Model-driven condition monitoring and diagnosis in digital machining process

Online monitoring and diagnosis are quite significant to ensure the normal operations of digital machining process and the reliability/safety of machine tools. Based on the cutting dynamic model, a digital machining process model which integrates product quality information and machining condition information is proposed, and the mapping relationship between the machine tool structural dynamics and the product quality information is analyzed in detail. On the basis of the digital machining process model, the condition monitoring and fault diagnosis of the machining process is investigated from another aspect, and an implementation flowchart of the system is proposed with explanations of critical techniques. The proposed model and method are applied in the condition monitoring and diagnosis of the machining process of a 3-axis end milling machine. Accelerometers are used to obtain the vibration signals of the machine tool spindle. The second-generation wavelet method is employed for processing the vibration signals and extracting the fault features. The defect of the bearing outer ring in the spindle is detected successfully, which provides the proof for the root cause identification of machining error in the digital machining process.