A novel feature selection method to boost variable predictive model–based class discrimination performance and its application to intelligent multi-fault diagnosis

Abstract

Effective and efficient incipient fault diagnosis is vital to the maintenance and safe application of large-scale key mechanical system. Variable predictive model–based class discrimination is a recently developed multiclass discrimination method and has been proved to be potential tool for multi-fault detection. However, the vibration signals from dynamic mechanical system always present non-normal distribution so that the original variable predictive model–based class discrimination might produce the inaccurate outcomes. An improved variable predictive model–based class discrimination method is introduced at first in this work. At the same time, variable predictive model–based class discrimination will suffer computation difficulty in the case of high-dimension input features. Therefore, a novel feature selection method based on similarity-fuzzy entropy is presented to boost the performance of the variable predictive model–based class discrimination classifier. In this method, the ideal feature vectors are optimized to acquire more accurate similarity-fuzzy entropies for the input features. And, the one with the largest similarity-fuzzy entropy value is removed to refine input feature subset. Moreover, the optimal input features are repeatedly evaluated using the improved variable predictive model–based class discrimination classifier until the expected results are achieved. Finally, the incipient multi-fault diagnosis model for a hydraulic piston pump is established and verified by experimental test. Some comparisons with commonly used methods were made, and the results indicate that the proposed method is more effective and efficient.