頁籤選單縮合
題名 | An On-line Fault Estimation in Distribution Substations Using Improved Cause-effect Networks=應用改良式因果網路於配電變電所之線上故障估測法 |
---|---|
作者 | 陳文輝; 劉志文; 蔡孟伸; Chen, Wen-hui; Liu, Chih-wen; Tsai, Men-shen; |
期刊 | Journal of the Chinese Institute of Electrical Engineering |
出版日期 | 20010800 |
卷期 | 8:3 2001.08[民90.08] |
頁次 | 頁267-277 |
分類號 | 448.132 |
語文 | eng |
關鍵詞 | 變電所故障估測; 因果網路; 能源管理/遙測監控系統; Substation fault estimation; Cause-effect network; CEN; EMS/SCADA system; |
中文摘要 | 本文提出一種適用於配電變電所故障診斷的新方法。它主要是利用改良式因果網路 (Improved cause-effect network),作為故障區段與保護設備間關係之表示,以及故障位置的推論與判斷。在過去的研究中,多數的學者,大都以規則庫為基礎 (Rule-based) 的專家系統進行推論,以求得故障發生的位置,這種方法的缺點在於須要建立一個龐大的知識庫,而且在推論時是相當耗費時間,無法滿足在即時環境下的要求。在本文中,我們使用改良式因果網路作為知識的表達及故障位置的推論。這種以因果網路為基礎的推論方式,可以得到快速的推論結果,大大縮短故障診斷的時間。此外,在實際應用的考量下,為了與目前自動化電力調度系統結合,我們也對原型因果網路提出一些修正,使原型因果網路可適用於配電系統中,並結合於實際電力公司的自動化遙測監控 (SCADA) 系統。 為驗証本文所提方法的實用性,我們以臺電公司二次變電所為例進行測試,結果顯示即使在發生多重故障且保護設備誤動作的情形下,均可得到正確的結果。值得一提的是,我們的方法,並非針對特定對象,不僅易於實現,也可以很容易地移植應用於其他型態的變電所。 |
英文摘要 | This paper present a novel approach to fault section estimation in distribution substations using improved Cause-Effect Neworks (CENs). Most traditional fault estimation methods are based on the rule-base expert system, which is usually not applicable in a real-time environment due to the conventional knowledge representation and inference mechanism. In this paper, a CEN is adopted to represent the functions of protective relays and circuit breakers to reduce the fault inference processing time. In addition, for practical considerations, we proposed some improvements to the original CEN to overcome its inherent insufficiencies for integrating with existing poer SCADA systems. The proposed approach has been verified by testing on a typical Taiwan Power Company's (Taipower) secondary substation. The experimental results demonstrate the effectiveness of the proposed method even for fault domains involving multiple faults and failure of protective devices. Moreover, it is easy to implement and transplant inot different substations. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。