PHOTOVOLTAIC CONVERGENCE TREND QUANTIFICATION METHOD BASED ON HIERARCHICAL CLUSTERING AND SCENARIOS
The continuous power curve can reflect the law of long-term fluctuation characteristics.By studying the known continuous power curve of photovoltaic clusters,a prediction model is established to reveal the convergence evolution law of clusters of different scales,and finally,the continuous power curve of the photovoltaic cluster to be built is obtained.Firstly,the hierarchical clustering algorithm is used to determine the hierarchical order of the aggregation scale of photovoltaic clusters,and the photovoltaic clusters with the installed capacity increasing layer by layer are obtained,and propose aggregation effect indicators to verify the effectiveness of the sequence.Secondly,in order to better predict the change trend of the photovoltaic continuous power curve,and divide the output scene of the continuous power curve.Finally,in order to avoid the prediction deviation of a single model,in each output scene,the improved information entropy combination prediction model is used to grasp the scale evolution law in the aggregation process and complete the prediction of the continuous power curve of the cluster to be built.The simulation results using the measured data in a certain area in Hebei show that the cluster hierarchical order obtained by verifying the clustering method can better reflect the convergence effect and effectively improve the prediction accuracy;the output scene division accurately describes the convergence trend of the continuous power curve of the cluster;and the improved information entropy combination prediction model can more accurately complete the quantitative analysis of the continuous power characteristics of the photovoltaic cluster to be built.
万方数据
维普
