小波分析和神经网络研究纯锌大气腐蚀  

PREDICTIVE MODEL BASED ON WAVELET TRANSFORMATION AND ARTIFICIAL NEURAL NETWORK FOR ZINC ATMOSPHERIC CORROSION

作　　者：王守琰[1] 宋诗哲[1,2] 

机构地区：[1]天津大学材料学院,天津300072 [2]金属腐蚀与防护国家重点实验室

出　　处：《中国腐蚀与防护学报》2005年第5期257-261,共5页Journal of Chinese Society For Corrosion and Protection

基　　金：国家自然科学基金重大项目(50499335);国家重点基础研究发展规划项目(G19990650)

摘　　要：将小波分析和神经网络技术用于纯锌大气腐蚀研究,用图像扫描方法获取纯锌大气腐蚀形貌图像.运用小波变换对图像进行分解并提取子图像的能量值作为特征信息.用典型相关技术分析了特征值和腐蚀深度之间的典型相关性,将提取的图像特征值与相关系数的乘积作为神经网络的输入,建立了加权能量值和腐蚀失重数据间的神经网络模型,运用该模型可以对纯锌大气腐蚀失重做出预测并具有较高精度.A computation approach using wavelet transformation and artificial neural network （ANN） to analyze and quantify the atmospheric corrosion of zinc by corrosion morphology image is described. A scanner is used to obtain the corrosion morphology images of zinc samples. Multi - resolution wavelet analysis is performed on the selected image to obtain energy values as feature vectors. The canonical correlation analysis is used to gain the correlative coefficient between the features and the corrosion loss. The weighted features were analyzed to obtain the material loss due to corrosion by using an ANN model. The results indicated that the computational methods developed for corrosion analysis could provide reasonable results for estimating material loss due to corrosion mor- phological image.

关 键 词：神经网络 小波分析 大气腐蚀 锌 

分 类 号：TG171[金属学及工艺—金属表面处理]