Research On Tool Wear Monitoring System Of CNC Machine Tools And Information Exchange Technology

In machining processes, cutting tool gradually wears out, loses its sharpness and becomes blunt due to thermal fracturing, attrition, abrasion, plastic deformation, diffusion, chemical wear, and grain-pullout. Tool wear or tool breakage affects the surface finish and dimensional accuracy of products, reducing the machining quality, working efficiency and stability of machine tool as well. Tool wear monitoring system CNC machine plays a decisive role, its major aim is a desire to decide the suitable machining parameters and to replace the broken tool at the right time, reduce operational costs and to increase performance.In this paper, a tool wear monitoring experimental system for CNC machine milling process is proposed, which selects vibration signal and cutting force as the original signal. The systematic theoretical analysis and experimental implementation procedures of denoising, feature extraction and tool state recognition is presented in detail as well.The filtering method based on mathematical morphology can remove background signal of the original cutting force signals and vibration signals, as much as possible to retain the original geometry of the signals, which are considered as nonstationary signals. By the signal time-domain analysis, frequency domain analysis and wavelet packet analysis of cutting force and vibration signals, the feature vectors which have good mapping relation with the cutter wear condition are obtained. Then, they are input into multi-class classifier based on support vector machine algorithm for training and testing. Experiments show that performance of recognition is good, to achieve tool wear estimation effectively. In the case of small samples, the tool monitoring system has a good learning performance and generalization ability.Taking the functions fusion of tool wear condition monitoring and CNC machine condition monitoring, data exchange theoretical basis between the monitoring units and numerical control systems into consideration, data exchange standards for monitoring units and numerical control systems, integrated architecture of monitoring and diagnosis system are presented. The implementation mode and specific functions also are detailed described. Modular data standards can solve the problems in the data transmission between different devices. The machine tool continuous dynamics status information, load parameters and processing parameters of manufacturing procedure are integrated as the raw data information to achieve the comprehensive status monitoring information.