基于数据挖掘及聚类分析的主动配电网简化等值

Simplified equivalence of active distribution network based on data mining and clustering analysis

针对可再生能源发电大量并入配电网,使配电网由被动受电转变为主动发电和受电多重功能,由此使主动配电网的分析和调度日益困难。对此,在配电网量测环境下,该文提出基于数据挖掘及聚类分析的主动配电网简化等值方法。首先,根据主动配电网中单一的风、光可再生能源发电具有不可控、波动性强的特点,提出电源、负荷的等值简化模型;其次,在此基础上,进一步给出配电网发电、负荷及其系统的等值简化模型有功、无功表达模型;最后,基于数据挖掘中的多元线性回归方法,给出主动配电网简化等值模型及其优化求解方法。研究通过对改进的IEEE9节点系统进行验证,验证了所提方法的有效性。

A large number of renewable energy power generation is continuously integrated into the distribution network, which makes the distribution network transfer from passive power to active power generation and multi-function, which makes the analysis and scheduling of active distribution network increasingly difficult. In this paper, a simplified equivalent method of active distribution network based on data mining and cluster analysis is proposed in the distribution network measurement environment. Firstly,according to the characteristics of uncontrollable and strong fluctuation of single wind and light renewable energy generation in active distribution network, the equivalent simplified model of power supply and load is proposed. Secondly, on this basis, the equivalent simplified model of power generation, load and its system in distribution network is further given, and the expression model of active and reactive power is given. Based on the multiple linear regression method in data mining, a simplified equivalent model of active distribution network and its optimal solution method are given. Finally, the improved IEEE9 bus system is verified to show the effectiveness of the proposed method.