Research On Econolic Operation Optimization Of Microgrid

As an effective organization form of distributed power and a powerful supplement of conventional power utility, microgrid has become one development direction of the smart grid all over the world. Through studying microgrid on economic operation optimization, scheduling mode and algorithm, reasonably arranging the working state of microsource and the demand response, can save operating costs and reduce pollutant emissions.In this thesis, by adopting the idea of centralized hierarchical control and studying on characteristics of distributed power source, a multi-objective nonlinear mixed integer programming model about microgrid’s economic operation is created. The model takes the minimum operation costs and minimum pollutant emissions as the target, take the electric quantity limit and the charge/discharge power limit of storage unit, maximum ramp rate limit and minimum on-off time limit of microsource into account as various constraint conditions. At the same time, by taking the on-off states of microsource as decision variables, it is more reasonable to describe the working state of microsource, also makes the model become a mixed integer programming model. For microgrid’s scheduling mode, this thesis analyzed the the grid-connected mode, the grid-isolated mode with controllable load reduction strategy, the priority scheduling mode of microsource with electricity price policy. To strengthen stability of system, this thesis raised the power reserve mode by reserving certain electric quantity of storage unit and its realizing method. Because the maximum ramp rate and minimum on-off time of microsource, the electric quantity of storage unit is limitted, the optimization of adjacent time sections has a strong coupling, this thesis raised an effective measure to put the microgrid’s scheduling from dynamic scheduling into static scheduling.Aiming at the nonlinear and mixed integer characteristics of microgrid’s economic operation multi-objective optimization, two solutions based on ∈-constrained method and NSGA-II is proposed in this thesis. In the solution based on ∈-constrained method, the branch and bound algorithm and interior-point algorithm are applied to solve the subproblem which has mixed integer and nonlinear characteristics, this thesis reduced the number of constraints by putting some equality constraints into process flow, and enhanced the uniformity of the Pareto optimal solution set by adopted two kinds of improved ∈ sequence generation method. In the solution based on NSGA-II, this thesis adopted the way by using piecewise function to build evaluation function, to deal with inequality constraints in the model, and removed the equality constraints in the model by equivalent transformation, improved the solving efficiency. This thesis raised two kinds of population initialization methods which are based on random method and ∈-constrained method to improve the diversity of initial population. This thesis also gave two kinds of decision-making methods based on Pareto optimal solution set. A satisfied solution is chosen from the Pareto optimal solution set, as the settings of MC/LC which are unit level controllers.Finally, this thesis takes a typical microgrid system as an example to verify the proposed microgrid’s economic operation optimization model and solving algorithm, and analyses the impact of different satisfied solutions on the operation of microgrid. The results show that the multi-objective optimization method which is proposed in this thesis is effective, and the dynamic scheduling method is reasonable.