Applying Neural Networks to Choose The Material of Steels

碩士 === 大葉大學 === 電機工程學系碩士在職專班 === 93 === We would like to ensure the product with more market competition, only choose the proper steel material. As usual to choose the proper steel material method are according to the past usage materials and the function of choosing steel, thus to choose the steel...

Full description

Bibliographic Details
Main Authors: JIUN-HUEI CHEN, 陳君輝
Other Authors: YUNG-NAN HU
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/79555970635236023020
Description
Summary:碩士 === 大葉大學 === 電機工程學系碩士在職專班 === 93 === We would like to ensure the product with more market competition, only choose the proper steel material. As usual to choose the proper steel material method are according to the past usage materials and the function of choosing steel, thus to choose the steel material’s character will direct to affect the product’s quality. So the alternative of material is the most important factor by the new product development by design engineer. Because the time of developing new product will affect the developing cost, so will take the time to find all variant specification and well done consideration as the company to accept their customer’s commitment. For usual, we are prefer to choose the past steel material then the property of that, then we can not match the demand of product character by customer. Thus, our research report will support to rapid and proper the right solution of the steel alternative. In general of 602 times’ testing data of the steel material, we are classify the character and analyze the value range with neural network to find the proper value of weight and bias. Additional, we will test the right value rate by 10 times’ testing data for the reference index of steel material by designer.