Addressing Heterogeneity in Federated Load Forecasting with Personalization Layers

TL;DR

This paper introduces PL-FL, a federated learning framework with personalization layers that improves load forecasting accuracy and reduces communication costs in heterogeneous energy consumption data scenarios.

Contribution

It proposes a novel personalization layer approach for federated load forecasting, enhancing model performance and efficiency over standard FL and local training.

Findings

PL-FL outperforms FL and local training in accuracy.

PL-FL requires less communication bandwidth.

Extensive simulations validate the effectiveness across multiple datasets.

Abstract

The advent of smart meters has enabled pervasive collection of energy consumption data for training short-term load forecasting models. In response to privacy concerns, federated learning (FL) has been proposed as a privacy-preserving approach for training, but the quality of trained models degrades as client data becomes heterogeneous. In this paper we propose the use of personalization layers for load forecasting in a general framework called PL-FL. We show that PL-FL outperforms FL and purely local training, while requiring lower communication bandwidth than FL. This is done through extensive simulations on three different datasets from the NREL ComStock repository.