场站输油管道风险智能评估系统研究

采用基于风险的检验分析方法和灰色关联分析方法,分析各参数对风险影响的关联程度和关联度值,对其进行排序并选取关联度值较大的参数作为关键参数,将关键参数作为风险智能评估系统的输入参数,选用支持向量机的机器学习方法进行场站输油管道风险智能评估系统的开发。场站输油管道风险智能评估系统开发的基础数据为白沙湾输油站输油管道的实测数据,智能评估系统计算值与原始实测值吻合良好,平均精度达到99.5%,验证了场站输油管道风险智能评估系统的正确性,有效提高了场站输油管道风险评估的智能化水平,这种开发思路和实践可作为其它场站输油管道风险智能评估系统研发时的参考。

Research on Risk Intelligent Evaluation System of Oil Transportation Pipeline in Field and Station

The risk based inspection analysis and grey relational analysis method were proposed.The relevancy and value of the typical parameters with the risk were analyzed and the parameters were ranked for selecting the key ones with greater relevancy value,which was taken as the input of the intelligent risk assessment system.As the final step,the support vector machine learning method was employed in developing the intelligent risk assessment system for the field and station pipelines.The practical testing data from Baishawan oil transmission station was set as the basic data source in the process of developing intelligent risk assessment system.The research shows that output of the intelligent system is identical with the original testing data with the average accuracy up to 99.5%,which verifies the correctness of the developed intelligent risk assessment system.The intelligent technology can be effectively improved with the practical application of the risk assessment,which provides reference for intelligent risk assessment for other similar field and stations.