Recognition of power quality disturbance based on artificial bee colony algorithm to optimize kernel extreme learning machine

Aiming at the problem of classification and recognition of power quality disturbance signals, this paper proposes an artificial bee colony algorithm to optimize the classifier model of the kernel extreme learning machine. First, the wavelet transform is used to extract the features of the power quality disturbance signal simulated by Matlab, and then the traditional back-propagation (BP) neural network feedforward neural network is used for classification and recognition. Due to the inherent limitations of the algorithm, the network training calculation is large, the calculation is complicated, and the time is long. Disadvantages, and the proposal of the extreme learning machine just solves this problem, but the accuracy of the classification cannot meet the requirements, and the network needs to be optimized. Therefore, this paper proposes to use the artificial bee colony algorithm to optimize the nuclear extreme learning machine, and classify and recognize the disturbance signal under the optimal parameters of the nuclear extreme learning machine. The simulation results show that after the artificial bee colony algorithm is optimized, the correct rate of classification and recognition has increased by nearly 20%, and the misjudgment rate has dropped to about 3%.