Intelligent method for solving optimal control of discrete stochastic dynamic systems with probability criterion

In this paper, the optimal control problem of discrete nonlinear stochastic systems with probability criterion is investigated, in which the criterion function is the product of probability of some events. It is to maximize the product of probability that the output belong to a given set in every period. Since this problem is an optimal control problem of a discrete stochastic system and more complex than the classical optimal control problem, the criterion function can not be calculated by analytic formulation. The traditional solving techniques are no longer valid. The stochastic simulation based genetic algorithm and artificial neural network is integrated with the dynamic programming, and an intelligent method of solving optimal control strategy is given. The intelligent solving method is programmed by Matlab and realized on computer. An example shows that this intelligent solving method is efficient.