A review on autoencoder based representation learning for fault detection and diagnosis in industrial processes

Process monitoring technologies play a key role in maintaining the steady state of industrial processes. However, with the increasing complexity of modern industrial processes, traditional monitoring methods cannot provide satisfactory performance. In the past decades, deep learning models have achieved rapid development in in-dustrial data analysis, especially autoencoder (AE), which has been widely used to deal with various challenges of process monitoring, and a number of related works have been proposed. This paper aims to present a comprehensive review of AE-based industrial applications, which mainly includes two parts: AE-based repre-sentation learning and monitoring strategies, which illustrate the entire design process of AE-based monitoring methods. In particular, AE, AE variants, and the encoder-decoder framework are briefly introduced first. Sec-ondly, AE-based representation learning is comprehensively reviewed from the aspects of industrial data char-acteristics. Then, the state-of-the-art studies of monitoring strategies, including fault detection strategies and fault diagnosis strategies, are reviewed and discussed. Finally, some prospects for future research are explored.

Automated summary

This paper presents a comprehensive review of AE-based industrial applications, including AE-based representation learning and monitoring strategies. The paper introduces the basic concepts of AE and the encoder-decoder framework, reviews the progress of AE-based representation learning from the perspective of industrial data characteristics, and discusses the latest research on monitoring strategies, including fault detection and fault diagnosis. Future research prospects are also explored.