The Statistical Characteristics Of Measured Date In The Load Modeling

The accuracy of load model has great effects on the validity of power system simulation.However,it is widely known that the load modeling which can reflect the actual characteristics of the load is still an unresolved problem due to its special properties such as time-variant characteristics of the loads.In this paper,based on composite load model,cluster analysis are used to classify the response space under different voltage.The use of statistical SPRSQ,RSQ,PSF,PST2 four targets in the cluster in the process of changing trends to determine the best number of classification and classified,the results of the classification under different voltage is very similar,to a certain extent,the model shows a very good generalization ability that can better describe the characteristics of the load.find the same data in the different groups and classified,then identify the hidden rules and made a method of measured data classification.Such classification also explain.The other measured data proves that the classification is feasible.