A kurtosis-based fastica and approximation solution domain for stlbo motor bearing fault diagnosis

Abstract

The invention relates to a kurtosis-based FastICA and an STLBO motor bearing fault diagnosis method approximating the solution domain. The method of approximating the solution domain is combined with the initial training of BPNN to determine the upper and lower bounds of network structure parameters and TLBO initialization population; finally, the TLBO correction algorithm of adaptive dynamic learning factor is integrated, and the optimal weights and thresholds are iteratively searched for input into BPNN. The invention can enhance signal characteristics, reduce noise interference, and improve the recognition rate of fault diagnosis; for unknown sources, it does not need too much prior knowledge and theoretical reserves, and has strong generalization; It is necessary to set specific parameters and dynamically change the learning factor according to the current number of iterations, which can improve the convergence speed of the algorithm, and at the same time avoid falling into local convergence, which is convenient for debugging and simple in calculation.

Description

technical field [0001] The invention relates to a motor fault diagnosis method, in particular to a kurtosis-based FastICA and an STLBO motor bearing fault diagnosis method based on an approximation solution domain. Background technique [0002] The motor is the core equipment that converts electrical energy into mechanical energy. With the development of science and technology, the normal operation of various fields is inseparable from the motor. However, in the complex industrial environment where the motor runs for a long time, it is prone to the influence of its own factors such as parts aging and excessive load pressure, coupled with the influence of various external environmental factors, various failures are prone to occur. Once a fault occurs, if it is not diagnosed and repaired in time, it will cause unpredictable damage to production and life, and even life. Therefore, it is of great significance to deeply study the methods of motor fault diagnosis and find more ac...

AI Technical Summary

Problems solved by technology

However, in actual research, most algorithms rely largely on the researchers' prior knowledge and practical experience as support, and the many parameters involved in the algorithm require a huge workload during the debugging process, and the generalization ability is low. There are also disadvantages of slow algorithm convergence, low reliability and practicability

Images