一种改进的风电功率预测误差分布模型

受限于现有系统的预测精度,风电功率预测值与实际值之间存在一定的误差,该预测误差的分布特性是风电功率预测精度评估和大规模风电并网电力系统优化调度的重要参考依据。已有的方法大多采用正态分布模型,就精确度而言,会在一定程度上偏离预测误差的实际概率分布。在正态分布模型的基础上,根据概率密度函数以及最小二乘法的相关理论提出了一种描述预测误差分布模型的新方法。对EirGrid风电功率预测数据的计算分析结果验证了该方法的有效性及相对于已有正态分布模型的更精确性。

An Improved Model for Wind Power Forecast Error Distribution

Due to the limited accuracy of present wind power forecasting system,error may occur between the wind power forecast value and the actual output.And the distribution characteristic of the forecast error provides important reference to the forecast accuracy evaluation and the optimal dispatching of power system with large-scale wind power integration.As for the currently-used normal distribution model,its accuracy,to some extent,deviates from the actual probability distribution of forecast error.Based on this normal distribution model,this paper proposes a new method of modeling wind power forecast error based on the theories of probability density function and least square method.The calculation and analysis results based on EirGrid wind power forecast data verify the validity of this proposed method,and compared with the normal distribution model,this method is more accurate.