Optimization Method Towards Economic Operation Of Microgrid

Microgrid is a new power network which contains distributed generations, energy storage devices, loads and control units. Microgrid is one of numerous means to solve the problems of traditional power system, which brings huge economic benefits. Therefore, economical operation of microgrid is the key to attract users and makes it generalized in the power system. However, with the increasement of the number of devices in microgrid, the optimization of economical operation of microgrid is becoming more complex. Existing optimization of scheduling algorithmn cannot meet the new trends of microgrid. In this paper, aiming at optimizing the economic operation of microgrid, new optimization algorithms are proprosed to improve the operating efficiency of the algorithm based on lagrangian relaxation, which reduces the cost of micro-grid operation. The main contents are as follows:1) The load scheduling of demand-side management is improved in microgrid. Existing algorithm cannot handle large-scale load scheduling problem. Therefore through improving bounded quadratic programming and Benders algorithm based on the framework of lagrangian relaxation, a new distributed algorithm is proposed to solve the problem of demand side management in microgrid which is suitable for large-scale computing. Through simulation, our algorithms is better than the existing central algorithm compared with computation time.2) The problem of joint scheduling of loads and thermal units is solved in microgrid. In this section, microgrid is in isolated mode which only contains thermal units and loads. The purpose of this section is that when the power grid fails, the microgrid can continue to run in a low-cost mod. The joint scheduling is decoupled by lagrangian relaxation. The subproblem of scheduling of single thermal unit is solved by dynamic programming. Through simulation, the proposed method can effectively reduce the cost of the operation of microgrid and shift peak load..3) Considering wind power in the grid, economical operation of microgrid under the grid-connected mode is studied. We use chance constrained programming to describe the randomness of wind power which solves the infeasible probability of expected value model. In addition, distributed implementation framework and a general algorithm of the economic operation of microgrid are proposed which reduce the cost in grid-connected mode. Simulation results show the effectiveness of this algorithm.