基于BP神经网络的输油管道内腐蚀速率预测模型  

BP Neural Network-based Prediction Model for Internal Corrosion Rate of Oil Pipelines

作　　者：胡松青[1] 石鑫[1] 胡建春[1] 任振甲[1] 郭爱玲[1] 高元军[2] 

机构地区：[1]中国石油大学(华东) [2]新疆吐哈油田建设有限责任公司

出　　处：《油气储运》2010年第6期448-450,共3页Oil & Gas Storage and Transportation

摘　　要：采用BP神经网络技术,以原油硫含量、酸值、温度、压力和流速作为输入参数,以管道内腐蚀速率作为输出参数,建立了输油管道的内腐蚀速率预测模型。预测了各因素对管道内腐蚀规律的影响,结果表明:硫含量和酸值是影响输油管道内腐蚀的主要因素。预测结果和模拟实验数据吻合较好,表明BP神经网络模型能较准确地预测管道内腐蚀速率。Based on BP neural network technique,taking crude oil's sulphur content,acid value,temperature,pressure and flow rate as input parameters and internal corrosion rate as output parameter,prediction model of internal corrosion rate for oil pipeline is built.Influence of different factors on internal corrosion rules of pipeline is predicted.Prediction results fit simulated test data perfectly,which indicates that BP neural network model can accurately predict internal corrosion rate of oil pipelines.

关 键 词：神经网络 输油管道 内腐蚀 含硫原油 腐蚀速率 预测模型 

分 类 号：TE973[石油与天然气工程—石油机械设备]