Nonparametric model learning adaptive control method of DC motor

Abstract

Nonparametric model learning adaptive control method (NMLAC) presented in this paper is based on new concepts called pseudo-partial-derivatives (PPD) for a class of nonlinear systems. No structural information, no mathematical model, no training process and no external testing signals are needed. The unmodelled dynamics do not exist. In this paper, nonparametric model learning adaptive control (NMLAC) approach of a class of SISO nonlinear discrete-time systems based on linearization of tight format is applied to DC motor rotate speed control. The design of controller is model-free, based directly on pseudo-partial-derivatives (PPD) derived on-line from the input and output information of the motor motion model using novel parameter estimation algorithms. Simulation experiment examples are provided for real nonlinear systems, which are known to be difficult to model, and control to demonstrate the correctness, effectiveness and advantages of the approaches proposed.