Rotating Machinery Remaining Useful Life Prediction Scheme Using Deep‐Learning‐Based Health Indicator and a New RVM

Abstract

Remaining useful life (RUL) prediction plays a significant role in developing the condition-based maintenance and improving the reliability and safety of machines. This paper proposes a remaining useful life prediction scheme combining deep-learning-based health indicator and a new relevance vector machine. First, both one-dimensional time-series information and two-dimensional time-frequency maps are input into a hybrid deep-learning structure network consisting of convolutional neural network (CNN) and long short-term memory network (LSTM) to construct health indicator (HI). Then, the prediction results and confidence interval are calculated by a new RVM enhanced by a polynomial regression model. The proposed method is verified by the public PRONOSTIA bearing datasets. Experimental results demonstrate the effectiveness of the proposed method in improving the prediction accuracy and analyzing the prediction uncertainty.

Transfer-Learning-Based Estimation of the Remaining Useful Life of Heterogeneous Bearing Types Using Low-Frequency Accelerometers

“…There are also current approaches, such as the one by Zhang et al [ 54 ], that are superior to the presented one. To the best of the authors’ knowledge, all superior approaches use features of the time–frequency domain as the input.…”

“…The results in the form of the relative error ( Er ), its mean, and the PHM score are presented in Table 4 . In addition, the results of Sturisno et al [ 51 ] (winner of the academics), Porotsky and Bluvband [ 52 ] (winner of the industrial), Zheng [ 53 ] (a current work), and Zhang et al [ 54 ] are also presented. The approach of Zhang et al is the best current one in terms of PHM score and mean relative error.…”

“…The few other current approaches that reach a higher PHM score, such as the one of Zhang et al [ 54 ], use the time–frequency domain. They consider the natural frequencies of the bearing components Hz, 6000 Hz, and 12,000 Hz, for their RUL estimation by using the whole bandwidth of the available frequencies.…”

Scalability, Explainability and Performance of Data-Driven Algorithms in Predicting the Remaining Useful Life: A Comprehensive Review

“…Zhang et al created a hybrid network consisting of a CNN and LSTM to create HIs. The RVM was then used to create the relevance vectors based on the data and their HIs, to create a polynomial model for RUL prediction [127]. Other researchers combined particle swarm optimization, extreme learning machine and RVM for RUL prediction [196].…”

“…[26,27,38,55,58,59,65,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128] …”