基于贝叶斯网络的气井井筒完整性风险评价

发布时间：2018-03-04 08:11

本文选题：贝叶斯网络　切入点：井筒完整性　出处：《中国安全生产科学技术》2017年09期 　论文类型：期刊论文

【摘要】：井筒完整性失效是气井生产中的主要风险,为有效评价井筒完整性风险,应用贝叶斯网络的推理与学习能力,建立了基于贝叶斯网络和Noisy-OR gate模型的井筒完整性失效概率计算方法和风险评价模型。由故障树分析将井筒分为管柱、水泥环密封性、井口装置、水力屏障和其他部件5个评价单元,确定了各单元的主要风险因素,建立了井筒完整性失效的贝叶斯网络拓扑结构;由Noisy-OR gate模型和历史数据,确定了贝叶斯网络的条件概率参数;将基于贝叶斯网络的失效概率与层次分析法相结合,确定了风险评价指标和等级划分标准;建立了气井井筒完整性风险评价方法。结果表明,该方法实现了井筒完整性失效概率的定量计算、风险的定量评价和主要风险因素的反向推理,可为预防和控制井筒完整性失效提供决策依据,有助于降低井筒完整性失效风险。
[Abstract]:Wellbore integrity failure is the main risk in gas well production. In order to evaluate wellbore integrity risk effectively, the reasoning and learning ability of Bayesian network is applied. Based on Bayesian network and Noisy-OR gate model, the failure probability calculation method and risk evaluation model of wellbore integrity are established. The wellbore is divided into pipe string, cement ring seal, wellhead device and so on by fault tree analysis. Five evaluation units of hydraulic barrier and other components are used to determine the main risk factors of each unit, and a Bayesian network topology for wellbore integrity failure is established, which is based on Noisy-OR gate model and historical data. The conditional probability parameters of Bayesian network are determined, and the risk evaluation index and classification standard are determined by combining the failure probability based on Bayesian network with the analytic hierarchy process (AHP). A risk assessment method for wellbore integrity of gas wells is established. The results show that the method realizes quantitative calculation of wellbore integrity failure probability, quantitative evaluation of risk and reverse inference of main risk factors. It can provide decision basis for preventing and controlling wellbore integrity failure, and help to reduce wellbore integrity failure risk.
【作者单位】：中国石油集团钻井工程技术研究院;中国石油勘探开发研究院;
【基金】：中国石油集团公司课题(2015E-400602,2015E-4003)
【分类号】：TE38;TE88

【相似文献】