改进混合半云模型在不规则风速概率分布拟合中的应用

发布时间：2018-07-16 22:01

【摘要】：针对基于概率密度峰值法的混合半云模型存在的模型精度易受扰动风速影响的问题,提出了一种基于范数理论的混合半云模型峰值求解方法。首先,由风速历史数据获得风速概率密度统计离散点;然后,采用范数理论选取风速概率密度离散点的拟合多项式;最后,求取多项式的峰值点,并将其对应的风速值作为风速概率密度区域划分的边界,从而建立混合半云模型。结果表明:与基于概率密度峰值法的混合半云模型相比,所提方法可以有效地避免扰动风速对模型概率峰值选取的影响,模型拟合度保持在99%以上。所提方法提高了混合半云模型的鲁棒性和拟合精度,有效降低了"峰值偏离"对半云模型区域划分的影响。
[Abstract]:In order to solve the problem that the model precision of mixed semi-cloud model based on probability density peak value method is easily affected by disturbed wind speed, a new method for peak value solution of mixed semi-cloud model based on norm theory is proposed. Firstly, the statistical discrete points of wind speed probability density are obtained from the historical wind speed data; then, the fitting polynomial of the wind speed probability density discrete point is selected by norm theory; finally, the peak point of the polynomial is obtained. The corresponding wind speed is taken as the boundary of the wind speed probability density region, and the mixed semi-cloud model is established. The results show that compared with the mixed semi-cloud model based on the peak probability density method, the proposed method can effectively avoid the influence of the disturbance wind speed on the selection of the probability peak value of the model, and the fitting degree of the model remains above 99%. The proposed method improves the robustness and fitting accuracy of the mixed semi-cloud model and effectively reduces the effect of "peak deviation" on the regional division of the semi-cloud model.
【作者单位】： 广西大学广西电力系统最优化与节能技术重点实验室;
【分类号】：TK81;TM614
，

本文编号：2127802