基于MATLAB人工神经网络的土壤腐蚀性评价模型  

Evaluation Model of Soil Corrosiveness Based on MATLAB Artificial Neural Network

作　　者：朱庆杰[1] 张建龙 陈艳华 雒振林 李雪[1] 刘亚婷 ZHU Qingjie;ZHANG Jianlong;CHEN Yanhua;LUO Zhenlin;LI Xue;LIU Yating(Key Laboratory of Oil and Gas Storage and Transportation Technology of Jiangsu Province,Changzhou University;College of Civil and Architectural Engineering,North China University of Science and Technology;Hebei Construction Engineering Quality Inspection Center Co.,Ltd.)

机构地区：[1]常州大学江苏省油气储运技术重点实验室 [2]华北理工大学建筑工程学院 [3]河北省建筑工程质量检测中心有限公司

出　　处：《油气田地面工程》2021年第12期10-15,共6页Oil-Gas Field Surface Engineering

基　　金：科技部国际合作司中国波兰政府间科技合作项目(2012-35-05);国家自然科学基金面上项目(51378172);河北省自然科学基金(E2020209072)。

摘　　要：土壤腐蚀是导致埋地燃气管道泄漏的主要原因。埋地燃气管道在土壤中的腐蚀破坏受多种因素的共同作用,根据土壤腐蚀等级指标,确定了土壤含水量、电阻率、氧化还原电位等5个风险评价因子。为确定风险评价因子与土壤腐蚀概率之间的非线性相关关系,需建立多元非线性模型。运用MATLAB中的人工神经网络工具箱,通过人工神经网络的计算结果以及神经网络的训练误差,对网络结构进行优化,最终建立了常州市埋地燃气管道土壤腐蚀性的人工神经网络评价模型。通过分析计算结果,得到了土壤腐蚀分布与特征,为常州埋地燃气管道敷设避开土壤腐蚀性较大的区域以及管道的安全防护提供了建议。Soil corrosion is the main cause of leakage of buried gas pipelines.The corrosion damage of buried gas pipelines in the soil is affected by many factors.According to the index of soil corrosion grade,five risk assessment factors such as soil water content,electrical resistivity,and redox potential are determined.In order to determine the nonlinear correlation relationship between risk assessment factors and soil corrosion probability,a multivariate nonlinear model should be established.Therefore,the artificial neural network toolbox in MATLAB is used to optimize the network structure through the calculation results of the artificial neural network and the training error of the neural network.Finally,the artificial neural network evaluation model of soil corrosiveness of buried gas pipeline in Changzhou city is established.Through the analysis and calculation results,the distribution and characteristics of soil corrosion in Changzhou are obtained,which provides suggestions for the laying of buried gas pipelines in Changzhou to avoid the areas with high soil corrosion,and the safety protection of pipelines.

关 键 词：埋地然气管道 土壤腐蚀性 评价模型 人工神经网络 评价因子 训练误差 

分 类 号：TU996.8[建筑科学—供热、供燃气、通风及空调工程]