Research On Operation Strategy Of Combined Wind-storage System Participating In Real-time Energy And Regulation Market

Due to the intermittent and uncertainties of wind power,there will be bias penalties in the process of participating in power market.It results in a decrease in the revenue of wind power.With flexible charging and discharging capability,energy storage systems can help in reducing wind power’s energy imbalance.Therefore,the joint participation of wind power and energy storage in electricity market can not only improve their revenues,but also improve the safety of power systems with a high proportion of wind power participation.The existing operation strategy research is mainly based on the combined wind-storage system to obtain revenue by providing energy and frequency regulation services to day-ahead power market.The optimization strategy is mainly based on the maximum expected market return,and a single-time stochastic optimization model is established to solve the optimal bidding and operation strategy.With the continuous improvement of electricity market in various countries,many real-time markets are opened in electricity market.So the combined wind-storage system can participate in real-time market to obtain greater benefits,and it can reduce the reserve capacity of the power system.Therefore,an optimization method based on robust model predictive control for the combined wind-storage system in real-time energy and regulation market is proposed.The main research work of this paper is as follows:Firstly,the model of the combined wind-storage system participating in power market is analyzed.The frequency regulation characteristics of wind power and energy storage system are introduced respectively.The model of the combined wind-storage system participating in real-time energy and regulation market is summarized.Secondly,the single-time stochastic optimization model of the combined wind-storage system participating in real-time energy and regulation market is studied.The optimization strategy which combines the low operating cost of wind power and the fast response speed and high precision of the energy storage is aim to maximize expected revenue and the scene probability method is used to express the uncertainty of wind power output and market price.And the operation strategy optimization model for the combined wind-storage system participating in real-time energy and regulation market was established.The linearization method was used to transform the nonlinear model into a mixed integer programming model that can be solved by CPLEX.Finally,the validity of the algorithm is verified by the data of the PJM market.Thirdly,the operation strategy optimization model for the combined wind-storage system that considers the impact of energy storage system’s SOC on the provision of frequency regulation services in real-time energy and regulation market is established.It is proposed that the combined wind-storage system can provide more regulation services by purchasing electricity in real-time energy market to maintain the storage SOC within the set range.It has been counted it into the original model to optimize the operation strategy.The effectiveness of the proposed method is validated with PJM market data.Fourthly,the operation strategy optimization model of the combined wind-storage system participating in real-time energy and regulation market based on robust model predictive control is established.The strategy complements the advantages of wind power and energy storage system.It is proposed that the combined wind-storage system can gain more revenue by purchasing electricity in real-time energy market to maintain the storage SOC within the set range.Its objective function is that the combined wind-storage system has the largest revenue in real-time energy and regulation market.Finally,the strong duality theorem is used to transform the two-layer robust model into a single-layer robust model which can be solved by CPLEX.And the effectiveness of the algorithm is verified by PJM market data and compared with the results of the single-time stochastic optimization model.The sensitivity analysis of each parameter were done.