Hysteretic chaotic operator network and its application in wind speed series prediction

A novel hysteretic chaotic operator network is constructed to improve the prediction performance of the wind speed series. The network is composed of three layers: the input layer, the chaotic operator layer and the hysteretic output layer. The hysteretic output can enhance the storage and memory capacity of the network, which can restrain the error change of the neuron state. Genetic algorithm is used to change the dynamic behavior of the network to follow that of the predicted system. Thus, the network can obtain the regular information contained in the training samples, and the dynamic prediction can be performed. Simulation results show that the network can be applied to perform the wind speed series prediction, and it can obtain better prediction performance than conventional methods. (C) 2015 Elsevier B.V. All rights reserved.