Small-Sample Partial Discharge Diagnosis for GIS Based on Improved Capsule Generative Adversarial Network and Dual-Mode Generator Optimization

Data-driven deep learning (DL) methods excel in gas-insulated switchgear (GIS) partial discharge (PD) diagnosis, but their performance falters in data-scarce scenarios. Constrained by the intricate operating environment of GIS and the transient and stochastic nature of PD, acquiring sufficient PD samples in practical applications remains challenging. Therefore, this article proposes an improved capsule generative adversarial network (GAN) for small-sample GIS PD diagnosis. First, a capsule network (CN) is introduced in both the generator and the discriminator to extract more comprehensive feature information from PD signals, thus avoiding the potential loss of crucial features due to pooling operations. Then, a dual-mode optimization method is used to update the model parameters of the generator, and sample quality and diversity are used as evaluation indicators to guide the training process of the generator, realizing automatic optimization of generator parameter-level model construction. Meanwhile, the discriminator stabilizes the GAN training by adopting the Wasserstein distance and incorporating a gradient penalty (GP) term. Experimental results show that the proposed method achieves diagnostic accuracy of 97.47% in small datasets, significantly outperforming other methods and exhibiting robust performance, providing a viable solution for high-precision and robust GIS PD diagnosis with limited samples.