變壓器油中氣體分析系統之設計及資料探勘技術於油中氣體分析之應用

臺灣電力公司於民國99年「變電設備維護管理系統」，各供電區營運處轄下相關變電設備於該系統內建立「設備基本資料」、「點檢試驗資料」及「維護歷史資料」等，該系統將管理變電設備試驗及點檢記錄，如可燃性氣體總量(Total Combustible Gases, TCG)、SF6及糠醛…等試驗進行建檔。為進一步提升該系統之效能，以落實變電設備狀態基準維護(Condition-based Maintenance, CBM)之目標，本論文參考IEC、IEC增量、IEEE及日本電氣協會(ETRAJ)之標準，開發網路版變壓器油中氣體狀態分析(Dissolved Gas Analysis, DGA)系統，將各項標準建立法則，以人機介面設計出一套變壓器油中氣體狀態分析系統，利於臺電資料分析各變壓器之狀態，提升變壓器狀態判定之準確性、提高供電品質、提升供電效率並建立變壓器完整狀態資訊。本論文另一重點，利用資料探勘決策樹之技術於異常故障資料進行分類，根據資料探勘技術於DGA故障案例，目的在於找出故障案例中，各氣體占總TCG之比例關係，除找出各氣體比值與故障狀態之隱藏關聯外，本論文將依故障案例設計決策樹，並建立屬於變壓器狀態之決策樹，以提升臺灣電力公司對於變壓器事故之預防，進而節省人力、物力、時間及資產等各項資源。

關鍵字

油中氣體分析；變壓器；預防性維護；資料探勘；決策樹

並列摘要

Taiwan Power Company (TPC) has designed and installed a system called Substation Facility Maintenance and Management System (SFMMS). The fundamental data, inspection results and maintenance historical records for all of TPC’s primary substations and distribution substations have been built up and stored in FFMMS. Various data inquiry and management can be easily conducted through SFMMS. Moreover, annual work scheduling can be generated with SFMMS. The facility testing records such as total combustible gases (TCG), SF6 and dissolved gas analysis can be saved and managed in the SSFMS as well. Dissolved gas analysis (DGA) is widely used to detect incipient faults in transformers. It is shown how the accuracy of DGA research results can affect the reliability of DGA diagnosis. The minimum gas levels in service above which diagnoses may be attempted are indicated, as well as the gas levels observed before fault. In order to further improve the effect and efficiency of SSFMS so that the goal of facility condition-based maintenance can be attained, a software system for dissolved gas analyses will be developed in SSFMS. The international standards such as IEC, IEC increments, IEEE and Japanese Electric Association standard will be utilized in the software system to be designed for transformer dissolved gas analyses aiming to improve the effect and efficiency of transformer preventive maintenance. In this paper, we designs data mining decision tree (ID3/C4.5) with the main gases include H2, CH4, C2H2, C2H4 and C2H6 called total combustible gas (TCG). It can be observed a power transformer’s condition. In the further, prevent an unexpected which is unable to avoid.

並列關鍵字

Dissolved Gas Analysis ； Power Transformer ； Preventive Maintenance ； Data Mining ； Decision tree

參考文獻

[69] 林桐斌，變電設備維護管理系統之建置與變壓器油中氣體模糊分析法則之研究，碩士論文，國立臺北科技大學，2010。

[1] T. J. Haupert and F. Jakob, Review of the Operating Principles and Practice of Dissolved Gas Analysis, Philadelphia:Z ASTM, 1988.

[2] A. M. Emsley and G. C. Stevens, “Review of chemical indicators of degradation of cellulosic electrical paper insulation in oil-filled transformers,” IEE Proc. Sci. Measur. Techn., vol. 141, pp. 324-334, 1994.

[6] M. Duval and A. D. Pablo, “Interpretation of gas-in-oil analysis using new IEC publication 60599 and IEC TC 10 databases,”IEEE Electr. Insul. Mag., vol. 17, no. 2, pp. 31-41, Mar./Apr. 2001.

[7] A. Mollman and B. Pahlavanpour, “New guidelines for interpretation of dissolved gas analysis in oil-filled transformers,” Electra, CIGRE France, vol. 186, pp. 30-51, Oct. 1999.

國際替代計量

變壓器油中氣體分析系統之設計及資料探勘技術於油中氣體分析之應用

全文下載

主題瀏覽