Research On Power System Bus Load Situation Prediction And Economic Dispatch Strategy

With the development of power systems,the load characteristics of busbars have changed significantly,and higher requirements have been placed on the decision-making of dispatching systems.It is urgent to study the operating state and development trend of busbar loads.On the other hand,the adjustable resources can not effectively enter the electricity market,which will lead to waste of resources.In order to solve the above problems,this paper mainly studies from the following three aspectsFirstly,the load condition of the busbar is analyzed,and the concept and structure of the system load situation are extended to the load situation of the busbar.The load state of the busbar is divided into the static potential characterizing its state parameter and the dynamic potential characterizing the change trend of the state parameter.The bus load situation prediction It is divided into three stages:situational information awareness,understanding and prediction.Based on the situational awareness and understanding stage,An improved K-means clustering method based on the elbow method and the maximum and minimum distance algorithm is used to cluster the historical load situation information of the bus.Secondly,the load state of the busbar is predicted,Substituting the daily dynamic potential data to be tested into the model,and the Fisher's discriminant model is established from the data of the daily dynamic potential data to be measured.The Fisher discriminant analysis criterion is used to classify and predict the daily situation,and the load state of the measured daily load is obtained.The clustering category,the historical bus load situation data of the category is substituted as a training sample into the fuzzy neural network(FNN)prediction model training,and the static load prediction result of the bus load on the day to be tested is obtained;Finally,a day-to-day economic dispatch strategy based on bus load situation prediction is constructed.For the bus load situation prediction results,the energy storage system is used to optimize it.Through the optimized load state curve of each bus in a certain area,the comprehensive load of the area is obtained,and then the peak-to-valley electricity price and load aggregator(Load Aggregator,LA)The scheduling model with the minimum compensation cost as the target.The simulation of the example is to separately participate in the electricity market by LA and the coordination between the LA and the peak-to-valley electricity price to participate in the two scenarios in the current electricity market.The results show that LA and the peak Mutual coordination of peak-to-valley electricity prices can cut peaks and fill valleys and reduce the cost of abandoned winds and the total cost of operating the power system.In this paper,the problem that the bus load difference between the bus and the load is increasing,the load rate is getting lower and the adjustable load can not effectively participate in the dispatching system,and the power system bus load situation prediction and economic dispatch strategy are studied.The simulation results show that the improved k-means clustering algorithm and the constructed scheduling model can effectively improve the prediction accuracy of the bus load situation,reduce the operating cost of the power system,and provide reference for the scheduling system decision.