| Natural gas pipeline network is an important link of natural gas supply chain,which plays a vital role in connecting resource supply and market demand.In order to effectively balance the relationship between resources and market demand for natural gas,and ensure the economy of pipeline network gas supply,this paper studies the relationship between resource supply and demand side loss on the basis of the uncertainty analysis of pipeline network and demand side,and proposes a method to determine the target dependence of natural gas pipeline network gas supply based on the investment benefit criterion.A hybrid prediction model using complete EEMD with adaptive noise(CEEMDAN)and wavelet threshold denoising is proposed,which integrates the whale optimization algorithm(WOA)and the long and short term memory neural network(LSTM).The model reduces the non-stationarity and noise of daily load data by using CEEMDAN and wavelet threshold denoising.The short-term gas demand of different nodes of natural gas pipeline network is predicted based on holidays,date types,weather and the interaction between nodes.The applicability of CEEMDAN-WOA-LSTM model is verified by conducting an error analysis.A reliability evaluation method based on demand-side economic loss risk is proposed to evaluate the reliability of gas pipe network from the economic perspective.The established reliability evaluation model reflects the risk of gas shortage in the process of gas supply to the demand side.Considering the supply priority of important users,the minimum cost and maximum flow algorithm of the graph theory is improved to achieve the reasonable distribution of different types of users’ flow under the gas shortage condition.The gas shortage risk of the pipeline network system can be obtained by calculating the gas shortage under each unit failure and combining the failure probability of each unit.According to the theory of energy substitutability,gas shortage is transformed into economic loss form to assess the economic risk of gas shortage for different types of users.A reliability evaluation method based on economic loss risk is established to correlate reliability with economy.In order to take into account the reliability and economy of the pipeline network in the process of gas supply to the downstream,the target reliability optimization model of the natural gas pipeline network is established based on the investment benefit criterion,taking the resource allocation of the natural gas supply chain system as the research object and the benefit maximization of the whole resource market as the goal.The decision variable is the peak regulating reserves of each gas source node under the scenario of pipeline network structure uncertainty coupled with market uncertainty.The model considers the costs generated in each link of the natural gas supply chain.The target reliability of interest balance is determined through the game between the supply and demand parties on input cost and economic loss cost.Particle swarm optimization algorithm(PSO)is used to solve the problem,and the benefits of each party are analyzed.Based on the analysis of the factors affecting the gas supply of the pipeline network system,the principle of reliability distribution of the gas supply of the system is determined,and the factors affecting the reliability of the pipeline network system are quantified.By using the comprehensive weighting method,the method of reliability distribution for the gas supply of the pipeline network system is established.The proposed method is applied to pipeline network system,and combined with the gas supply characteristics of the pipeline network system,analyzing the influence of various influencing factors on the gas supply and completing the distribution of gas supply reliability. |
