Wind Speed Distribution Forecasting Model Using The Exponential Smoothing Technique And Its Application In Reliability Evaluation Of Electrical Distribution Systems

Owing to the increasingly serious problems of environment pollution and energyshortages, wind turbine generators (WTG), as a distributed generation, has beenwidespread concerned. As the structure of power grid has beening improved, and thecustomer demand for electricity service quality has beening increased, the reliability ofthe electrical distribution networks (EDN) is more and more important. Considering therandomness and volatility of wind energy, it’s necessary to analyze the characteristics ofwind speed distribution, wind speed distribution forecasting and the establishment theWTG output model, for the purpose to assess the reliability of EDN containing WTGaccurately.In order to improve the accuracy of wind speed simulation and forecasting,describe the distribution characteristics of wind speeds, and consider the property ofrecent wind speeds having a greater effect on the current wind speed, an exponentialsmoothing model for forecasting wind speed distribution is used in this thesis. Theinitial value and the smoothing coefficient of the proposed model are discussed basedon the properties of exponential smoothing method, and the optimal smoothingcoefficient is formulated as an optimization problem with the objective function ofminimizing the sum of absolute errors between the forecasted and statistic wind speeds.Then, the model can be used to forecast the wind speed distribution of a wind farm.Case studies show that the proposed model has a higher accuracy of forecasting thewind speed distribution compared with the traditional methods.Multi-state output model of WTG is established and the reliability of EDNconsidering WTG is assessed based on wind speed distribution exponential smoothingmodel. Combined with the characteristics of complex distribution network, thereliability of EDN considering WTG is evaluated by the reliability assessment blockalgorithm of complex medium voltage distribution network, the node types of system isclassified and the load reduction strategy considering WTG is established according tothe load type and size. The algorithm is applicated on the RBTS-BUS6test systemreliability assessment, the different assessment results based on different forecastingmodels is comparaed. The results show that the EDN containing WTG improved thesystem reliability. Compared with the WTG outputs using different forecasting methodsfor wind speed distribution, the proposed model, which is usefull and posseses efficiency, provides a more accuracy results.Several sensitivity analysis are conducted to analyze the impacts of parameters ofWTG on EDN containing WTG. The result shows that the change of the model statesnumber of WTG output, the reliability parameters, the cut-in, rated and cut-out windspeed of wind turbine and the installation location have varying degrees influence onthe reliability indices of distribution network.