福建省沿海风电场短时预报风速的订正  

Revision of Short-time Wind Speed Forecast in Fujian Coastal Wind Power Plant

作　　者：林智琛 陈晓梦 李白良 谢嫔 Lin Zhichen;Chen Xiaomeng;Li Bailiang;Xie Pin(Fujian Meteorological Service Center,Fuzhou 350001,China)

机构地区：[1]福建省气象服务中心,福州350001

出　　处：《气象与环境科学》2023年第1期23-31,共9页Meteorological and Environmental Sciences

基　　金：国家重点研究发展计划项目(2018YFC1506905)。

摘　　要：为了进一步提高福建沿海风电场短时风速预报的准确性,以福建沿海某测风塔2017年9月、10月和2018年1月、4月的数据为例,通过对预报因子的筛选,设定3种不同的卡尔曼滤波量测方程方案,对ECMWF模式的预报风速进行订正,并通过误差分析来构建福建沿海风电场短时(012 h)预报风速的订正模型。结果表明,方案2(以地面风速和海平面气压作为预报因子)适用于风速变化较小的秋冬季节,方案1(以地面风速作为预报因子)和方案3(以地面风速、海平面气压、地面温度、地面湿度作为预报因子)对风速突变较频繁的春夏季节的预报风速订正更有效果。经过订正模型订正后,各代表月在未来012 h预报风速的平均绝对误差和均方根误差都有显著的减少。从订正效果来看,卡尔曼滤波法存有一定局限性,风速变化越平缓其订正效果越好,同时在风速突变较大时易造成较大的误差,产生反相位订正结果。To improve the accuracy of short-term wind speed prediction in Fujian coastal wind power plant,four sets of wind data collected by one wind measuring tower in September 2017,October 2017,January 2018 and April 2018 are used in this paper to revise the ECMWF wind speed forecasts through selecting predictors and setting up three different Kalman-filter schemes.Besides,the error analysis is used to construct the correction model for the short-term(012 h)wind speed forecast in Fujian coastal wind power plant.The result shows that Scheme 2(taking the surface wind speed and sea-level pressure as predictors)is suitable for autumn and winter when wind speed changes little,while Scheme 1(with surface wind speed as predictor)and Scheme 3(which has sea-level pressure,surface wind speed,temperature and humidity to be predictors)are more effective in correcting wind speed forecast in spring and summer when wind speed changes more dramatically.After the model’s revision,the mean absolute error and root mean square error of the forecast wind speed in the next 012 h of each representative month are significantly reduced.Judging from the correction effect,the Kalman-filter method has some limitations.When the wind speed changes more gently,the correction effect is more accurate.At the same time,when the wind speed suddenly changes,Kalman-filter is prone to generate large errors,resulting in out-of-phase correction results.

关 键 词：卡尔曼滤波 风速订正模型 误差分析 

分 类 号：P456[天文地球—大气科学及气象学]