西安石油大学学报（自然科学版）2017，Vol.32Issue(3)：78-82,5.DOI:10.3969/j.issn.1673-064X.2017.03.012
基于BP神经网络的天然气集输管道结垢预测
Scaling Prediction of Natural Gas Gathering Pipeline Based on BP Neural Network
摘要
Abstract
The ion composition and concentration of the produced water from Qiongxi block of Northwest Sichuan Gas Field were determined,and based on this,the natural gas pipeline scaling prediction model based on BP neural network was established,and the layer number,the input vectors,the output vectors,the training function,the hidden layer node number,the parameter weights and thresholds of the model were determined.The accuracy of the established BP neural network model was verified by field test data.It is shown that the scaling tendency of the natural gas gathering pipeline in Qiongxi block can be accurately predicted by the BP neural network whose input parameters are the ion concentration,temperature,pressure,pH value,salinity and flow rate of oilfield produced water.关键词
天然气集输管道/管道结垢预测/结垢影响因素/BP神经网络Key words
natural gas gathering pipeline of pipeline/scaling prediction of pipeline/scaling influence factor/BP neural network分类
能源科技引用本文复制引用
袁兆祺,李长俊,杜强,贾文龙..基于BP神经网络的天然气集输管道结垢预测[J].西安石油大学学报（自然科学版）,2017,32(3):78-82,5.基金项目
国家自然科学基金项目“高含硫天然气集输管道硫沉积机理与预测方法研究”(编号:51674213) （编号:51674213）
