Unsupervised and Interpretable Synthesizing for Electrical Time Series Based on Information Maximizing Generative Adversarial Nets

TL;DR

This paper introduces an unsupervised, interpretable generative model for electrical time series data that automatically extracts physically meaningful features and generates realistic synthetic data without needing labeled datasets.

Contribution

The paper presents the application of infoGAN to electrical time series, enabling controllable, interpretable data synthesis without labeled training data, which is novel in power system data generation.

Findings

Successfully extracts features with physical meanings

Generates realistic synthetic electrical time series

Controls generation results via extracted features

Abstract

Generating synthetic data has become a popular alternative solution to deal with the difficulties in accessing and sharing field measurement data in power systems. However, to make the generation results controllable, existing methods (e.g. Conditional Generative Adversarial Nets, cGAN) require labeled dataset to train the model, which is demanding in practice because many field measurement data lacks descriptive labels. In this paper, we introduce the Information Maximizing Generative Adversarial Nets (infoGAN) to achieve interpretable feature extraction and controllable synthetic data generation based on the unlabeled electrical time series dataset. Features with clear physical meanings can be automatically extracted by maximizing the mutual information between the input latent code and the classifier output of infoGAN. Then the extracted features are used to control the generation results similar to a vanilla cGAN framework. Case study is based on the time series datasets of power load and renewable energy output. Results demonstrate that infoGAN can extract both discrete and continuous features with clear physical meanings, as well as generating realistic synthetic time series that satisfy given features.