Research On Assessment Method Of Supply Reliability Of Natural Gas Pipeline Networks

Natural gas pipeline networks,the critical infrastructures connecting sources and demands,are the best choice to transport natural gas on land.Several unexpected natural gas supply interruptions have occurred in the last decade with severe consequences on society stability and economy around the world.Insufficient knowledges of the uncertainty and the complexity of gas pipeline network systems is one of the main reasons of this problem.In general,the research of supply reliability of natural gas network is still in the initial stage.This work aims to analyze several problems of the supply reliability of natural gas pipeline networks from the methodology perspective,and to develop a method to help to evaluate the supply reliability.A systematic framework to analyze the supply reliability is developed from three perspectives which include uncertainty analysis,system complexity modeling and consequence analysis.To describe the uncertainty,the reliability theory is used to quantify the failure probability and a stochastic system evolution model is developed,based on stochastic process modeling and graph theory.The model is further embedded in an optimization algorithm,based on the max-flow algorithm and the shortest path algorithm,to simulate the transform process of uncertainties due to the system behaviors and to analyze the consequences of failures in the system.In the case study,the effectiveness of the developed framework is verified,and the results of the numerical simulation indicate that it can be applied to analyze the supply reliability from multiple perspectives.This framework can be used as the basis of further researches of supply reliability analysis in the real-world applications.Considering the defect of probability-dependent reliability analysis method,the concept of vulnerability analysis is adopted to supplement the analysis of supply reliability.The key point of the vulnerability analysis is to analyze the system ability to maintain a reliable supply to its customers,and to identify the potential threats to its supply capacity.Considering that,a systematic framework of vulnerability analysis of natural gas pipeline networks,including global vulnerability analysis,customers’ robustness analysis and critical pipeline identification,is developed in this work.In the case study,the effectiveness of the developed framework is proved by the numerical simulation.In this verification part,a Monte-Carlo based global vulnerability analysis is developed,whose results provide a comprehensive picture of the system ability to withstand strains imposed on the system.The network-flow based critical pipeline identification method shows good effectiveness and efficiency on measuring pipeline importance and locating critical pipelines.Besides,two indices are developed to quantify the customers’ robustness,and topology properties are used to explain the differences of the robustness between customers.Considering the demand fluctuation and the peak-shaving capacity,a short-term supply reliability analysis method is developed.To further manage this supply reliability,a demand side management framework is developed,by integrating deep-RNN-based demand forecasting,Deep-Learning-based system condition prediction,the UtilityTheory-based customer behavior analysis and the Genetic-algorithm-based multiple objective optimization process.The results of the simulation show that the developed Demand Side Management framework is able to effectively reduce the risk of shortage of supply and fluctuation of aggregative demand,and to enhance the profit of the supply side.This work provides a novel idea to analyze and manage the short-term supply reliability of natural gas pipeline networks.