Approximate Model Predictive Control for Microgrid Energy Management via Imitation Learning

TL;DR

This paper presents an imitation learning framework that approximates economic model predictive control for microgrid energy management, enabling faster decision-making with comparable economic performance.

Contribution

The authors develop a neural network-based imitation learning method that replicates EMPC control actions, ensuring real-time applicability and robustness in microgrid management.

Findings

Achieves economic performance similar to EMPC.

Reduces computation time by about ten times.

Ensures constraint satisfaction through a constraint-tightening approach.

Abstract

Efficient energy management is essential for reliable and sustainable microgrid operation amid increasing renewable integration. In this paper, an imitation learning-based framework to approximate mixed-integer Economic Model Predictive Control (EMPC) is proposed for microgrid energy management, considering fuel generators, renewable energy resources, a unified energy storage unit, and curtailable loads. Within the proposed framework, a neural network is trained to imitate expert EMPC control actions from offline trajectories, thereby enabling fast real-time decision making without solving online mixed-integer optimization problems, which often exhibit highly variable solution times across instances and do not scale well to large problem sizes; in particular, worst-case solve times can be excessively large and therefore unsuitable for real-time deployment. In contrast, the learned policy provides predictable and consistently low computation times. To enhance robustness and generalization, the learning process incorporates noise injection during training to mitigate distribution shift and explicitly accounts for forecast uncertainty in renewable generation and demand. Furthermore, a constraint-tightening approach combined with a projection layer is proposed to ensure recursive feasibility and constraint satisfaction of the learned controller. Simulation results demonstrate that the learned policy achieves economic performance comparable to EMPC, while reducing computation time by approximately one order of magnitude relative to the optimization-based EMPC.