Research on fault identification method based on multi-resolution permutation entropy and ABC-SVM

Abstract

Aiming at the difficulty of pipeline blockage fault identification, a fault identification method based on multiresolution permutation entropy and artificial bee colony (ABC) algorithm optimized support vector machine (SVM) is proposed. Firstly, the acoustic active detection method is used to collect the signals under different blocking conditions. Then, the extracted acoustic impulse response signal is decomposed into different scale components by using a wavelet transform algorithm, and the arrangement entropy of each component is extracted based on the arrangement entropy theory. And then, based on the permutation entropy theory, the permutation entropy in each component is extracted to effectively characterize the different levels of nonlinear features in the signal. Secondly, the artificial bee colony algorithm is used to optimize the penalty factor and kernel function of support vector machine, and the optimized parameters are used to construct the fault identification model. The results show that the proposed method can improve the identification accuracy of pipeline blockage fault. Meanwhile, compared with the traditional SVM and feed forward neural network with backpropagation (BP) model, it has a better effect.

Keywords

• multiresolution analysis
• permutation entropy
• support vector machine
• pipeline
• blockage
• fault identification