摘要
传统油气集输过程安全预测预警技术,由于未考虑过程信息与过程机理的结合而导致其具有一定的局限性。本文通过斯皮尔曼相关系数分析,基于动态模拟提取油气集输过程的关键机理特征变量,建立半监督的油气集输过程安全状态深度学习模型,实现对过程关键安全参数的智能预测。该方法可以提升油气集输过程安全管理水平,有效化解安全风险,减少企业损失。
The traditional safety early prediction and warning technology of oil and gas gathering and transportation process has some limitations because it does not consider the combination of process information and process mechanism.In this paper,Spearman correlation coefficient analysis is used to extract key mechanism characteristic variables of oil and gas gathering and transportation process based on dynamic simulation.A semi-supervised deep learning model for safety state of oil and gas gathering and transportation process is established to realize the intelligent prediction of key safety parameters of the process.This method can improve the safety management level of oil and gas gathering and transportation process,effectively resolving safety risks and reducing the losses of enterprises.
作者
马珍福
陈鲁
瞿健
田文德
Ma Zhenfu;Chen Lu;Qu Jian;Tian Wende(Hekou Oil Production Plant,Shengli Oilfield Branch Company of SINOPEC,Dongying 257200,China;College of Chemical Engineering,Qingdao University of Science&Technology,Qingdao 266042,China)
出处
《山东化工》
CAS
2021年第17期166-168,172,共4页
Shandong Chemical Industry
基金
国家自然科学基金项目(21576143)。
关键词
油气集输
动态模拟
深度学习
安全预测
oil and gas gathering and transportation
dynamic simulation
deep learning
safety prediction
