Day-ahead Coordinated Optimal Dispatching Of Source-load-storage For Power System Containing Wind Turbines

As more and more wind turbines are connected to the grid,the uncertainties in power system gradually increase.Demand response and energy storage equipment are effective ways to solve the uncertainties of wind power and realize wind power consumption.Considering the uncertainty of wind power,it is of great significance to study the day-ahead coordinated dispatching of source-load-storage for power system containing wind turbines to realize the safe and economic operation of power system.This thesis focuses on day-ahead coordinated optimal dispatching of source-loadstorage for power system containing wind turbines,and the main works are as follows:(1)This thesis summarizes the operational characteristics of generators,loads and energy storage equipment,selects time-of-use tariffs to guide power users to transfer their power consumption periods,and selects interruptible loads and battery energy storage equipment to participate in day-ahead dispatching,which lay the foundation for realizing day-ahead coordinated dispatching of source-load-storage for power system containing wind turbines.(2)This thesis proposes the idea of updating peak and valley hours and time-ofuse tariffs according to the daily variation.Firstly,the synchronized optimization model of time period division and time-of-use tariffs is proposed,which introduces the state variables of peak,flat and valley hours to realize the synchronized optimization of peak and valley hours and time-of-use tariffs.Particle swarm optimization is used to solve the synchronized optimization model,and a method for generating the initial particle swarm is given,which improves the convergence performance and search ability of particle swarm optimization.Secondly,a model of clustering-based time period division and tariffs optimization is established.This model constructs a tariffs optimization model to optimize the time-of-use tariffs based on the time period which is divided by clustering according to the net load.The analyses of the case studies show that both optimization models above can reduce the peak-to-valley difference of the net load.Among two models,the synchronized optimization model proposed in this thesis,which achieves simultaneous optimization of peak and valley hours and time-of-use tariffs,gets smaller peak-to-valley differences and fluctuations of net load.In addition,it can obtain the optimal peak and valley hours and time-of-use tariffs which meet the customer’s demand by adjusting the results of time period division according to the customer’s satisfaction demand,thus has better effectiveness in guiding power users to transfer their power consumption periods.(3)This thesis applies the non-parametric probabilistic prediction results of wind power to the day-ahead dispatching model of power system,which describes the probability distribution of wind power more accurately.Based on the probabilistic optimization method,a source-load-storage day-ahead probabilistic optimal dispatching model is constructed according to the load data obtained after the implementation of time-of-use tariffs,which is formulated by the synchronized optimization model of time period division and time-of-use tariffs proposed in this thesis.The model avoids the generation and reduction of a large number of scenarios while considering the probability distribution of wind power,and can take into account the adjustment of conventional units after the wind power deviates from the predicted value to achieve the optimal allocation of reserve capacity among units.When battery energy storage equipment participate in the day-ahead dispatching,the upper limits of charging and discharging power of energy storage equipment are dynamically adjusted according to the variation of the state of charge,which improves the mathematical model of energy storage equipment and avoids overcharging and overdischarging of battery energy storage equipment.The validity of the constructed day-ahead probabilistic optimal dispatching model is verified by simulation for IEEE 6-bus system and IEEE 39-bus system with actual wind power data from the Belgian grid.