Wind Power Prediction Based On Gated Recurrent Unit Neural Networks And Economic Dispatch

With the decline of traditional fossil energy,renewable energy has received more and more attention worldwide.As an important form of renewable energy,wind energy has great development potential and has been developed and utilized by many countries.However,the randomness and intermittent nature of wind energy bring many problems to large-scale grid connection of wind power.Especially in the two aspects of wind power forecasting and power system economic dispatching,due to imperfect technological development,wind power utilization rate is not high,and even wind abandonment occurs.Therefore,research on wind power forecasting and economic dispatching of power systems is particularly important,and it has practical significance for improving wind power consumption.This paper studies the relevant models and methods for this topic.Wind power forecasting is based on historical data from wind turbines to predict trends in wind power changes over time.Based on the research of existing wind power prediction technology,this paper proposes a wind power prediction model based on Complete Ensembel Empirical Mode Decomposition with Adaptive Noise-Gated Recurrent Unit Neural Network-eXtreme Gradient Boosting.In order to improve the accuracy of wind power prediction,the first model adopts signal analysis technology,and uses CEEDAN algorithm to smooth the wind power signal,and decomposes the fluctuation or change trend of different scales in wind power signal to reduce wind power signal.Nonlinearity and non-stationarity,a set of wind power signals with different frequency bands are obtained;then the wind power signals of each frequency band are predicted by GRU neural network respectively;finally,the input is added to the limit lifting tree for correction and fusion,and the final wind power prediction value is obtained.Based on the previous research,an economic dispatching model for power systems with wind turbines and gas turbines is established.The fourobjective scheduling model considering the minimization of power generation cost,maximizing wind power utilization,minimizing pollutant emissions,and the stability of thermal power units is considered.Then the Unified algorithm is used to solve the scheduling model.The solution of the model is obtained by an example.The validity and feasibility of the proposed model and the improved algorithm are verified.