Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

88 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Robustness and Post-Earthquake Recovery of the Los Angeles Power Grid at Census-Tract Scale

Team: Yinchen Yi, Yutong Li

A census-tract–scale modeling pipeline to (i) simulate earthquake-induced substation damage, (ii) propagate disruption to tract-level service via a tract–substation mapping matrix, and (iii) compare post-earthquake repair prioritization strategies under three objectives:

  • Network robustness (connectivity / fragmentation proxies)
  • Population service restoration
  • Equity-aware restoration across census tracts (e.g., SVI-weighted objectives)

The pipeline supports both an unconstrained baseline and a logistics-aware restoration setting (multi-crew, multi-depot, travel-time constraints). We further perform PCA + K-means to group census tracts into resilience typologies using simulated outage/recovery metrics and socioeconomic indicators.

Scope note: robustness metrics here are connectivity-based (e.g., largest connected component) and do not enforce power-flow feasibility unless operational constraints are added.


Workflow

  1. Topological network construction from transmission line and substation data using endpoint snapping, line splitting at substations, and graph extraction of substation-to-substation direct links
  2. Tract-substation weighting matrix (W) construction based on Inverse Distance Weighting
  3. Centrality and percolation analysis of topological network
  4. Scenario-based stress testing using ShakeMap PGA fields (1933 Long Beach, 1971 San Fernando, 1994 Northridge, 2%-in-50yr)
  5. Probabilistic substation damage-state modeling using lognormal fragility with Monte Carlo sampling
  6. Tract-level service estimation through tract–substation weighting matrix (W)
  7. Restoration simulation under:
  • Baseline (no constraints)
  • Logistics-aware rule-based and GA optimisation-based scheduling (multi-crew, multi-depot, travel-time constraints)
  1. Strategy comparison (Theoretical Limit / Random Baseline / Betweenness (Bridges) First / Impact (Population) First / Degree (Hubs) First / Hospital First / Impact λ2 (Grid) First / Closeness First / Balanced GA / HospFirst GA / Efficiency GA) using service restoration metrics and connectivity-based robustness metrics
  2. Tract resilience typology via PCA + K-means using simulated + socioeconomic features

Repository structure

  • Topology_and_Weight.py — Construct substation-level transmission topology (snapping/splitting + direct links) and export tract–substation influence weights (W / mapping CSVs)
  • IDW.py — Interpolate scenario PGA grids to substations (IDW/KDTree), producing per-scenario PGA_* columns for downstream simulations
  • build_travel_matrices_osm.py — Precompute base→substation and substation→substation travel-time matrices from OSM road network for crew scheduling
  • C257H_Project_Main.py — End-to-end pipeline (hazard→damage MC→network impact→restoration scheduling/GA→KPIs→typology clustering), outputs organized by Stage folders
  • Project_Visualizer.py — Post-processing visualizer: maps and comparative figures from Stage outputs (e.g., supply maps + histograms, logistics heatmaps, KPI bars, Gantt charts, cluster profile plots)

Quickstart (suggested run order)

  1. Topology + W-matrix

    python Topology_and_Weight.py
  2. Intensity measure interpolation

    python IDW.py
  3. Travel matrix

    python build_travel_matrices_osm.py
  4. Main pipeline (simulation + restoration + KPIs + typology)

    python C257H_Project_Main.py
  5. Visualization

    python Project_Visualizer.py

Inputs

This repository expects external datasets, all of which included in the "data" folder.


Outputs

Depending on enabled stages, the pipeline can export:

  • Substation-level hazard/intensity sampling outputs (e.g., PGA assigned per substation per scenario)
  • Substation-level damage-state probabilities (MC) and recovery trajectories
  • Tract-level service and recovery trajectories (via tract–substation mapping W)
  • Network robustness / topology metrics
  • Logistics: travel-time matrices (base→task, task→task) and derived heatmaps
  • Strategy-level KPI tables (system / population-weighted / equity-weighted), plus efficiency summaries (e.g., T50/T80/AUC)
  • Comparative restoration curves across strategies
  • Scheduling: rule-based schedules and Gantt plots, GA schedules and Gantt plots
  • PCA & clustering outputs: PCA stats/loadings/scree plots, k-means diagnostics, final cluster labels, and tract typology summaries
  • Maps and figures generated from Stage outputs (recovery curves, KPI comparisons, clusters)

Limitations

  • Connectivity-based robustness metrics are graph proxies and do not guarantee operational feasibility (e.g., AC/DC power flow, voltage constraints) unless such modules are added.
  • Service estimation depends on the tract–substation mapping matrix (W); results are sensitive to how W is constructed (distance decay, assignment rules, normalization, etc.).

Citation

  • Cheng, B., Nozick, L., Dobson, I., Davidson, R., Obiang, D., Dias, J., & Granados, M. (2024). Quantifying the earthquake risk to the electric power transmission system in Los Angeles at the census tract level. IEEE Access. https://doi.org/10.1109/ACCESS.2024.3408797
  • Çağnan, Z., Davidson, R. A., & Guikema, S. D. (2006). Post-earthquake restoration planning for Los Angeles electric power. Earthquake Spectra, 22(3), 589–608. https://doi.org/10.1193/1.2222400
  • Xu, N., Guikema, S. D., Davidson, R. A., Nozick, L. K., Çağnan, Z., & Vaziri, K. (2007). Optimizing scheduling of post-earthquake electric power restoration tasks. Earthquake Engineering & Structural Dynamics, 36(3), 265–284. https://doi.org/10.1002/eqe.623
  • Cavdaroglu, B., Hammel, E., Mitchell, J. E., Sharkey, T. C., & Wallace, W. A. (2013). Integrating restoration and scheduling decisions for disrupted interdependent infrastructure systems. Annals of Operations Research, 203(1), 279–294. https://doi.org/10.1007/s10479-011-0959-3

About

Census tract-level earthquake resilience modeling for the City of Los Angeles power grid that compares substation repair prioritization for network connectivity, service restoration, and equity, using network centrality metrics and k-means census-tract resilience typologies.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages