A Fault Diagnosis Method for Drilling Pump Fluid Ends Based on Time–Frequency Transforms

Drilling pumps are crucial for oil and gas operations. Timely diagnosis troubleshooting of fluid end faults is to ensure the safe stable operation drilling prevent further deterioration faults. Hence, from a data-driven perspective, this study proposes fault method based on generalized S transform (GST) convolutional neural networks (CNN), using vibration signal end. To address issue noise pollution in resulting unclear feature information difficult extraction, transformed into time–frequency diagram GST, which more accurately characterizes characteristics signal. An AlexNet model, improved by introducing batch normalization optimizing number neurons fully connected layer, used analyze recognition performance model normal, minor damage, severe damage states pump. Finally, results compared other methods, with showing that proposed has highest accuracy. With an average rate 99.21% nine types end, provides way diagnose failures, thus supporting efficient pumps.