Evaluation Methods Research On Green Degree Of Grinding Process Based On Clustering And ANFIS

Green manufacturing is the main way to solve the problem of manufacturing environment pollution and realize sustainable development of manufacturing industry in our country.In recent years,specialists and scholars from different countries have done extensive research on the green manufacturing evalution.However,there is little research on evaluating green mechanical manufacturing process process.Grinding is one of the most widely used machining technologies in the machining field.It is a main method to obtain the final size,the shape precision and the surface roughness of the parts.But at the same time,the grinding process has greatest influence on natural resources and environment.Implementation process of the greening grinding has attracted more and more attention.Therefore,for establishing green degree evaluation system of grinding process and improving the green degree of grinding process,it is important theoretical and practical significance to optimize the existing grinding process and realize green grinding.Based on the research about the evaluation methods and the evaluation tools of the green manufacturing,the green features and influencing factors of grinding process are analysed.The green degree evaluation index system of grinding process is established.The construction of sample training set with new kernel possibilistic fuzzy C-means clustering(NKPFCM)and the green degree evaluation method based on adaptive neuro-fuzzy inference system(ANFIS)are discussed emphatically.The development process of evaluation tools is presented.The contents of the paper mainly include the following aspects:(1)Based on the in-depth study on the characteristics and the decision goal of the green process planning,green characteristics of grinding process are studied by analyzing resources,environment,energy and social.The evaluation goal,evaluation criteria and scheme design are put forward,with the help of the existing evaluation method.(2)The green degree evaluation index system has three layers,including the properties layer,general character layer and index layer.The evaluation indexes and influencing factors are thoroughly analyzed from the environment attribute,resources attribute,energy attribute,economic attribute and human-machine attribute.In view of the quantization of evaluation index,evaluation index can be divided into qualitative indexes and quantitative indexes.At the same time,semi-quantitative index quantitative method is proposed,and a quantitative method of evaluation index is presented.(3)Through deep analysis of clustering algorithm,this paper proposes a new possibility fuzzy clustering algorithm based on nuclear(NKPFCM).By MATLAB,Iris flower data set and wine data set are used to test.The simulation results show that this algorithm has good robustness and accuracy.This algorithm is used to carry out the cluster analysis of sample data in grinding process,realizing adaptive optimal classification of the data sample and constructing high quality evaluation sample training set.(4)Based on the research on ANFIS,the green degree evaluation system of grinding process is established by fuzzy C-means(FCM)and least mean square(LMS)hybrid algorithm.The evaluation system has the characteristics of autonomous learning,fuzzy information expression and reasoning.It can scientifically reflect the nonlinear mapping relationship between the evaluation index input and the evaluation results and effectively eliminate the influence of fuzziness and randomness of the evaluation index on the evaluation results.The results show that the evaluation model can make objective and effective evaluation of process scheme.(5)For improving the green degree of grinding process,the software FLUENT is used to study the method of increasing grinding fluid utilization efficiency.Bsed on the analysis on the mechanism of the grinding process,the useful parameters of grinding fluid jet are obtained.Finally,combined with the related research results,the optimization methods of the grinding process green degree are summarized.That can provide guidance for technology personnel to optimize the process scheme.(6)According to the green degree evaluation method principle of grinding process based on clustering and ANFIS,in the Windoes environment,using C++ language and SQL Server2005 database management system,through the modular design method,the modular programming is adopted to develop the grinding process green degree evaluation system software.And the key technology in the development process,ADO data interface technology is presented.Grinding process green degree evaluation system software has the characteristics of friendly interface,easy operation,stable and quick data access,and science evaluation with high intelligent degree,etc.It can be used as an effective evaluation tool for grinding process design,optimization and decision-making in the enterprise.