A weak fault diagnosis scheme for common rail injector based on MGOA-MOMEDA and improved hierarchical dispersion entropy

Abstract

Aiming at the problem that the common rail injector’s early fault characteristics are very weak and susceptible to random noise and other signal interference, this paper proposes a new common rail injector weak fault diagnosis method based on multipoint optimal minimum entropy deconvolution adjusted based on modified grasshopper optimization algorithm optimization algorithm (MGOA-MOMEDA), improved hierarchical dispersion entropy, and least square support vector machine. First, the fault period T is determined using the multipoint kurtosis spectrum. Through the MGOA optimization algorithm, the optimal filter length L of MOMEDA is obtained adaptively, and the optimal performance filter is used for filter processing. Then, improved hierarchical discrete entropy is used to measure the complexity of the filtered fuel pressure signal to extract weak fault features. Finally, the fault feature vector is input into the LS-SVM multi-classifier to realize the weak fault diagnosis and recognition of the common rail injector. Through experimental verification, the proposed method can effectively achieve the weak fault diagnosis of the common rail injector.