短期风电功率预测误差的混合偏态分布模型

Mixed Skew Distribution Model of Short-term Wind Power Prediction Error

风电功率预测目前仍存在较大误差,决策需结合其不确定性信息。分析实际运行系统中短期风电预测的误差,得到不同规模及出力区间下的误差分布呈有偏、重尾、多峰等特性。为此,提出混合偏态分布模型估计短期风电预测误差分布,并通过概率密度曲线非线性拟合进行参数估计。通过一个实际运行的区域风电预测系统数据,验证了混合偏态分布模型能很好的描述风电预测误差分布特性。相比较其它分布模型,该模型应用范围更广,拟合精度更高,灵活性更强,能够用于描述不同情况下风电预测误差分布。

The wind power prediction uncertainty information should be referred to make decision due to low prediction precision. The error distributions of different sizes and value intervals present characteristics of biasness, heavy tail and multi-modality through analyzing an actual operated system’s prediction and operation data. Therefore, the mixed skew distribution was proposed to model short-term wind power prediction error distribution for depicting the features, and nonlinear fitting of the statistical probability density curve was used to estimate the model parameters. The case study of a regional wind power prediction system data manifests that wind power prediction error distribution characteristics can be described by the mixed skew distribution model more accurately compared with other distribution models with wider applications, higher fitting precision, and more flexibility in different situations.