Real-time adaptive control for nonlinear stochastic systems

An efficient model of nonlinear stochastic systems that can use on initial batch of data is developed using orthogonal estimation including the error reduction test and other monitoring modifications. A recursive identification on-line algorithm is implemented to track the nonlinear time variable process. The ELS algorithm is proposed for the parameters, and the nonlinear adaptive controller is designed by the ELS algorithm. The convergence rate of the parameter estimation and the adaptive are tracking are established.