论文部分内容阅读
国内大部分油气田进入高含水开发期,油气田管道结垢已经成为危害管道正常运行、影响油气田生产的重要问题。文章建立了混输管道污垢监测装置,采集了数据样本,利用BP人工神经网络理论对油田地面集输系统管道结垢情况进行预测和评价,通过数据的比较分析发现人工神经网络预测的数据和实测数据之间具有较小的误差。同时它不用建立数学模型,学习过程通过自动调节神经元之间的连接权值完成,具有较好的预测性。
Most domestic oil and gas fields have entered a period of high water cut. Pipe scaling in oil and gas fields has become an important issue that endangers the normal operation of the pipeline and affects the production of oil and gas fields. In this paper, a device for monitoring fouling in mixed pipelines was set up and data samples were collected. Based on the theory of BP artificial neural network, the scale fouling of the surface of the oil gathering system was predicted and evaluated. The data of artificial neural network There is a small error between the data. At the same time, it does not need to set up a mathematical model, and the learning process is completed by automatically adjusting the connection weights between neurons and has good predictability.