Adaptive Pole-Placement Control Incorporating Neural Network for Pneumatic Servo System.

In pneumatic servo system, an adaptive control method is useful since change in load mass and air compressibility cause parameter variation of the plant. When the discrete time model of the plant becomes non-minimum phase, the adaptive pole-placement control should be applied. However, it is difficult to obtain satisfactory control performances by only using the adaptive pole-placement control when the non-linearity of the plant is not negligible. On the other hand, it is well-known that neural network is effective for such nonlinear plant. This paper presents a design scheme of the discrete-time adaptive pole-placement control incorporating neural network for a pneumatic servo system. In this experiment, we compare the proposed design scheme with a design scheme only using adaptive pole-placement control. As the results, proposed design scheme shows superior control performance in both the transient and steady-state response suppressing the effect of the non-linearity.