Study On The Risk Assessment And Decision Making Of Wind Power Ramp Events

With the increasing penetration of wind power,the power system's operation risk is increasing because of the fluctuation and uncertainty of wind power.When extreme weather condition(such as tropical cyclone,tornado,explosive temperate cyclone,severe convection weather and so on)occurs,the wind turbine will slow down or shut down for safety.And this will cause wind power ramp events whose power output drop sharply in a short period of time.The wind power ramp events will cause great impact and operation risk to the power grid which seriously threaten the operation safety of the power system.So this paper conducts a systematic study of the wind power ramp events' risk from three aspects including risk prediction,risk assessment and risk decision making.Firstly,this paper proposes a new wind power output forecasting method.This method fully considers the meteorological influence on the wind power output capacity including wind speed,wind direction,temperature,atmospheric pressure,humidity and so on.First this paper adopts the wavelet-BP neural network method to make deterministic wind power prediction.And the comparison between the wavelet-BP method and the traditional BP neural network at the difference of wind power prediction accuracy is made.Then based on the deterministic prediction results,the probability prediction is made adopting the kernel density estimation technique.It is more convenient to model the wind power.At last this paper recognizes the wind power ramp events from the wind prediction results based on the definition of wind power ramp events.So the wind power ramp events can be predicted qualitatively.Secondly,the wind power ramp events risk assessment model is established based on the prospect theory.Different from the traditional wind power fluctuation,wind power ramp events usually caused by extreme weather conditions,its probability of occurrence is small.But once happen,the wind power ramp events will trigger serious consequences such as load loss and frequency instability,even lead to the system blackout.In order to solve this problem,at first,the wind power ramp evens model is established.Then build the risk assessment indicator based on the prospect theory.At last the risk assessment process is also given.The algorithm case study proves that the risk indicator proposed in this paper can highlight the wind power ramp events whose probability is small but the risk is high.Compared with the traditional assessment methods,the proposed methods in this paper can fully consider the risk aversion attitude of the system decision maker.It is of great significance to the system operation personnel.Finally,the reserve decision model to deal with the wind power ramp events is built.On the basis of traditional reserve decision model,in the objective function of the model,the loss of load and the cost of abandoned wind are considered.This paper analyzes the influence of the different system reserve cost and the loss of load on the system scheduling decision when the wind power ramp events occurs.