Research On Dispatching Model Of Direct Power Purchase By Large Consumers Involved System Considering Wind Power Prediction Error

Modern society is facing many problems such as energy crisis,environmental pollution and economic development.With the maturity of wind power technology,large-scale wind power integration has become an inevitable trend.On the other hand,as an important measure of the electricity market reform,direct power purchase by large consumers will be actively carried out throughout the country.However,the inherent volatility and randomness of wind power determine that there are prediction errors which are difficult to eliminate,and there is a large demand for the spinning reserve.At the same time,in order to ensure the full delivery of direct purchase contracts,power generation enterprises tend to let the thermal power units which with stable and controllable output sign direct purchase contracts,that will reduces the spinning reserve space of power system.With the increased capacity of wind power integration and the continuous promotion of direct purchasing power,the larger reserve demand of wind power and the less reserve space of the system will certainly limit their further development.Therefore,it is necessary to consider the impact of wind power integrated system on system dispatch when participating in direct power purchase.In summary,this paper establishes a dispatching model for the participation of wind power integrated system in direct power purchase and conducts research,which provides a theoretical basis for the participation of wind power integrated system in direct power purchase.Firstly,this paper uses BP neural network to predict wind power output by combining wind speed and wind power curve,uses Beta distribution to fit the probability density function of wind power output,and finally calculates the prediction error by using probability theory.This method of considering prediction errors is the basis of the following two parts.That is to say,in the first part,the spinning reserve is used to deal with the prediction error of wind power,and finally the optimal direct purchase form is obtained;in the second part,the energy storage system and spinning reserve are used to deal with the prediction error,and finally the optimal integration capacity of wind power is obtained.specifically as follows:The first part studies the economic dispatch of direct-purchased power system with a certain capacity of wind power integrated and considering wind power prediction error.Using spinning reserve to deal with the prediction error of wind power.At the same time,because thermal power participates in direct power purchase,it has an impact on the reserve capacity.Spinning reserve is used as the combination point of wind power forecasting error and direct power purchase.A power system scheduling model considering wind power forecasting error and direct power purchase is established.Chance constrained programming is used to deal with random variables,and penalty function method is used to deal with multiple constraints.Finally,the improved particle swarm optimization algorithm is used to solve the model in the IEEE 30 bus system of wind power integrated,and the best way for the power system with wind power to participate in direct power purchase by large users is determined.The another part studies the optimization of integration capacity of wind power utilizing energy storage and spinning reserve in the context of direct power purchase.This part of the research is that the integration capacity of wind power is increased reasonably on the basis of the above part,and the energy storage system is added to deal with the situation that the reserve space is insufficient due to the increase of integration capacity of wind power.It is proposed that large consumers directly absorb wind power in the form of direct power purchase.Based on game theory,a stackelberg game model is established,in which the power generation enterprise is the leader and the wind farm is the follower.Finally,the optimal integration capacity is obtained by solving the model in the IEEE 30 bus system with wind power.