基于风速误差校正和ALO-LSSVM的风电功率预测

风资源的随机波动性引起的相位滞后性问题,导致风电功率预测精度不高,尤其是风速变化较快时,滞后性引起的预测误差较大。考虑到风速波动与风功率变化密切相关,提出一种非参数核密度估计和数值天气预报(NWP)相结合的方法,并对预测风速误差进行校正,改善了预测风速的相位滞后性;然后将校正后的风速和风功率作为输入数据进行风电功率预测;采用蚁狮算法(ALO)优化最小二乘支持向量机(LSSVM)参数,从而建立基于风速误差校正和ALO-LSSVM组合的风电功率预测模型。算例结果表明,所提方法风功率预测精度更高。

WIND POWER FORECASTING BASED ON WIND SPEED ERROR CORRETION AND ALO-LSSVM

The problem of phase lag caused by the random fluctuation of wind resources lead to low wind power forecasting accuracy,especially when the wind speed changes rapidly,and the prediction error caused by hysteresis is large.Considering that wind speed fluctuation is closely related to wind power variation,a method combining non-parametric kernel density estimation and numerical weather prediction(NWP)is proposed,and the predicted wind speed error is corrected to improve the phase lag of predicted wind speed.The corrected wind speed and wind power are then used as input data for wind power forecasting.Using an ant lion optimizer(ALO)is proposed to optimize the parameters of least squares support vector machine(LSSVM).Thus,a combined super-short-term wind power forecasting model based on wind speed error correction and ALO-LSSVM is established.The results of the example show that the proposed method wind power forecasting accuracy is higher.