Hyper-Chaotic Neural Network Based on Newton Iterative Method and Its Application in Solving Load Flow Equations of Power System

摘要

Most of load flow equations of power system are multi-variable nonlinear equations set which need to know the initial value for solving, and the initial value is very difficult to choose. Neural network is a kind of highly complex nonlinear dynamic system and chaotic phenomenon is found in it. By utilizing the simulated annealing mechanism to eliminate transiently chaotic neuron, this paper presents a kind of chaotic neuron which can permanently sustain chaotic search. The topology of chaotic neural network composed of four chaotic neurons in which hyper-chaos exists is studied. For the first time, a novel method to find all solutions of nonlinear equations is proposed in which initial points are generated by hyper-chaotic neural network. The numerical example shows that the new method proposed in this paper is correct and effective, and it lays a good engineering foundation for finding all the solutions of load flow equations of power system.