Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Risk Modeling Analytics topic
No spam. Unsubscribe anytime.
SDG&E details WINS risk-model upgrades, lifecycle-cost analysis and mitigation effectiveness
Summary
SDG&E told the Office of Energy Infrastructure Safety that it has consolidated risk‑modeling tools, increased Monte Carlo simulation scale, and added life‑cycle costs and risk‑aversion factors to evaluate wildfire mitigations over a 55‑year horizon.
San Diego Gas & Electric told the Office of Energy Infrastructure Safety that it has substantially upgraded its wildfire risk-analytics platform, merging planning and operations models, expanding Monte Carlo simulations, and adding life‑cycle costing and risk‑aversion inputs to inform mitigation choices.
“We've been working really hard the last probably 6 months to align requirements from WMP, RAMP, and 84,” Joaquin Sebastian, SDG&E’s wildfire risk analytics manager, said. “When we started running, the probabilistic calculations, we ran into a computational limitation. We were only able to run hundred thousand simulations. Then we moved to 1,000,000, and now we are in 5,000,000. And, hopefully, pretty soon, we can run up to 10,000,000 simulations.”
Model changes and capabilities described by SDG&E: - Platform consolidation: SDG&E said it integrated the WINS planning tool and the WINS OPS operational tool on a shared probabilistic model so planning and real‑time de‑energization decisions use consistent analytics. - Monte Carlo and tail‑risk estimation: The company said it now produces expected-value and tail‑risk estimates to compare mitigations at a feeder/segment level under many weather/fuel scenarios. - Lifecycle-cost analysis: SDG&E said it compares mitigations over a 55‑year window per the guidance it cited, factoring installation, ongoing maintenance, and shared foundational costs. “When we are thinking to deploy, sustained mitigation for 55 years, is not only the CapEx or the installation cost where we need to look at it. It's also all the ongoing cost that certain mitigations need to happen to maintain a level of effectiveness,” Sebastian said. - Validation and reproducibility: presenters said inputs, outputs and model snapshots are stored in AWS to enable replication, and third‑party reviews are part of the validation plan.
Mitigation‑effectiveness estimates presented: - SDG&E said its subject‑matter‑expert and data‑driven method yields very high modeled effectiveness for strategic undergrounding; presenters gave a system‑level figure of roughly 99% effectiveness for undergrounding in reducing the modeled ignition pathway. - For combined covered conductor (including falling‑conductor protection and early fall detection), SDG&E presented a combined effectiveness of about 58.1% in the company’s analysis.
How SDG&E combines metrics: SDG&E said the model evaluates expected value, residual/tail risk, life‑cycle cost and risk aversion to choose mitigations. Joaquin Sebastian described subject‑matter weighting where low counts of historical ignitions require combining data review with SME judgment to estimate effectiveness, and then applying those estimates in a probabilistic simulation.
What SDG&E did not claim: The company presented internal model outputs and scenario comparisons; it did not present a final regulator-approved project list converting model outputs into mandatory work orders. Presenters also said climate‑change adjustments and population changes are not yet included in all models but are under evaluation.
Implication: SDG&E’s modeling upgrades signal stronger integration between planning and operational decision tools and a shift to lifecycle and tail‑risk metrics in selecting expensive mitigations such as undergrounding versus covered conductor.
