基于小波-原子分解的超短期.docVIP

基于小波-原子分解的超短期 风电出力预测模型 曲正伟 张坤 （燕山大学电力电子节能与传动控制河北省重点实验室　秦皇岛　066004） 摘 要：为提高含风电场电网经济调度能力以及降低电力系统规划决策的保守性，本文提出了基于小波-原子分解（WD-AD）的风电出力超短期预测模型。该模型采用小波分解（WD）作为前置环节，以基于原子表达式的自预测和基于最小二乘支持向量机（LSSVM）的残余分量预测为基础构建原子分解（AD）预测模型，分别对风电出力的高低频分量进行预测，并将结果相加得到最终预测值。AD分解过程由衰减线性和Gabor原子库交替分解完成，可自适应匹配不同类型分量。同时，本文提出将细菌群体趋药性和正交匹配追踪算法相结合（BCC-OMP）优化的原子分解法，进一步增强了原子分解能力。实际风电场算例验证了所提方法的自适应性、快速性及有效性。 关键词：风电出力预测； Short-term wind power forecast model based on wavelet decomposition and atomic decomposition Qu Zhengwei, Zhang Kun, Wang Yunjing, Wang Yakun, Cui Zhiqiang （Key Lab of Power Electronics for Energy Conservation and Motor Drive of Hebei ProvinceYanshan University Qinhuangdao 066004 China）Abstract: A short-term wind power output forecast model based on wavelet decomposition and atomic decomposition (WD-AD) is proposed in this paper. This approach is aimed to improve the economic operation ability of wind power system and reduce the conservatism of power system planning and decision making. Wavelet decomposition (WD) is used as the prepositive step. The self prediction and the residual component prediction are based on atomic expression and the least squares support vector machine (LSSVM) respectively. And on the basis of them, atomic decomposition prediction model are established. Then the high and low frequency components of wind power output are predicted. At last the two results are added together to get the final predicted value. The atomic decomposition process is composed of damped liner dictionary and Gabor dictionary, thus different types of components can be matched adaptively. In addition, bacterial colony chemotaxis (BCC) and orthogonal matching pursuit (OMP) algorithm are combined as BCC-OMP optimization algorithm to optimize atomic sparse decomposition, and the decomposition ability can be enhanced further. The proposed approach is verified by the actual wind power farm example and simulation results show the adaptivity, rapidity and effectiveness of this method. Keywords: