Font Size: a A A

Study Of Grinding Feature Identification, Process Optimization And Monitoring Methodology On Surface Quality

Posted on:2021-04-03Degree:DoctorType:Dissertation
Country:ChinaCandidate:W C GuoFull Text:PDF
GTID:1361330623478688Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
The rapid development of the aeronautics and astronautics high-end equipment industry has increasingly demanded the processing quality of parts.Grinding,as an important precision and ultra-precision machining method,plays a key role in the manufacture of high-end parts.Its processing quality and stability determine the service performance and reliability of products.Grinding is a complex multi-variable coupling process.During the manufacturing,it will be affected by various factors such as grinding parameters,grinding wheel characteristics,workpiece material properties and dynamic stability of the process system,which resulting in great uncertainty in processing quality.The accurate and effective monitoring of the grinding process and its processing optimization has been a bottleneck in improving grinding quality and efficiency.This dissertation aims at the processing optimization and process quality monitoring of high performance inertial navigation key component-integral double balance ring flexible joint(constant elastic alloy steel 3J33)and carry out some research,such as basic theory of grinding monitoring,extraction and identification of signal features,multi-objective process optimization,process and surface quality monitoring and so on.The main work and innovations of this dissertation are as follows:(1)An experimental platform for grinding process monitoring based on cyberphysical system and status data-driven(CPS-SDD)is constructed.Evaluation indicators and test methods have been defined according to the grinding surface quality objectives,and the physical signal generation mechanisms such as force,temperature,vibration,and acoustic emission during the grinding process have been mastered.The monitoring techniques are studied to prepare intelligent monitoring of surface quality control.A mapping relationship between grinding process information-process,processing status and processing quality is established.(2)A feature identification method based on physical signal in griding process and relevant surface quality(PSRSQ-FI)method is proposed.The original signal is preprocessed using wavelet packet decomposition and ensemble empirical mode decomposition methods.The mean,standard deviation,and kurtosis in different frequency bands are extracted,and the signal characteristics based on the RMS,skewness,and crest factor of power spectral density estimation are extracted as well.According to different grinding surface quality evaluation indicators,the correlations between time-frequency characteristics and surface quality indicators are determined.These correlations are sorted and features with high correlation are used to represent the quality of grinding.(3)The influence rules of the grinding process parameters and the force and heat load on the surface quality of grinding are revealed.With the surface roughness,surface residual stress,and material removal rate as the optimization targets,and the process parameters such as grinding wheel speed,workpiece speed,and grinding depth as constraints,the optimized process parameters for grinding were obtained based on the non-dominated multi-objective genetic algorithm.The optimized parameters guarantee the quality as well as the efficiency of grinding.Such a new approach is provided for surface quality optimization and a process basis for intelligent grinding monitoring(4)A high relevant feature fusion based montoring of surface quality(HRFF-MSQ)method is proposed.Based on the monitoring objectives of grinding wheel wear,surface roughness,residual stress and grinding burn,a high correlation of grinding characteristics for quality monitoring is achieved,and corresponding monitoring methods and control decisions are given.Surface quality control for high-performance integral double balance ring flexible joint is complished and supported by the CPS-SDD grinding process monitoring platform.The above research results have been practically applied in the manufacture of integral double balance ring flexible joint and have played a role in completing major national scientific and technological projects and improving the international competitiveness of China's aeronautics and astronautics high-end equipment.
Keywords/Search Tags:grinding, surface quality, feature identification, machine learning, multiobjective process optimization, process monitoring
PDF Full Text Request
Related items