Font Size: a A A

Research On Multi-index Fusion Evaluation Of The Abrasive Media

Posted on:2021-09-02Degree:MasterType:Thesis
Country:ChinaCandidate:X YangFull Text:PDF
GTID:2481306113451544Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
The surface quality of the parts has an important influence on the effect and service life of the parts and the assembled machine.The barrel finishing technology is designed to improve the surface quality of the parts.The technology has the characteristics of good processing effect,strong applicability of parts,high cost performance,and environmental friendliness.The grinding tools used in barrel finishing processing are collectively referred to as the abrasive media.The abrasive media are one of the main factors that affect the effects and efficiency of barrel polishing processing.The reasonable evaluation of the abrasive media is of great significance to objectively understand the performance of the abrasive media and to guide the selection of the abrasive media in the barrel finishing processing.Because of the complexity of the composition,preparation and use process of the abrasive media,the evaluation indexes of the abrasive media are complicated and numerous.For users and researchers,there is no unified evaluation indices of the abrasive media,and there are few studies on multi-index fusion evaluation of the abrasive media.In response to the above problems,this paper proposes "Research on multiindex fusion evaluation model of the abrasive media".By constructing the evaluation indices of the abrasive media,the multi-index fusion evaluation of the abrasive media is carried out.At the same time,in order to provide users with a friendly,intuitive and easy-to-operate human-machine interface,the abrasive media evaluation platform is designed in C# language.The main research contents of this paper are:(1)Through in-depth analysis of actual cases of barrel finishing processing,the abrasive media preparation processing,existing research literature and opinions of consulting experts,preliminary selection of the evaluation indexes of the abrasive media was carried out,and the data collection methods of each evaluation index were specified description.(2)In order to remove the redundant evaluation indexes in the primary evaluation indexes of the abrasive media,and make each evaluation index independent and significant,adopt the analytic hierarchy process and the principal component analysis method to build the evaluation indices of the abrasive media;In order to provide an intelligent evaluation method for the production quality and processing ability of the abrasive media manufacturers,according to the constructed evaluation indices,the multi-index fusion evaluation of the historical abrasive media was carried out based on the support vector data description.Simulation results show that this method can reasonably evaluate the abrasive media,It also verifies the rationality of the constructed evaluation indices.(3)For users who have high requirements for the abrasive media,they are no longer satisfied with "qualified products" when purchasing.On the basis of evaluating the qualified products of the abrasive media according to the support vector data description,based on the constructed evaluation indices,a multiclassification and multi-index fusion evaluation model of the abrasive media based on whitening weight function weighted clustering is provided for manufacturers.The simulation results show that this method can reasonably and flexibly perform multi-class multi-index fusion evaluation on the abrasive media.(4)On the basis that the research group already has a database for barrel finishing processing,C # language is used to design the evaluation platform for the abrasive media,which can process the input information of the abrasive media to obtain the evaluation result and store the results and analyze the results for the user’s convenience.
Keywords/Search Tags:The abrasive media, Analytic hierarchy process, Principal component analysis, Support vector data description, Whitening weight function weighted clustering, Multi-index fusion evaluation
PDF Full Text Request
Related items