基于灰色系统理论的风速—风功率预测研究

发布时间：2018-06-16 05:27

本文选题：风速预测 + 风电功率预测　；参考：《华北电力大学》2017年硕士论文

【摘要】：由于风能的非平稳性和波动性,风电的大规模并网将会给电网带来非常大的冲击。对于风速与风电功率的精准预测不仅有利于电网的调度,还可以提高风电功率的利用效率以及保证电网的稳定运行。本文采用了基于优化的灰色GM(1,1)模型与应用BP神经网络修正的灰色残差模型分别对风电场的风速与风电功率进行预测。首先,由于传统灰色GM(1,1)模型在对非线性数据进行预测时误差较大,本文采取了数值逼近算法对灰色GM(1,1)模型进行优化。选取风电场实际的风速数据,应用优化的灰色GM(1,1)模型对风速预测时,预测误差明显减小,相对误差降低了35%。在预测风速的基础上,对风电场周边的温度及风向也分别进行了仿真预测,并对结果进行了误差分析,更加验证了此预测方法的可行性与稳定性。其次,由于BP神经网络对非线性的预测有很好的效果,因此结合GM(1,1)模型与BP神经网络模型各自的优点,提出了利用BP神经网络修正的灰色残差预测模型。分别采用灰色GM(1,1)模型、BP神经网络模型以及构建的BP神经网络修正的灰色残差模型对风电场风电功率进行短期预测。再次,对预测结果进行误差分析后得出,修正后模型的预测精度和稳定性都比单一模型有了较大的提高,证明了建立的模型在实际应用中的有效性,可以为电力调度部门提供参考依据。最后,作为今后工作的研究方向,本文对风电场风速分布特性进行了研究,通过研究风速的年变化与日变化规律,得出风电场风速分布具有明显的季节性的特点。
[Abstract]:Due to the nonstationary and fluctuating of wind power, large-scale grid connection of wind power will bring great impact to the power grid. The accurate prediction of wind speed and wind power is not only conducive to the dispatch of power grid, but also can improve the utilization efficiency of wind power and ensure the stable operation of power grid. In this paper, the wind speed and wind power of wind farm are forecasted by using the grey GM1 / 1) model based on optimization and the grey residual model modified by BP neural network, respectively. Firstly, because the traditional grey GM1 / 1) model has a large error in forecasting nonlinear data, the grey GM1 / 1) model is optimized by the numerical approximation algorithm in this paper. The actual wind speed data of wind farm are selected and the prediction error is obviously reduced and the relative error is reduced by 35% when the optimized grey GM1 / 1) model is applied to predict the wind speed. Based on the prediction of wind speed, the temperature and wind direction around the wind farm are simulated, and the error analysis of the results is carried out, which verifies the feasibility and stability of the prediction method. Secondly, because BP neural network has good effect on nonlinear prediction, combining the advantages of GM-1) model and BP neural network model, a grey residual prediction model modified by BP neural network is put forward. The grey GM1 / 1) model and the BP neural network model and the modified grey residual model are used to predict the wind power of wind farm in the short term. Thirdly, after error analysis of the prediction results, it is concluded that the prediction accuracy and stability of the modified model are much higher than that of the single model, which proves the validity of the established model in practical application. It can provide reference for power dispatching department. Finally, as the research direction in the future, the wind speed distribution characteristics of wind farm are studied in this paper. By studying the annual and diurnal variation of wind speed, it is concluded that the wind speed distribution of wind farm has obvious seasonal characteristics.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM614;N941.5

【参考文献】