A Low-Complexity Data-Driven Algorithm for Residential PV-Storage Energy Management (Papers Track)

Mostafa Farrokhabadi (University of Calgary)

Paper PDF NeurIPS 2024 Recorded Talk Cite
Power & Energy Time-series Analysis

Abstract

This paper uses the principles of online convex learning to propose a momentum-optimized smart (MOS) controller for energy management of residential PV-storage systems. Using the self-consumption-maximization application and practical data, the method's performance is compared to classical rolling-horizon quadratic programming. Findings support online learning methods for residential applications given their low complexity and small computation, communication, and data footprint. Consequences include improved economics for residential PV-storage systems and mitigation of distribution systems' operational challenges associated with high PV penetration.