DE-ICA优化算法在工作模态参数识别的应用

提出一种差分进化(DE)改进的独立成分分析(ICA)优化算法,解决工作模态参数识别时容易陷入局部最优,难以识别出高阶模态参数的问题.通过对悬臂梁的ANSYS仿真数据对比可知:相对于传统的ICA方法,结合差分进化算法的ICA识别的模态参数精度更高,且能分离出更多的高阶模态,更适合于高阶模态参数的识别.

Application of DE-ICA Optimization Algorithm in Operating Modal of Parameter Identification

An independent solve the problem of the parameter identification in operating model, in which component analysis(ICA)algorithm modified by differential evolution(DE)is proposed. To process the operating modal analysis problem, which is easy to go into local optima and difficult to identify higher order modal parameters. ANSYS simulation result on cantilever shows that, comparing to the traditional ICA, the ICA combined with DE identified the modal parameter is more accurately, and is able to extract more higher-order modal, hence is more suitable for solving higher modal parameter identification.