摘要
设计了一个应用于大停电事故恢复的基于案例推理系统。系统纵向采用分层结构,抽象层体现恢复过程的一般策略,具体层案例包含恢复操作的具体措施。系统横向分成黑启动模块、网架和负荷恢复功能模块。黑启动模块中的抽象层用一个定性进程理论模型表示,模型考虑了黑启动中的主要限制条件和各类机组的启动特性,发电容量调度算法用于具体层的修正。网架、负荷恢复模块的抽象层和具体层由详细程度不同的系统潮流图表示。开发了串并行送电阶段优化算法、系统网架及负荷恢复优化算法,用于具体层的修正。该系统以基于案例推理作为主框架,把定性分析和定量分析、理论研究和电力部门的实际经验很好地结合起来,多种在线优化算法提高了系统的灵活性。
This case-based reasoning (CBR) system for power system restoration has a hierarchical structure. The cases in abstract level present the general strategies of restoration, and the cases in the detail level contain the operations during restoration. The CBR system has two functional modules: one is the black-start module, another is the system reconstruction and load recovery module. In black-start module, a qualitative reasoning (QR) model is developed by qualitative process theory (QPT). The QR model is used to revise the abstract cases, and can act as an explaining tool for detail black-start cases. In system reconstruction and load recovery module, the abstract cases and detail cases are represented with power flow maps in different details. Various on-line optimal algorithms were developed to revise the detail cases for both function modules. This study uses the CBR as the main frame, which combines qualitative analysis and quantitative analysis, theoretic research and field experience well. This system was used to constitute the restoration plan of a steel plant.
出处
《电力系统自动化》
EI
CSCD
北大核心
2007年第18期87-90,共4页
Automation of Electric Power Systems
基金
国家自然科学基金资助项目(50307006)~~