Detection And Monitoring Recognition And Learning Optimization Method Of Time-Varying Factors In Milling Condition

Process parameters are the basic elements of milling condition,selecting the appropriate process parameters is important to control the process,improve the efficiency and quality of machining.The cutting process parameters are currently determined based on the experience,and the fixed parameters are used in the whole process.These parameters can only apply to a specific condition,once the machining condition changes they are invalid.Using scientific methods,optimizing and controlling the process parameters for real machining condition,are the urgent problems in the efficient and quality machining of aerospace complex part.In this paper,the milling force modeling method of worn tool,the working condition modeling theory of real process,the online recognition method of time-varying process condition,and the optimizing and controlling method of process parameters for real process condition,are studied in-depth,for the roughing CNC milling process.The main contents and innovative achievements are as follows:1.In the milling force modeling of worn tool,for flat end mill,the engaged part of tool and material was divided infinitesimal along tool axis.The infinitesimal cutting force was divided into the shearing forces on rake face and friction effect forces on flank face.The shearing forces were expressed by the product of cutting load and shearing force coefficients.The friction effect forces were expressed by the integrations of normal stress and shear stress on the tool flank face.The contact area of worn flank and workpiece is divided into plastic flow region and elastic contact region,the friction effect forces could be obtained by studying the distribution of stress.According to the force model,the shearing force coefficients and various friction and press forces on unit flank length are calibrated,and then the distributions of stress were obtained.Finally,this force model was verified through experiment.This provided the basic model for process parameters optimization.2.In the process condition modeling,for complex surface part milling process,the factors affecting machining were analysed.According to the process needs,the part characteristics,and milling process information,the factors were divided into four sub-process conditions,process system,workpiece information,tool state,and dynamic parameters.According to the actual needs,process condition sub-vectors were established.Then,based on the force model,the influence relationship of time-varying process condition factors on milling force was established.Based on the parameterized tool life curve,the influence relationship of time-varying process condition factors on tool wear rate was established.Finally,based on the process condition model,the process knowledge database model was provided.The influences of process condition factors on machining process were stored into the process knowledge database as knowledges,and the method of obtaining process knowledge was proposed,to provide the reliable process knowledge for recognizing process condition and optimizing process parameters.3.In the online recognition of time-varying process condition,for real milling process,the detection and monitoring machining theory and achieving method were proposed.Based on existing NC system,the acquisition scheme of spindle speed and feedrate was designed,and the function of this scheme was realized through independent development.Based on the influence relationship of time-varying process condition factors on milling force and tool wear rate,the recognition methods of milling load and tool wear state were proposed.Finally,the recognition methods were verified through experiments.This provided the basic support for optimizing and controlling process parameters.4.In the process parameter optimization and controlling,based on the above research,the process marameters multi-criteria optimization method of roughing milling based on learning cycle was studied.Firstly,the characteristics of online and offline optimization were analysed,the iterative optimization strategy of milling parameters based on combining the solving online and the learning offline was proposed.The optimization objectives and constraints guidelines were analysed,the multiple criteria optimization model was established.Then,the online solving method of optimization model was studied,and the feedback controlling scheme of feedrate was proposed.Finally,the offline updating method of process parameters was studied.In mass production,the learning cycle was defined,the process parameters were accumulated and optimized through learning cycle,to provide better NC program for the follow machining.5.In the building of experiment platform,based on the above research,the verification experiment platform was designed.The software and hardware system of experiment platform were developed.Finally,the proposed theory and methods were verified systematically on this experiment platform.Eventually,the feasibility and effectiveness of process identification,parameters optimization,feedrate controlling were verified.