Prediction of the surface characteristics of shot-peened stainless steels by statistical design of experiments

In order to develop the quantitative model to predict the surface characteristics of austenitic stainless steels from shot peening conditions, the response surface model which represents the quantitative relationship between the shot peening conditions and the surface characteristics of austenitic s...

Full description

Bibliographic Details
Main Authors: Masatoshi KURODA, Koichi AKITA, Yuji KOBAYASHI, Toshiya TSUJI
Format: Article
Language:Japanese
Published: The Japan Society of Mechanical Engineers 2018-08-01
Series:Nihon Kikai Gakkai ronbunshu
Subjects:
Online Access:https://www.jstage.jst.go.jp/article/transjsme/84/865/84_18-00150/_pdf/-char/en
Description
Summary:In order to develop the quantitative model to predict the surface characteristics of austenitic stainless steels from shot peening conditions, the response surface model which represents the quantitative relationship between the shot peening conditions and the surface characteristics of austenitic stainless steels has been constructed by using statistical design of experiments, and the validity of the model has also been discussed. As a result of Fisher's F tests in analysis of variance (ANOVA), the response surface models representing surface roughness parameter RSm, surface hardness (Hv) and surface residual stress (σ), which were constructed in the present study, were statistically significant. It was also found that shot diameter and air pressure of the shot peening conditions were statistically significant factors for the response surface models of RSm and σ, while the shot diameter was the statistically significant for Hv. The predicted values of the surface characteristics estimated from the response surface models of RSm, Hv and σ agreed well with the measured values. It was concluded that the surface characteristics of RSm, Hv and σ for austenitic stainless steels could be predictable from shot peening conditions by using the response surface models constructed in the present study.
ISSN:2187-9761