Application of the Transiently Chaotic Neural Network to Nonlinear Constraint Optimization Problems

To deal with the deficiencies of the neural network model based on Hopfield neural network (HNN) for nonlinear constraint optimization problems that is easily trapped in local minimum, a novel optimization network model based on transiently chaotic network (TCNN) is proposed in this paper. Because TCNN has richer and more flexible dynamics compared to HNN, this network model that combined with Lagrange multiplier theory has higher ability of searching for globally optimal solutions to the nonlinear constraint optimization problems. Its asymptotic stability is proved and its equilibrium point is the optimal point of the original problem. The simulation results illustrate the effectiveness of this optimal network algorithm