Fault Detection of Servo Motor Bearing Based on Speed Signal Under Variable Speed Conditions

As a crucial component in servo motors, rolling bearings are susceptible to malfunctioning, and the diagnosis of bearing faults is imperative to guarantee the dependable operation of equipment. In this innovative alternative, the paper proposes a bearing fault diagnosis algorithm based only on the speed signal, which enables automatic alarm functions under variable speed conditions. Firstly, the algorithm converts the unfiltered speed signal into the equal-angle interval speed signal based on the cubic spline interpolation algorithm. Secondly, the optimized maximum correlation kurtosis deconvolution algorithm suitable for variable speed conditions is proposed to extract bearing fault features. Finally, an automatic alarm strategy is proposed based on the order spectrum characteristics of bearing failures. The experimental results demonstrate the effectiveness of the proposed method in mechanical fault diagnosis under variable speed conditions, which highlights the potential applications in the field of safety monitoring for servo motors.