Detection Method for Bolt Loosening Based on Summation Coefficient of Absolute Spectrum Ratio

A bolt loosening detection method based on the summation coefficient of the absolute spectrum ratio technique is proposed to address the prevalent issue of bolt loosening in mechanical connections. This proposed method involves initially collecting vibration and rotation speed signals of the motor bolt connection structure, acquiring the baseline spectrum curve of a healthy structure and the spectrum curves of non-healthy structures under different degrees of bolt looseness through chirp Fourier transform (CFT). Subsequently, the spectrum ratio curves between healthy and non-healthy structures are calculated for different degrees of bolt loosening, and then the Summation Coefficient of Absolute Spectrum Ratio (SCASR) is defined to indicate the looseness. In the mathematical model, a linear relationship is observed between the SCASR and the frequency shift of the resonance peak. To standardize the results of bolt loosening detection, the SCASR could be divided by the number of points in the fixed frequency band to obtain the average of SCASR as the detection index for bolt loosening. Finally, a linear fitting equation is established between the bolt torque and the average of SCASR, so that the change in the average value can be used to determine whether the bolt is loose and evaluate the severity of bolt looseness. The detection performance of this proposed method has been effectively validated through both simulation and experiments. Experimental results indicate that the proposed method can effectively detect bolt loosening, particularly in its early stages, using low-frequency band data in three axes.