Study On The Optimization Of Large-Scale Weak Rigid Parts Processing Process And The CNC Process Monitoring Methods

Aluminum alloy has high specific strength and high specific stiffness,good corrosion resistance and excellent fatigue strength,and is widely used in lightweight aerospace equipment.Aerospace parts are characterized by large size and thin wall,which are prone to deformation or uncontrollable deformation during processing.For example,the lower end frame of a typical large-scale rotary thin-walled part requires high quality and performance when forming the part.The features of thin wall thickness,low rigidity,and large size of this part result in poor processing technology,and it is difficult to ensure processing accuracy and quality.In the current production,in order to meet the quality requirements,relying on experience to reduce the cutting amount for processing,greatly reducing the processing efficiency.At the same time,there is no complete relevant research and conclusion on the influence of process parameter combination changes on processing.The industry urgently needs to conduct basic process experiments and analysis on it to improve its processing quality and efficiency.In addition,the processing cycle of such parts is long,and uncertain factors in the processing,such as abnormal operation of the machine tool,will affect the processing quality.Therefore,it is necessary to monitor and analyze the processing process to ensure the normal progress of the processing process.This subject takes the milling process optimization and process monitoring of the lower end frame of typical aerospace thin-walled structural parts as the research object,and carries out the finite element simulation analysis of the basic process problems to control the processing deformation and improve the material removal rate.Process parameter optimization,signal feature extraction and analysis in the milling process,and a process monitoring method based on the internal signal of the machine tool.The main research contents completed in this project are as follows:(1)A simulation model for milling processing of large weak rigid thin-walled parts based on ABAQUS is constructed.According to the structural characteristics of the research object,the model size,simulation model establishment method and process parameter combination in simulation analysis are determined.According to the established finite element model,the simulation of milling processing is carried out,and the law of the change of milling force and processing deformation with process parameters is clear.(2)Optimized the milling processing method and processing parameters of thinwalled parts.Based on the display dynamics simulation model,the simulation analysis of different processing methods is carried out at different processing heights of the workpiece.The deformation after processing was used as an index to evaluate the pros and cons of the processing method,and the processing method with less deformation should be clarified.Taking the deformation and material removal rate in milling processing as the objective function,taking the process parameters such as spindle speed,radial and axial depth of cut and feed per tooth as decision variables and constraints,based on non-dominated sorting multi-objective genetic algorithm The optimized combination of process parameters in the milling process was obtained.On the premise of ensuring the processing quality,the processing efficiency is improved.(3)An indirect monitoring method of milling force and machining status based on the internal signal of the machine tool is proposed.Fast Fourier transform,wavelet packet decomposition and other methods are used to preprocess the internal and external signals in the milling process,and the time-frequency domain eigenvalues of the processed signals in different frequency bands are extracted.The correlation between the milling force and the internal signals of the machine tool is determined by calculation,and the features with high correlation are used as the monitoring variables in the milling process.Finally,through neural network training,a milling force prediction model based on the internal signals of the machine tool was established to realize indirect monitoring of milling force and processing status.