抄録
In the fluid power field, a number of applications of the adaptive control theory have been reported in recent years. The adaptive control theory can control the system of which the parameters are unknown or vary during system operation. Especially, the model reference adaptive control theory is very popular. In addition, the controller is generally designed in discrete time. The reason why the discrete-time controller is used is that the model reference adaptive control must perform many calculations. It is difficult to make a continuous-time controller which performs many calculations precisely. In addition, the decrease in cost and progress of performance of a digital computer are helpful for designing the controller for a digital computer.
The application of the model reference adaptive control theory needs two requirements. One is that the plant is of a minimal phase system. This is a very serious problem for the discrete-time system. Most of the discrete-time system composed of a zero-order hold, a continuous-time plant and a sampler in series becomes a nonminimal phase system even if the continuous-time system is a minimal phase system, when the sampling periods of the system is decreased. Its application to electrohydraulic servo systems is no exception.
This paper deals with the application of a generalized minimum variance adaptive control theory to the electrohydraulic servo system. This theory can control the nonminimal phase system. However, there is a difference between the plant and model output. Thus, we tried a combination of a generalized minimum variance adaptive control and additional feedback. As a result, the advantage of the present method is clarified by the simulation and experiment in this paper.
