Study On Optimization Of Wind Farm Short-term Dispatch Scheme Based On Imperialist Competitive Algorithm

In recent years,with the appearance of the energy crisis and the pressure of the global warming,developing the sustainable energy sources has become an emergent issue.Among the different kinds of sustainable energy,the wind power is the most widely applied sustainable energy.Therefore,the wind power has drawn the worldwide attentions.However,due to the highly stochastic and fluctuate nature of the wind resources,the electrical power generated by the wind farm is highly fluctuate.The highly fluctuated wind power will thus threaten the stable operation of the power grid.Currently,two main approaches are mostly applied and studied by both of the industry and academic circle to reduce the stochastic and fluctuation of the wind power.1.Using wind speed forecasting techniques to predict the amounts of the energy production of the wind farms,and do the wind farm operational planning based on the predicted wind energy to reduce the impact of the stochastic nature of the wind power;2.Adopting the battery energy storage system to coordinate operate with the wind farm.By using the battery energy storage system,the energy generated by the wind farm can be stored in the battery,and the energy in the battery can also be output to the power grid.Since the energy stored in the battery is controllable,the total energy injected into the power grid from the wind farm will become controllable.Based on this scheme,the impact of the fluctuation nature of the wind power on the power grid can be effectively reduced.The short-term dispatch of wind farm aims to schedule the total power output of the wind farm of a certain future time horizon,so as to achieve a specific objec tive.The research works in the literature mainly focus on the controller design of the wind turbines and some rule-based short-term dispatch strategies.Currently there is no work to model the wind farm dispatch problem as a mathematical optimization problem.In this thesis,a non-linear optimization model for the short-term dispatch of wind farm is proposed.The model considers both of the operational constraints of the wind farm and battery energy storage systems,as well as the economic profit of the wind farm.The intelligent optimization technique is employed to solve the proposed model.The main research works completed in this thesis is as follows:1.Establishing the operational model of the wind farm.The operational constraints of the wind farms are fully considered in the model;2.Establishing the model of the battery energy storage system.The operation characteristics and the operational constraints are fully considered in the model;3.A mathematical optimization model for the short-term dispatch of wind farm is proposed.The proposed model is a combinatorial optimization problem.The optimization objective is to decide suitable charging actions of the battery energy storage system to maximize the total profit of the wind farm over the whole di spatch horizon,based on the operational constraints;4.Based on the imperialist competitive algorithm,this research proposes a short-term dispatch framework for the wind farm with the battery energy storage system.The proposed framework searches out t he optimal dispatch scheme of the wind farm,based on the self-adaptive evolution mechanism of the imperialist competitive algorithm;5.Comparing the imperialist competitive algorithm with the other widely used intelligent optimization algorithms(includi ng the differential evolutionary algorithm(DE),the particle swarm optimization algorithm(PSO),etc.).The algorithm efficiencies on the studied short-term wind farm dispatch problem are compared.