Forecast Of Wind Power Consumption Ability Based On Artificial Intelligence And Optimization Of Consumption Measures

As the international community pays more and more attention to environmental protection and energy security,all the world reach a consensus in reducing fossil fuel consumption and accelerating the development of renewable energy.In recent years,wind power has become one of the fastest-growing clean energy sources in the world.The penetration rate of wind power increase with the increase of wind power installed capacity and grid-connected capacity,which making the operation mode on the power side more complicated and variable,bringing great challenges to the safe and stable operation of the power grid and seriously affecting the wind consumption of the power grid.Therefore,assessing wind power consumption capacity,identifying key factors affecting wind power consumption and optimizing wind power consumption measures are the focus of current research,which have important practical significance for ensuring the safe and stable operation of the power grid,improving the utilization rate of renewable energy and protecting the ecological environment.This paper conducts research on the generation of typical scenarios for assessment of wind power consumption capacity of power systems,prediction methods of wind power consumption capacity and optimization of wind power consumption measures,as follows:Aiming at the problem of randomness and uncertainty in the selection of typical scenarios of wind power and load,this paper proposed a method for generating typical scenarios of wind power and load oriented to assessment of wind power consumption capacity.Considering the correlation between wind power and load output,the characteristic indicators for wind power consumption capacity evaluation are extracted.And then constructing the characteristic curve of wind power output and analyzing and comparing the similarity between the original curve and the characteristic curve.Based on the layer-based thinking,a wind power-load typical scene generation method based on AHC and improved K-means two-level clustering is established.The upper layer uses the AHC clustering algorithm to cluster the wind power characteristic curve and the lower layer uses the improved K-means algorithm to cluster the wind power-load coupling curve.The effectiveness of the proposed algorithm is verified according to evaluating and analyzing the cluster results of wind power and load.The result shows that the two-layer clustering model is superior to single-layer clustering in algorithm performance and its error is much lower than that of single-layer clustering algorithm.When using the typical scenes replace all the scene to analysis wind power consumption capacity,the calculation time is greatly reduced,and the calculation efficiency is improved.So typical scenarios are conducive to short-term scheduling and mid-and long-term planning of power system to improve the system’s wind power consumption capacity.In order to make full use of historical data to predict the capacity of wind power consumption,and analyze the key factors affecting the capacity,a combined forecast model of wind power capacity based on DNN-XGBoost is proposed.Establish DNN wind power consumption prediction model and XGBoost wind power consumption prediction model respectively and optimize the hidden layer and neuron parameters of DNN model and Booster parameter of XGBoost model by grid search method.Combining the prediction accuracy of the DNN model and the interpretability of the XGBoost model,the reciprocal error method is used to establish a combined prediction model and predict the wind power consumption capacity,and then analyze the influencing factors.The combined model is compared with the other eight regression algorithms,and the accuracy of the proposed model and the degree of influence of various influencing factors on wind power consumption capacity are verified by scoring the model and error analysis.The prediction of wind power consumption and the analysis of key factors will help the power grid propose different improvement measures for different scenarios to enhance the system’s wind power consumption capacity.In order to fully analyze the potential of the power system flexibility,a joint optimization planning model for the flexibility reform of the thermal power unit and energy storage was proposed.Based on the key factors of influencing wind power consumption capacity,the flexibility potential of the power and load side is analyzed.Using the current thermal power plant flexibility transformation technology,a power system planning and operation model considering the cost of unit transformation and the cost of low load life loss is proposed.At the same time,through the selecting of energy storage and the planning of its power and energy,a joint optimization planning model considering thermal power units and energy storage was established.This paper has carried out simulation calculation analysis of various scenarios based on the improved IEEE30 node system.The analysis results show that the model can improve the operation economics of the power system and promote wind power consumption by configuring different depths of thermal power unit transformation and energy storage capacity.