低碳背景下基于改进通用生成函数法的随机生产模拟  

Power System Probabilistic Production Simulation Based on Improved Universal Generating Function Methods in Low-Carbon Context

作　　者：王洪涛[1] 刘旭[1] 陈之栩[2] 房光华 傅磊[3] 

机构地区：[1]电网智能化调度与控制教育部重点实验室(山东大学),山东省济南市250061 [2]华北电网有限公司电力调度通信中心,北京市西城区100053 [3]山东电力调度中心,山东省济南市250001

出　　处：《电网技术》2013年第3期597-603,共7页Power System Technology

基　　金：国家863高技术基金项目(2011AA05A101);国家自然科学基金项目(51177092)~~

摘　　要：通用生成函数法(universal generating function,UGF)具有计算简单,运算快等优点,但其传统的应用中忽略了时序性,评估结果粗糙。为此,提出基于改进UGF法的随机生产模拟算法。首先,采用主成分分析法(principal componentanalysis,PCA)和分层聚类法相结合的方法获得典型风电出力时序样本,然后采用均摊法建立风电出力UGF模型。在建立UGF模型时,通过选择合适的公因子,将状态值转化为公因子的整数倍,以有效抑制状态数指数增长,减缓了状态数爆炸问题。其次,对聚类后的时序风电出力模式进行UGF建模,得到时序UGF模型,该模型既可以体现风电出力的波动性,也可以体现风电出力的不确定性,克服了传统UGF法因缺乏时序性所致的风电特征代表性差的问题。最后,采用修正的IEEE-RTS79算例验证了所提算法的有效性。To utilize the advantage of traditional universal generating function（UGF） method in fast and simple calculation and remedy the defect of dissatisfied accuracy of evaluation results by this method due to neglecting the influence of time-sequence in the application,based on improved traditional UGF method a probabilistic production simulation algorithm is proposed and applied in probabilistic production simulation of power system connecting with wind farm.Firstly,a method combining principal component analysis（PCA） with hierarchical cluster analysis is adopted to obtain typical time-sequential samples of wind power output;then a UGF model for wind power output is established by apportionment in average.During the establishment of the UGF model,by means of choosing proper common factor the state values are turned into integral multiples of common factor to effectively restrain the exponential growth of state number and alleviate the explosion of state number.Secondly,the UGF modeling of post-clustering time-sequential wind power output mode is performed to obtain time-sequential UGF model of wind power output,which can incarnate both fluctuation and uncertainty of wind power output,thus the insufficient representativeness of wind power features based on traditional UGF method due to neglecting time-sequence characteristic is overcome.Finally,the effectiveness of the proposed algorithm is verified by modified IEEE-RTS 79 system.

关 键 词：聚类技术 通用生成函数法 随机生产模拟 风电 低碳 

分 类 号：TM721[电气工程—电力系统及自动化]