Title: A multivariate grey RBF hybrid model for residual useful life prediction of industrial equipment based on state data
Authors: Xiaoshuang Chen; Hui Xiao; Yejun Guo; Qi Kang
Addresses: Department of Control Science and Engineering, Tongji University, Shanghai 201804, China ' Department of Control Science and Engineering, Tongji University, Shanghai 201804, China ' Department of Control Science and Engineering, Tongji University, Shanghai 201804, China ' Department of Control Science and Engineering, Tongji University, Shanghai 201804, China
Abstract: In order to enhance the prediction precision of performance degradation characteristic parameters, we consider the condition parameters. This paper proposes an improved hybrid model named MVG+RBF which combines multivariate grey model and neural network. This hybrid model has the advantages of both grey model and RBF neural network. Based on the definition of residual useful life (RUL), this paper also gives a forecast model of RUL. Finally, through a case study, we can draw a conclusion that the MVG+RBF model has a better prediction ability and a lower relative error than the traditional MVG model. The proposed model MVG+RBF also has extensive application.
Keywords: multivariate grey modelling; RBF neural networks; state data; residual useful life forecasting; performance degradation; industrial equipment; equipment lifetime.
DOI: 10.1504/IJWMC.2016.075230
International Journal of Wireless and Mobile Computing, 2016 Vol.10 No.1, pp.90 - 96
Received: 13 Jun 2015
Accepted: 12 Jul 2015
Published online: 08 Mar 2016 *