FedREP: Towards Horizontal Federated Load Forecasting for Retail Energy Providers

Abstract

As Smart Meters are collecting and transmitting household energy consumption data to Retail Energy Providers (REP), the main challenge is to ensure the effective use of fine-grained consumer data while ensuring data privacy. In this manuscript, we tackle this challenge for energy load consumption forecasting in regards to REPs which is essential to energy demand management, load switching and infrastructure development. Specifically, we note that existing energy load forecasting is centralized, which are not scalable and most importantly, vulnerable to data privacy threats. Besides, REPs are individual market participants and liable to ensure the privacy of their own customers. To address this issue, we propose a novel horizontal privacy-preserving federated learning framework for REPs energy load forecasting, namely FedREP. We consider a federated learning system consisting of a control centre and multiple retailers by enabling multiple REPs to build a common, robust machine learning model without sharing data, thus addressing critical issues such as data privacy, data security and scalability. For forecasting, we use a state-of-theart Long Short-Term Memory (LSTM) neural network due to its ability to learn long term sequences of observations and promises of higher accuracy with time-series data while solving the vanishing gradient problem. Finally, we conduct extensive data-driven experiments using a real energy consumption dataset. Experimental results demonstrate that our proposed federated learning framework can achieve sufficient performance in terms of MSE ranging between 0.3 to 0.4 and is relatively similar to that of a centralized approach while preserving privacy and improving scalability.

Advancing Power System Services With Privacy-Preserving Federated Learning Techniques: A Review

“…Hence, the characteristics of horizontal federated learning can be concluded to be similar features and different samples. In [92], a privacy-preserving approach is developed to forecast energy demand for retail energy providers using a horizontal federated learning framework to handle the residential household energy data collected from the smart meters. [47] provides a horizontal federated learning approach for household load identification.…”

“…Demand forecasting [8], [9], [19], [25], [26], [27], [30], [31], [32], [33], [34], [40], [51], [55], [63], [65], [66], [67], [69], [70], [73], [74], [77], [79], [81], [90], [91], [92] Achieving Generation forecasting [7], [35], [38],…”

“…In [114], different machine learning algorithms for building load prediction are analyzed. However, these centralized machine techniques cannot overcome challenges such as data privacy and security [19], [31], [32], [51], [74], communication overhead [25], [31], [32], [34], computational ability [63], data insufficiency [66], [77]; hence, the federated learning based model has been introduced, which can also be used to improve the model's scalability [74], [92] and the model generation ability [30], and to mitigate the problems of data heterogeneity [81], [92].…”

Comparative Analysis of Data-Driven Algorithms for Building Energy Planning via Federated Learning

“…Renewable energy integration at the prosumer end enabled the bidirectional flow of electricity, and distributed energy providers played a crucial role in energy trading, demandside management, load shifting, and infrastructure development. Husnoo et al [21] proposed an FL architecture as a FedREP for retail energy providers to address the scalability issue of a centralized system through a privacy preserving distributed network. It showed compromising results with a mean square error (MSE) of 0.3 to 0.4, comparable to the centralized system with the advantages of a possible network extension and preserving the privacy of connected households.…”