Identification method for disturbance signal of power quality based on improve Harris Hawks optimization-support vector machine

Aiming at the problems that the penalty factor and kernel function parameters of SVM（support vector machine） are easy to fall into the local optimal solution in the optimization process and the Harris Hawks optimization algorithm is easy to fall into the local optimal solution， a method of using IHHO （improved Harris Hawks optimization） algorithm to optimize the penalty factor and kernel function parameters of SVM and constructing IHHO-SVM classifier to identify disturbance signals of power quality is proposed. By adding 0 dB，20 dB and 30 dB Gaussian white noises to nine different disturbance signals of power quality， the improved empirical mode decomposition algorithm of adaptive noise complete set is used to decompose the signal， and the energy entropy and sample entropy of the first three intrinsic mode function components of the signal are extracted as a set of feature vectors. The feature vectors are normalized and input into nine classifiers for comparison. The simulation results show that the recognition accuracy of IHHO-SVM classifier is 99.11%， 97.78% and 97.33%， respectively， when the signal is added with 0 dB，20 dB and 30 dB Gaussian white noises. The classification effect of IHHO-SVM classifier is the best among all classifiers， which proves the accuracy， superiority and noise immunity of its classification.