摘要：The sparse decomposition of vibration signal is an important part of the fault diagnosis of Double Cavity Jaw Crusher. But calculation of sparse decomposition is very large, and it is difficult to fulfill signal processing. After analysing characteristics of Double Cavity Jaw Crusher, this paper proposes to apply the hybrid algorithm, DEPSO which mixes the advantages of particle swarm optimization (PSO) and difference evolution (DE) algorithm to extract signal feature of Double Cavity Jaw Crusher and deal with the search of the best atoms during the signal decomposition. With the combination of PSO and DE, this method avoids falling into the partial optimal solution. Besides, after the algorithm imports the chiasm and variation operations, its adaptability has made a great improvement. The test result shows that applying DEPSO to extracting signal feature of Double Cavity Jaw Crusher improves the search speed, the efficiency and accuracy of decomposition, and the calculation has also dropped down dramatically.
