BCU Classifiers for On-line Dynamic Contingency of Electric Power Systems: Design, Developments and Applications

本提案提出了一个在线DSA的架构，并试图开发一个基于直接方法的动态应急筛选方案，使在线DSA在当前或不久的将来的电力系统运行环境中可行。特别是，本提案将开发BCU分类器，以满足在线DSA的以下基本要求：(1)（可靠性度量）不稳定偶然事件的绝对捕获；即没有不稳定（单摆或多摆）偶发事件被遗漏。换句话说，捕获的不稳定偶然事件与实际的关键偶然事件之比为1。(2)（效率测度）筛除稳定偶然性的高产率，即检测到的稳定偶然性与实际发生的稳定偶然性的比值尽可能接近于1。(3)（在线计算）为了适应不断变化和不确定的运行条件，几乎不需要离线计算和（或）调整。(4)（速度测量）速度高，即对每个突发情况进行快速分类。(5)（性能度量）对电力系统运行条件变化的鲁棒性。为了满足这五个基本要求，我们建议首先完成以下任务：(1)对数值指标的发展进行一些理论研究；(2)构建改进的数值能量函数；(3)开发更可靠的BCU方法的数值实现；(4)开发用于检测偶然的、无效的数值能量函数的数值指标；(5)开发用于检测偶然控制u.e.p.收敛问题的数值指标。然后，将这些任务的结果整合到BCU分类器的初步版本中，形成BCU分类器的改进版本，该版本将满足进行在线动态偶然性筛选的五个要求。该项目的成功与否直接影响到在当前或不久的将来的电力系统运行环境中执行在线DSA的任务是否可行。

ECS-9812139 Chiang This proposal presents an architecture for on-line DSA and seeks to develop a dynamic contingency screening scheme based on the direct method to make on-line DSA feasible for current or near future power system operating environments. In particular, this proposal will develop BCU classifiers to satisfy the following essential requirements for on-line DSA: (1) (reliability measure) absolute capture of unstable contingencies; i.e. no unstable (single-swing or multi-swing) contingencies are missed. In other words, the ratio of the captured unstable contingencies to the actual critical contingencies is 1. (2) (efficiency measure) high yield of screening out stable contingencies, i.e. the ratio of the stable contingencies detected to the actual stable contingency cases is as close to 1 as possible. (3) (on-line computation) little need of off-line computations and/or adjustments in order to meet with the constantly changing and uncertain operating conditions. (4) (speed measure) high speed, i.e. fast classification for each contingency case. (5) (performance measure) robust performance with respect to changes in power system operating conditions. To meet the five essential requirements, we propose to first perform the following tasks: (1) to conduct some theoretical study for development of numerical indices, (2) to construct improved numerical energy functions, (3) to develop a more reliable numerical implementation of the BCU method, (4) to develop numerical indices for detecting the occasional, invalid numerical energy functions, and (5) to develop numerical indices to detect the occasional controlling u.e.p. convergence problem. Then, the results of these tasks will be integrated into the preliminary version of BCU classifiers to form an improved version of BCU classifiers which will satisfy the five requirements for performing on-line dynamic contingency screening. The success of this proposed project has a direct impact on making the task of performing on-line DSA feasible for current or near-future power system operating environments.