パターン認識法による圧延加工表面粗さの特徴抽出

概要

論文の詳細を見る
The topographical feature of behavior that surface asperities are deformed under rolling process has been clarified by the method of pattern recognition. First of all, it is confirmed that the topographical feature of surface roughness can be extracted with six models of geometrical shape and two models of slope setting on the present study. Next, the relationships between the rolling conditions and the number of pattern matchings are investigated. It is found that the results are summarized as follows. The number of pattern matchings on triangle model and trapezoid model are increasing along with the increasing number of rolling passes. In case the back tension is a high level, the number of pattern matchings on the trapezoid model are increasing along with the increasing front tension in the right-angled direction to the rolling direction. Moreover, in consideration of accidental factors such as friction and wear, it can be assumed that the random surface roughness curves are deformed in accordance with Markov processes. With further advancing the investigation, it is clarified that the state probabilities of slope model, geometrical shape model of peak and valley are already in steady-state and about 4 : 3 : 3 when the number of rolling passes is three.
公益社団法人精密工学会の論文
1989-02-05