Research On Gas-power Coupled System Economic Dispatch Based On Distributionally Robust Optimization

With the introduction of the concept of Energy Internet and the development of gas unit and P2G technology,the coupling relationship between power system and natural gas system is gradually deeper,the influence between the two systems can not be ignored anymore.In addition,due to the rapid growth of the installed capacity of renewable energy generators worldwide and the proposal of China "3060" carbon peak carbon neutral target,renewable energy power generation will become the main power generation method in the future,and wind turbines will be one of most important power generation methods,the problem of insufficient accuracy of its predicted output has always been a concern.Therefore,this paper considers the wind power day-ahead forecast output error,and studies the day-ahead and real-time scheduling optimization problems of the gas-electric coupling system,aiming to put forward an effective balance of wind power day-ahead forecast by constructing an economic dispatch model of the gas-electric coupling system.First of all,this paper summarizes the current research status of gas-power system optimization and uncertain dispatch optimization problems.By comparing the similarities and differences between the power system and the natural gas system,the value of the coordinated and optimized operation of power system and natural gas system is analyzed.This paper describes the operation principles of wind turbines,gas turbines and P2G equipment in the gas-power system and constructs mathematical models.The power system network power flow model,the pipeline flow model of the natural gas system and the natural gas system power flow model are also established.Considering the application of energy storage equipment and demand response mechanism in the gas-power system,adding batteries and gas storage tanks,and considering the electric load demand response and the gas load demand response,the basic framework of the gas-electric coupling system is finally built.Secondly,this paper uses the Wasserstein based distributionally robust optimization model to deal with the uncertainty of wind power output.the fuzzy set of wind power day-ahead forecast error is first established,and the wind power historical forecast error data is used to construct the support set of wind power day-ahead forecast error.Combining the basic framework of the gas-power system,with the cost minimization as the optimization goal,an economic dispatch model including day-ahead and intra-day is established,and the model is transformed into an easy-to-solve MILP form through linearization,affine transformation,and duality.Finally,this paper uses the test system and test data to perform in-sample analysis and out-of-sample verification of the model results obtained under different sample sets.The results show that the constructed wind power day-ahead forecast error fuzzy set will gradually converge to the true distribution.Compared with stochastic optimization and robust optimization,the method has superiority,and the proposed model can propose a two-stage scheduling strategy that effectively reduces system costs and reduces wind power curtailment.In addition,this paper analyzes the results of the calculation examples in the scenarios with or without energy storage equipment and with or without demand response mechanism,and compares the effects of energy storage equipment and demand response mechanism in the system scheduling process,and explains the energy storage equipment and the demand response mechanism.Demand response can all play a role in reducing cost and stabilizing wind power fluctuations in a gas-power system,and the role of demand response is greater than that of energy storage equipment,and the role of electrical load demand response in demand response is greater than that of gas load demand response.