An improved B-spline-based energy operator method for fault detection of motor bearings

Motor bearings play a crucial role in the smooth operation of motors. However, they are often subjected to various external factors such as load, transmission, and impact during their operation. These factors can lead to bearing failure, which ultimately results in mechanical breakdown of the entire motor. To address this issue, an improved B-spline-based energy operator technique is proposed in this study for fault detection of motor bearings. This technique offers several advancements over traditional methods. Firstly, it utilizes an improved grey wolf optimizer (I-GWO) to search for the optimal solution of the B-spline order that matches the input signal. This eliminates the need for blind selection of parameters and ensures accurate results. Secondly, instead of relying on the Hilbert transform (HT) for estimating instantaneous amplitude and frequency (IF and IA), a novel IF and IA tool is employed. By doing so, it establishes precise nodes for the B-spline function, significantly enhancing its precision. The improvements made in this study have demonstrated superior performance in detecting faults compared to existing techniques. Through simulations and case studies conducted during our research, we have observed that the improved demodulation technique outperforms other methods in terms of accuracy and effectiveness.