基于神经网络的抽油机井下故障诊断专家系统

本课题是针对油田边远井、孤立井创新的提出了手持自动化功图诊断系统。本文以抽油机作为研究对象,从油井的示功图入手,利用自组织竞争网络对抽油井示功图进行智能识别分类,实现油井故障的自动诊断。本文所研究的基于神经网络的故障诊断系统在手持Android终端得到了实现,并且江苏油田得到了广泛应用。

Oil pumping motor-pumped wells based on neural network fault diagnosis expert system

This article proposal hand-held automated fault dagnosis systems innovatively is mainly for the remote oil field wells. This paper presents a fault diagnosis of self-organizing competitive neural network. Pumping as the object of study, indicator diagram is used to acquire to the fault characteristic, using the self-organizing competitive network to achieve intelligent classification and diagnose fault diagnosis automatically. Indicator diagram evalution researched in this paper has achieved hand-held automated fault dagnosis systems in and has been widely used in Jiangsu Oilfield.