Research On Electric Real-time Balance Dispatch Via Bi-level Coordination Of Source-grid-load Of Power System With Risk

The clean change of energy has brought great challenges to the traditional power dispatch. Recently, in order to cope with the dual pressures of energy shortage and environmental degradation, countries around the world have accelerated the development of distributed new energy generation plants and electric vehicles. They will interact with main network based on their own interest, which will break traditional operation management mode and bring new challenges to power dispatch. This paper carries out research on the electric real-time balance dispatch, and the main work is as follows:(1) A novel framework of electric real-time balance dispatch is proposed. First, considering the characteristics of modern power grid with more and more distributed new energy generation plants and electric vehicles, two key problems named interest coordination and risk coordination were extracted, and then the novel framework of electric real-time balance dispatch was established based on the concept of virtual power plant.(2) An electric real-time balance dispatch method based on bi-level stochastic chance constrained programming is proposed. The models of economic dispatch in the grid layer and source-load layer were built according to their own interests and the various uncertain factors, respectively. For translating the stochastic chance constrains into the deterministic constrains, the support vector machine(SVM) and stochastic simulation are utilized simultaneously. Then the bi-level optimization method composed by the pattern search(PS) algorithm and artificial bee colony(ABC) algorithm is used for solving the bi-level real-time economic dispatch model with the deterministic form.(3) An electric real-time balance dispatch method based on bi-level fuzzy chance constrained programming is proposed. Considering the difficulty of describing the uncertainty of distributed new energy power plants and loads with probability, the uncertainty are further described with fuzzy parameters, then an electric real-time balance dispatch model based on bi-level fuzzy chance constrained programming is established. For the solution of the proposed model, it is first transformed into a deterministic form by the crisp equivalent method, support vector machine and fuzzy simulation. Then it is solved by combining pattern search algorithm with artificial bee colony algorithm.(4) An electric real-time balance dispatch method considering carbon emission cost is proposed. Considering the demand for energy saving and emission reduction at this stage, this paper carries research on carbon emission characteristics and carbon emission trading cost of various types of power source in electric real-time dispatch. By introducing carbon emission cost into the proposed electric real-time dispatch model based on bi-level fuzzy chance constrained programming, a more comprehensive electric real-time dispatch method is formed. The simulation analysis results demonstrate the effectiveness of the proposed method.