each other's plans as we look at them as well. Could you go on to the next slide please? So on the risk assessment mapping and resource allocation methodology, Fair Valley follows the small multi jurisdictional utilities process that was provided in CPUC decision back in April 2019.
You can go to the next slide. This is basically a seven by seven matrix frequency versus consequence or likelihood versus consequence logarithmic has the six step process similar to what Liberty discussed. And you look at the various risks, you look at mitigations and the output is a risk spend ratio and a risk reduction. This chart sort of demonstrates what comes out of that. So on the x axis is are all the mitigations and there we have alternatives too.
So you'll have installing covered wire, you'll have undergrounding and then you can make a choice on which provides the best risk spend ratio. As we've had this model in place since 2017 actually and we x axis just seems to grow because we add mitigations and look at alternatives. So it gets a little complicated, but it does give us a good sense of how much risk do you reduce versus the for the cost. We can go to the next slide. Okay.
So that method is that I discussed is a little less satisfying because it's not it doesn't have that bow tie. It doesn't show how different mitigations interact with each other. And so that's probably the least satisfying part of that model. And so as an interim measure, we developed what we call the fire safety matrix circuit matrix. And basically, you're not expected to actually read the lines in that spreadsheet.
That's just showing you the model. But what the model does is it lists for each circuit. It takes into account what are those things that make a circuit high risk to wildfire. And then what are those things that would mitigate the risk? And so those are those colored boxes off to the right.
And then finally, and this is tracked in the matrix, but not you don't get a score for these items. It's one of those things that lessen the impact of a wildfire, okay? For example, evacuation route hardening. It doesn't reduce us the risk of the utility causing a wildfire, seeing any points for it, but it's important to the community. So we want to track it here.
And this is the model where we track things. So we incorporate some of these items that are important to just wildfires in general. And so the way we apply this model is we take the output from the previous model, the risk spend ratios and we decide, okay, if we apply if we want to apply this mitigation, which circuits should we apply this mitigation to? And so that's where this gets allows us to get that granularity down to the circuit level. And we would like to get as in the future, we want to look to get even more granular, but we also temper that with the fact that, like I said, we're 32 square miles.
So circuits really aren't that long. And in general, many circuits, you're pretty you're as granular as you're going to get. And so but we want to take a look at that as we develop our processes. We go to the next one. So one of the things I really like about this model is when we decide to do different mitigations, we can then fast forward and plug them in the model and say, okay, if we apply that mitigation, what will that what will the risk value of that circuit be?
If we look at our plan holistically and step through it at the three year point, what's our system going to look like? And then at the ten year point, what are we going to look like? And you can see how there's that progression. You can also, which is really neat is you can play with the mitigations and you say, well, if I only if I install covered wire, but I only need to install a certain amount on this particular circuit, I already get it down to the green. So those are ways that we can prioritize or get circuits down to medium or moderate risk and then focus on getting those other reds down to moderate and then go on to the green.
So those are some items that the model allows us to do, allows the decision makers to determine where we're going to put our resources. Important to note, what the algorithm does, it's really heavily weighted towards bare wire, overhead bare wire, okay. So and one of the things that we also did in model, we tried to we rely a lot on subject matter experts, right. So in the early days, subject matter experts would say, hey, that's an area of high winds and heavy vegetation density. Now we've quantified those things.
For example, with the weather stations dispersed throughout our service area, we can say, okay, we know what the winds are at different points on our circuits. And so and we have history now. So we can say if the winds are less if the highest winds are less than 20 miles per hour on dry days, then the wind intensity of blow. And we can quantify these so that we're no longer relying on an SME to tell us, oh, that's a windy area. So and before this effort, you need to understand the only weather station in Big Bear Lake was at the airport.
Now we have a bunch throughout the service area. So we get very granular. The other thing is also on vegetation, same thing with the advent of putting all the trees in our GIS. We're able to quantify the vegetation density. So for example, if on average Span has less than 10 trees that might be classified as low and I'm just picking some arbitrary numbers here, 10 to 20 trees would be medium and then more than 20 trees is high density.
So, those are ways that we get around not relying so much on someone's judgment in particular, but being quantified across the board and providing some reasonableness to our assessments. Go on to the next one. So we're a small utility and one of the important pieces in resource allocation is staff. And we did a lot of things recently. One, we rewrote a lot of the job descriptions for people and for people who are there to really emphasize their role in the wildfire mitigation plan.
And in Section one, we identify who's who and what they do in this wildfire mitigation plan. So we want to make sure that works with their job description too. So that their his personal accountability to me is absolutely essential in addition to all the risk assessments and so forth. It's also that personnel side, making sure there's true accountability by the individuals involved in the plan. We did in our job description roles, have a utility engineer and wildfire mitigation supervisor that just used to be the engineering supervisor now very specific role in wildfire mitigation.
We added wildfire mitigation and reliability engineer who runs all these programs and gets the data. We brought on a dedicated project coordinator because we have quite a we have like any utility, you have two functions going on at the same time. You have execution and you have planning. And in the past, Bear Valley Electric was doing both with the same group of people and we found that we really had to break that piece up. And then we've contracted a forester to provide us a third party support in our vegetation and risk assessments in that area.
And then the other area that is really bringing our GIS and data management up to speed because the data really drives our decisions and shows us the results of where we're making a difference, where we're maybe not making a difference and where we ought to be heading. So it's really important we allocate resources to that piece and that's where we're moving. So the big thing that we're our next step is to transition to calculating ignition probabilities at different points and also bringing on transitioning to a bow ties style of risk assessment model. And our goal is to have that in operation sometime mid-twenty twenty two. We'll be kicking off pretty soon a project to take us to that next level in the risk assessment modeling and bring us on to that granularity of different ignition points along our in our service area.
And that's it for this section for me.
Speaker 3
Great. Thank you for the presentation, Paul. And now we're going to enter into a short break. We are all going to come back for the question and answer segments of today's workshop for risk assessment mapping and resource allocation. So we will all meet back here at 10:40 and members of the audience, this is a great time to submit your questions via chat.
So yes, everyone, let's take a short break, and we'll all be back at 10:40. Thank you. Session. So one thing I wanted to note here is we're going to ask questions in order of the different parties. So we have WSD here represented by me and then we also have the Green Power Institute.
We also have the Public Advocates Office, and we also have Mr. Abrams, and then we'll also have Coco Tomasian to read out the questions. So at this point here, let's have all of these stakeholder parties use the raise hand feature in the WebEx and we will leave the hand raised and then what I'll do is I'll cycle through and ask all the questions. And for all the panelists when asking questions, let's ask one question at a time, totally fine if the question has multi parts, but, we will go, go with order. And what we'll do here is if a question is addressed to all three utilities, we'll have the utilities answer in the order of Pacific Corp, Liberty and then Bear Valley just like how we went through the presentation.
So great. The first question here is going to be from WSD, and this is a question for all utilities. Please describe at a high level the mitigation initiative selection process and pinpoint where the outputs from the risk models and this could be RSEs, this could be identification of high risk segments and more are used in the decision making process.
Speaker 5
So this is Heidi Cowswell from PACIFICCORE and hopefully some of the graphics that were in the presentation helped make it a little bit tied to the question well. So we use again that combined fire risk, which is the model output of the climate and fire climatology history to help identify areas that due to their location and, you know, environmental conditions, etcetera, are are the where a utility ignition could result in fire spread. So then on top of that, we lay that the utility ignition piece and the output is signaling to us at that zonal level what the what the absolute value is for that particular circuit zone. With that, we then look to see what the drivers for that specific risk are. So is it something to do with how the ARC energy manifests itself, how the vegetation you know, what the vegetation history is, what the historic outages might be.
And all of those help us then understand what particular symptoms that segment experiences. Then to the mapping that was identified where we outline what the cause history is that helps us get into an understanding of what kind of events that particular area would be susceptible to. And then we look to see whether fault response, inspection and maintenance programs, vegetation management or asset hardening would be the best way to mitigate for that particular risk event driver. And with that, then we start to look at how much any one of those measures potentially buys down our risk. One of the areas that we have work to do is quantifying that risk spend efficiency.
That is kind of the next target that we're striving for is some of the incorporating the effectiveness measures to identify how much that risk is improved at a quantitative level instead of the qualitative level that we've used in the past. But that's generally the approach that we've been taking and then in terms of its impact, the probability and the impact of the event, it helps to identify then the prioritization for that particular zone being advanced further.
Speaker 7
All right. Yes. So Liberty, the approach that we use is looking at those RFCs. And as part of those RFE calculations, there are lot of factors that go into that. So during the presentation, I mentioned that we do look in many cases at the segment span level of our particular circuits.
And at that particular segment and span level, there were three different risks that we looked at in addition to the factors that REAC scored for us at that particular segment or span. So those three risks were our vegetation risk, our asset condition risk and our performance risk. And that was in the form of what was the type of driver that caused the forced outage. So a lot of the factors we saw in the WMP are like if an animal chews our circuit or if a tree branch cuts our wire, anything that we can attach to a risk driver, what we try to do is look in our outage management system output, trace that to the location and record those details. So next, when we've looked at particular mitigations that address certain risk drivers, we can look at the statistics in terms of the number of forced outages by type and also the number of those forced outages.
And we can look at if these mitigations reduce these risk drivers from occurring. And if they can, then that is what factors into our RSE calculations. So that is the main basis. We recently just completed our modeling to reach those RSEs. And again, this is in the same form as the MAVF in the MARS methodology that the larger IOUs have completed previously.
So we really haven't had much time to incorporate the RSE values into our decision making in terms of wildfire risk. But that's certainly what we've already started to discuss in terms of how we will be factoring that in. And then, of course, there are other factors that you look at in addition to RSE. So this also has to do with timing, cost, resource constraints. For example, a lot of times permitting is an issue of doing a lot of work in the Tahoe area.
So there are other considerations that we do take into consideration when selecting a particular mitigation. But now that we do have the tool to produce those RSEs, we're going to incorporate that into our forward looking capital investment and operating and maintenance decision making.
Speaker 6
Yes. I'll be brief here and what I'll do is walk you through an example. We were looking at evacuation hardening and there are a number of alternatives. Obviously you could underground, you could put fire resistant composite, lightweight steel poles, the wood poles or you can also install what's called a wire mesh wrap that wraps around the base of the pole and goes up a certain height. And those are all good alternatives.
Probably the least satisfying is the wire mesh, but it does do the job, okay? At $950 a pole, that's ROC goes up quite high, whereas when compared to replacing the pole or even undergrounding, those are much more expensive options. And so get an immediate grip on the problem and solve it, we have over 800 poles on our evacuation routes. Our ROCs drove us to output drove us to, hey, let's get the wrap on the poles in the next two years and then implement a policy that whenever we replace a pole for whatever reason, we will replace it with a lightweight steel or fire resistant composite pole. And so you can get that balance, achieve the risk reduction in a very efficient and cost effective manner.
And then in the long run, as you replace poles for other reasons or development and so forth, you can use that RFC to drive your policy in the future.
Speaker 3
Great. Thank you for that Paul. And next, let's go to the panelists for Green Power Institute.
Speaker 1
the Green Power Institute. Thank you for all these presentations. My question is related to essentially dataset size. So compared to the IOUs, the SMJUs have much smaller risk event outage and ignition datasets. And this can really make essentially the predictive modeling approach very difficult.
So I was hoping you could each talk a bit about how you actually overcame this modeling limitation. Thank you.
Speaker 5
So this is Heidi Caswell from PACIFICCOR. Great observation. We actually ran into this issue as a fairly substantial one when you tried to create some kind of risk event to ignition relationship. And that is actually the reason that we didn't try to do anything mathematically in those two pie charts other than to draw the correlations, the observation and not try to turn it into a number essentially. The hope that we have over the next kind of model development is to try to extend our reach in terms of application of the ignition probability for specific fire risk or for risk events so that we become more quantitative with that.
And that might be by reaching out to the large IOUs to determine whether there's any kind of extensible data that exists within their data sets that we could kind of attempt to leverage or, you know, by continuing to just work with smaller data sets that we have extended into other we have other service territory, so perhaps there's some opportunity for outside of California informing some of the relationships. Hopefully that's helpful.
Speaker 7
Yes, that was an excellent question. I'm glad you asked that. So that is absolutely a challenge for Liberty, perhaps in most applying most to Liberty because we are such a new utility in this space. So one example of that challenge that it did present to us in addition to the lack of real depth of history of our data sets is that we did try to use a neural network approach that the larger IOUs and I believe also Heidi mentioned that Pacific Corp was using. The problem that we ran into is once we were utilizing that approach, found that we really didn't have enough history of data points to make it that reliable compared to using a more standard approach, which would basically be weighting the each outage type equally in the prediction of a wildfire risk event.
So in a neural network, there's a lot of feedback in terms of if this occurs, then that's and a lot of information is basically looped for each observation. But unfortunately, we weren't able to look at those results as reliable as the standard approach. So we did go with the standard approach to come up with our RSEs. In terms of looking outside of our own data sets, that's something that we are definitely exploring. So we've talked with Bear Valley about doing that too last summer, how are ways that we can help each other.
I think we're still in the discussion phase of just what that looks like, but it's definitely on our radar and it's something that Liberty looks forward to doing.
Speaker 6
Yes, it's a great question. Reminds me of my statistic professor figures lie and liars figure statement. So you do have to look at what you're doing in statistics. You can quantify the amount of error, right, based on the amount of data you have. So when you do look at the numbers, you do look at them with eyes wide open and caution that.
We have the benefit, although we're small, of having a database that goes back now about twenty years. So we look back further in the past to sort of glean and increase our database. The other thing I would just say that as much as a heavy list that is at least for small utilities to do these quarterly data reports and so forth, they do give us a glimpse at the data from other utilities. And so that does pay off in increasing our ability to assess and look at a broader picture. And then once again, collaborating between the utilities would be beneficial in resolving that issue.
Speaker 3
All right, thank you for that question, Zoe. Next, let's move on to the Public Advocates Office.
Speaker 10
Okay. This is Henry Burton. Can you hear me okay? Yes. Okay, great.
So I'll start with Pacificor. A question for Pacificor. Mentioned using a Cal Adapt tool to estimate or model climate change trends over time. Can you elaborate a bit on that and how you used it for your risk analysis?
Speaker 5
Yes. So in the model, there was the expression or the demonstration, I think, of the different layers that are incorporated for the as additional inputs for the calculation process. And with the Cal Adapt data, what we did was looked at, I think, four different scenarios for the climate change out to 2030 at this point. And basically, we're looking to see the variation that occurs for the contemporary climate aspect of the model with regard to the fire weather indices and the dryness. So it was specifically focused on what would we expect with at that zonal level that we've been talking about with the change in local dryness during the extreme periods of the year.
And then at the same time, how are the models suggesting that fire weather indices would change? And so those are around temperature extremes and wind extremes as well, so gusts specifically. And with that, we then were able to identify the zones that had the most substantial impact as far as ramping up their combined fires or combined scores. So, was the output. There's actually it's a little tough to tell the specifics because it's really best viewed in the model way, but there's a graphic that demonstrates that distribution of climate variation as a function of time at the zonal level in our fire risk assessment or in the risk modeling section in 4.5.1 in RWMP.
And we anticipate refreshing that as cal adapters future climate change models are produced. And we have begun some conversations, I think it's mentioned in the document as well, with the climate work stream that's part of the Pyrogens project.
Speaker 10
Great. And for the other two utilities, did you use Cal Adapt or a similar tool for the same purpose?
Speaker 7
Henry, could you repeat that question one more time? Think I just want to make sure I can answer it fully.
Speaker 10
Sure. So the question was, did you use the Cal Adapt tool to understand climate trends over the coming decade or two? And or do you use a similar tool? And could you just describe a little bit how you look at climate trends?
Speaker 7
Sure. That's a good question. Thanks, Henry. So as part of looking at our climate trends and anything that would be related to climate change and how it does affect our wildfire risk, that would be part of the REACTS analysis that we used. So whatever trends that they identified or whatever input variables are related to climate change that was part of their fire propagation models that we incorporated into our analysis, that's where that would be captured.
Speaker 6
Yes, for Bear Valley, I don't think we used the Caldap, but we did look at climate change studies. Our area is very small, so it's almost a pixel in the picture. But what we did look at for example, right now due to the winds during dry periods are typically pretty historically are not high for us to a point where we would initiate a PSPS. But we did look at, hey, if we did nothing, okay, to harden some of our systems and you look at the climate change occurring, we would in the future be in PSPS scenarios. And therefore taking a lead angle on this problem, we did use that kind of information as we planned mitigations and so forth looking towards the future.
Speaker 3
All right. Thank you, Henry, the question. Next, we're going to move on to Mr. Abrams.
Speaker 3
Let's give it a few more seconds here. If not, we can move on to the chat questions if we have any, and then we can circle back to Mr. Abrams.
Speaker 6
Hey, Alan, this is Coco. While we're waiting for Mr. Abrams' audio, just wanted to let you know we have not received any questions yet in the chat, so you can circle back.
Speaker 3
Great. Thank you so much, Kogar. Well, let's go into the next question for WSD. And Mr. Abrams, when you're able to get the audio working, go ahead and let us know, we'll put you up next for questions.
So this next question here is, targeted towards, Pacific Core, but I would also like to hear, Liberty and Bear Valley's approaches, as well. So Pacifico, in the 2021 WMP update, we have Figure 4.2-three, which is the breakdown of annual distribution of risk events by cause categories. And from 2015 to 2020, even though the amount of risk events fluctuates, the category of unknown is staying proportional and it's one of the higher brackets. So I'm wondering what is the utility doing to better understand the unknown category?
Speaker 5
That is a super question. So for us unknowns, the natural response to that unknown would be to give guidance to people to help them decide how to code it, but we do not want to corrupt the data that we have. So it's important for us to make sure that we build the investigative competency so that we actually are able to turn an unknown into a certain known. So the oftentimes what ends up happening is whatever the evidence would have been about what the fault or the risk event was is gone by the time a patrolman gets there or it was, you know, it was a transient sort of event and didn't actually leave any indications. With the eye, you wouldn't necessarily be able to detect what that was.
As we are moving forward with the deployment of many of the relay replacements that are part of our plan as well as some recloser control technology, we'll be in a better position to actually diagnose the specific locations and have better targeted inspection of those particular areas so that those unknowns can be made known but not necessarily by the immediate responders. So the not the it'll be on the follow-up action as opposed to on the outage restoration that we're able to determine what the known is. And we do that actually through some post processing that happens or we will be doing that through some post processing that happens with mining the outage history and triggering notifications to people to conduct those analyses.
Speaker 7
Yes, that's a great question. So as I referenced earlier in the presentation, we have upgraded our outage management system. And as part of that upgrade, it will do a couple of things. The first thing is that it does produce a report that contains a lot more details in terms and data points in terms of what the commission was looking for. And some of those data points were if it was energized equipment, how many splices that were there, what type of equipment was default on, the location.
So some of those data points are now features as part of this new report that we have. A couple other things that the company is doing is as part of creating this new report, there were more options for the communication between our dispatch and our trouble men to log in our outage management system. So what we did was we retrained and went over each of these new options with our dispatch teams as well as our field workers in order to properly capture those data points and have those data points be accurate in the report because that will eventually feed into our risk models. So I think that that's going to have some improvement over a lot of these data points that got listed as unknown. So we'll to see going forward what that looks like, but it's our expectation that the unknown category will decrease with these new options and with us producing this report and retraining our staff.
Speaker 6
So I don't think our unknown is that large a number, but we do have and we've been doing this now for about four years, a very detailed prescriptive report that the patrolmen use. So they don't have a lot of options. Have to check the most appropriate. And then the next day there's follow-up. And if we have questions, we sit down with the patrolman, get pictures.
So being on top of the data gathering is important and making sure it's done real time rather than at the end of the quarter, you're trying to do your quarterly upload and now you're trying to reach back a couple of months and figure out what happened that night. So that's important. There are genuinely events that are unknown. I mean sometimes an OCV will open, you'll patrol it, you'll patrol it again. Okay, we test it, it's fine.
We patrol it again just to see if we miss something and you don't it's not there. So there are a few that are going to be in that unknown category and just the fact of life.
Speaker 5
Can I add to my answer, please? Kind of precipitated by what Paul just mentioned, which is historically and I think Zoe's question really tipped this one off, is that we have long history of outage records and those systems may have changed over time. The goal of the outage was to solve and restore power. Now what we're doing is we're using that outage as a proxy for a risk event. And so where in the past an unknown, as long as you had the power restored, was kind of, you know, I looked for everything, everything looked fine, I couldn't actually detect what it was.
It was an acceptable outcome to leave it at that point. Now it's less acceptable, and so we have to build different processes in order to actually go deeper into that investigation. And that's what we're really trying to recognize in our tool sets and move forward with so that we have better answers for those that can be made known if we have more data to inform what probably happened or what maybe conclusively happened when you have that some of that electronic data.
Speaker 3
Great. Thank you for all the answers. At this point here, I'm Mr. Abrams, are you able to patch through with audio?
Speaker 11
Great. Thanks. I'm glad I was able to fix that. So I appreciated the presentation and really just had some questions around sort of transparency of the information and the degree to which collaboration, particularly with local agencies, contributes to how this information is used. So Pacific Corp.
Talked about you talked about how there's this circuit prioritization map. And so just wanted to understand how much of that information is transparent. And so if I lived as an example in an area where there was one of those red dots, might I be able to sort of contribute towards solutions to sort of reducing the risk around that circuit? And how does that type of collaboration go back and forth? And similarly, sort of around the sectionalization devices that are within your grid, how much of that information is shared specifically about what types of sectionalization devices are in what areas?
And so does that also then empower emergency managers or other folks to be able to help when there might be a de energization or there might be decisions around how to mitigate that they're able to contribute in a way because they have this information? Thank you.
Speaker 5
So good questions there, Mr. Abrams. First off, to this point, we haven't made the zonal quantification of risk part of our conversation. We started with the sharing of our mitigation plans with the PSPS impact that were really reliant on the HFTD designation and attempted to be very transparent with customers and public safety partners in that regard. In terms of that sectionalization question that you're asking, we went the other direction and it's interesting to hear your approach to it.
With us, what we did was we used the identification of where critical customers or medically vulnerable customers were at that section level, had the conversations with our emergency or with the public safety partners and then went backwards to try to determine both mitigation and sectionalization opportunities. We've never identified, I guess, facility at point level to our public safety partners where each of those kind of control points is, and maybe that's worthy of a conversation with them. What we tried to do was to use that whole PSPS impact and risk impact as a conversation. I don't know that we have yet figured out how to make the model usable if you don't have some of the GIS data behind it, But certainly we'll have some of those conversations with public safety partners, hopefully through our Wildfire Safety Advisory Board.
Speaker 6
I think community engagement is really important and we're a small enough area, but there's always why is that street getting covered wire or when am I getting covered wire on my street, small town issue. And so what we did was we briefed the city manager for Big Bear Lake, gave him a very detailed on how we arrived at the mitigations we selected and then what are how we were prioritizing them. And a good number of them fall outside the air in the unincorporated areas, which he really cares about the incorporated areas, but he understands why we're doing that those areas first and how it's all linked. The other area that I use, I'm not sure if you're familiar with fire safe councils throughout California, but I briefed those personally, take our plan slides brief the fire safe councils because those are individuals and if you're willing to join a fire safe council, you're definitely a concerned person in your community about fire safety and wildfires. And so those are people that I like to reach out to because they spread the word around the community as well.
Speaker 7
And for Liberty, we're still in the planning phase on that. But we do have emergency plan as well as PS and PS protocols that we follow. So that's about the stage where we're at right now.
Speaker 11
Thanks very much. I appreciate the answers. Really, think my question, which I think was addressed is really to sort of have more collaboration around decision making in terms of the mitigation measures that are put forward so that public officials are empowered to be able to say, well, if you move this sectionalization device here and we're able to gate in this way, we protect this telecommunications infrastructure or limit the effects in this way or, oh, I know the landowners that are around this circuit where you've identified that there's a risk, I can work with you to address it. So I guess just I really appreciate the answers. I guess I just would say that it sounds like that's an area for further development so that these partners in the public can really be active in terms of working with you in mitigations going forward.
But thank you very much.
Speaker 3
Thank you for that question Mr. Abrams. And we have about five minutes left here. So Henry, go ahead and take it away and this will likely be the last question that we'll
Speaker 10
Great. Okay. Thank you. I have a question mainly intended for Liberty. Liberty, it appears, is relying on REACT for risk assessment and modeling.
And so I have a couple of questions on that. One is, how did you develop the polygons that you used to do risk analysis of different parts of your service territory? I'm just trying to understand like how you drew those polygons to identify discrete regions? And then secondly, how did you validate the results that REACTS provided in terms of risk modeling? Like how did you ground truth it or test it in the field to make sure that that risk modeling was providing a reliable output?
Speaker 7
Yes. Thank you for that question, Henry. So part of how they segmented off those areas, if you remember those green lines, some of those areas would follow where those circuits are located. So that was part of it. Another part of it is an elevation change, what type of vegetation, ground fuels that were out there.
So these are a lot of factors that go into their fire propagation models. And what they were able to do as part of that modeling, at areas where if a significant wildfire change. So for example, if we start moving closer to Lake Tahoe, the fire spread would end because there's no vegetation, there's no wood, there's nothing that's going to have the fire, let's say, if there was a forest there in Woodcarry. So that's one example. Another example too is that if we looked in areas that were heavily commercialized, A lot of times, these buildings aren't necessarily great conductors of wildfires, so the spread isn't going to escape containment as easy as it was as easy as it would, say, in a very heavily forest area, particularly with vegetation or ground fuels that can really be conducive to wildfire spread.
So there's a lot of things that are taken consideration how to area those off and segment it off. But those were some of the different risks that they looked at and that we talked about how to change from one category to the next and where one border would start and the other would end. In terms of validating out in the field what their outputs were, I think we're still in the collection. We're still sort of in the beginning stages of that. I'm not sure we really have a long enough history in terms of really validating.
I think anytime you do something like that, need to have a reliable sample size to look backwards on and say, okay, yes, this is where we think it is. Maybe this needs to change. So I think as time goes on, we're going to be able to do that validation like you just mentioned. The analysis that they completed with for us wrapped up in Q3. So really, we haven't really had that much time to really let with the scoring and rating that they've identified and looking backwards to any types of changes, whether we need to say something's more risky or less risky.
But this is something that we do have our eye on and this is something that we know is going to happen. We know that our first generation models the information that feeds into these first generation models will change. We know not everything will be static. And so we do plan to keep an eye on that and adjust as we see the data come in and test it, like you mentioned.
Speaker 3
Great. Thank you for that question, Henry. At this point, we have reached the end of the time slot for our question and answer session. To all three of the presenters, thank you so much for presenting your guys' risk assessment and mapping and resource allocation. We will now take a short break here.
We are going to resume at 11:35. The next session will be moderated by my colleague, Andy Biggs.
Speaker 9
As part of properly assessing risk, it's vital that the utilities have frequent and thorough inspections to understand asset health and accurately identify where to conduct repairs, replacements, and system hardening projects. Utilities need to demonstrate how they are identifying advancements to the traditional from the ground visual inspections, whether it be through drones, lidar, or infrared, in order to augment inspections and properly catch issues in field before they lead to an ignition. Integrating an understanding of equipment and asset health, utilities should be vetting various mitigation measures to implement to reduce the risk posed by their systems, including PSTS risk. One of the higher bill options for mitigation will be digging into deeper discussion of grid hardening, which includes covered conductor, undergrounding, and microgrid development, given the largely rural nature from much of the SMJU's customer basis. With climate change and weather conditions worsening over the past few years, utilities should be building their systems to be able to withstand known local conditions to ensure systems won't fail or lead to an ignition event at high wind speeds and focus on maintaining a reliable system that minimizes PSPS events and keeps power and lights on for customers.
With that, I'll now turn it over to the utilities with ten minutes each, starting off with the same order as earlier with Pacific, we're starting off then Liberty's and Bear Valley. We'll also include the two independent transmission operators this session being Horizon West and Transbay Cable. Even though they have a much smaller footprint, they've still undergone the work to understand and mitigate wildfire risk along their systems. As a reminder, if you have any questions, please enter them into the chat for our moderator Ryan Arba to go through and we'll have them answered during the Q and A session this afternoon.
Speaker 1
Yes, okay. Sorry. I'm so paranoid about that. Thank you. So thanks for that, Andy.
I assume you want me to go ahead and kick it off with Pacificaure?
Speaker 1
Yes. Okay, great. Thank you. So my name is Amy McCluskey. I'm the Director of Asset Management at Pacific Core.
And today I will be presenting to you on three key elements of our wildfire mitigation plan. These are grid hardening, inspections and emerging technology. Next slide, please. Here they are again with a few images to give a little color. Specifically grid hardening, this is the area where we're talking about changing our grid.
A huge portion of that is implementation of covered conductor for us. Inspections, we're going to focus on both standard as well as enhanced inspections. And then the last year is emerging technology. And we've added this in because this is an area where we're testing out some new things and trying out some new things, and there's a high potential that what comes out of this area can strongly influence how or what we do in grid hardening and inspections. Throughout the presentation, I may refer to these as pilot projects or research projects.
You may have seen them in the WMP also referred to as different names, but here we are going to talk about them as emerging technologies. Next slide, please. So, before we dig, we dive in, I did want to provide a quick refresher into our plan objectives. I'll walk through this. It's not meant to, I would say, replace the detailed work that Heidi walked you through this morning on risk modeling.
It's really here to just provide a simplified view of what our plan hopes to accomplish and how our initiatives are designed to meet those objectives with an overall eye on reducing wildfire through the implementation of our WMP. If you take a look here, have got four objectives on the slide, so greater resilience through fewer fault events, rapid fault response, facilitate situational awareness and operational readiness, and consider and mitigate impacts to customers and communities. When I think about why certain mitigation activities, it really boils down to meeting these objectives and reducing risk by observing the risk drivers that Heidi talked about this morning and trying to really change the way our system operates. What I'd like you to take away from this slide is that each of our initiatives is designed around meeting a plan objective and therefore can either reduce risk or enable the reduction of risk. While each individual component can be very effective, where we really start to see benefit is when we combine initiatives or layer them on in these high risk areas.
As an example, I'll just walk through one. When I think about that first plan objective, greater resilience through fewer fault events. If you remember from this morning, Heidi walks through our fire risk event data that tells us we have got two fairly large categories. One is contact with objects and another is equipment failure. When I think about grid hardening, I think about trying to reduce those fault events, it certainly checks that first box of reducing the impact of contact, but it doesn't necessarily check the box of reducing risks due to equipment failure.
However, that's where our asset inspections can really come into play and handle that second risk reduction. So when you're in areas where you have multiple risk drivers, you need multiple initiatives. The last row there, emerging technology, certainly we're still in test phase with a lot of our different pilot projects. So they're all kind of under construction. But I did want to point out that when we do develop these pilot projects, we're thinking about how these projects can meet these objectives and how these projects can inform other initiatives.
That's what it's all about, sticking with our vision and our plan. Next slide, please. Now that I've covered a little bit of the why,
Speaker 1
on the where, just in case there are those who may not be particularly familiar with our service territory, I've got a map here on the left, just a few of the general stats. We provide electricity to approximately 45,000 customers via 63 substations, just over 2,500 miles of distribution and about 800 miles of transmission, and our service territory spans around 11,000 square miles. So it's a fairly low customer density as compared to some of the other California utilities. Specific to the HFPD or higher risk assets, have the stats here on the right hand of the slide. Approximately 1,200 miles or just over a third of our overhead lines are within the HFPD, eight fifty of those distribution and three fifty transmission.
So this will become a little bit important when I'll get to a later slide to talk about where we deploy our different initiatives, what we look for. Certainly these HFTD high risk areas as well as the work Heidi talked about this morning become very, very critical to the deployment. Next slide, please. All right, so we've talked about the why and the where, and so now we'll start to get into the what. So this slide here is all about our grid design and system hardening initiatives.
We've got, as we put here on the slides, about four main areas where we focus on system hardening. First, being installation of covered conductors second, proactive replacement of poles, installation of system equipment and replacement of small diameter conductor. I've also included in this view the installation of weather stations. It's not strictly a grid design and system hardening initiative, but it's such an important part of how we deploy, how we manage the work that we have included it here because it's such a critical component of the overall program. However, if you were to dig into the WMP, you would see it under situational awareness.
So I did just want to note that in case you're trying to follow along with the document. It's not going to follow in order with the other components here. When I think about the initiatives here and why covered conductor, why replacement of holes, the first one, as I touched on earlier, covered conductor is really around limiting the impact of contact. So it's preventing the event altogether. The other three on here focus on a few different things.
So they focus more on controlling the event or making sure the event doesn't get worse. So they mitigate risk slightly differently. For example, the proactive replacement of poles. For that program, Pacific Oar is targeting high risk locations that might be susceptible to a brush fire where we think we could use some additional structural resilience to make sure that restoration can be faster to support ingress, egress. So it's not really a program that prevents an event, but it helps control it and makes sure that it can be recovered from quickly.
The last two, installation of system automation equipment and replacement of small diameter conductors, they have a similar goal in mind, which is our ability to implement advanced protection and control schemes. So system automation equipment, that's all your widgets, your relays, your reclosers, your circuit breakers, your advanced communication equipment that allow us to identify faults faster, be smarter, isolate quicker and really reduce the potential energy that can be released. Replacement of small diameter conductor is an enabler to that program. So we have a few areas where we need additional capabilities and in order to do that we have to upgrade the small diameter conductor first. I have included some numbers on here, program targets as well as annual program targets.
As opposed to going through those, because sometimes it can be challenging to figure out what they mean and if they are good, I have included a qualitative phasing and deployment graph on the right. And I just want to point out just a few things. There are no units, it is qualitative. And along the x axis are the years. And then the colors do correspond to the programs.
And you can kind of see in 2019, we started focusing on weather stations and system automation and plan development. In 2020, began to shift efforts and ramp up our covered conductor program focused on design and engineering and scope. And now we're pretty much fully ramped up and focusing on high risk PSPF mitigation, which includes pairing different initiatives in a specific location, which I'll walk you through in the next slide. Our plan is to continue with that, but I did want to point out sorry, the can you go back one slide, please? Towards the very right, you see it says risk shift.
To date, we've been focused on PSPS locations, but as Heidi walked you through the risk modeling this morning, as that evolves and we get more granular with our risk and exactly where our high risks are and what location specific drivers are, there is the potential that we may change how we target programs and where. But right now you can kind of see where we are in the graph. We are really focused on PSPS mitigation. Okay, now next slide. This is just a quick snapshot of what that looks like.
I don't expect you to read everything on the left here, but I am going to point out a couple of things. So this is an example of focusing in on a high risk PSPS location, and there is some color coding that is important. The color coding on the circuitry kind of describes how we might bucket the different work, So where we have isolation or where things are electrically connected. And what we do is we look color by color and say, are the things we can do here to mitigate risk? Come up with an implementation plan, which is what you see on the left, which is kind of a color section by color section how we're attacking that portion or that circuit.
And they did once you complete your construction, then you would have mitigated the PSPS risk in this location. So as opposed to looking at all the places you might have covered conductor or all the places you might have reclosers, we're looking at what list, what menu items of our system hardening do I need in this TSPS location to mitigate the PSPS risk. Next slide, please. Now I will shift gears a little bit, so moving from grid design and system hardening to inspections. I've got here I'm going talk to you through our standard inspection programs as well as our enhanced inspection programs.
On the left, in 2021, our goal is to complete 63,000 cyclic inspections. These are typically driven by compliance requirements. So for us, it's just around 3,000 intrusive inspections where we also typically perform a detail at the same time, so it's kind of a double inspection, 10,000 additional detail and around 50,000 visual assurance versus the inspection. In addition to just performing the body of the work, our goal is to complete the Tier two and Tier three inspections prior to fire season, making sure we focus on those high risk areas. Additionally, identify fire threat conditions and accelerate corrections.
If 2021 is anything like 2020, we think we will be finding around 8,800 conditions, 800 of which will be fire threats that we need to accelerate correction for. That's what we are planning on. We will see how the numbers shake out when it's all done. Finally, another goal we have is to bring a refreshed focus to our QAQC of inspections. This is an area we received feedback on and things like we could it was important that we really focus on this area.
So for us that means incorporate fire risk identification into annual training, prioritize the Tier three and Tier two locations, complete desktop reviews. We are planning for around 6,500 of those this year, as well as around just under 700 field inspection audits as well. So that's our standard program. Let me move over to enhanced program here. For us right now, that means our infrared inspections of overhead transmission lines.
This was something we kicked off in 2019, has evolved in 2020, and is now a full program in 2021. I believe to date we've found seven different hotspots on our transmission lines. At first it was one or two, and I think we have learned a lot over the years that we really need to target the inspections when the lines are more heavily loaded, peak or near peak loading, so that gives us the best possible diagnosis or the best possible chance to see that extra hotspot. An example of one of those is here on the slide, that was abnormal heating on a jumper on a 69 kV structure. And one of the big things in 2021 that we've introduced the new program management processes and procedures, so we have better tracking to describe program effectiveness, program costs, We do see this as something that is supplemental and in addition to our standard programs, and we want to make sure we are measuring the effectiveness as we go.
Next slide. So, is the last topic I'll cover today and I really hope it's a lot on the slide, so I'll kind of touch on a few things. This is emerging technology or our pilot project. You will find a lot more detailed information in the 2021 WMP, specifically Section 4.4.1. I picked a six here to just throw on the slide and talk through a little bit.
With each, we have got the name, a very, very brief description, the application, which is really how I see this pilot project influencing something else, the plan objective, so does it align with one of our plan objectives I walked you through, and then the status. The first one, distributed fault anticipation and waveform analysis. This is a partnership with Texas A and M to install new devices on circuits and see if we can get smarter about predicting faults. So obviously if you think about the plan objectives, it's predicting faults and ensuring rapid fault response. The LiDAR Pole Loading Assessment Program, we've got two LiDAR programs up here.
One focuses on assets. The other one focuses on vegetation. So the first one is really around leveraging data to see if we can improve how we model circuits, how we understand risk, and how we identify equipment for replacement. The LIDAR vegetation, which I believe will be touched on a little bit later this afternoon, is around informing risk mapping and potential threats that exist on the system. The fourth pilot here, radio frequency and infrared for line patrolmen, it takes a lot of the positive things we've learned from our existing enhanced inspection program and explores whether or not we can incorporate those elements into our standard inspection program.
Certainly still in the early stages there, but hoping that we can take a lot of our existing lessons learned and apply them. The ARC Energy Fault Modeling Project, this has been really critical as the company has advanced the risk modeling efforts. So being able to simulate and calculate the potential ARC Energy certainly helps us understand where our risks are. And then finally, the Advanced Weather Station Modeling project, which is about comparing different types of weather stations. So the company is certainly adamant about deploying weather stations for situational awareness.
We're just not sure exactly which one yet. And so with all of these, we're trying to keep a close eye on how they inform other programs, how they change our risk modeling, and how they meet the different plan objectives. One more slide. So with that, would say thank you and I appreciate the attention and thanks for listening.
Speaker 4
Is Dylan Harris with Liberty. Could you bring the slide deck up, please? Great. Thank you. And I believe it is halfway roughly through this presentation is where my piece starts after this.
Right there. Perfect. Hello everyone. Thanks for being here today. My name is Dylan Harris.
I'm an Electrical Engineer with the Operations Department here at Liberty. I'm here today to discuss with you Liberty's grid design, system hardening, inspections and mitigation choices within the 2021 WMP update. Obviously, that's a mouthful and just a lot of topics that cover a pretty wide breadth of information here. So I'm just going to try and hit on the key points of each of these. We really have a wide array of approaches for grid design and system hardening, everything from covered conductors to a microgrid project we just started this year as well as some substation hardening efforts.
And then we've also made great strides in improving our inspections in 2021. Next slide please. So in 2020, we completed our first covered conductor projects. We now have seven miles of covered conductor roughly installed within our service territory. The majority of that is this ACS or Aerial Conductor System type construction that you see here where the conductor is sort of bundled together and supported by a messenger wire.
In future years, we plan to use risk mapping and modeling to drive where we really want to implement covered conductor and other grid hardening initiatives. So far, some challenges we have faced with the covered conductor program are permitting issues, visual impact and initially we had some unfamiliar maintenance practices and tools due to it just being a new construction type within our company. We have since held some trainings and gotten the right tools and we have largely bridged that gap of the big question is how do we if a tree totally falls through this and breaks everything, how do we address that type of catastrophic style of outage. Another area we are targeting with the use of covered conductor is the creation of resiliency corridors. So the idea here is we have some existing generation assets specifically within our North Lake Tahoe service territory that in the event of a PSPS, we can create this resiliency corridor with covered conductor, shut down portions of the grid that we need to, but have confidence that a key area of the service territory is going to have power during these PSPS events and give the community a place to be during one of these events and have critical services available.
And cover conductor is a good technology, but we are continuing to evaluate alternatives. Next slide, please. So among those alternatives is microgrids. In 2020, we were successful in constructing and commissioning our first microgrid solution to a remote mountain research station. It is the Sage Hen Creek Field Station, of Northwest Of Truckee, if you are familiar with the area.
This project saved customers over $2,000,000 by replacing high fire risk distribution line with a containerized solar plus battery storage microgrid. The line is still physically in place, but we have the ability to totally de energize it during fire season. Additionally, are undergoing feasibility studies for microgrids in other locations. We have planned covered conductor projects that we are looking taking a step back and saying, does it make more sense to do a microgrid here? What is the feasibility from a cost perspective and what is that bias in terms of resiliency and value to the customer?
Next slide please. So in 2020, we did a system wide inventory of all our overhead assets. This kind of serves the baseline of our system condition and we can use this to develop programs to proactively replace infrastructure. So this system inventory really allows us rather than just having a time based maintenance program to evaluate the condition today of every piece of equipment in our overhead inventory and kind of create a more condition based maintenance program, which is something that is really exciting for us. Go ahead and go to the next slide.
So another outcome of the system inventory was also application based inspection. So believe it or not, in 2019 Liberty was using paper based GO165 inspection forms and that just doesn't have the granularity and quality of data that you need for what the Commission is asking for. So we drastically improved that in 2020 and also completed inspections on the entire overhead system. So this improved inspection practice, like I was mentioning in the previous slide provides prioritization for mitigation of at risk structures. So when you take this asset condition combining with the risk mapping and modeling effort that Greg spoke about this morning, you really get a really solid way to prioritize system maintenance.
And in future years, are looking at expanding our inspection program to include infrared imagery and seeing the applications that that can have for us. Next slide please. So a couple of emerging technologies that we want to talk about today. Pacific Court mentioned distribution fall anticipation. And so the idea here is you look at high fidelity current and voltage waveforms and Texas A and M has developed algorithms that can reliably detect This nuance looks like it might be a branch hitting the tree or might be a failed connector or something days or weeks before an actual outage or failure occurs, which obviously has really big impacts.
So we don't have this deployed yet within our service territory. Within 2021, we will have approximately 20% of the feeders covered and this is starting in our very high wildfire risk areas. So we are using that risk mapping to prioritize this work. And kind of since we don't have this technology deployed yet, we sort of have to look at other resources for, hey, how effective is this? And so we have looked at SCE's wildfire mitigation plan and they have deemed it successful enough to drastically expand this program.
So that is something we are looking at as an indication that is a promising technology. Additionally, the initial risk analysis is favorable and cost effective based on the fact that it is relatively easy to install. Hardware wise, just a single piece of equipment for each feeder. And so we're looking to expand that as a cost effective mitigation solution. Next slide please.
Lastly, I will touch quickly on our distribution automation program that we target the designed to be complete within this year. Again, are targeting feeders in the very high fire risk areas and in terms of reducing the risk of and scope of PSPS events as part of this, we are adding a lot of sectionalizing options to really limit that scope as much as we can combined with data from our weather stations and circuit monitoring, we can really make the most informed decision possible to try and limit the scope of those events. Distribution automation, like I mentioned, is going to leverage some new technology as well as existing reclosers and isolation devices that we have in the field. And it also is a great improvement to system reliability. And with that, that is all I have today.
Thank you.
Speaker 6
Most of my slides will talk about the grid hardening aspects. I'll talk about inspections although a lot of slides are in the next section where I can explain those further. And then on emerging technologies, I'll talk about that as well at the end. You go to the next slide. You know, once when you look at grid design and hardening, it's important to understand the environment.
It's a small area, 32 square miles. It's up about 7,000 feet is where most of the facilities are. So it's in the Heavy Loading District for those
Speaker 6
GO95, construction requirements and then it's also all within the High Fire Threat District mostly about 90% Tier two, about 10% Tier three. And key jurisdictions, it's really three principal ones, the County Of San Bernardino, City Of Big Bear Lake and U. S. Forestry Service. And most of our voltages are either sub transmission or distribution sub transmission at 34,500 volts and distribution is 40 kV.
You go to the next slide. So I'm going to highlight some of the projects we've embarked on. Covered wire conductor installation is something we've embraced. We did pilot the program last year and selected two vendors, Priority Wire and the Southwire products. We tested others and our decision was to go with these two as the principal solutions.
And we do have a pretty ambitious program. We got about eight circuit miles last year and we plan on doing about 13 circuit miles per year going forward and that split between 4.3 circuit miles per year on the sub transmission side and 8.6 certain miles distribution. And this is all targeted on a prioritized list using that fire safety matrix. So, to decide where to go and where to do it. So, that's how that's driven.
If you go to the next slide. Just one comment, the covered conductor is a long term project, but as we target those high risk areas, obviously we start getting start seeing early payoffs. Fuse replacement program, this project started in 2019. The objective was to remove conventional expulsion fuses. As you can see fuse events are spark producing events that could potentially result in fire or wildfire.
And the nice thing about this is an early win I think. We had a little over 3,200 fuses in our system. We came up with a strategy to use current limiting fuses and trip savers and a strategy on where
Speaker 6
each and we have as of thirty one December, we had nine zero one remaining to replace and our goal is to have those out of the system by June 2021. And one of the things like for example, last year we had I think 22 fuse events and because of the progress we had made in this project, I think only three of those events were conventional fuses. So it just shows we reduced that risk by getting this done and getting it done quickly. So that to me has been a good project and reasonable cost. You go to the next slide.
As you can see evacuation hardening, I talked a little bit when I answered one of the questions last panel. In particular, there's basically three ways off the mountain on paved roads. And so it's very important that we keep the poles from falling into the road and obviously there's different ways of handling it. So we piloted to make sure we were familiar and understood what's involved with the different solutions, different types of poles. We've obviously we've our underground before.
In fact, a good portion coincided with an evacuation route. We look we evaluate what's the cost of relocating infrastructure. And then we've looked at this wire mesh cladding wrap. And we looked at what other utilities are doing. And once we saw that they had adopted certain solutions, we went out and tried them ourselves or worked talked to them to make sure we understood so that we could get the benefit of what they had learned.
So as I mentioned, our plan is to harden the three main evacuation routes over two years and using the wire mesh wraps that quickly gets a handle on the solution on the problem. And then we develop a policy to replace the wood poles when they're being replaced for any other reason we do it with something that's fire resistant or and so forth. Just it's anecdotal, but I think it's key to understand if you did have a wildfire in Big Bear Lake, for any reason, generally you'll find at least one of the evacuation routes is closed down due to the fire. Just if you look at the geography, that's a fact that's generally true. And most fires that have threatened us in recent years have threatened at least one evacuation route.
Last summer there was a fire that entered our territory but it shut down Highway 38 and a number of the scenarios that the CAL FIRE and the incident commander ran, if the fire had spread up the backside of the mountains up to Big Bear Lake, it would have taken away the area near the dam over to the left on that chart, Highway 18 over there. So we would have been left with one route down to Lucerne. So it's very important that we focus on these evacuation routes and understand the weaknesses that they cause and that we contribute positively to making them safe. Go to the next slide. So obviously, we use our risk modeling to develop our prioritized list of mitigations.
But at a higher level, like I said, we're mostly Tier two, a small portion Tier three. So we sort of have to prioritize even we don't unlike the larger utilities, we have to get even more granular than the high fire threat districts. And so we're a little more surgical. We've identified high risk circuits. And we consider these things as we develop our plans.
One of the things that I don't have it up here on the slide because we've already sectionalized our high risk areas. And so from a PSBS perspective, we've already got that in place. The next level is automating those particular switches, which is what we're doing. We have a grid automation project, which is in progress right now. They're installing a fiber network that will allow us to do that.
But once again, 32 square mile area, it's not hard for our crews during high risk fire weather to be out in the field and get those switches. We haven't had to do that. We do exercise heavy mountain fields in order to do that, but we are fully sectionalized to where we want to be right now. On the inspection front, first of all do the obviously the GEO165 required periodicity due to our location, do annual ground patrols, we do five year detailed inspection, we do the intrusive pole inspections. We also in addition to the ground patrol, we do an annual a second annual ground patrol by a third party.
So this is another contractor independent of our field inspector. And we make sure that contractor has no construction work with us. So they're completely focused on inspection. And that's a good check and it helps us get that outside view. Then we also use LiDAR.
We've now completed two LiDAR surveys and we're trying to do a LiDAR per year and I think that's a very good effective program. We did do demography throughout our entire area from using ground mounted two years ago and it didn't yield a whole lot, so we have increased the periodicity on that. And right now, we're looking we've got an RFP out to contract out UAV where we'll get some high definition imagery and thermal imaging as well. So we're excited about using that inspection technique to help identify areas for improvement. Likewise, we're in the process of implementing a inspection app for patrol, our field inspector so that we get highly accurate and automate the process of identifying deficiencies and getting into our GIS system.
One of the things too that I it's not in our wildfire mitigation plan this time around, but it's something we're talking about and considering detailed inspections by GEO-ninety five or GEO-one 165 are required every five years on all circuits. It doesn't matter if they're underground overhead, bare conductor overhead, covered conductor. It's a five year inspection. So we split it up each year we do a certain number. Having worked in other communities such as the nuclear field and petrochemical field, those areas they do what's called risk based inspection.
So, you know, we might adopt a plan where you more frequently inspect the higher the bare wire and high risk areas and so forth. So you develop a periodicity based on that. So that's something we're trying to focus on so that you know treating underground systems which are fairly stable and reliable the same as bare wire in Tier three area is probably not the correct answer on detailed inspections. And so that's something we're going to address in the future. Regarding emerging technologies, we pilot programs that have already shown promise at other utilities.
Being a small train, the technology risk for our customer base is a lot. And so we don't want to buy into something that's going to turn out to not work. So we tend to leverage work with the big utilities and then apply it to our unique circumstances. And we think that's a very cost effective way of getting these new technologies into our system in a fairly responsible and yet efficient and fast manner. We also are open to folks who want to do research on our environment and our system.
We certainly are open to collaboration in those areas. That's all I have.
Speaker 9
Great. Thank you so much, Paul. Now we're going to transition over into the ITO starting off with Verizon West.
Speaker 1
Thank you, Andy. This is Elena on this Horizon West. Can you can you hear me?
Speaker 1
Okay. Great. So I know we're running a little short on time, so I'll be pretty brief here. But like I said, my name is Elena Sias. I'm Director of Strategy and Business Development and I represent Horizon West Transmission.
Speaker 1
In terms of the agenda for today, so before we get into the wild time mitigation strategies and measures and inspection cadence that we are employing at Horizon West Transmission, I wanted to spend just a few minutes to give an overview of Horizon West Transmission given that we're a very new independent transmission owner in California, and I wanted to give an understanding of our footprint and our current scope of operations in the state. And then after that, we're going to talk through the specific measures we're taking that we detailed in our 2021 WMP to further harden the assets that we have in the state.
Speaker 1
So like I said, we are a very new utility. We're a transmission only utility and we were the first non incumbent to be awarded two competitive transmission projects by the Kitesville back in 2015. And those two projects are the Suncrest FCC project, which is in Southern California, interconnecting the G and E. And then the second project is the Estrella substation project that's interconnecting in DG and E in Northern California. So both of these projects were awarded to us as part of our Quarter one thousand process.
Both of these are very contained substations, so very limited transmission lines and no distribution lines. Next page, please. So like I said, the only project we currently have in operation is our Suncrest project which is in Southern California and it's located in Tier 3 wildfire area. It's interconnecting with SVG And E and it's about 40 miles East of San Diego in San Diego County near the town of Alpine. It's a very new asset.
It's a little bit over a year old, so we energized it late last or early last year. And we have not had any utility cost technicians, near misses, or PSBS events at the SunTrust SEC site to date. And the project was selected to provide reactive power support and enable more renewables being transmitted from Imperial Valley into the rest of the system.
Speaker 1
And then once again, this is the location of the project in relation to the high fire threat district area. As you can see, it's in a higher fire district area since tier three, and we've had a history of fast spreading wildfires, especially last year with the Valley Fire being very close to to Arafat as close as four miles away.
Speaker 1
I won't spend a ton of time here. This is just another deep dive into the project location. Let's go to the next slide. Okay, so on this slide you can see exactly the project overview here. On the right hand side this is the project location.
Sorry. It looks like the font is a little blurry, but that's specifically thank you. That's the project location itself. The blue line that you can see from that project location is the underground, about one mile line going into the G And E Substation, and then the tiny green line is a very short span of overhead wire transmission line that we're currently working on undergrounding. Like I said, this is just to give you sort of an understanding of the scope of operations that we have to make sure that our wildfire and we're making sure that our wildfire mitigation plan is really tailored to our current scope and footprint of operations in California.
Thank you for sending in.
Speaker 1
And then in terms of to give you a visual of what the project looks like, this is the main facility, static bar compensator. This is the substation itself. Like I said, it's it's being connected to the DG need through the about one mile underground transmission line. And as you can see, there's really no vegetation around around the substation to make sure that the original design was hardscaped, taking into account the high risk of wildfire in the area where we have the assets. So we've really done great job of just thinking through our original design and making sure that we, make those modifications early on to minimize any contact miseducation that we could have to minimize, any risk recognition.
Let's go to the next page. Okay, great. So now that everybody has a better understanding of our scope and operations in California, wanted to talk through our approach to identifying further hardening measures and making sure that we are still keeping wildfire risk top of mind and making sure that we are protecting the infrastructure as best as possible. So we primarily employ cellular modes and effects analysis, FMEA methodology, to inform appropriate wildfire hardening. This methodology was pioneered by NASA to identify potential failure modes and proactively address them.
Really what we do, we use this five step methodology to understand how what is the underlying risk, quantify the risk, and then based on that quantification, identify modifications that we should be implementing and implementing those. So based on that, what we do is the first step is the risk identification. So for each major component of the equipment that we have, it's being the underground transmission line or the substation site itself or transformer. We go through every single element trying to identify, what are all the potential ways in which that equipment component may fail. And then based on that, we brainstorm what could be potential drivers of the failure of that component, whether it's contact with education or just equipment failure or a seismic event, trying to be very holistic in our assessment.
And then based on that we prioritize, we assign a score in this third step risk prioritization. We think through how likely that event to occur, what would be the impact of that event occurrence, the severity of that impact, and then what measures or capabilities do we have in place to detect that event in case it does occur so we can respond very proactively and quickly. And then based on that we calculate a risk priority number. And after we've done that exercise for each element, we go through the risk mitigation. So for any element that has starting with the highest risk priority number, we try to brainstorm what potential mitigation measures we should be deploying to minimize the risk of that element failing and potentially creating a utility cost ignition.
After we go through that entire exercise for the entire scope of our assets, we do the risk assessment of reprioritization, sort of going back and asking ourselves based on that mitigation measure that we're going to implement, what is the resulting risk return priority number and do we need to do anything further to further reduce the risk. So that is the primary methodology that we deploy to identify the right initiatives for GRD hardening.
Speaker 1
And I think because we're running a little short of time, I'll probably skip this one. Let's go to the following one where we go through the actual initiatives we have identified through this. Thank you. So through that, like I said, because the asset is new and we've had the benefit of learnings and best practices from all the other utilities who've been doing work on wildfire mitigation for much longer than we have, We really took a lot of these mitigation measures and implemented them in the original facility design. So we thought through the original design, made sure that we are, for example, under we are underground at the line because year three, we wanna make sure we have fuel free vegetation setbacks and there's no vegetation within the site itself.
The project is located on land that has been previously discharged, so there's really very little vegetation immediately in the surrounding area of the project. We also contracted fire support to to have on-site during construction, so anytime you have a spark or hot work, you really minimize risk of ignition. And then in terms of the hardening measures themselves that were implemented, As you saw from the picture, we erected a concrete perimeter wall to minimize any ignitions or any arcs from the substation itself leaving the site and igniting vegetation off-site. We're also implementing transformer gas monitoring and containment hardening to make sure that we have that real time visibility to the health of our transformers as potential oil leaks if they were to occur. Also, because the project is in a seismically active area, we install intertwined on the seismic pads.
Then similar to a lot of other utilities, we're installing a weather station to just further enhance our real time situational awareness of the site and cameras as well. So
Speaker 1
I have a couple more pages but I think I'll end on this page. In terms of inspections and maintenance, we have the site is remotely monitored 20 fourseven by qualified personnel through those all the sensors that I talked about, but also the overhead cameras. In addition to that, we do a detailed inspection of the sites on a monthly basis, and that includes all facility, all equipment, all fuel mitigation areas just to make sure that everything's in in top shape. And then we also have regular vegetation patrols to make sure that there's no vegetation that's encroaching on the on the site and can cause contact with energized equipment. Then in terms of operational practices, we have the RASK Live Warning Protocol where we really don't do any non critical construction activities during RASK Live Warning days given that those are much higher risk.
We have hot work safety program where any hot work that involves grading or parts on the site is supervised and we have fire safety training for all personnel including contractors on-site.
Speaker 1
And then lastly, I'll briefly touch on these. Like I said, in terms of condition awareness, we have twenty four seven monitoring. We have equipment health sensors that help us really be more proactive about understanding the health of our critical pieces of equipment and rather than reactive as problems come up. We have the periodic inspections that I talked about. Then in terms of response and recovery, we provided specific tailored training to local fire agencies.
They're very familiar with the site. They know how to respond to electrical fires that were to happen there. We have an emergency response plan, in place, we have all of our key personnel trained on that. And, in addition to that, we have a private fire brigade that's dedicated to the SunTrust project in case there were an admission at the site to make sure that we have dedicated access to fire suppression resources that are trained on electrical fires and are familiar with the layout of the site. And like I said, because we're transmission only, we do not have end use customers.
We do not foresee ever deploying PSDS, but we do have some protocol to communicate with the stakeholders like CISO, like CDC, SDG and E in this case we do have a need for emergency communication procedure. So that's all I have for Horizon West. Thank you again for the time and thank you for all your input.
Speaker 9
Thank you so much, Joanna. I know we're running a little bit over but we just have one more presentation with Transbay Cable and then we'll break for lunch.
Speaker 13
Well, thank you for having us here. So I'll be doing a quick intro to Transpay Cable and then I'll hand it off to Laniel Gardner, who is our Regulatory and Business Manager who's going to talk about some of our hardening and risk mitigation strategies. If could go to the next slide. I'll try to keep this short because we're running a little bit late on time. But basically, Transbay Cable is an HVDC converter station which operates in the San Francisco Bay Area.
Next slide, please. We have 53 miles of submarine cable and two converter stations which are interconnected with PG and E. So, both of these interconnections with the PG and E substations are using an underground cable. So we're capable of providing 400 megawatts of real power transmission into the city of San Francisco. And each station can also provide a significant amount of reactive power support to support the network out in Contra Costa County and in San Francisco.
Speaker 13
Here's a picture showing our cable route through the bay and generally speaking our converter stations located at each end. Next slide. So, our converter station in San Francisco is the Potrero converter station. We are primarily located in urban city environment in the city. And if you go to the next slide, we have our Pittsburgh Converter Station.
Pittsburgh Converter Station is located right on the edge of the Delta, right where the San Joaquin And Sacramento Rivers intersect.
Speaker 13
So, I'm going talk briefly about some of our wildfire mitigation strategies and measures here.
Speaker 13
So, as you can see, we've identified our proximity to the risk areas for the wildfires in California. And our Pittsburgh Converter Station is located closer or closest to some of these fire risk zones, so approximate to the tier two high fire threat district. We are still not located in any wild lands. So, next slide. But we are located close to some proximate vegetative fuels.
The marshlands located behind us and to our west. On the next slide, we have a more zoomed out view which shows our proximity to those vegetative fuels and then also the nearby residential communities. And next slide. We are also taking into consideration our location in the Bay Area and our proximity to seismic fault lines. We are located near several faults that run through the Bay Area with the San Andreas Fault being out in the San Francisco Peninsula, the Hayward, Diablo And Greenville Fault.
So although unlikely, we do want to take into consideration that seismic activity could be a precursor event that causes an ignition. Next slide. I think here I will turn it over to Laniel for our overview of our FMEA process, which is similar to the one used by Horizon's Web.
Speaker 8
Thank you, Michael. And just for realizing that I'm standing between everyone at lunch, I will skip over this slide. As Michael stated, it's very similar to the process that Horizon West utilizes for risk assessment and prioritization of wildfire risk and drivers and medication measures. So I'll skip to the next slide and discuss briefly TBC's overall wildfire risk management overview. As Michael described earlier, TBC has a very limited footprint and as a result, we assess wildfire risk as part of our overall fire prevention program and operational risk reduction.
And as stated here, we developed site specific fire prevention plans. These plans are reviewed annually by local fire departments. If you move to the next slide, I'll kind of discuss some of our facility design, construction, etcetera. So just to highlight here, TBC's cables are all underground or submerged beneath the Bay waters. All of the above ground infrastructure is fully contained within our converter station sites.
They're surrounded by 12 foot concrete walls with inward and outward facing security cameras. Inside those walls, there's fire lanes that are approved by the local department departments and dock boxes. TBC, our buildings are meet California building code in addition to California building codes for seismic resiliency. Some of the infrastructure is actually designed to meet IEEE seismic resiliency which is higher than the California Building Code. The electric system employs controls which implement protective blocking which can shut the system down within microseconds.
And so if there is a fault on the system, the system will trigger an immediate shutdown. Below that, you see there's a list of some of the incremental hardening projects that we've undertaken this upcoming year. So that's transformer seismic pads. We're increasing the resiliency of our transformers. We've got eight transformers at each site.
We are adding some fire suppression capability to some of the auxiliary buildings that we have on-site in Pittsburgh including additional thermal barriers for auxiliary rooms. Next slide please. Thank you. This slide see a description of our maintenance practices. We conduct weekly and monthly inspections of our facilities.
This includes sidewalks, inspecting cable faults. And we also on the next page, I'll discuss we have real time monitoring of our cable, which includes the ability to monitor in past big notices. So we've got 20 fourseven real time monitoring of our cable and of the system through our human machine interface. Operational practices,
Speaker 8
a number of fire prevention and emergency policies or hot work fire safety program and all operators and engineers receive training on all of these procedures. Next slide please. Thank you. I kind of hit earlier on some of the situational and additional awareness capabilities that TBC has. And lastly, just the response and recovery, one of the things that I'd like to highlight this year, we purchased Class B foam trailers, which are designed to fight and suppress transformer oil fires.
So that's a capability that local fire departments did not have and so TPC purchased two trailers. There's both at our Potrero and our Pittsburgh sites. So those are on-site for depression in case the situation arises. And other than that, think I've kind of hit all the highlights that I'd like to. So thank you much for this opportunity to present.
And that's all for us. Thank you.
Speaker 9
Great. Thank you so much, Sunil and Michael. Thank you also to all of our presenters that we've had so far. I know has to fit a lot of material within ten minutes. It's understandable that this is over.
With that, we'll meet back here at 01:30 to go into the Q and A session. And apologies for eating a little bit into your lunchtime.
Ask anything about this video — I analyse its transcript to answer.