Model-free learning adaptive control for nonlinear systems with multiple time delay

The model-free learning adaptive control and its parameter adaptive prediction with parameter control length for a class of nonlinear discrete-time systems with multiple time delay based on dynamic approximate linearization increment minimized model and recursive predicting model were described. The use of model-free learning adaptive control for nonlinear systems with heavy multiple time delay was introduced. There is no need for structural information, mathematical model, external testing signals, training process, Diophantine equation solution and matrix operation. It has the advantages of small on-line computation amount, excellent real-time operation, simple design (only by using I/O data of the controlled systems) and no unmodelled dynamics. The simulation results demonstrate that the approach is correct and effective.