Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Cost Attribution And Risk-Averse Unit Commitment In Power Grids Using Integrated Gradient

Rene Carmona, Ronnie Sircar, Xinshuo Yang

TL;DR

The paper tackles uncertainty in power system operation due to forecast errors in load and renewable generation by introducing an Integrated Gradients–inspired cost attribution method. It treats the day-ahead UC/ED cycle as a differentiable mapping to system cost and attributes the resulting difference between actual and forecasted costs to individual assets, ensuring completeness and symmetry properties. Building on this, it implements a risk-averse unit commitment framework that adjusts renewable capacities based on Monte Carlo scenario attributions to mitigate risk while maintaining competitive operating costs; simulations on the RTS-GMLC grid show reduced load shedding and robust cost performance. This approach offers a practical, attribution-driven mechanism to manage uncertainty in high-renewable grids and can guide more reliable, economics-aware UC decisions.

Abstract

This paper introduces a novel approach to addressing uncertainty and associated risks in power system management, focusing on the discrepancies between forecasted and actual values of load demand and renewable power generation. By employing Economic Dispatch (ED) with both day-ahead forecasts and actual values, we derive two distinct system costs, revealing the financial risks stemming from uncertainty. We present a numerical algorithm inspired by the Integrated Gradients (IG) method to attribute the contribution of stochastic components to the difference in system costs. This method, originally developed for machine learning, facilitates the understanding of individual input features' impact on the model's output prediction. By assigning numeric values to represent the influence of variability on operational costs, our method provides actionable insights for grid management. As an application, we propose a risk-averse unit commitment framework, leveraging our cost attribution algorithm to adjust the capacity of renewable generators, thus mitigating system risk. Simulation results on the RTS-GMLC grid demonstrate the efficacy of our approach in improving grid reliability and reducing operational costs.

Cost Attribution And Risk-Averse Unit Commitment In Power Grids Using Integrated Gradient

TL;DR

The paper tackles uncertainty in power system operation due to forecast errors in load and renewable generation by introducing an Integrated Gradients–inspired cost attribution method. It treats the day-ahead UC/ED cycle as a differentiable mapping to system cost and attributes the resulting difference between actual and forecasted costs to individual assets, ensuring completeness and symmetry properties. Building on this, it implements a risk-averse unit commitment framework that adjusts renewable capacities based on Monte Carlo scenario attributions to mitigate risk while maintaining competitive operating costs; simulations on the RTS-GMLC grid show reduced load shedding and robust cost performance. This approach offers a practical, attribution-driven mechanism to manage uncertainty in high-renewable grids and can guide more reliable, economics-aware UC decisions.

Abstract

This paper introduces a novel approach to addressing uncertainty and associated risks in power system management, focusing on the discrepancies between forecasted and actual values of load demand and renewable power generation. By employing Economic Dispatch (ED) with both day-ahead forecasts and actual values, we derive two distinct system costs, revealing the financial risks stemming from uncertainty. We present a numerical algorithm inspired by the Integrated Gradients (IG) method to attribute the contribution of stochastic components to the difference in system costs. This method, originally developed for machine learning, facilitates the understanding of individual input features' impact on the model's output prediction. By assigning numeric values to represent the influence of variability on operational costs, our method provides actionable insights for grid management. As an application, we propose a risk-averse unit commitment framework, leveraging our cost attribution algorithm to adjust the capacity of renewable generators, thus mitigating system risk. Simulation results on the RTS-GMLC grid demonstrate the efficacy of our approach in improving grid reliability and reducing operational costs.
Paper Structure (15 sections, 16 equations, 9 figures, 4 tables)

This paper contains 15 sections, 16 equations, 9 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Cost Attribution
  3. Integrated Gradient
  4. Production Cost Model
  5. Cost Attribution of PCM
  6. Computation of the Gradient
  7. Case Study: the RTS-GMLC Grid Model
  8. Risk-Averse Unit Commitment
  9. Scenario-Based Risk Allocation
  10. Reliability Adjustment
  11. Simulation results
  12. Load Loss Abatement
  13. Conclusion
  14. Declaration of Competing Interest
  15. Acknowledgments

Figures (9)

  • Figure 1: Generation deficiency, cost attributions and shadow prices for sub-areas in RTS-GMLC.
  • Figure 2: Scenarios and cost attributions for Abel at 2020-07-08 04:00.
  • Figure 3: Scenarios and cost attributions for 317-WIND-1 at 2020-07-08 17:00.
  • Figure 4: Scenarios and cost attributions for 324-PV-1 at 2020-07-08 15:00.
  • Figure 5: Scenarios and cost attributions for 317-WIND-1 at 2020-07-08 19:00.
  • ...and 4 more figures