Application of complexity and approximate entropy on fault diagnoses

In connection with the nonlinear dynamic characteristics shown from the performance of fault rotating mechanical system, based on the research and analysis, complexity and approximate entropy can be used to characterize the system state of motion and non-degree rule. The authors propose to apply complexity and approximate entropy to the feature extraction of fault signal. From the analysis and calculation on simulation of different fault signals, it shows that under different rotating machinery fault conditions, its complexity and approximate entropy are significantly different, which verifies that the two quantities are effective parameters for fault information and they are excellent parameters in terms of extraction and recognition of fault feature. Studies have shown that, the complexity and the approximate entropy value can reflect the nonlinearity of the system. If combine these two parameters, it will be more conducive to recognize and analyze fault signal recognition, enhance the reliability, and thus to study the fault diagnosis of complexity rotating machinery in a more effective way.