Int J Performability Eng ›› 2019, Vol. 15 ›› Issue (5): 1453-1461.

### Analog Circuit Fault Prognostic Approach using Optimized RVM

Chaolong Zhanga,b,*, Yigang Heb, Shanhe Jianga, Lanfang Zhanga, and Xiaolu Wangc

1. a School of Physics and Electronic Engineering, Anqing Normal University, Anqing, 246011, China
b School of Electrical Engineering and Automation, Wuhan University, Wuhan, 430072, China
c School of Information Technology, Jiangsu Vocational Institute of Commerce, Nanjing, 211168, China
• Submitted on  ;
• Contact: * E-mail address: zhangchaolong@126.com
• About author:Chaolong Zhang received his Ph.D. from Hefei University of Technology in 2018. He is currently an associate professor in the School of Physics and Electronic Engineering at Anqing Normal University. He is also a postdoctoral researcher in electrical engineering at Wuhan University. His current research interests include fault diagnostics and prognostics of analog and mixed-signal circuits and battery capacity prognostics; Yigang He received his Ph.D. from Xi'an Jiaotong University in 1996. He is currently a professor and doctoral supervisor in the School of Electrical Engineering at Wuhan University. He is one of the winners of the National Distinguished Young Scientists Foundation. His research interests include circuit theory and its applications, testing and fault diagnosis of analog and mixed-signal circuits, smart grid, radio frequency identification technology, and intelligent signal processing; Shanhe Jiang received his Ph.D. from Jiangnan University in 2015. He is currently a professor in the School of Physics and Electronic Engineering at Anqing Normal University. His main research interests include computational intelligence, power system economic dispatch, multi-objective optimization, and artificial intelligence; Lanfang Zhang received his Master's degree from Hefei University of Technology in 2009. He is currently an experimentalist in the School of Physics and Electronic Engineering at Anqing Normal University. His main research interests include fault diagnostics and prognostics; Xiaolu Wang received her Master's degree from Nanjing University of Aeronautics and Astronautics in 2012. She is currently a lecturer in the School of Information Technology at Jiangsu Vocational Institute of Commerce. Her research interest is computer applications.
• Supported by:
This work was supported by the National Natural Science Foundation of China (No. 51607004) and Anhui Provincial Natural Science Foundation (No. 1608085QF157) and Natural Science Research Key Project of Education Department of Anhui Province (No. KJ2018A0369).;

Abstract: In this paper, a novel analog circuit fault prognostic approach is presented. The Pearson product-moment correlation coefficient (PPMCC) is used to calculate the circuit's health degree on the basis of the extracted output voltages. The relevance vector machine (RVM) algorithm with kernel function optimized by the quantum-behaved particle swarm optimization (QPSO) algorithm is utilized to estimate the remaining useful performance (RUP). A leapfrog filter is used in a fault prognostic experiment to verify the prognostic approach, and the experimental results reveal that the presented approach can forecast the analog circuit's RUP precisely.