Research On Multi-time Scale Optimal Scheduling Considering Wind Power Frequency Modulation Capability

In the past,the excessive use of fossil energy has highlighted the problem of environmental pollution,and fossil,as a non renewable energy,has gradually dried up in recent years.In this context,renewable energy has been developed and utilized,among which wind energy is one of the important clean energy and has great development potential.With the large-scale grid connection of wind power,China’s power system is also developing in a complex and diverse direction.The safety and stability of high permeability wind power grid operation are threatened by the randomness and uncertainty of wind power.With the large-scale grid connection of wind power,the technology of wind power participating in power grid frequency modulation control is also developing.Based on wind speed prediction technology and uncertainty analysis,it is of great significance to study how to consider frequency index in optimal dispatching and how to consider wind power frequency modulation capacity in power system optimal dispatching model,so as that wind energy is fully utilized to improve the reliability and the economy of the system.Firstly,the primary frequency modulation ability of wind power is explored and analyzed,and the wind fire joint frequency modulation strategy is designed in this paper.Starting from the dynamic formula,according to the fan output is mainly related to wind speed,speed and pitch angle,the fan operation area is divided into maximum power tracking area and power constant area according to the rated speed of wind.The primary frequency modulation principle of traditional synchronous generator is simulated through overspeed and pitch control in different areas,the maximum power tracking area adopts overspeed control,and the constant power area adopts pitch angle control.Thus,the quantitative calculation formula between wind speed,angular velocity,angle of pitch and additional power for frequency modulation of wind turbine is established.The difference adjustment coefficient with the same effect as the traditional synchronous machine is set in the two areas.According to the difference adjustment coefficient,the doubly fed wind turbine and the traditional synchronous machine can participate in the primary frequency modulation of the system together;According to the change of system frequency,the joint frequency regulation of DFIG and thermal power unit can improve the consumption of wind power in the power grid.When the system frequency decreases,they participate in frequency modulation together in the above way,and when the system frequency increases,only the thermal power unit adjusts the output.The simulation results show that the fan can participate in frequency modulation well in the load shedding operation of the two zones,and the combined frequency modulation strategy of wind and fire can not only ensure a low air rejection rate,but also better improve the steady-state frequency of the disturbed system.This lays a foundation for integrating wind power frequency modulation into optimal dispatching.In this paper,the day ahead-day rolling-real-time three-stage coordinated scheduling model is adopted to gradually correct the wind power prediction error,so as to suppress the impact of wind power fluctuation on the system.Considering the impact of wind power prediction error on the optimal dispatching,the deviation left by the upper level is corrected from the lower level.In terms of wind power prediction,the dispatching model uses the characteristics related to the distance between the prediction information and the prediction point,the time scale and the modeling method of wind power prediction results（When the wind speed prediction is closer to the prediction point,the shorter the prediction time scale,and considering the uncertainty of wind power,the higher the prediction accuracy）.In the previous stage,the time series Arima method is used to predict the wind speed with historical data,while in the intra day stage,the wind speed prediction is based on the latest data.The uncertainty of wind power is considered in the real-time stage,and the typical scenes are analyzed by using the multi scene probability method,which makes the wind power prediction in the next stage more accurate than the previous stage,and can better reflect the impact of wind speed prediction error on the system,so as to correct it.In terms of optimization objectives,each stage of the scheduling model is optimized for different objectives.In the previous stage,the unit start-up and shutdown plan is optimized with the objective of minimizing the comprehensive cost of unit start-up and shutdown and operation;In the day rolling stage,the output of conventional generator set is corrected according to the day wind power and load fluctuation,and the interior point method is used to solve it,so as to minimize the operation cost of conventional thermal power unit again;In the real-time dispatching stage,the uncertainty of wind power is taken into account,the influence of power fluctuation on the frequency quality of the system is considered,and the minimum network loss and frequency offset is taken as the goal.The dynamic power flow algorithm is used to integrate the frequency regulation capacity of thermal power units and loads and the participation of fans in primary frequency regulation mentioned above,so as to jointly share the unbalanced power of the system.In this stage,the multi-objective particle swarm optimization algorithm is used to solve the multi-objective real-time scheduling model and generate the Pareto optimal solution set.