Multi Energy Local Area Network Optimization Based On Data Driven Multi-stage Robust Optimization

At present,the penetration of new energy such as wind power and photovoltaic in the power system is increasing.Because the output of new energy is subject to the influence of weather and other factors,its output has great volatility and intermittence,which brings a huge challenge to the safe and stable operation of the power grid.In this thesis,the current power system to deal with wind and other new energy output uncertainty optimization theory is analyzed and compared,and finally the robust optimization is used as the core optimization model of this thesis.Through the idea of data-driven,the uncertain set of robust optimization is optimized to reduce the conservatism of robust optimization.Finally,a robust optimization model is built in energy local area network for simulation analysis,which verifies the correctness and effectiveness of the model.The main work and achievements of this thesis are as follows:1)Two methods of data-driven robust optimization are described:data-driven theory under distance estimation and data-driven theory under probability information.Both methods are to reduce the improved theory derived from the conservatism of the robust optimization model,but the data-driven theory under the distance estimation is to optimize the uncertain set twice,while the data-driven theory under the probability information is to reconstruct the existing uncertain set.Although the purpose of the two methods is the same,the difficulty of solution and the idea of modeling are totally different.In addition,the general framework of data-driven theory under distance estimation and the general mathematical model of data-driven theory based on probability information are given.2)In this thesis,the process and mathematical model of data-driven theory based on distance estimation in power system are described.Based on the idea of norm,the data-driven theory of distance estimation optimizes the model.Compared with the original model,the improved model is simpler and more accurate.Finally,the correctness and validity of the model are verified by the simulation analysis in the integrated energy LAN.3)The application of data-driven theory in comprehensive energy system based on probability information is given.A new robust stochastic optimization model is proposed.Compared with the general distributed robust optimization model,this model has a strong universality.The model is further optimized to make it more suitable for power system related problems.In the simulation analysis,the correctness and feasibility of the robust stochastic optimization model are verified by comparing with the distributed robust optimization model under Wasserstein distance and the distributed robust optimization model under generalized moment information.