基于FAHP-ICV的燃气管网风险评估方法

Risk assessment method of gas pipeline networks based on fuzzy analytic hierarchy process and improved coefficient of variation

可靠的风险评估结果有助于提升燃气管网安全管理水平。该文提出一种基于模糊层次分析和改进变异系数(FAHP-ICV)的燃气管网量化风险评估方法。结合某省燃气管网实际情况构建了包含3个一级指标、9个二级指标的燃气管网风险评估指标体系。采用FAHP与ICV相结合的组合赋权法确定指标的综合权重,基于统计学中K-means聚类、抽样技术及概率分析方法对指标评分分值进行调整,结合专家经验确定评分标准。采用线性综合评估方法计算管段相对风险值,实现风险排序及风险分级。以某省12个城市燃气数据为基础验证了该方法的可行性与适用性。该方法提升了燃气管网量化风险评估的客观性与准确性。

[Objective]Reliable risk assessment results can help to improve the efficiency of safety management in gas pipeline networks.The Kent method is widely used as an accepted risk assessment method.However,relevant literature suggests that the Kent method is inadequate in the determination of weights,scoring items,and scores,and the determination of index weights and scoring criteria requires expert experience,which is highly subjective.Therefore,the traditional Kent method needs to be improved to comply with the risk assessment of different gas pipeline networks.To improve the objectivity and accuracy of risk assessment of gas pipeline networks,a quantitative risk assessment method based on the fuzzy analytic hierarchy process-improved coefficient of variation(FAHP-ICV)for gas pipeline networks is proposed.[Methods]In this work,based on data from the gas pipeline networks and their surroundings,the traditional risk assessment method for gas pipeline networks was improved in terms of index system and weighting and scoring criteria determination using statistical methods.First,a risk assessment index system comprising three primary indicators and nine secondary indicators was constructed while considering the actual operation of the gas pipeline networks in a province.Second,the subjective weighting method represented by the hierarchical analysis method and the objective weighting method represented by the coefficient of variation method were improved.The fuzzy hierarchical analysis method was used instead of the traditional one,and the improved coefficient of variation method was used to modify the weighting results of the original coefficient of variation method.The two methods were combined to determine the comprehensive weights of the evaluation indicators based on expert experience and the inherent rules between the indicator data.Next,based on the K-means clustering and sampling techniques in statistics,the sample data for the pipe section were determined and pre-processed through probability analysis to determine the upper bound of the scores of evaluation indicators.Then,the final scoring criteria were determined by integrating expert reports.Finally,a linear integrated assessment method was used to calculate the relative risk values of the pipe sections to achieve risk ranking and classification.[Results]To analyze the distribution of risk classes across gas pipe sections in the cities,the relative risk values of gas pipe sections in 12 cities were calculated and compared with the risk class classification criteria.For example,in city four,a comparison between the distribution of risk classes across gas pipe sections and the local map showed that the overall risk of the city was relatively high.The average risk values for gas pipe sections and different level indicators were compared between 12 cities;four cities were found to have great risk,and one city was found to have significant risk.Furthermore,cross-analysis was carried out on the city where the inspection and maintenance indicators suggested not fulfilling the requirements of gas pipe inspection regulations.[Conclusions]The feasibility and applicability of the method were verified through examples,providing new ideas and methods for the quantitative risk assessment of gas pipeline networks.