框架结构松动EEMD损伤识别实验研究
Study on Bolt Looseness Detection in Frame Structures using Ensemble Empirical Mode Decomposition

针对连接结构在振动环境下易发生松动的问题，进行了框架结构模型连接松动损伤识别实验研究。根据螺钉在不同扭矩下的结构稳态响应信号，分析了信号功率谱差异和松动损伤引起的非线性特征，对响应信号进行了总体平均经验模式分解（EEMD），利用第1阶固有模式函数（IMF）构造能量损伤指标进行螺钉连接松动识别。结果表明，基于高频固有模式函数所构造的能量损伤指标可以有效表征不同扭矩下的连接松动所引起的结构非线性损伤，能够较好地反映螺钉连接结构的松动情况。
In the vibration environment, bolted joints are tending to be loosed in service. Based on a bolted frame structure random vibration testing, a vibration-based bolt looseness detecting method is developed by analyzing response signal from bolt joints of various torque states in this paper. Through the discussion about differences of the power spectral density of signals, the nonlinearity of response caused by bolt looseness is analyzed. Using the high-frequency intrinsic mode functions which contain the frequency modulation ingredient based on ensemble empirical mode decomposition (EEMD), an effective energy-based damage index is established to detect the presence of the damage. The experimental result demonstrates that the proposed energy-based damage index based on high-frequency intrinsic mode functions can be used to accurately detect the nonlinear damage caused by bolt looseness under various torque states