Robust control strategies for motor servo systems

The demands for high performance motor servo systems have been increasing in a wide range of industrial applications. A good control strategy should be able to provide satisfactory control performances in the presence of plant parameter variations, external disturbances, unmodeled dynamics and actuator nonlinearities. The objective of this research is to develop fast, effective and practical robust speed/position control strategies for motor servo systems.;Fuzzy control is capable of yielding robust control performances but there are no systematic design procedures for fuzzy controllers. A Genetic Algorithm Based Fuzzy Control (GABFC) is proposed to overcome the above problem and to yield optimal scaling factors. Sliding mode control is also a salient robust control approach but suffers from annoying chattering. An Equivalent-Current-Filtered Sliding Mode control (ECFSM) is developed which uses an inner current control loop as an internal filter to reduce chattering. The GABFC and ECFSM also form a basis of the advanced hybrid robust control strategies.;Motivated by the adaptive model following concept, a novel robust model following scheme Nominal-Model Following Control (NMFC) is developed. The control architecture of NMFC is quite different from and much simpler than those of prevailing adaptive model following control schemes. NMFC has three main features: the use of a nominal model of the plant as a reference model, the introduction of a model controller which fulfills the reference tracking requirement and the inclusion of a simple PID correction mechanism which copes with all dynamic deviations of the real plant from its nominal model. With its novel robust control architecture, NMFC separates the reference tracking and robustness fulfillment into two independent problems. Both simulation and experimental results illustrate its effectiveness and remarkable robustness against parameter variations, load disturbances and unmodeled dynamics. Another important contribution of NMFC is that it also provides an attractive robust control architecture. Based on this architecture, many other robust control schemes can be developed.;By combining GABFC with NMFC, a Fuzzy Nominal Model Following Control (FNWC) is developed where the model controller is a GA based fiizzy controller. Simulation results show that FNMFC can retain satisfactory system performance even in the presence of significant plant parameter variations and load disturbance. The comparative study between FNMFC. GABFC and NMFC demonstrates the superiority of FNMFC.;To improve load disturbance rejection of ECFSM during non-sliding transient, ECFSM and NMFC are combined together and a Variable Structure Nominal Model Following Control (VSNMFC) is developed. VSNMFC uses a sliding mode controller as the model controller. Simulation results illustrate the effectiveness of VSNMFC and its superiority over ECFSM and NMFC.;To tackle deadzone and other unmodeled nonlinearities more effectively, a deadband-relay compensator is proposed to combine with the simple PID correction mechanism and a Deadband-Relay Nominal Model Following Control (DRNMFC) results. Both simulation and experimental results demonstrate that DRNMFC not only yields satisfactory robustness against deadzone nonlinearities and plant parameter variations but also speeds up the error convergence.