基于拟蒙特卡罗模拟和核密度估计的概率静态电压稳定计算方法  

Probabilistic Static Voltage Stability Calculation Based on Quasi-Monte Carlo and Kernel Density Estimation

作　　者：赵建伟[1] 李禹鹏[1] 杨增辉[2] 严正[1] 徐潇源 冯楠[2] 崔勇[2] 

机构地区：[1]电力传输与功率变换控制教育部重点实验室(上海交通大学),上海市闵行区200240 [2]国网上海市电力公司电力科学研究院,上海市虹口区200437

出　　处：《电网技术》2016年第12期3833-3839,共7页Power System Technology

基　　金：国家电网公司科技项目(52094015001S)~~

摘　　要：概率静态电压稳定计算能考虑电力系统中的随机因素,获得稳定裕度的统计数字和概率分布,是传统确定性的电压稳定计算方法的有效补充。文章研究了基于模拟法的概率静态电压稳定评估方法,通过引入拟蒙特卡罗模拟获得输入随机变量样本,以提高模拟法的计算效率,并采用基于扩散方程的核密度方法以准确获得稳定裕度的概率分布函数。将所提算法应用于IEEE 118节点系统和某省级电网的静态电压稳定分析。计算结果表明:相比于常用的模拟法,所提算法仅需要较少的采样规模即可获得较高的计算精度;随着外部直流馈入功率的增加和内部发电机开机的减少,该省网在低谷时段的电压稳定裕度低于峰荷时段的稳定裕度,即系统轻载时更加逼近稳定运行临界点。In probabilistic static voltage stability analysis, uncertainties in power systems are considered and statistical indices and probability distributions of stability margins can be obtained. Therefore, probabilistic method is an effective complement of traditional deterministic methods. In this paper, probabilistic static voltage stability evaluation based on Monte Carlo simulation（MCS） is studied. In order to improve MCS efficiency, Quasi-Monte Carlo is adopted to obtain samples of input random variables. Besides, diffusion-based kernel density method is designed to accurately estimate probability distribution function of stability margins. IEEE 118-bus system and an actual provincial power grid are studied and simulation results indicate that compared with commonly used MCS methods, the proposed method maintains a high degree of accuracy with relatively smaller sample sizes. With increase of power import and reduction of online generators, stability margin of the provincial system in off-peak period is smaller than that in peak period, i.e. system with smaller loads is closer to critical stable operating point.

关 键 词：电力系统 概率评估 静态电压稳定 拟蒙特卡罗模拟 核密度估计 

分 类 号：TM721[电气工程—电力系统及自动化]