小波分析与Kohonen神经网络方法在埋地管道防护层缺陷现场检测中的应用  

APPLICATION OF WAVELET ANALYSIS AND KOHONEN NEURAL NETWORKS FOR FIEID DETECTION OF SPOTS ON UNDERGROUND PIPELINE COATING

作　　者：高志明[1] 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,沈阳110016

出　　处：《腐蚀科学与防护技术》2001年第z1期464-466,共3页Corrosion Science and Protection Technology

基　　金：国家重点基础研究发展规划项目 (G19990 65 0 )

摘　　要：对埋地管道防护层进行现场测试 ,以连续小波变换提取正弦电流激励响应中特定频率的信息 ,建立了Ko hoen神经网络方法评价防护层状态的适合于现场检测的智能模型 ,并对埋地模拟管道及大港油田港沧输气管线管道防护层状态进行判断 。A method has been developed for field detection of underground pipeline coating.The continuous wavelet transform was introduced to pick up the useful information from the response of small amplitude sine current with high disturbance.A diagnosis model based on Kohonen neural network was established. The method had been used in detection of simulated pipeline and actual pipeline coating of Dagang Oil Field. The result was satisfactory.

关 键 词：小波分析 KOHONEN神经网络 防护层 

分 类 号：TG174.1[金属学及工艺—金属表面处理]