Intelligent Monitoring Method for Gear Grinding Machine Spindle Based on Multi-source Information Fusion

As a high-precision gear machining process, tooth surface grinding can be applied to gears that require high precision and smoothness. However, there are widespread anomalies in grinding processes that affect tooth surface accuracy and load-bearing capacity due to unreasonable process parameters, grinding wheel threshing, and other reasons, such as tooth surface burns, off center grinding, and vibration lines. In response to the problems of fault diagnosis relying on manual experience, difficulty in effective data collection, and low data utilization in the machining process of gear grinding machines, the key technology of intelligent detection of gear grinding machines based on multi-source information fusion is studied. Firstly, by deploying various types of sensors on the gear grinding machine to collect its multi-source data; Secondly, based on the characteristics of its spindle vibration signal, local discriminant wavelet packets are used to achieve spindle vibration state recognition and health assessment; Finally, based on the obtained multi-source signals such as vibration, acoustic emission, current, and power, fusion signal feature extraction is carried out. The improved D-S evidence theory based on entropy weight method is used to complete spindle state recognition, achieving efficient and high-quality machining of the gear grinding machine.