摘要
风电具有很大的随机性、间歇性和不可控性,为了电网可靠、经济运行需要对风机发电功率进行预测,其中的关键环节是风速预测,风速预测的方法分为使用数值气象预报和不使用数值气象预报两种方式。本文基于中尺度气象数值预报模型,简要介绍了其特点,对其预报风速的误差组成进行了分析,并提出用线性回归方法和BP神经网络模型对风速的预测值进行修正的方案,用实际数据进行了分析验证。研究结果表明,提出的修正方法对风速预测精度的提高有所帮助。
With the characteristics of the wind power's randomness, intermittence and uncontrollability, the wind power prediction is greatly significant for the reliability and economic operation of power system. The wind speed forecasting is the most important part during the wind power forecasting course. The method of wind speed prediction is divided into two ways: one is using numerical weather prediction while the other one is not. Based on the WRF mesoscale weather forecast model, the paper introduces its characteristics briefly and analyzes the error of wind speed of the forecasting. And then proposes the linear regression and BP neural network model to correct the prediction value of the wind speed. The effect of each correction model is obtained by validating and analyzing the actual data. The test results show that the proposed method improves the prediction accuracy of wind speed.
出处
《电测与仪表》
北大核心
2013年第3期33-36,59,共5页
Electrical Measurement & Instrumentation
关键词
风速
预测
误差分析
修正
wind speed, prediction, error analysis, correct