基本信息:
- 专利标题： 一种基于Siamese网络模型确定电力系统的临界切除时间的方法及系统
- 专利标题（英）：A method and system for determining a critical cut-off time of a power system based on a Siamese network model
- 申请号：CN201811074883.5 申请日：2018-09-14
- 公开(公告)号：CN109376990A 公开(公告)日：2019-02-22
- 发明人： 史东宇 , 陈柏任 , 严剑峰 , 周二专 , 陈颖 , 陈鎏凯 , 冯东豪 , 凡航 , 关慧哲 , 于之虹 , 吕颖 , 鲁广明 , 侯金秀 , 李刚
- 申请人： 中国电力科学研究院有限公司 , 北京科东电力控制系统有限责任公司 , 清华大学 , 国家电网有限公司 , 国网山东省电力公司电力科学研究院
- 申请人地址： 北京市海淀区清河小营东路15号
- 专利权人： 中国电力科学研究院有限公司,北京科东电力控制系统有限责任公司,清华大学,国家电网有限公司,国网山东省电力公司电力科学研究院
- 当前专利权人： 中国电力科学研究院有限公司,北京科东电力控制系统有限责任公司,清华大学,国家电网有限公司,国网山东省电力公司电力科学研究院
- 当前专利权人地址： 北京市海淀区清河小营东路15号
- 代理机构： 北京工信联合知识产权代理有限公司
- 代理人： 胡秋立
- 主分类号： G06Q10/06
- IPC分类号： G06Q10/06 ; G06Q50/06 ; H02J3/00
The invention relates to a method and a system for determining a critical cut-off time of a power system based on a Siamese network model. The method comprise the following steps: determining a twin network model by using samples obtained from a historical sample data set including power flow data and a critical cut-off time result; When the new power flow data is received, the twin network modelis used to obtain the high-level features corresponding to the current power flow data; Calculating distances between high-level features corresponding to each sample in the historical sample data setand high-level features corresponding to the current tide data, and determining the closest sample; The average value of the critical cut-off time closest to the sample is calculated as the criticalcut-off time corresponding to the predicted current tidal current data. According to the characteristics of the transient stability and the on-line data, the invention determines a twin network modelwhich is more suitable for the power system, increases the adaptability of the method, is faster than the simulation, is suitable for on-line analysis, and can meet the speed and precision requirements of the on-line analysis system, and is particularly suitable for small sample sets.
公开/授权文献:
- CN109376990B 一种基于Siamese网络模型确定电力系统的临界切除时间的方法及系统 公开/授权日：2022-03-04