Model-based adaptive feedforward compensation for disturbance caused by overturning moment in 2-dimensional shaking table systems

Abstract

This paper presents a disturbance compensation approach using a model-based feedforward control with an adaptive algorithm for the two-dimensional (2D) shaking table systems. In the system, the overturning moment due to a specimen largely deteriorates the motion performance of the table, thereby decreasing the reproducibility of the desired earthquake acceleration. To solve the problem, first, as one of the disturbances, the overturning moment is modeled for the target shaking table system. In the modeling process, a physical model is derived based on the geometrical arrangement and the equation of motion, wherein the moment is modeled as a disturbance acting on each actuator. Based on this model, the feedforward compensators based on the mathematical disturbance model are adopted to cancel the disturbance. In addition, an adaptive algorithm is employed to reduce the effect of the modeling error and/or parameter variation. The designed compensators are validated by conducting experiments using a 2D laboratory prototype of the shaking table system.

Keywords

• shaking table systems
• disturbance compensation
• modeling
• feed forward compensation
• recursive least-squares algorithm
• acceleration reproducibility