Fault diagnosis method for hydraulic pump based on EMD-AR (empirical mode decomposition-auto-regressive) and MTS (mahalanobis taguchi system)

Abstract

The invention provides a fault diagnosis method for a hydraulic pump based on EMD-AR (empirical mode decomposition-auto-regressive) and MTS (mahalanobis taguchi system). The fault diagnosis method includes firstly collecting time-domain signals of the hydraulic pump in three states of normality, original rotor wear-out failure of a valve plate and wear-out failure of sliding shoes and a swash plate; then subjecting each group of the time-domain signals to EMD, and subjecting each acquired basic model component to energy-normalization processing; establishing an AR model according to each energy-normalized basic model component to acquire a characteristic matrix corresponding to the AR model, and subjecting the AR model to singular value decomposition; constructing a benchmark space by singular values in the normality, optimizing the same by an orthogonal table, calculating Mahalanobis distance between test data in the states of the normality and failures and the benchmark space, and finally judging the state of the hydraulic pump and performing failure recognition according to the Mahalanobis distance. The fault diagnosis method is applicable to processing non-stationary signals, can acquire more complete failure characteristics, so that effect of failure recognition is better, and cost is lower.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis of hydraulic pumps, and specifically relates to a method based on EMD-AR (EMD: Empirical Mode Decomposition, empirical mode decomposition; AR: Auto-regressive, auto-regressive model) and MTS (MTS: Mahalanobis Taguchi System, Martin System) hydraulic pump fault diagnosis method. Background technique [0002] The hydraulic pump is a key component in the hydraulic system, and its performance has an important impact on the reliability of the entire hydraulic system operation. Once the hydraulic pump fails, the vibration and noise will increase, reducing work efficiency; and the hydraulic pump will not work. , and even cause serious accidents. Therefore, the performance detection and fault diagnosis of hydraulic pumps are of great significance in industrial applications. Due to the incompressibility of the fluid, the fluid-solid coupling between the pump source and the hydraulic circuit, and ...

AI Technical Summary

Problems solved by technology

[0010] In order to overcome the shortcomings of the traditional hydraulic pump diagnosis method, such as not being suitable for processing non-stationary signals, complex reasoning, and high cost, the present invention proposes a hydraulic pump fault diagnosis method based on EMD-AR and MTS

Images