Characteristics And Prediction Of Seasonal Average Wind Speed In Yangtze River Delta Region

The development of wind energy resources plays a crucial role in reducing fossil fuels,lowering carbon dioxide emissions,mitigating anthropogenic global warming,and enhancing drought resilience.Studying wind speed changes is one of the most commonly used methods to evaluate global or regional wind energy density.Based on the carding and analysis of previous researches,this thesis uses daily wind speed data at the 44 weather observing stations in Shanghai,Anhui province,Jiangsu province and Zhejiang province to analyze the distribution and trend of average wind speed in spring,summer,autumn and winter from 1959 to 2016 in the Yangtze River Delta region.Based on the sea surface temperature and geopotential height,climate variables that highly correlated with the average wind speed in different seasons in the Yangtze River Delta region were identified,which were used to build statistical models to predict seasonal average wind speed.By assessing the goodness-of-fit and the prediction performance of the models,the models that can effectively predict the average wind speed in different seasons in the Yangtze River Delta region are selected.The main research work and conclusions are as follows:(1)From 1959 to 2016,The largest annual average wind speed in the Yangtze River Delta region is in Shanghai,followed by Zhejiang,Jiangsu,and Anhui province.In Shanghai,Anhui province,and Jiangsu province,the maximum and minimum average wind speed occur in spring and autumn,respectively.The differences between summer and winter average wind speed are relatively small.In Zhejiang province,the average wind speed has little difference across the four seasons.For temporal trends,the average wind speed of the entire region for all four seasons has significant decreasing trends in the past 58 years,with a larger decreasing trend in winter and spring,and a smaller decreasing trend in summer and autumn.Shanghai has the largest decreasing trend in the region for all four seasons,while Anhui province has the smallest decreasing trend.(2)When looking at the days with the wind speed below 3m/s from 1959 to 2016in the Yangtze River Delta region,decreasing trend was found in the seasonal average of those days in Jiangsu and Zhejiang province for all four seasons,while number of days has a significant increasing trend for all four seasons in the entire region.This indicates that the number of days with low wind speed has increased in the Yangtze River Delta region in recent years.This increasing rate is the largest in Shanghai,while Anhui province is the smallest.From the perspective of wind energy development,due to the increasing occurrence of low wind speed in Shanghai,the use of wind energy resources is becoming more and more difficult.(3)By calculating the correlation between the field of the sea surface temperature/geopotential height and the average wind speed in the four seasons,the regions with significant spatial correlation and temporal consistency were identified.Empirical Orthogonal Functions(EOFs)were used to extract the leading principle components from the identified regions,and the first principal component of those regions are selected as climate covariates for building the prediction model.(4)Given the significant trend in the average wind speed and the selected climate variables,three linear temporal regression models were built to predict wind speed.According to the fitting result,the effects of temporal trend and climate variables on the average seasonal wind speed were analyzed for each season.It was found that,according to both AIC values and Adj.R~2,the models considering climate predictors are superior to those only consider temporal trend.(5)Leave-one-out-cross-validation and mean squared error are applied to compare the prediction performance of all candidate models,according to which the model with the best predictive performance is selected for each season and region.Data are separated into calibration period(1959-1998)and validation period(1999-2016)to evaluate the prediction skill of each selected model.In the end,we find that the selected models can predict the spring,autumn and winter average wind speed in Shanghai with a 3 to 6-month lead time,the spring and summer average wind speed in Anhui province with a 3 to 4-month lead time,the spring,summer,autumn and winter average wind speed in Jiangsu province with a 4 to 6-month lead time,and the autumn average wind speed in Zhejiang province with a 4-month lead time.The research of seasonal wind speed prediction can be effectively used to assess wind energy resources,especially when wind speed increases or decreases,resulting in conflicts between energy supply and demand,and precautionary actions can be taken to improve energy system planning and operation.