风电场风速及风电功率预测方法研究综述

Overview on Prediction Methods of Wind Speed and Wind Power

由于风能的随机性以及电力系统的非线性等原因,预测风电功率时需要考虑众多的不确定因素影响。现有预测方法主要包括物理预测方法、统计预测方法以及学习预测方法、综合预测法等。基于数字天气预报(NWP-numerical weather prediction)的物理预测方法模型复杂、计算量大,较少用于短期预测;统计预测方法模型简单,数据需求量少,较适合于数据获取有一定困难的情况;人工智能预测方法不依赖于对象的精确模型,适合于随机非线性系统;综合预测方法可一定程度地扬长避短。本文主要就风电场风速及风电功率预测方法研究进行了综合阐述,并在总结前人研究的基础上提出了一些可进一步研究的问题。

Due to the intermittency of wind energy and the non-linearity of power system, there exist many uncertain variables which should be considered in the wind power prediction. The current prediction methods include the physical method, statistical method, learning method and the comprehensive one combining all the other methods. Based on accurate numerical weather prediction （NWP）, the physical method is seldom used in the short term prediction, as its model is complicated and deals with large quantities of calculations. The model of the statistical method is simple and requires a small amount of data. It can be applied in those situations where data acquisition is difficult. The AI method is suitable in the random or non -- linear system as it does not rely on the accurate mode of the objective. The comprehensive method maximizes favorable factors and minimizes unfavorable ones as contained in above-mentioned methods. This paper presents a brief overview on prediction methods of wind speed and wind power, and raises further issues worth further research on the basis of summarizing the previous studies.