| With the rapid reduction of fossil energy storage and the rapid deterioration of the environment,developing renewable energy and improving energy efficiency have become the only way out for sustainable energy development.The social energy supply system has transformed from independent planning and operation of electricity,heat,and gas to joint planning,design and operation,forming an energy Internet for efficient energy use.The park is a multi-energy system built on user terminals and the technology carrier of the energy Internet,involving the production,transmission and consumption of multiple energy sources such as electricity,heat,and cold.For multiple parks in the same regional power grid,the parks are interconnected through tie lines,which can improve operating economy.At the same time,the uncertainty of renewable energy in the park will have a great impact on scheduling decisions.When the operating entities are different,centralized optimization scheduling cannot protect the private information of each park.Therefore,it is necessary to study a decentralized coordinated optimization scheduling method of multiple parks that considers the uncertainty of renewable energy.This thesis studies the method of multi-park decentralized optimal scheduling under uncertainty.The specific research content is as follows:1)A robust optimization scheduling method for the park is proposed.First,the energy transmission relationship of the equipment in the park and the information needed for park scheduling decision-making is analyzed.Secondly,an optimal scheduling model for the park is established,with the goal of minimizing the operating cost of the park and considering the operational constraints of various equipment in the park.Finally,a polyhedral uncertainty set is introduced to consider the uncertainty of new energy output,and a single-stage robust optimization model of the park is established to find the optimal scheduling scheme under the worst scenario of renewable energy output based on the deterministic model.2)A distributed optimization scheduling method for multiple parks is proposed.First,a coordinated and optimal dispatch model for multiple parks that considers the power interaction is established,taking the sum of the operating costs of multiple parks as the goal,considering the internal equipment constraints and energy transmission constraints of each park.Secondly,the centralized model is transformed into a decentralized model through the alternating direction multiplier method(ADMM),and the decentralized optimal scheduling of multiple parks is realized through the communication between adjacent parks and the parallel iterative calculation of all parks.At last,simulation cases are used to analyze the optimal scheduling results of each park,the effectiveness of the model,the influence of the network loss coefficient on the coordinated scheduling of multi-parks,and the computing performance of multi-park systems of different scales.3)A decentralized robust optimization scheduling method of multiple parks is proposed.The ADMM is used to solve the multiple parks’ scheduling problem in a decentralized framework,and the column-and-constraint generation(C&CG)algorithm is used to solve the park’s robust optimization model considering power interaction.In the decentralized solution process,only the expected interactive electric power information needs to be transmitted between the parks.The optimal scheduling of multiple parks is realized through the parallel iterative calculation of each park.The robust optimization model of the park is divided into the master problem and sub-problem,which is solved alternately and iteratively.The master problem solves the expected interactive electric power between the parks,and the sub-problem solve the worst scenarios of renewable energy in each park and the internal scheduling scheme.Finally,the effectiveness of the proposed method is demonstrated by comparing the simulation result with the centralized robust model. |
PDF Full Text Request