Intelligent tracking control of a DC motor driver using self-organizing TSK-type fuzzy neural networks

Abstract

In this paper, a self-organizing TakagiSugeno-Kang (TSK) type fuzzy neural network (STFNN) is proposed. The self-organizing approach demonstrates the property of automatically generating and pruning the fuzzy rules of STFNN without the preliminary knowledge. The learning algorithms not only extract the fuzzy rule of STFNN but also adjust the parameters of STFNN. Then, an adaptive self-organizing TSK-type fuzzy network controller (ASTFNC) system which is composed of a neural controller and a robust compensator is proposed. The neural controller uses an STFNN to approximate an ideal controller, and the robust compensator is designed to eliminate the approximation error in the Lyapunov stability sense without occurring chattering phenomena. Moreover, a proportional-integral (PI) type parameter tuning mechanism is derived to speed up the convergence rates of the tracking error. Finally, the proposed ASTFNC system is applied to a DC motor driver on a field-programmable gate array chip for low-cost and high-performance industrial applications. The experimental results verify the system stabilization and favorable tracking performance, and no chattering phenomena can be achieved by the proposed ASTFNC scheme.

Intelligent complementary sliding-mode control with dead-zone parameter modification

“…In this paper, L298 motor driver and HCTL-2032 decoder are used to improve system performance in digital closed loop motion control. Meanwhile, the DC motor driver can be represented in the following form [37,38] u JR…”

Intelligent total sliding‐mode control with dead‐zone parameter modification for a DC motor driver

“…The threshold s th is usually selected according to the desired control performance specified by the designers. In the case of s th = 0, the modified parameter adaptation laws in (35)-(39) are the same as the traditional ones in (24)- (28), which have been widely used in the existing literature. If s th is too large, then the parameter adaptation laws can be easily stopped even if the network parameters are not well learned.…”

“…DC motor has been widely applied to automotive, treadmill, magnetic bike, vending machine, massage tool, auto shutter, auto mah‐jongg table, medical equipment etc. The dynamic equation of a DC motor driver can be represented as [27, 28] where θ is the rotor position, is the system dynamics, g = K t / JR a is the control gain, u is the armature applying voltage used as the control input, J is the moment of inertia of the motor shaft, B is the damping coefficient, R a is the armature resistance, K t is the torque constant and K b is the back electromotive force coefficient. The control objective is to find a control law so that the rotor position θ can track a position command θ c as closely as possible.…”

A New Adaptive and Self Organizing Fuzzy Policy to Enhance the Real Time Control Performance

“…Generally, SOFLC systems have demonstrated their usefulness in applications that do not present high level of complexity, since it does not require more than tuning the consequents of the rules. On the other hand, more complex systems require designing more sophisticated controllers, that's because, SOFLCs are only capable of a coarse tuning of the controller parameters [7][8][9][10][11].…”