Quantitative Risk Assessment Methodology And Application Research For Gas Pipeline Networks

In recent years,with the accelerated urbanization in China,the scale of gas pipeline networks has been expanding and gas accidents have been frequent,making it urgent to carry out risk assessment of gas pipeline networks.This paper takes city gas pipeline networks as the research object,in response to the shortcomings of traditional gas pipeline network risk assessment methods,and addresses three key technologies in the process of risk assessment,namely,the reasonable construction of the index system,the quantitative allocation of index weights,and the scientific establishment of scoring criteria,based on the data and actual situation of gas pipeline networks in Province A,to provide new ideas and methods for quantitative risk assessment of gas pipeline networks.The main research elements are as follows.1)A gas pipeline network risk assessment index system is constructed.Based on the basic data collected from research and survey,the Delphi method is used to determine the risk level of some gas pipelines combined with the experience of frontline personnel.The gas pipeline network assessment data is used as input,and the factor analysis method is used to reduce the dimensionality,and a gas pipeline network risk assessment index system is constructed containing three primary indicators for pipeline own factors,external environment and inspection and maintenance,and subdivided into nine secondary indicators.2)The weights of the gas pipeline network risk assessment indicators are quantified.To address the shortcomings of the single weighting method,the hierarchical analysis method and the coefficient of variation method are selected as representatives of the subjective and objective weighting methods respectively,to overcome their shortcomings of requiring consistency tests and considering only the volatility of indicators.Fuzzy hierarchical analysis is used to replace the traditional hierarchical analysis method,and the improved coefficient of variation method is used to modify the original coefficient of variation method to find the weighting results,and the final indicator weights are determined by combining the weights based on the idea of game theory.The results of the factor analysis are analysed using the ordered multiclassification logistic regression method,and the risk factors are ranked in terms of importance according to the absolute value of the regression coefficients,and the possible reasons for deviations from the weighting results are analysed.3)The scoring criteria for gas pipeline network risk assessment indicators are established.To address the problem of the current scoring criteria relying on expert experience,based on the basic data of gas pipeline network,K-means clustering is used to mine the risk patterns of pipeline segments,stratified and simple random sampling is used to determine representative pipeline sample data,the upper bound of the scores of assessment indicators is determined through probability analysis,and the scoring criteria for each indicator of gas pipeline network risk assessment are determined by combining expert experience.4)Example application of gas pipeline networks in 12 cities in Province A.In order to verify the feasibility and applicability of the method,the above results are applied to 12 city gas pipeline network examples in Province A.The linear integrated assessment method is used to calculate the relative risk values of the pipeline segments and to achieve risk ranking and risk classification against the risk classification criteria.Taking city 4 as an example,the distribution of the number of different risk classes of pipe sections and the local risk map indicate that city 4 as a whole is at greater risk.Four cities with greater risk and one city with significant risk are also identified,providing a theoretical basis for city gas risk prevention and control.