摘要
基于压电智能结构,构成冲击荷载监测系统。冲击样本的应力波信号经Hilbert变换变为连续的瞬态时序信号后,提取时间和峰值等特征信息训练BP神经网络。在线监测时由神经网络根据压电响应信号映射冲击荷载的位置。实验表明,该方法定位精度高,稳定性好,用于层合板结构低能量冲击的健康监控是可行的。
A system to monitor impact forces on piezoele-cric smart structures is constructed. The signals of the impact stress wave in time domain were transformed into momentary sequential time signals using Hilbert transform. Neural networks were established to extract different characteristics from these signals for automatic identification of impact location. Experimental results show that the method enjoys good accuracy and stability, which may find applications in the health monitoring of laminated structure damages caused by not very energetic impacts.
出处
《力学与实践》
CSCD
北大核心
2003年第6期37-40,共4页
Mechanics in Engineering
基金
国家自然科学基金(10072026)
航空科学基金(01G52041)
