Font Size: a A A

Research On Prediction Method Of Output Power Of Photovoltaic Power Generation System

Posted on:2018-01-09Degree:MasterType:Thesis
Country:ChinaCandidate:P ChengFull Text:PDF
GTID:2392330599962410Subject:Engineering
Abstract/Summary:PDF Full Text Request
Since twenty-first Century,with the large consumption of coal,oil and other fossil fuels,energy problems have become increasingly prominent,the environmental problems have become more severe,and the use of renewable,non-polluting new energy for power generation is imperative.Because of its simple structure,no pollution,simple maintenance,easy transportation and easy installation,photovoltaic(PV)power generation technology plays an important role in various new energy power generation technologies.However,as photovoltaic power generation itself is restricted by various environmental meteorological factors,its output power is uncertain,which has a certain adverse impact on the power grid.Therefore,the prediction of the output power of PV power generation system has become an important research subject in the field of power generation.Firstly,the factors affecting the output power of the PV power generation system is analyzed,specifically,the effects of solar intensity,atmospheric temperature,relative humidity,wind speed,weather type and season type on the output power are studied.Through qualitative comparison of linear relationship between the output curve of power curve and the influencing factors,and quantitatively using the path analysis method to calculate the correlation degree of influence factors,the input variables needed for the prediction model are determined.After that,the support vector machine(SVM)theory and the(IFA)are studied,and the two methods are combined together effectively,which are the theoretical basis for the prediction of the output power.On this basis,the prediction model is established.By using the prediction model,the output power of total 14 days in the forecasting area for two weeks in February and May are forecasted respectively.Compared with the traditional particle swarm optimization algorithm method,the prediction effect of the prediction model is verified.The average absolute percentage errors of the output power prediction results of the PV power generation system meet the actual forecasting requirements.In addition,the interval prediction of the output power of PV power generation system is studied.The range of the output power in December 31,2015 is predicted.Then,the feasibility of the proposed interval prediction method is verified by comparing with the interval range of the actual output power value.To sum up,the results of this research have certain guiding function and practicalvalue for the prediction of the output power of PV power generation system.
Keywords/Search Tags:Photovoltaic power generation system, The output power prediction, Support vector machines, Improved firefly algorithm, Interval forecasting
PDF Full Text Request
Related items