Remaining useful life prediction techniques for electric valves based on convolution auto encoder and long short term memory

• Published in: ISA transactions Vol. 108; pp. 333 - 342
• Main Authors: Wang, Hang, Peng, Min-jun, Miao, Zhuang, Liu, Yong-kuo, Ayodeji, Abiodun, Hao, Chengming
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.02.2021
• Subjects: Convolutional auto-encoder; Electric valve; Long short term memory; Remaining useful life prediction; Remaining useful life prediction; Convolutional auto-encoder; Long short term memory; Electric valve
• ISSN: 0019-0578; 1879-2022
• DOI: 10.1016/j.isatra.2020.08.031

To optimize the operation and maintenance of nuclear power systems, this study presents a remaining useful life (RUL) prediction method for electric valves by combining convolutional auto-encoder (CAE) and long short term memory (LSTM). CAE can extract deeper features and LSTM is efficient in dealing with time-series data. Moreover, by designing a parallel structure between the outputs of CAE and the original data, features fed into the LSTM are enriched. Also, network structure and corresponding hyper-parameters are compared to obtain a more suitable model. Moreover, the accuracy of the proposed method is tested and compared with other machine learning algorithms. This work also serves as a critical innovation to enhance the safety and economic operation of nuclear plants and other complex systems. •The benefits of condition-based maintenance for nuclear power systems are analyzed.•CAE and LSTM are combined for RUL prediction to compensate for each other.•The parallel structure of CAE and original data is adopted, which enriches the features.