Methods And System Research Of Milling Tool Condition Monitoring

With the development of modern manufacturing automation technology processing,CNC machine tools are widely used in product processing and production processes.As one of the most vulnerable parts of CNC machine tools,the wear degree of tool directly affects product quality.Replace the severe wear tool in time not only helps to improve product quality and production efficiency,but also helps to improve tool utilization and reduce production costs.Therefore,it is especially important to establish a reliable online tool condition monitoring system.Tool condition monitoring system mainly consists of three parts: signal acquisition,signal processing and online monitoring.In this paper,the tool condition monitoring methods and system are studied based on numerical control milling tool.Aiming at the problem of signal classification under the background of strong noise and periodic pulse interference in milling process,a classification method of tool wear condition based on intrinsic scale decomposition and kernel extreme learning machine is proposed.The intrinsic scale decomposition technique is introduced to decompose original signals,then extracts the proper rotation component which has high similarity to the original vibration signal to feature extraction,and then put into kernel extreme learning machine model to classification.The results of tool condition monitoring experiments in milling process show that the method can classify the tool wear condition reliably.Aiming at the instability problem of tool wear estimation by extreme learning machine method,a genetic algorithm and particle swarm optimization algorithm is proposed to enhance extreme learning machine.Genetic algorithm and particle swarm optimization algorithm are used to optimize the connection weight between extreme learning machine input layer and hidden layer and the threshold of hidden layer neurons.The results of tool condition monitoring experiments in milling process show that the method can estimate the tool wear value effectively.In view of the difficulty in measuring tool wear during machining,this paper developed an online system based on embedded system(TMS320C6748F)platform for tool wear monitoring.The above two methods are applied to the online tool wear monitoring system.Experimental results show that the system can recognize the current tool condition with the vibration signals during machining process,which provides an effective basis for operator to change the severe wear tool in time.