Economic Dispatch Of Microgrid Based On Bi-level Programming

In the power industry, the combination of power grid and distributed power is a method to solve the contradiction between the growth in electricity demand and environmental pollution, energy supply. As a local area network, micro grid can make full use of the technical and economic benefits which was brought by distributed power supply to users and public grid, and effectively solve the contradiction between the power grid and Distributed Generation.Optimal dispatching plays an important role in the research of microgrid, which is mainly address how to optimize the distribution of energy, making microgrid run more economical and environmentally friendly, improving the efficiency of the use of energy. Currently, the main content of the research on microgrid scheduling optimization at home and abroad is use a single-objective or multi-objective to solve the minimum cost of power generation, the minimum cost of environmental impact. However, the major industrial countries began to implement the electricity market since from the last century 90's, layered coordinated decision have become an important feature of the system to optimize decision-making, then take advantage of the traditional single goal or multi-objective optimization methods can not solve the economic dispatch of such systems, bi-level programming start coming into view. The main work of this paper is as follows:First of all, the problem of optimal supply function for power producer with Energy Storage System(ESS) is studied. In order to improve the reliability and security of the power supply, We often need to consider a certain spinning reserve for the power system. Traditional method is generally considering gas turbines or diesel units at the power supply side, but with the integration of distributed energy resources more, spinning reserve need more. This way is not the economy, and also has increased the pollution. Energy storage system has the advantage of rapid absorption or release energy, and can effectively compensate for the shortcomings of renewable power volatility, which provides some new idea for large-scale renewable energy centralized grid. A bi-level optimization model for power producer and Energy Services Provider(ESP) was constructed, in which the upper layer is considered diesel supplier's maximum profit, the sublayer is considered the total profit of energy services provider, and the sublayer is considered energy storage system as the spinng reserve. Numerical experiments analysis the influence of parameters of the model, and verify that the proposed model has significant economy than when no storage system.Secondly, in order to deal with the electric power system scheduling problem caused by large-scale wind power,we use the energy management strategy of demand response and establish a bi-level optimal scheduling model that takes into account the user and management side. According to the operating characteristics and characteristics of the controllable load user side, the upper layer divide the controllable load into four categories in all, and the goal is to minimize the total cost of the household controllable load, and the characteristics of the mathematical expression are described; the sublayer optimization objective function is to make the microgrid power supply to meet the total minimum cost under the condition of power supply reliability, and make use of the method of overestimate and underestimate to deal with the uncertainty of the wind power output. Numerical experiments show that the proposed model has obvious economic benefits, and enhance the absorptive capacity of the wind power.At last, according to the piecewise function in the objective function of optimal model, we use the processing method of smoothing to deal with it. The Original bi-level model is transformed into an equivalent single-level program by its KKT(Karush–Kuhn–Tucker) condition.With the deepening of the reform of the power market, the characteristics of hierarchical coordination decision become power system will be more and more obvious. At this point, the use of bi-level programming to solve scheduling optimization problem will be more and more widely. In this paper, two optimization models are established by using bi-level programming, and the economy and validity of the proposed model and the algorithm are verified by numerical experiments. The work of this paper provides a theoretical basis for the wide application of the power system.