Validation and Evaluation of the China Meteorological Administration Wind Energy and Solar Energy Forecasting System(CMA-WSP)in Short-Term Wind Resource Forecasting
[Introduction]To test the reliability of the CMA-WSP wind speed product in short-term wind resource forecasting,the CMA-WSP 3 d wind speed forecasting product with a wind speed of 100 m is tested and analyzed.[Method]This research was based on the measured data of 100 m wind speed in three wind farms in Zaoyang Zhoulou,Macheng Caijiazhai and Central Dajin.[Result]The results are as follows:(1)CMA-WSP has a good overall forecasting performance on the wind speed in Zaoyang wind farm within three days.The forecasted results are consistent with the measured wind speed change trend.The correlation between the forecasted and the measured wind speed on the first day are 0.728,0.74 and 0.86 for 15 min intervals,hourly averages,and daily averages,respectively.(2)The relative error between the CMA-WSP forecast and the measured wind speed shows a strong regularity.The relative errors for the forecasted wind speed on the first day are 68％,70％and 92％for 15 min intervals,hourly averages,and daily averages,respectively.The forecasted wind speed is higher than the measured wind speed.The hourly average wind speed and relative error are characterized by low wind speed during the day and high wind speed at night.The change of monthly average wind speed is opposite to the change of MRE value,and it is the lowest from January to June and from October to December,and the highest from July to September.(3)When considering regional differences,CMA-WSP has the best forecasting effect on the wind speed in Zaoyang wind farm.The correlation between the forecasted and the measured wind speed on the first day can reach 0.728,and the correlation on the second to third days is also more than 0.6.The correlation between the forecasted wind speed and the measured wind speed in Caijiazhai and Central Dajin by CMA-WSP is less than 0.6.[Conclusion]CMA-WSP forecasting performance is favorable as a whole,and the relative error has strong regularity.It is beneficial to revise the product and reduce the error level in the next step.
万方数据
维普
