Data-Driven Power System Optimal Decision Making Strategy under Wildfire Events

Hong, Wanshi; Wang, Bin; Yao, Mengqi; Callaway, Duncan S.; Dale, L.A.; Huang, Can

doi:10.24251/hicss.2022.436

Cited by 8 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous work has investigated approaches to optimize a deenergization plan that minimizes wildfire ignition risk while maximizing the load demand that can be met [27], [28]. Data driven and machine learning methods attempt to accelerate the solution time of these optimization problems, which can be very difficult to solve to optimality [29]- [31]. Other research uses the Optimal Power Shutoff problem as a basis to plan infrastructure upgrades that minimize the impact of PSPS or reduce the need for PSPS in the future.…”

Section: A Related Workmentioning

confidence: 99%

Recursive restoration refinement: A fast heuristic for near-optimal restoration prioritization in power systems

Rhodes

Coffrin

Roald

2022

Electric Power Systems Research

View full text Add to dashboard Cite

Section: A Related Workmentioning

confidence: 99%

Recursive restoration refinement: A fast heuristic for near-optimal restoration prioritization in power systems

Rhodes

Coffrin

Roald

2022

Electric Power Systems Research

View full text Add to dashboard Cite

“…Other papers focus on methods for operating power systems during wildfire-prone conditions. For instance, both Nazemi et al and Tandon, Grijalva, and Molzahn study the impact of dynamic line ratings to increase operational flexibility [32], [33], Hong et al propose data-driven techniques for minimizing load shedding after switching off high-risk lines while considering the possibility of cascading failures [34], Zhou et al use data-mining techniques to assess and mitigate wildfire ignition risks [35], Haseltine and Roald analyze how recloser operation affects both wildfire risks and system reliability [36], and Kadir et al describe a reinforcement learning approach to line deenergization and other operational decisions [37]. However, none of these papers incorporate an infrastructure investment model.…”

Section: Introductionmentioning

confidence: 99%

Optimizing Transmission Infrastructure Investments to Support Line De-energization for Mitigating Wildfire Ignition Risk

Kody¹,

Piansky²,

Molzahn³

2022

Preprint

View full text Add to dashboard Cite

Wildfires pose a growing risk to public safety in regions like the western United States, and, historically, electric power systems have ignited some of the most destructive wildfires. To reduce wildfire ignition risks, power system operators preemptively de-energize high-risk power lines during extreme wildfire conditions as part of "Public Safety Power Shutoff" (PSPS) events. While capable of substantially reducing acute wildfire risks, PSPS events can also result in significant amounts of load shedding as the partially de-energized system may not be able to supply all customer demands. In this work, we investigate the extent to which infrastructure investments can support system operations during PSPS events by enabling reduced load shedding and wildfire ignition risk. We consider the installation of grid-scale batteries, solar PV, and line hardening or maintenance measures (e.g., undergrounding or increased vegetation management). Optimally selecting the locations, types, and sizes of these infrastructure investments requires considering the line de-energizations associated with PSPS events. Accordingly, this paper proposes a multi-period optimization formulation that locates and sizes infrastructure investments while simultaneously choosing line de-energizations to minimize wildfire ignition risk and load shedding. The proposed formulation is evaluated using two geolocated test cases along with realistic infrastructure investment parameters and actual wildfire risk data from the US Geological Survey. We evaluate the performance of investment choices by simulating de-energization decisions for the entire 2021 wildfire season with optimized infrastructure placements. With investment decisions varying significantly for different test cases, budgets, and operator priorities, the numerical results demonstrate the proposed formulation's value in tailoring investment choices to different settings.

show abstract

“…In [10], only the location of transmission lines was included, and the study considered a small area near Monticello and Winters, California. Previous work on wildfire risk estimation in power system overlayed power lines on a general fire risk map to identify highrisk power lines [11,12], without considering the specific wildfire risk of electric components. While some utilities have built their own wildfire risk models, there are no such models in the public domain.…”

Section: Introductionmentioning

confidence: 99%

Predicting electricity infrastructure induced wildfire risk in California

Yao

Bharadwaj

Zhang

et al. 2022

Environ. Res. Lett.

Self Cite

View full text Add to dashboard Cite

This paper examines the use of risk models to predict the timing and location of wildfires caused by electricity infrastructure. Our data include historical ignition and wire-down points triggered by grid infrastructure collected between 2015 to 2019 in Pacific Gas & Electricity territory along with various weather, vegetation, and very high resolution data on grid infrastructure including location, age, materials. With these data we explore a range of machine learning methods and strategies to manage training data imbalance. The best area under the receiver operating characteristic we obtain is 0.776 for distribution feeder ignitions and 0.824 for transmission line wire-down events, both using the histogram-based gradient boosting tree algorithm (HGB) with under-sampling. We then use these models to identify which information provides the most predictive value. After line length, we find that weather and vegetation features dominate the list of top important features for ignition or wire-down risk. Distribution ignition models show more dependence on slow-varying vegetation variables such as burn index, energy release content, and tree height, whereas transmission wire-down models rely more on primary weather variables such as wind speed and precipitation. These results point to the importance of improved vegetation modeling for feeder ignition risk models, and improved weather forecasting for transmission wire-down models. We observe that infrastructure features make small but meaningful improvements to risk model predictive power.

show abstract

Data-Driven Power System Optimal Decision Making Strategy under Wildfire Events

Cited by 8 publications

References 21 publications

Recursive restoration refinement: A fast heuristic for near-optimal restoration prioritization in power systems

Recursive restoration refinement: A fast heuristic for near-optimal restoration prioritization in power systems

Optimizing Transmission Infrastructure Investments to Support Line De-energization for Mitigating Wildfire Ignition Risk

Predicting electricity infrastructure induced wildfire risk in California

Contact Info

Product

Resources

About