Application of SOM Neural Network and C Means Method in Load Classification

The time-variable and dispersive characteristics of load limit the application of the load model,and the load characteristics classification is an effective approach to solve this problem.In this paper,a new classification method based on SOM neural network and C means algorithm was proposed.Load model parameters are chosen as the characteristic vectors in the cluster analysis.The initial training samples are classified using SOM neural network,and the obtained clustering number and clustering center are given as initial value of C means method for further clustering.Simulation results show that the proposed method can automatically obtain the best classification number,and it is practical and effective for load characteristics classification.