Research On Load Profile Classification And Load Model Aggregation Methods For Power System

The simulation of power system is an important tool for the stability analysis,designation and operation of power system.The credibility of simulation results and the power system operation based on this are directly affected by the accuracy of each component model.Especially,the load model is still relatively simple and difficult to accurately reflect the complex process of load changes.Therefore,establishing realistic load model has very important practical significance.Under load nodes there is a huge complex types and different parameters load,however in the process of power system operation,the model established by the regional power grid should be as simple as possible and the model number should not be too large.At the same time,affected by operational complexity,there is a contradiction between the refined load model and engineering application and it is impossible to build a detailed model for each load node.To this end,this article attempts to solve this problem from two aspects,one is the load profile classification,and the other is the load model aggregation.The purpose of load profile classification is to classify the load nodes with similar characteristics into one category according to the load data and the nature of the load.So that the loads belonging to the same category can approximate use the same load model.Simultaneously,directly using all load data to classify faces the challenge of computational efficiency.In this regard,this paper presents a novel dimension-reduced clustering method for load profiles based on singular value decomposition.Combined with the actual power grid load data,the results show that this method not only improves the validity of clustering,but also improves the efficiency of clustering.The purpose of load model aggregation is to aggregate a large number of induction motors into a small number of aggregated induction motors and reduce the simulation complexity.Therefore,this paper presents a method that for induction motor group aggregation considering static critical stable characteristics.The method takes into consideration the physical characteristics and static critical stable characteristics of each motor,and then aggregates the motor group.Taking the WSCC-9 node system as an example,10 different characteristics motors are aggregated.The results show that under different operating conditions,the equivalent motor obtained by the method can reflect the dynamic process of the original motor group more accurately both system steady state and critical instability.