Research on Bearing Fault Diagnosis Algorithm of Shift Invariant Dictionary Matching Pursuit Based on Objective Optimization
As a core rotating component,rolling bearings are widely used in high-speed trains.Fault diagnosis of rolling bearings is of great significance to maintain the safety and comfort of trains.In the process of fault diagnosis,it is very important to realize the detection of incipient weak fault and the separation and extraction of compound fault.Therefore,a shift invariant dictionary matching pursuit algorithm based on dung beetle optimization algorithm is proposed in this paper.The thought of pre-defined dictionary structure and cyclic extraction are adopted in this algorithm.Firstly,the dung beetle optimization algorithm is introduced and a novel objective optimization function based on correlated kurtosis is established to adaptively construct an impulse dictionary of pre-defined shift invariant structure.Meanwhile,in order to obtain the optimal sparsity in the matching pursuit framework adaptively,an optimal sparsity acquisition criterion based on Hoyer index is proposed.In order to verify the superiority of the algorithm,the fault signals of gearbox bearing and axle-box bearing commonly used in high-speed train transmission system are analyzed and compared.The results show that the algorithm is more robust to the incipient weak fault detection and has better adaptive detection ability to the compound fault,which has certain engineering application value.
NETL
NSTL
万方数据
