【摘要】 讨论了振动控制中采用试验模态分析技术实现模态滤波器，以获取模态坐标时传感器的优化配置问题。为了使由频响函数所构成的矩阵条件数最小，以减小模态滤波器误差，提出了一种次优方法。它逐步消除那些对该矩阵条件数贡献最小的传感器可选位置，一直到只剩下希望的传感器数目的位置为止。这种方法简单易行，效率较高。并通过一个空间智能桁架结构的传感器优化配置研究，说明该方法的有效性更多还原

【Abstract】 The problem of optimal sensor placement is discussed when the sensor outputs are used to extract the modal coordinates by means of Modal Filters (MF) implemented on the basis of the experimental modal analysis theory. To make the MF errors as small as possible, the condition number of the matrix formed from frequency response functions is minimized by the proposed successive reduction approach. The approach reduces successively the number of candidate sensor locations which contribute less to the condition number (the objective function) until the number of remained candidate sensor locations is equal to the number of required sensors. It is efficient and easy to use. Furthermore, the approach is applied to place sensors optimally on an intelligent space truss structure. Simulations demonstrate that the approach is effective.更多还原