基于最小二乘支持向量机的结构地震响应时滞控制算法

Least squares support vector machine based time-delay control algorithm for reducing seismic responses of structures

针对时滞会导致控制系统效果降低、控制性能恶化甚至系统不稳定,将线性二次型控制(Linear QuadraticRegulator,LQR)与最小二乘支持向量机(Least Squares Support Vector Machines,LSSVM)进行集成,提出时滞LSSVM-LQR智能控制算法。该算法采集结构状态响应数据后,将用LQR算法计算出的结构最优控制力输入到LSSVM中,以训练并回归预测出时滞后时刻的最优控制力;由作动器对结构提供控制。基于MATLAB平台编写计算程序,并用一幢三层框架结构进行数值验证。结果表明,LSSVM-LQR算法能有效降低时滞对结构控制系统的不利影响。

The existence of time-delay always lowers control effects of control systems, results in worse control performance and even causes system instability. Through integrating the least squares support vector machine (LSSVM) and the linear quadratic regulator (LQR), a new intelligent control algorithm, referred to as the time-delay LSSVM-LQR, was proposed. The algorithm includes the following four steps: acquiring the state-space response data of the structure; using the LQR algorithm to calculate the optimal control forces; inputting the force data to LSSVM for training and giving out the prediction of optimal forces after time-delay; providing the control on the structure with the help of actuators. Based on MATLAB, a program was compiled. Subsequently, the LSSVM algorithm was verified by using the example of a 3-storey frame structure. The results show that the LSSVM algorithm can effectively reduce the adverse impact of time-delay on the control system of structures.