TY - GEN
T1 - Enhancement of power transformer state of health diagnostics based on fuzzy logic system of DGA
AU - Aburaghiega, Ehnaish
AU - Farrag, Mohamed Emad
AU - Hepburn, Donald
AU - Haggag, Ayman
N1 - Acceptance in SAN
AAM query, author confirmed remove footer, re-uploaded. ET 1/10/18
No ISSN found for output, out of policy scope. ST 13/11/19
PY - 2019/2/7
Y1 - 2019/2/7
N2 - Dissolved Gas Analysis (DGA) of liquid insulation is an effective means for diagnosing power transformers. The gas contents in insulating oil can be gathered on-line and off-line to indicate the health condition of the transformers, thereafter there are many interpretations of the gas contents. In this work, Seven-fuzzy interpretation modules are individually established, tested and lately combined to monitor power transformers’ health. The developed method incorporates trending of the concentration of the dissolved gases over the operating life. The approach processes current and/or historical DGA data, using the 7-developed logic modules, to determine the current state of a transformer, provide information regarding the fault type, fault probability, fault severity and recommended future sampling interval in addition to operating procedure, consistent with industry standards. The developed diagnosis system has been validated using 1290 samples from fresh and previously tested mineral oil filled transformers. The proposed system is proved, based on field data, to be 99% accurate in identifying transformers being in normal or abnormal operation. For the cases where a transformer fault was known, the proposed technique has less than 2% inaccuracy in recognizing the fault’s type in comparison to other approaches discussed in literature.
AB - Dissolved Gas Analysis (DGA) of liquid insulation is an effective means for diagnosing power transformers. The gas contents in insulating oil can be gathered on-line and off-line to indicate the health condition of the transformers, thereafter there are many interpretations of the gas contents. In this work, Seven-fuzzy interpretation modules are individually established, tested and lately combined to monitor power transformers’ health. The developed method incorporates trending of the concentration of the dissolved gases over the operating life. The approach processes current and/or historical DGA data, using the 7-developed logic modules, to determine the current state of a transformer, provide information regarding the fault type, fault probability, fault severity and recommended future sampling interval in addition to operating procedure, consistent with industry standards. The developed diagnosis system has been validated using 1290 samples from fresh and previously tested mineral oil filled transformers. The proposed system is proved, based on field data, to be 99% accurate in identifying transformers being in normal or abnormal operation. For the cases where a transformer fault was known, the proposed technique has less than 2% inaccuracy in recognizing the fault’s type in comparison to other approaches discussed in literature.
KW - oil insulation
KW - gases
KW - fault diagnosis
KW - power transformer insulation
KW - monitoring
KW - oils
U2 - 10.1109/MEPCON.2018.8635154
DO - 10.1109/MEPCON.2018.8635154
M3 - Conference contribution
SN - 9781538666524
T3 - 2018 20th International Middle East Power Systems Conference, MEPCON 2018 - Proceedings
SP - 400
EP - 405
BT - 2018 Twentieth International Middle East Power Systems Conference (MEPCON)
PB - IEEE
ER -