Variable-parameter feedforward control for centrifuge shaking table based on nonlinear frequency characteristic model

Abstract

The electro-hydraulic centrifuge shaking table is an effective equipment to investigate the seismic response of large scale structures. However, the shaking table system possesses inherent nonlinearity. In addition, the frequency and acceleration of the motion to be simulated in centrifuge environment are larger than those in normal gravity according to the scaling law. The vibration frequency required in the rigid structure testing should be especially high. Such operation condition enlarges the influence of system nonlinearity on the acceleration tracking accuracy, making it difficult to achieve high-fidelity control of the centrifuge shaking table. In this paper, an additional variable is introduced to the frequency characteristic model to formulate the system nonlinearity. A variable-parameter feedforward control method is proposed to improve acceleration tracking accuracy of the centrifuge shaking table for rigid structure testing. The method can adjust control parameters offline to adapt to various operation conditions. Compared with the traditional command shaping method, no iteration is required in the proposed method, which saves manpower and resources. Experimental verification shows that the proposed method possesses better phase characteristic and higher tracking accuracy than conventional fixed-parameter control methods.