【摘要】 通过对结构响应进行连续小波变换将多自由度模态参数识别转化为多个单自由度模态参数识别。建立小波骨架理论公式与由结构输出信号计算而得的小波骨架之差为目标函数的优化问题,通过搜索包含于小波骨架理论公式中的模态参数的取值而使目标值最小,从而将优化问题转化为模态参数识别问题。量子粒子群算法是一种基于群体智能理论的优化算法。将量子粒子群算法应用到上述方法中一次性识别出结构的频率、阻尼和振型。最后采用数值模拟的简支梁对该方法进行有效性验证。结果表明,量子粒子群算法结合连续小波变换可以有效地识别环境激励下的结构模态参数。更多还原

【Abstract】 Multi-DOF structural modal parameter identification was converted into several single-DOF structural modal parameter identifications by treating structural output data with continuous wavelet transformation. An optimization with an objective function of the difference between theoretical formula of wavelet skeleton and the wavelet skeleton calculated from structural output data was performed. The minimum objective value was gained through searching reasonable modal parameters included in the theoretical formula of wavelet skeleton. And the optimization was turned into structural modal parameter identification. Quantum-behaved particle swarm optimization, as a swarm intelligence optimization algorithm,was used in the structural modal parameter identification above to identify the structural modal parameters( frequencies,damp ratios and modal shapes) simulataneously under ambient excitation. Finally,the modal parameter identification method based on quantum-behaved particle swarm optimization combined with continuous wavelet transformation presented herein was verified with a numerical simulation of a simply-supported beam. The results showed that the methodology herein can effectively be used to identify structural modal parameters under ambient excitation.更多还原