Sparse Diagnosis Method for Rolling Bearing Faults

Abstract

The invention discloses a rolling bearing fault sparse diagnosis method based on average random weak orthogonal matching tracking. Firstly, an over-complete dictionary is constructed according to the collected vibration signals of the rolling bearing, the initialization setting of the algorithm parameters is completed, and the sparseness of the original signal is estimated; Secondly, use the average random weak orthogonal matching pursuit algorithm to update the sparse dictionary and residuals; finally, use the obtained sparse dictionary to calculate the sparse representation coefficients, so as to reconstruct the fault signal; repeat the above process N times, and obtain the final processing by set averaging result. This method avoids the influence of artificial sparsity on the decomposition results through the atomic number estimation and the improved residual update method, and the improved simulated annealing algorithm increases the possibility of small-amplitude fault components being extracted, and solves the weak periodicity The problem that the impact features are difficult to be extracted effectively is of great significance for the early weak fault diagnosis of rolling bearings.

Description

technical field [0001] The invention belongs to the technical field of signal processing, and relates to a rolling bearing fault sparse diagnosis method based on average random weak orthogonal matching tracking, which can realize effective extraction of fault features in rolling bearing vibration signals. Background technique [0002] In recent years, new energy, as a renewable and clean energy that can be recycled, has been highly valued by countries all over the world, and some new energy utilization technologies have made great progress. With the characteristics of wide distribution and large reserves, wind energy has achieved sustained and rapid development in countries all over the world. As a country with great economic growth, my country has always used wind power as the main power generation. It is estimated that the installed capacity of wind power in my country will reach 1 billion kilowatts in 2050. As the key equipment for converting wind energy into electrical ...

AI Technical Summary

Problems solved by technology

This method can well overcome the shortcomings of the orthogonal matching pursuit, such as the difficulty of adapting the sparsity and the difficulty of extracting weak impact features, and then realize the effective extraction of weak fault features.

Images