Identification of a hydraulic simulator using the modified signal compression method and application to control

It is difficult to identify a nonlinear system using a least square method, adaptive filter, and so on. In the case of a nonlinear system, it is known that the signal compression method is able to estimate uncertain parameters of linear element in a nonlinear system. However, the signal compression method cannot be directly applied to an unique system with hysterisis characteristic because this system includes two different dynamic properties according to its motion direction. So, it is also difficult to identify a motion simulator with several single-rod cylinders with hysterisis characteristic, because single-rod cylinder includes expansion and retraction dynamic properties according to its motion direction. Therefore, this paper proposes the modified signal compression method which is able to estimate uncertain parameters of the motion simulator dynamics. The dynamic properties of this system are identified by separating expansion and retraction properties when applying the signal compression method And, as its application, the identified parameters are applied to a design of a sliding mode controller for the simulator