Chiller Fault Diagnosis Based on VAE-Enabled Generative Adversarial Networks

Abstract

Artificial intelligence (AI)-enhanced automated fault diagnosis (AFD) has become increasingly popular for chiller fault diagnosis with promising classification performance. In practice, a sufficient number of fault samples are required by the AI methods in the training phase. However, faulty training samples are generally much more difficult to be collected than normal training samples. Data augmentation is introduced in these scenarios to enhance the training data set with synthetic data. In this study, a variational autoencoder-based conditional Wasserstein GAN with gradient penalty (CWGAN-GP-VAE) is proposed to diagnose various faults for chillers. A detailed comparative study has been conducted with real-world fault data samples to verify the effectiveness and robustness of the proposed methodology. <i>Note to Practitioners</i>&#x2014;This work attacks the fact that faulty training samples are usually much harder to be collected than the normal training samples in the practice of chiller automated fault diagnosis (AFD). Modern supervised learning chiller AFD relies on a sufficient number of faulty training samples to train the classifier. When the number of faulty training samples is insufficient, the conventional AFD methods fail to work. This study proposed a variational autoencoder-based conditional Wasserstein GAN with gradient penalty (CWGAN-GP-VAE) framework for generating synthetic faulty training samples to enrich the training data set for machine learning-based AFD methods. The proposed algorithm has been carefully designed, implemented, and practically proved to be more effective than the existing methods in the literature.