Online Energy Storage Arbitrage under Imperfect Predictions: A Conformal Risk-Aware Approach

TL;DR

This paper introduces a conformal risk-aware control method for energy storage arbitrage that dynamically manages downside risk from imperfect price forecasts without relying on distributional assumptions.

Contribution

It develops an online calibration approach using conformal decision theory and temporal difference errors to control risk and improve arbitrage decisions under forecast uncertainty.

Findings

Proves bounded long-term risk with convergence guarantees.

Achieves better risk-opportunity balance than benchmarks.

Effectively manages profit loss risk in case studies.

Abstract

This work proposes a conformal approach for energy storage arbitrage to control the downside risk arising from imperfect price forecasts. Energy storage arbitrage relies solely on predictions of future market prices, while inaccurate price predictions may lead to significant profit losses. Based on conformal decision theory, we describe a controller that dynamically adjusts decision conservativeness through prediction sets without distributional assumptions. To enable online calibration when online profit loss feedback is unobservable, we establish that a temporal difference error serves as a measurable proxy. Building on this insight, we develop two online calibration strategies: prediction error-based adaptation targeting forecast accuracy, and value error-based calibration focusing on decision quality. Analysis of the conformal controller proves bounded long-term risk with convergence guarantees in temporal difference error, which further effectively manages risk exposure in potential profit losses. Case studies demonstrate superior performance in balancing risk and opportunity compared to benchmarks under varying forecast conditions.