Synthetic Data Generation for Residential Load Patterns via Recurrent GAN and Ensemble Method

TL;DR

Comprehensive evaluations demonstrate that the developed ERGAN method consistently outperforms established benchmarks in the synthetic generation of residential load data across various performance metrics, including diversity, similarity, and statistical measures.

Abstract

Generating synthetic residential load data that can accurately represent actual electricity consumption patterns is crucial for effective power system planning and operation. The necessity for synthetic data is underscored by the inherent challenges associated with using real-world load data, such as privacy considerations and logistical complexities in large-scale data collection. In this work, we tackle the above-mentioned challenges by developing the Ensemble Recurrent Generative Adversarial Network (ERGAN) framework to generate high-fidelity synthetic residential load data. ERGAN leverages an ensemble of recurrent Generative Adversarial Networks, augmented by a loss function that concurrently takes into account adversarial loss and differences between statistical properties. Our developed ERGAN can capture diverse load patterns across various households, thereby enhancing the realism and diversity of the synthetic data generated. Comprehensive evaluations demonstrate that our method consistently outperforms established benchmarks in the synthetic generation of residential load data across various performance metrics including diversity, similarity, and statistical measures. The findings confirm the potential of ERGAN as an effective tool for energy applications requiring synthetic yet realistic load data. We also make the generated synthetic residential load patterns publicly available.

Figures (6)

• Figure 1: The ERGAN framework for generating synthetic residential load patterns via an ensemble of recurrent GANs and K-means clustering. Note that this framework involves clustering original load patterns using K-means and Davies-Bouldin score, training separate Bi-LSTM GAN models for each cluster, generating synthetic patterns from the trained generators, and combining the synthetic clustered datasets to form the final synthetic load pattern dataset while preserving cluster structures.
• Figure 2: Selecting the best number of clusters based on the Davies-Bouldin score.
• Figure 3: Pattern and correlated autocorrelation comparison of original and synthetic residential load patterns via different generation methods.
• Figure 4: Comparative histograms of original and synthetic residential load patterns via different generation methods.
• Figure 5: Hourly comparative boxplots for original and synthetic load patterns generated by ERGAN and benchmark methods.