Oil-Filled Transformer Malfunction
Malfunction in oil-filled transformers can be assessed by detection of certain key gases which might be present in the transformer oil and collected Buchholz relay. Sampling and analysis of these gases is essential in testing and evaluating the health of a transformer. Formation of gases within operating transformers is caused by either thermal or electrical disturbances and if not corrected may result to system failure. The electrical disturbances may be categorized as corona or electrical arc (HARLOW, 2012, p.188-90). The interpretation of the test data is upon the basis of the type of individual gases liberated and their quantity present in the transformer (GILL, 2009, p.203-210).
Electrical equipment and apparatus use fluids and insulating oils as dielectrics. Mineral oils and synthetic fluids are used in transformers since they are less flammable (GILL, 2009, p.203-210). Effectiveness of insulating fluids is adversely affected by their deterioration which is brought about by overheating, oxidation, contamination and electrical stress. High temperatures from environmental conditions or from increased load are responsible for accelerating the deterioration process (IEEE, 2009, p.3; HARLOW, 2012, p.188-90). Therefore, to guarantee continued service, maintenance, and safety, a condition monitoring or testing program comprising of chemical and electrical analysis is essential for the dielectrics (PURKAIT, 2017, p.211-15; SCHRIEBER, WILLIS & PHILIPS, 2012, p.203-07). Mineral oils (hydrocarbon oils) have high chemical stability and dielectric strength which enables them to be used as transformer insulating fluid. Sampling of oil in the transformer and regular inspection is necessary to ensure that the transformer oil is free of contaminants (GILL, 2009, p.203-210).
An insulation system for an oil-immersed transformer comprises of cellulose and insulating oil materials (HARLOW, 2012, p.188-90). Normally, this insulation degrades and produces certain flammable and non flammable gases. However, their presence becomes further evident if the transformer dielectric is subjected to elevated temperatures (Sun, Huang and Huang, 2012, p.1220-25). Thermal decomposition of cellulose insulation at temperatures of 140 0C liberates, carbon dioxide (CO2), carbon monoxide (CO), and some methane (CH4) or hydrogen (H) (TANG & WU, 2011, p.125-26). The frequency at which the key gases are emitted is exponentially dependant on the temperature as well as proportionally aligned to quantity of the insulation at the temperature (GILL, 2009, p.203-210; PURKAIT, 2017, p.211-15).
Thermal decomposition of oil between 150oC and 500oC liberates large volumes of methane and hydrogen gases, and traces of ethane and ethylene (IEEE, 2009, p.3; Ward, 2003, p.463-68). Higher oil temperatures results to more hydrogen gas been emitted than methane as well as higher capacities of ethane (C2H6), and ethylene (C2H4) (LOO & ZULKURNAIN, 2014, p.48; GILL, 2009, p.203-210). Increased volumes of hydrogen and ethylene are liberated as the temperatures rise (IEEE, 2009, p.3). Partial discharges (corona) and low-intensity arcing emits mainly hydrogen gas coupled with minimal volumes of methane (CH4) and traces of acetylene (C2H2) (GILL, 2009, p.203-210). Acetylene gas becomes more pronounced during high-intensity arcing which occurs at (700oC – 1800oC) inside the transformer tank (TANG & WU, 2011, p.125-26; PURKAIT, 2017, p.211-15)
Insulation System for an Oil-Immersed Transformer
Detecting, analyzing and identifying of individual gases dissolved in transformer oil can be of great importance while evaluating the operating condition of an oil-impregnated transformer (Ahmed, Hassan, Ahmed & Kassir, 2016, p.35; GILL, 2009, p.203-210; SCHRIEBER, WILLIS & PHILIPS, 2012, p.203-07). Baseline data is necessary for establishing a reference point for newly adopted transformers for comparison with subsequent routine maintenance sampling results (Morais and Rolim, 2006, p.673-80; Huang, 2003, p.1257-61). However, monitoring and assessment of the transformer condition can commence at anytime even if the reference data is not available (GILL, 2009, p.203-210). The two methods for investigating the key gases liberated include Dissolved Gas Analysis (DGA) and Total Combustible Gas Analysis (TCGA). Inspection of the individual gases present in oil gives a convenient tool for testing the functionality of the transformer (CHAKRAVORTI, DEY & CHATTERJEE, 2013, p.118-22).
DGA method of investigating flammable gases is more informative than TCGA method as it provides the earliest probable diagnosis of any abnormality in the transformer. On the contrary, TCGA is widely applied and fast in continuously monitoring the health of a transformer (GILL, 2009, p.203-210; ELLEITHY, 2010, p.226-28). However, the demerit associated with TCGA is that it gives a single value of the oil-combustible gases hence not quantifying the individual gases present in the oil (Sun, Huang and Huang, 2012, p.1220-25). Additionally, it is important to note that temperature and pressure affects solubility of the gases in oil.
Detection of the key gases at the earliest possible time as well as undertaking the appropriate measures to rectify the transformer condition leads to the success of the fault gas analysis (SCHRIEBER, WILLIS & PHILIPS, 2012, p.203-07). The four-level criterion developed by IEEE for classifying risks in transformers is crucial for analyzing condition of a transformer if no previous dissolved gas history for transformers exists (IEEE, 1992, p16). This criterion applies both concentrations for individual gases and concentrations of all combustible gases. Since we have no prior dissolved gas history for the 10 installed transformer units, we shall adopt this criterion in establishing the risks that may be facing the transformers.
Condition 1: TDCG under this quantity shows that the transformer is working normally or satisfactorily. Individual combustible gases above specified limits should elicit further analysis or investigation (IEEE, 1992, p.9-12).
Condition 2: The normal combustible gas level has been exceeded. All combustible gases above the specified limits should give rise to additional investigation. Further measures ought to be undertaken to establish trends. There is high probability that faults might be present in the transformer (IEEE, 1992, p.9-12).
Analysis of Key Gases
Condition 3: TDCG at this limits suggest increased level of decomposition. Further analysis should be carried out for each individual flammable gas exceeding its specified. Quick response is desirable to establish trend. There is possibility of faults in the transformer (IEEE, 1992, p.9-12).
Condition 4: This level indicates extreme decomposition of oil or cellulose insulation. Unit must be changed since further operation of the transformer may result to its failure. Need to retest. If both the individual and TDCG gases are rising significantly (>30ppm/day), transformer should be de-energized. The fault is active (IEEE, 1992, p.9-12).
Transformer A1
Calculating the total dissolved combustible gases (TDCG) we have;
11.7 + 9.98 + 6.40 + 6.19 + 0.05 + 323.8 = 358.12
From the above TDCG result, 358.12 falls under condition 1 which is below 720 hence it is operating normally. Production of carbon dioxide may indicate cellulose ageing, thermal faults in cellulose or leaks in the oil expansion systems.
Transformer A2
TDCG = 19.68 + 92.93 + 362.06 + 3.57 + 0.05 + 243.94 = 722.23. The TDCG result shows that it is operating under condition 2. Production of ethane exceeds the specified range (66-100) in condition 2. The cause of the higher levels of ethane may be as a result of mineral oil decomposition or thermal faults in oils at temperatures of 150oC – 300oC and 300oC – 700oC. Carbon dioxide may be due to cellulose ageing or thermal faults in cellulose.
Transformer B1
TDCG is 18.04 + 75.02 + 306.76 + 7.54 + 0.05 + 204.15 = 611.56
The transformer is operating under condition 1. However, production of ethane needs to be further investigated to evaluate whether it is from thermal faults in oil or mineral decomposition of oil. There is production of carbon dioxide above the specified range an indication that there might be thermal faults in cellulose or cellulose ageing.
Transformer C1
TDCG = 46.23+366.55+391.87+464.15+0.05+247.41 = 1516.26
This lies between the range (721 and 1920), under condition 2. The production of methane, ethylene, ethane and carbon dioxide exceeds the specified limits. Methane production may be due to thermal faults in oils at 1500C, 300oC or 700o; decomposition of mineral oils and partial discharge. Ethylene may be due thermal faults in oils at 700oC, mineral decomposition of oil and arcing.
Transformer C2
TDCG = 150.81+365.87+183.97+899.42+6.48+35.72 = 1642.27.
This transformer is operating under condition 2. Investigation of some of the key gases such as hydrogen, methane, ethane, ethylene and acetylene must be carried out to test the abnormality in this transformer. Carbon dioxide and carbon monoxide are below the specified limits in condition 1 and condition 2. The faults associated from emission of those key gases may be due to thermal faults in oil, mineral decomposition of oil, arcing and partial discharge.
Condition Monitoring or Testing Program
Transformer D1
TDCG = 12.86+55.84+233.14+5.47+0.79+192.49 = 500.59. This transformer operates under condition 1. Emission of ethane is above the specified limit meaning there might be a thermal fault in the oil or decomposition of mineral oil. Carbon dioxide mission is normal meaning that no minimal ageing in cellulose and thermal faults in cellulose.
Transformer D2
TDCG = 25.04+134.39+637.14+23.67+0.05+218.5 = 1038.79.
The value for the TDCG is in condition 2. Ethane exceeds the specified level. This may be as a result of thermal faults in oils or decomposition of mineral oils.
Transformer D3
TDCG = 17.86+123.22+582.37+25.27+0.59+199.96 = 949.27.
Operation of the transformer is in condition 2. Investigation of ethane needs to be carried out to establish the cause that may be leading to its high emission. Thermal faults in oils and decomposition of mineral oil are the probable causes that are responsible for its high emission.
Transformer E1
TDCG = 13.75+86.17+350.73+5.35+0.05+127.26 = 583.31.
Transformer is operating under satisfactorily under condition 1. Carbon dioxide in the gas lies with the expected normal value but ethane presence in the oil is beyond the specified limit. Investigation of its high presence in the oil may indicate thermal faults in oil or decomposition of the mineral oil.
Transformer E2
TDCG = 15.95+94.17+223.02+8.85+0.05+201.2 = 543.24.
This indicates that the transformer is operating satisfactorily under condition 1. The carbon dioxide level is within the normal limit but ethane level exceeds the specified limit. This may be due to thermal faults in oil and decomposition of mineral oil.
CO2/CO ratio can be used to analyze the health of an oil-immersed generator. However, for the ratio to be significant; the individual gases, that is, both CO2 and CO, should be above 5000/500 ppm. For this reason, we shall not apply this ratio in investigation (IEC, 2015, p.12).
For transformer C2,
R1= 365.87/150.81 = 2.426, R2= 6.48/899.42 = 0.0072, R5= 899.42/183.97 = 4.889
From Roger’s ratio, see appendices – Table 2, the fault associated with this ratio is thermal at 700oC. Additionally, from the DGA Interpretation Table; see Appendices – Table 4, the ratio of C2H4/C2H6 in transformer C2 is 4.889. Since the ratio is >1, it indicates that there is thermal fault in oil.
Following the observations from the analysis of the DGA results, it can be seen that there is a thermal fault in oil witnessed in all the transformers except in transformer A1 where all the key gases are within the specified limits. The thermal fault in oil has led to the higher presence of the key gases in the transformers. Transformers which are working under condition 1 should be let to continue with normal operation. These transformers include: A2, C1, C2, D2 and D3. However, caution must be exercised while operating those transformers. Investigation and monitoring of individual gases must be carried to ensure that unplanned faults or failures are prevented. Asset manager team should therefore determine the load dependence of those transformers. For those operating under condition 2, asset managers must exercise caution as well as investigation for the individual gases. Load dependence of the transformer should be determined. Since these transformers have not had any previous dissolved gas history, the asset managers must ensure the frequently investigate the dissolved gases in the transformers to identify any trends in the emission of gases. Planned investigation is vital in establishing any gas presence which might cause unplanned failures and faults in the transformers. De-energized of transformers must also be carried out if those existing show signs of failure. The nature of the problem associated with all the transformers except transformer A1 is thermal fault in oil or decomposition in mineral oil. However, the intensity at which emission of the key gases takes place in transformer C2 is worrying. As a result, I would advise the asset manager team to consider replacing that transformer.
Conclusion
Dissolved gas analysis plays a vital role in establishing the cause of gas emission in oil-immersed transformers. Frequent investigation and evaluation of the gases emitted and the condition of the transformer is essential in ensuring that preventive measures are taken. Such investigations help in establishing the root cause of the presence of gases in transformer oil.
References
Ahmed, M.A.A., Hassan, T.I.H., Ahmed, A.A.M. and Kassir, E.A.A.E., 2016. Dissolved Gas Analysis for Transformer Oil(Doctoral dissertation, Sudan University of Science and Technology).
CHAKRAVORTI, S., DEY, D., & CHATTERJEE, B. (2013). Recent trends in the condition monitoring of transformers: theory, implementation and analysis.
ELLEITHY, K. (2010). Technological developments in networking, education and automation. Dordrecht, Springer. https://public.eblib.com/choice/publicfullrecord.aspx?p=645234.
GILL, P. (2009). Electrical power equipment maintenance and testing. Boca Raton, CRC Press. https://www.crcnetbase.com/isbn/9781574446562.
HARLOW, J. H. (2012). Electric power transformer engineering. Boca Raton, FL, CRC Press. https://www.crcnetbase.com/isbn/9781439856369.
Huang, Y.C., 2003. A new data mining approach to dissolved gas analysis of oil-insulated power apparatus. IEEE transactions on power delivery, 18(4), pp.1257-1261.
IEC. (2015). Mineral oil-filled electrical equipment in service: guidance on the interpretation of dissolved and free gases analysis.
IEEE,1992. IEEE guide for the interpretation of gases generated in oil-immersed transformers. New York, Institute of Electrical and Electronics Engineers.
IEEE. (2009). IEEE Std C57.104-2008 (Revision of IEEE Std C57.104-1991): IEEE Guide for the Interpretation of Gases Generated in Oil-Immersed Transformers. [S.l.], IEEE. https://ieeexplore.ieee.org/servlet/opac?punumber=4776467.
LOO, Y. T., & ZULKURNAIN ABDUL MALEK. (2014). Power transformers condition monitoring using dissolved gas analysis and hidden markov prediction model.
Morais, D.R. and Rolim, J.G., 2006. A hybrid tool for detection of incipient faults in transformers based on the dissolved gas analysis of insulating oil. IEEE Transactions on Power Delivery, 21(2), pp.673-680.
PURKAIT, P. (2017). Transformer ageing: monitoring and estimation techniques. Wiley-IEEE Press.
SCHRIEBER, R. R., WILLIS, H. L., & PHILIPS, E. (2012). Aging Power Delivery Infrastructures, Second Edition. Abingdon, CRC Press. https://proquest.safaribooksonline.com/9781439896914.
Sun, H.C., Huang, Y.C. and Huang, C.M., 2012. A review of dissolved gas analysis in power transformers. Energy Procedia, 14, pp.1220-1225.
TANG, W. H., & WU, Q. H. (2011). Condition monitoring and assessment of power transformers using computational intelligence. London, Springer. https://public.eblib.com/choice/publicfullrecord.aspx?p=666864.
Ward, S.A., 2003, October. Evaluating transformer condition using DGA oil analysis. In Electrical Insulation and Dielectric Phenomena, 2003. Annual Report. Conference on (pp. 463-468). IEEE.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
