Fault prognosis of industrial robots in dynamic working regimes: Find degradation in variations. (March 2021)

• Record Type: Journal Article
• Authors: Yang, Qibo; Li, Xiang; Cai, Haoshu; Hsu, Yuan-Ming; Lee, Jay; Yang, Chun Hung; Li, Zong Li; Lin, Ming Yi
• Is Part Of: Measurement. Volume 173(2021)
• Publication Date: 2021-03
• Subjects: Remaining useful life -- Domain generalization -- Neural network -- Industrial robots -- Dynamic working regimes; ANN Artificial neural network -- CL Center line -- CNN Convolutional neural network -- DANN Domain-adversarial neural network -- DGALSTM Domain-generalization-adversarial long short-term memory -- GRL Gradient reversal layer -- KNN K-nearest neighbors -- LCD Liquid crystal display -- LCL Lower control limit -- LSTM Long short-term memory -- MAE Mean absolute error -- PCA Principal component analysis -- PCA-SPE Principal component analysis-squared prediction error -- PHM Prognostics and health management -- RF Random forest -- RMS Root mean square -- RUL Remaining useful life -- SVM Support vector machine -- SVR Support vector regression -- UCL Upper control limit; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2020.108545
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: RUL prediction using domain-generalization-adversarial long short-term memory. Two-stage health assessment of PCA-SPE and p -chart to reduce outliers. A workflow of fault prognosis to reduce complex variations. Abstract: Industrial robots are widely used in modern factories. Robot faults lead to the inevitable suspension of production lines. The prediction of robot failure can improve production capacity. However, it is challenging due to the variations of robots in dynamic working regimes. This paper presents a methodology of fault prognosis of industrial robots, including (1) a modeling approach of remaining useful life prediction using domain-generalization-adversarial long short-term memory to reduce the robot-to-robot variations, (2) an approach of two-stage health assessment based on principal component analysis-squared prediction error and p -chart that can reduce the disturbance of outliers in normal operations, and (3) a workflow containing feature extraction using wavelet packet decomposition, feature smoothing using exponential smoothing, feature normalization using z-score and feature selection using Pearson correlation coefficient. The case study of liquid crystal display transfer robots shows the methodology can effectively reduce the variations and improve the prediction of RUL.