An improved satin bowerbird optimization-based error-tracking compensation algorithm for active control system considering large time delay

The issue of time delay remains prominent in the domain of active vibration control, owing to the dependence on external actuators for force generation. However, while most currently available methods can achieve small time delay compensation, their effectiveness in compensating for large delays is limited. This paper proposes a large time-delay compensation algorithm for active control systems. Based on improved satin bowerbird optimization (ISBO), it facilitates the acquisition of appropriate compensation control. The ISBO adopts adaptive weights and the Cauchy inverse cumulative distribution function to update and obtain the ideal compensation control force. Numerical simulations of a 20-story benchmark model were performed to evaluate the proposed algorithm, and the classical phase-shifting and third-order polynomial algorithms were included for performance comparison. Furthermore, the platform of the hybrid test was constructed and the proposed algorithms was validated by hybrid tests, wherein the 20-story benchmark model was regarded as the analytical substructure while the active mass damper (AMD) device was considered as the experimental substructure. The consistent results of the numerical simulations and hybrid tests demonstrate the effectiveness of the proposed error-tracking large time delay compensation algorithm. Classical phase-shifting and third-order polynomial algorithms lose their compensation capacity when the time delay is greater than 100 ms. In contrast, the proposed method remains stable and can achieve satisfactory control performance even under a delay of 250 ms, resulting in a reduction in the average acceleration on the top floor from 6.705 to 1.911.