Decision-focused learning for optimal PV-Battery scheduling

2026-05-27

AI Summary

This study addresses the limitation of existing photovoltaic (PV) power forecasting models, which, despite achieving high statistical accuracy, fail to align with the cost-minimization objective of battery dispatch, thereby yielding suboptimal energy savings. To overcome this, the authors propose a decision-focused learning framework that end-to-end jointly optimizes an LSTM-based PV forecast model with a mixed-integer linear programming (MILP)-driven optimal battery scheduling policy, directly training the predictor to minimize electricity costs rather than prediction error. Evaluated on 14 months of data from 20 households, the approach reduces electricity bills by 3.6% on average; when augmented with a warm-start strategy, savings increase to 8% (p<0.001), significantly outperforming baseline methods—even those with lower RMSE in power prediction.

Abstract

The use of residential photovoltaics has increased dramatically in recent years. With battery systems becoming more affordable, the optimal operation of a photovoltaic-battery system can bring significant savings to households. Optimal control requires correct forecasts of underlying parameters, such as photovoltaic power generation, to schedule the battery. While forecasting models have become increasingly accurate due to algorithmic advances and data availability, accuracy is typically measured in generic metrics which might not align with the downstream application. This study proposes a decision-focused learning framework that integrates optimization and prediction by training a Long Short-Term Memory photovoltaic energy forecaster on the downstream optimal scheduling of a battery system. The proposed methodology is compared against a standard two-phase approach. Across a 14-month evaluation period, the decision-focused method reduced average electricity costs across twenty buildings by 3.6% when normalized against performance bounds defined by a perfect forecast and a baseline of no optimization. Critically, this financial improvement was achieved despite the model exhibiting a root mean squared error of 19.9%, significantly higher than the decoupled model's 8.2%. Warm-starting the decision-focused model further improves results, lowering average cost by approximately 8%, while also mitigating the negative impact on statistical accuracy (root mean squared error of 13.7%). The findings are statistically significant at the 0.001 level across the twenty households and for each household individually. These results demonstrate that aligning forecast models with optimization goals is key for achieving cost advantages in PV-battery systems. Future research should replicate these findings on other datasets, alternate forecasting models and alternate optimization algorithms.

Problem

• decision-focused learning
• PV-battery scheduling
• forecast optimization alignment
• energy forecasting
• residential photovoltaics

Innovation

• decision-focused learning
• PV-battery scheduling
• integrated prediction and optimization
• LSTM forecaster
• warm-starting