基于神经网络的滑移隔震结构智能半主动控制

考虑上部结构的刚度和阻尼,使用神经网络控制算法计算基底摩擦力的大小,研究了滑移隔震结构的半主动控制。对计算实例的分析表明,通过半主动控制的滑移隔震结构不但具有较好的隔震效果,且能有效地减小基底的最大滑移量及残余位移。为对比各种控制方法的控制效果,文中还利用Bang-Bang控制和瞬时最优控制算法对滑移隔震结构进行了半主动控制。对比分析表明,基于神经网络控制算法的控制效果优于其它控制算法,具有反馈量少,稳健性强等特点。

Semi-active intelligent control of sliding structure based on artificial neural network

Considering the stiffness and the damping of super structure,this paper proposes an intelligent semi-active strategy to control the base friction force of sliding structures by artificial neural network.Numerical results show that the friction-controllable sliding structures not only have good isolation effect,but also reduce the maximum slippage and residual base displacement.In order to compare the control effect of different methods,the authors use Bang-Bang control method and instantaneous optimal control method to control the sliding structures.Simulation results clearly indicate that the artificial neural network control method has some advantages when comparing with other methods,such as small quantity to be fed back,robust characteristic and so on.