Fault diagnosis of abnormal vibration of coal mine underground transport machine bearings based on wavelet packet energy entropy

In the process of diagnosing abnormal vibration faults in transport aircraft bearings, sensors are used to directly measure normal signals, and those that do not match the normal signals are divided into abnormal signal sequences and diagnosed in the angle domain. There are cases of weak signal loss, which increases the diagnostic error of transport aircraft bearing faults and reduces the Acc value, affecting the accuracy of diagnosis. Therefore, a fault diagnosis method for abnormal vibration of coal mine underground transport machine bearings based on wavelet packet energy entropy was designed. Perform multi-layer wavelet packet decomposition on the abnormal vibration signal to obtain the decomposition sequence, and calculate the energy entropy of the wavelet packet of the abnormal vibration of the underground transport machine bearing. Using wavelet packet energy entropy as diagnostic weight, determine the division measure of normal and abnormal signal sequences, obtain the signal amplitude of each component angle domain, and reconstruct the equivalent angle domain of bearing abnormal vibration. By combining weak signals, noise, and nonlinearity, a stochastic resonance diagnosis equation for the vibration angle domain of transport aircraft bearings is established to achieve accurate diagnosis of abnormal bearing vibration faults. The final diagnosis result shows that after 5 fine adjustments, the diagnostic error of abnormal bearing vibration fault tends to approach 0; The Acc value approaches 1.0, indicating high accuracy in fault diagnosis and playing an important role in improving the stability of underground coal mine transport machines.