LOAD FLOW ANALYSIS WITH LARGE WIND GENERATOR

Abstract

More and more wind turbines are connected to electrical power system.The detailed modelling of wind farms for load flow will take lots of computation time.Hence in order to lessen the computation .time, aggregated model of wind turbine generator (Large Wind Generator) is needed. Common assumption is that wind speed reaching at all the wind turbine generating units is same resulting in single machine equivalent representation for the entire wind farm but this is not the case in reality due to wake effect.The wind speed reaching each WTGU may differ or may be same resulting in multimachine equivalent representation for the entire wind farm.The difference in wind speed reaching each WTGU is expressed by wake loss model. In a wind farm, it is obvious that two or more than two wind turbines are experiencing the same wake effect depending upon their locations.In order to find such groups a clustering method known as K Means clustering has been used.To find the average wind energy output of a turbine, Weibull wind speed distribution of a site together with the power curve of the turbine has been used. The impact of wind farm has been tested both by balanced as well as unbalanced load flow analysis.For the balanced case, a 69 bus distribution system has been taken and for the unbalanced case, an IEEE 37 bus distribution system has been taken.In order to validate the aggregated model, the load flow analysis has been carried out both with the aggregated model as well as with the detailled modelling of wind farm.The result shows that the aggregated model used is very accurate.Also it is found that the load flow with the aggregated model is several times faster than that carried out with the detailed modelling of the wind farm.