期刊論文

學年 108
學期 2
出版(發表)日期 2020-06-01
作品名稱 Wear Value Prediction of CNC Turning Tools based on v-GSVR with A new Hybrid Evolutionary Algorithm
作品名稱(其他語言)
著者 Min-Liang Huang; Jr-syu Yang; Jheng-yu Wu; Shih-Hsing Chang
單位
出版者
著錄名稱、卷期、頁數 淡江理工學刊 23(2),頁369-378
摘要 The dimensional accuracy of the workpiece will exceed the tolerance, therefore, to predict how many workpieces have been cut, the turning tool must be replaced is the important issue in machining field. To deal well with the normally distributed random error existed in the wear value prediction of CNC turning tools, this paper introduces the ν-Support Vector Regression (ν-GSVR) model with the Gaussian loss function to the prediction field of short-term wear value. A new hybrid evolutionary algorithm (namely CCGA) is established to search the appropriate parameters of the ν-GSVR, coupling the Chaos Map, Cloud model and Genetic Algorithm. Consequently, a new forecasting approach for the short-term wear value prediction of CNC turning tools, combining ν-GSVR model and CCGA algorithm, is proposed. The forecasting process considers the wear value prediction of CNC turning tools during the first few time intervals, the turning tool wear value for the spindle revolution, cutting depth and feed rate. It is used to verify the forecasting performance of the proposed model. The experiment indicates that the model yield more accurate results than the compared models in forecasting the short-term wear value on the turning tools. In this way, we can figure out how many turning tools to prepare for similar workpieces, which can reduce the stock of turning tools, and reduce the labor costs on quality inspection of workpieces during this period.
關鍵字 Wear Value Prediction;CNC Turning Tools;Support Vector Machine;Support Vector Regression
語言 en_US
ISSN
期刊性質 國內
收錄於 EI
產學合作
通訊作者 M. L. Huang
審稿制度
國別 TWN
公開徵稿
出版型式 ,電子版,紙本
相關連結

機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/119773 )