Early Weak Fault Diagnosis of Rolling Bearing Based on Duffing System and LMCD

Abstract

Abstract The early weak fault signals of bearings are usually hidden in a large amount of noise, and the components of the noise are more complicated. Therefore, it is difficult to extract the early weak fault features of bearings using conventional time-frequency methods. In this paper, the traditional Holms-Duffing system has been improved to enhance the anti-noise performance of the system, and then the Duffing system can be used to detect the signal in a broadband frequency through time transformation. At the same time, the four Duffing systems array model has been used to solve the misjudgment problem of the Duffing system caused by the phase difference. When the Duffing system is in a large period state, the complexity of the displacement sequence is smaller than that in the chaotic state. According to this characteristic, a method for diagnosing weak bearing faults based on the Duffing system and LMC difference entropy(LMCD) is proposed. In this paper, simulation signal with strong noise components and the real fault signals have been used to verify the suggested method. The proposed methods in this paper have certain practical value in the field of fault diagnosis.