The Transformer Age Index Model provides a visual indication of the Health Index of the power transformer fleet in relation to the “normal” designed life of the transformer.
With everything in life, as we age, our health deteriorates accordingly over time. We know that with humans some age well and outlast the average expected lifespan but most people, however, follow the trend of the normal bell curve distribution where the majority lead a life of balance of probabilities and live the average expected lifespan. On the one side, there are premature deaths related to bad genetics (design and manufacture), lifestyle (unfavourable operating and maintenance conditions) or unforeseen events (external stresses and events). On the other side of the spectrum, there are some that exceed 100 years!
I find making a correlation between human life and that of power transformers interesting, especially when it comes to the concepts of health. I would make a hypothesis that a power transformer also has a lifespan ageing bell curve distribution and may be represented in a Transformer Age Index Model (TAIM).
Transformer Age Index Model (TAIM)
The Transformer Age Index Model provides a visual indication of the Health Index of the power transformer fleet in relation to the “normal” designed life of the transformer.
The two main measures are Chronological Age and Biological Age.
The assumption is made that most transformers are manufactured for an average design life span of 35 years.
As a power transformer ages, it may follow the normal ageing profile where one chronological year is equivalent to one biological year. Figure A provides the concept of the Transformer Age Index Model (TAIM)
Chronological Age vs Biological Age
Chronological Age is ageing as equivalent to a calendar year.
Biological Age is the equivalent age according to the health of the transformer which is usually based on the condition of the paper and oil insulation, major components (tap changer, bushings), and the general status (faults, unforeseen events or diseases like the effects of Corrosive Sulphur).
This is based on the Health Index Factor which is then applied to the Chronological Age. For transformers that age well this Health Index Factor will be less than one and for transformers that have premature ageing or any risks to the health status the Health Index Factor will be more than one.
Regions in the TAIM
There are five regions within the Transformer Age Index Model; Healthy Ageing, Normal Ageing, Premature Aging, Old Aging and Risk of Failure
Healthy Ageing is the region where the transformer is ageing better than expected with respect to its expected health status. These are transformers that are designed and manufactured well, operated and maintained well and have no exposure to external stress or predisposition to diseases like corrosive Sulphur.
Normal Ageing is where the chronological age is equivalent to the Biological age. Any deviation from this based on the health condition will be reflected in the TAIM. Due to minor variations, this is taken within a ± 15% variation of the normal ageing curve (35 years). After reaching the designed age of 30 years the upper limit tapers off as the transformer approaches 40 years and moves into Old Ageing.
Premature Ageing is indicative of transformers that are ageing unsatisfactorily with respect to its expected health status when compared to the chronological age. This may be due to indications of faults as a result of design or manufacturing, inadequate operating or maintenance and frequent exposure to external stresses or predisposition to diseases like corrosive Sulphur. Transformers in this region may be able to recover if the problems identified early are managed effectively.
Old Ageing is the region where the transformer is ageing as expected but due to the normal ageing of materials, it may lose certain design characteristics and will require monitoring for any sudden changes or exposure to external stresses.
Risk of Failure is the region where action needs to be taken as soon as possible. It is indicative of a condition that is at high risk of failure.
Biological Age Explained
Biological Age (ABIO) is the current health status of the transformer based on the Health Ageing Factor (AFACTOR) which is effectively an assessment based on parameters measuring different aspects of the health condition of the transformer, multiplied by the Chronological Age (ACHRON).
Biological Age (ABIO) = Chronological Age (ACHRON) x Ageing Factor (AFACTOR)
Ageing Factor
The Ageing Factor (AFACTOR) then becomes the most important parameter as it reflects the health status (Index) of the transformer. This may be calculated using the following methodology which takes the weighted average of key parameters affecting the transformer life.
Ageing Factor (AFACTOR) = (APAPER x KPAPER) + (AOIL x KOIL) + (ATAP x KTAP) + (ABUSH x KBUSH) + (AFAULT x KFAULT) + (ADIS x KDIS)
Due to the different contributions of each of these factors, its influence is weighted. This weighting may also be adjusted if required as relevant for a different fleet of transformer designs and operating conditions. The following weighting was tested in this family of power transformers.
Paper Ageing Factor (APAPER)
The APAPER factor is based on the condition of the cellulose insulation. This is usually measured with the degree of polymerization or the furfural levels in the oil. Another factor that affects the insulation condition is the level of water within the insulation [1]. The following weighting was tested in this study but can be adjusted accordingly.
APAPER = (0.4*ADP ) + (0.6*APW)
Degree of Polymerisation Factor (ADP)
ADP may be calculated based on the following simple relationship due to the DP of the paper being acceptable up to 600 where higher DP values are usually an indication of good condition [2].
New transformers usually have a DP greater than 1000 so the ADP would be 0.6 and less for good condition. Transformers with end-of-life DP values of 200 and less would have an ADP value of 3 and more.
Paper Water Factor (APW)
The moisture content in cellulose is highly dependent on the temperature where APW may be calculated from generally accepted methods such as RVM or using moisture equilibrium charts [1]. The result can be estimated as the percentage of water in cellulose for transformers in operation being acceptable up to 1.5% and the lower the water content this is usually an indication of good condition.
New transformers usually have a Water Content (%) far less than 1% so the APW would be 0.5 and less. Transformers with Water Content (%) values of 3% and more would have an APW value of 2 and more.
Oil Ageing Factor (AOIL)
The AOIL Oil Ageing Factor is based on the condition of the oil insulation which usually deteriorates under thermal, electrical and chemical stress. This usually results in byproducts in the oil which can be measured in the form of Acidity (AACID), Water Content (AOW), Dielectric Strength (ADS) and Interfacial Tension (AIFT) [3]. The following weighting was tested in this study but can be adjusted accordingly.
AOIL = (0.4*AOW ) + (0.3*ADS ) + (0.2*AIFT) + (0.1*AACID )
Oil Water Factor (AOW)
AOW may be calculated based on the following simple relationship due to the water in oil being acceptable up to 10 ppm where lower water content is usually an indication of good condition [4, 5].
New transformers usually have a Water Content (%) less than 5 ppm so the AOW would be 0.5 and less. Transformers with Water Content (ppm) values of 20 and more would have an AOW value of 2 and more.
Oil Dielectric Strength Factor (ADS)
ADS may be calculated based on the following simple relationship due to the Dielectric Strength of the oil being satisfactory up to 50 kV/2.5mm for running transformers where higher Dielectric Strength values are usually an indication of good condition of the oil [4, 6].
New transformers usually have a Dielectric Strength greater than 70 so the ADS would be 0.714 and less. Transformers with Dielectric Strength values of less than 40 would have an ADS value of 1.25 and more.
Oil Interfacial Tension Factor (AIFT)
AIFT may be calculated based on the following simple relationship due to the Interfacial Tension of the oil being acceptable above 30 for running transformers where higher Interfacial Tension values are usually an indication of good condition [4, 7].
New transformers usually have an Interfacial Tension greater than 45 so the AIFT would be 0.67 and less. Transformers with Interfacial Tension values of less than 20 would have an AIFT value of 1.5 and more.
Oil Acidity Factor (AACID)
AACID may be calculated based on the following simple relationship due to the acidity of the oil being acceptable up to 0.07 KOH/mg where lower acidity content is usually an indication of good condition.
Newer transformers usually have an Acidity level less than 0.03 so the AACID would be 0.42 and less for good condition [4]. Transformers with Acidity values of 0.1 and more would have an AACID value of 1.43 and more.
Tap Changer Ageing Factor (ATAP)
The ATAP factor is based on the condition of the oil insulation, DGA faults, and Inspections. This is usually measured with Water Content (ATOW), Dielectric Strength (ATDS) and DGA Faults (ATFAULTS). The following weighting was tested in this study but can be adjusted accordingly.
ATAP = max (ATOW , ATDS , ATFAULTS )
Tap Changer Water Factor (ATOW)
ATOW may be calculated based on the following simple relationship due to the water content in the oil of the tap changer (in operation) being acceptable up to 30 ppm where lower water content is usually an indication of good condition [4].
New tap changers usually have a Water Content of less than 12 ppm so the ATOW would be 0.4 and less. Transformers with Water Content (ppm) values of 40 ppm and more would have an ATOW value of 1.33 and more.
Tap Changer Dielectric Strength Factor (ATDS)
ATDS may be calculated based on the following simple relationship due to the Dielectric Strength of the oil in the tap changer (in operation) being acceptable up to 40 kV/2.5mm where higher Dielectric Strength values are usually an indication of good condition [4].
New transformers usually have a Dielectric Strength greater than 60 so the ATDS would be 0.67 and less. Transformers with Dielectric Strength values of less than 30 would have an ATDS value of 1.33 and more.
Tap Faults Factor (ATFault)
ATFAULTS may be calculated based on the DGA ratios for tap changers. The Stenestam Ratio is used to assess the condition [8]. It is noted that this is only for non-vacuum tap changers using mineral oil. For other types of tap changers similar assessments can be used but for this study, the following relationship was used.
The Stenestam Ratio = (CH4 + C2H4 + C2H6) / C2H2
- < 0.5 No Overheating
- 0.5 – 3 Take frequent samples
- ≥ 3 Overheating – Investigate
New transformers usually have a Stenestam Ratio less than 0.5 so the ATFault would be 0.5 and less. Transformers with Stenestam Ratio values between 0.5 and 3 would require monitoring and has an ATFault value of 2. Transformers with Stenestam Ratios greater than 3 have an ATFault value of 3 and must have an immediate investigation.
Bushing Ageing Factor (ABUSH)
The ABUSH factor is based on the condition of the transformer bushings which is usually measured with Tan Delta (ABTAN) for OIP bushings but other factors can be used for other types of bushings. Inspections (ABINSP) of the bushings are another important measure and can include thermal scanning for hot connections.
ABUSH = max(ABTAN , ABINSP)
Bushing Dissipation Factor (ABTAN)
High oil conductivity, ageing, and increased water content are symptoms of ageing and insulation degradation which usually result in an increase in losses. This can be measured by the power factor or dissipation factor which is an important factor in oil-filled condenser core bushings.
Typical values for Power Factor / Dissipation Factor of bushings at line frequency and at 20°C/68°F according to international standard IEC 60137 are provided in the attached figure [9].
New transformers usually have a Tan Delta less than 0.3 so the ABTAN would be 0.42 and less. Transformers with Tan Delta values of 1.0 and more would have an ABTAN value of 1.42 and more.
Bushing Inspections Factor (ABINPS)
ABINSP may be assessed from field inspections such as Thermal scanning to assess hot spots, cracks and leaks. Scoring can be adjusted accordingly.
Fault Ageing Factor (AFAULT)
The AFAULT factor is based on the presence of any faults within the transformer (tank structure, core and windings) as identified by dissolved gas analysis. The faults are based on the assessment of individual Gas Limits (AGL), Duvals Triangle (ADT) and LEDT – Low Energy Degradation Triangle LEDT (ALEDT). Other techniques can also be used accordingly.
AFAULT = (0.2*AGL ) + (0.4*ADT ) + (0.4*ALEDT)
Gas Limits Faults Factor (AGL)
AGL may be assessed from DGA Analysis based on the specific limits for each of the combustible gases (Hydrogen (H2), Methane (CH4), Ethane (C2H6), Ethylene (C2H4), Acetylene (C2H2), and Carbon Monoxide (CO) [10]. New transformers usually have gas limits within specifications. Due to some gases having higher values each of the combustible gas factors is limited to a maximum of 3.
AGL is taken as the maximum of any of the combustible gas limits that require further or urgent investigation.
Duval’s Triangle 1 Analysis Factor (ADT)
ADT may be assessed from the specific regions of the Duvals Triangle 1 [11]. The scoring in the table is used for the specific fault types.
LEDT Faults Factor (ALEDT)
ALEDT may be assessed from the specific regions from the Low Energy Degradation Triangle [12]. The scoring in the table is used for the specific fault types.
Disease Ageing Factor (ADIS)
The ADIS factor is based on any predisposing factors that will cause failure over time. This includes the presence of Corrosive Sulphur (ACS) or any other such factors identified. In this study, only the effects of corrosive Sulphur were considered but other such parameters can be included accordingly.
ADIS = ACS
Corrosive Sulphur Factor (ACS)
ACS may be assessed from analysis for the presence of Corrosive Sulphur [13]. It was considered that not having corrosive Sulphur should not influence the transformer’s health condition positively so a factor of 1 was selected. For transformer oils with the presence of Corrosive Sulphur, the factor of 2 was selected as there are significant negative influences on the health of the transformer.
Case Application
The Transformer Age Index Model was applied to a fleet of 36 GSU transformers which ranged in chronological age from 4 years (new) to 45 years (old). The power ratings range from 65 MVA to 350 MVA. Figure B provides the results from the TAIM fleet assessment:
Firstly, it is noted that the TAIM provides a clear picture of the transformer fleet with regard to the ageing profile.
Secondly, there is now a clear picture of the transformers that may need special attention. There are 2 transformers (32, 19) in the “Risk of Failure” region and 8 transformers (22, 27, 13, 35, 34, 21, 36, 33) in the “Premature Ageing” region. These transformers would require immediate attention. On closer inspection, it was noted that:
Transformer 22: had an AFACTOR = 1.233 which meant that the ACHRON of 12 resulted in an ABIO of 14.79 years pushing the transformer into the “Premature Ageing” region. Inspection of the results revealed that this transformer had an ATAP = 3 due to high discharges and the ADGA = 1.80, due to presence of presence of corrosive sulphur (ACS = 2) with active faults (AFAULT = 2.2)
Transformer 27 had an AFACTOR = 1.228 which meant that the ACHRON of 17 resulted in an ABIO of 20.88 years pushing the transformer into the “Premature Ageing” region. Inspection of the results revealed that this transformer had an ATAP = 3 due to high discharges and the ADGA = 1.80, due to presence of presence of corrosive sulphur (ACS = 2) with active faults (AFAULT = 2.2)
Transformer 13: had an AFACTOR = 1.292 which meant that the ACHRON of 13 resulted in an ABIO of 15.48 years pushing the transformer into the “Premature Ageing” region. Inspection of the results revealed that this transformer had an ATAP = 3 due to high discharges and the ADGA = 1.80, due to presence of presence of corrosive sulphur (ACS = 2) with active faults (AFAULT = 2.2)
Transformers 35, 34, 33 and 36 respectively had relatively elevated AFACTOR values when compared to their age and being new transformers which were commissioned in the last 4 years. These transformers were now sitting in the “Premature Aging” region of the TAIM. The main reason for these transformers having a high AFACTOR was due to the tap changer ATAP having high values due to the oil dissolved gas tests indicating high discharge levels. This risk was also exasperated due to the first major service not being done after 4 years of operation. These transformers were also from a pumped storage power station and the tap changer worked frequently on a daily basis. Services were subsequently planned with high urgency.
Transformer 32: had an AFACTOR = 1.150 which meant that the ACHRON of 43 resulted in an ABIO of 49.44 years. The important aspect, in this case, is to identify what has resulted in the AFACTOR value of 1.150. Closer inspection revealed that this transformer had the presence of corrosive sulphur (ACS = 2) with active faults (AFAULT = 2.2)
Transformer 19: had an AFACTOR = 1.066 which meant that the ACHRON of 45 resulted in an ABIO of 47.98 years. Closer inspection revealed the following factors; APAPER = 1.11, AOIL = 2.5 (AOW = 1, AIFT = 2.5, AACID = 1.14), with defects found on bushings (ABUSH = 2)
Conclusion
From the theory and the results achieved for the fleet of 36 transformers, the TAIM provides a simple way to integrate the asset health index with the transformer fleet data to identify specific focus areas for transformers.
Another advantage is that the TAIM provides an asset manager with information on which ageing transformers to focus on for replacement, repair and maintenance strategies. This is very important due to current economic challenges where key financial decisions have to be made.
If you require an assessment of your fleet of Power Transformers using the TAIM please feel free to make contact at Powertransformerhealth@gmail.com
References
- T. V. Oommen and T. A. Prevost, “Cellulose insulation in oil-filled power transformers: part II maintaining insulation integrity and life,” in IEEE Electrical Insulation Magazine, vol. 22, no. 2, pp. 5-14, March-April 2006, doi: 10.1109/MEI.2006.1618996
- CIGRE Brochure 323: “Ageing of cellulose in mineral-oil insulated transformers.” TF D1 1 (2007).
- IEC 60422, “Mineral insulating oils in electrical equipment–supervision and maintenance guidance”. BS EN, 60422.
- Condition Assessment of Power Transformers, CIGRÉ Technical Brochure 761, March 2019
- ASTM D1533-12, Standard Test Method for Water in Insulating Liquids by Coulometric Karl Fischer Titration, ASTM International, West Conshohocken, PA, 2012.
- IEC 60156, “Insulating liquids – Determination of the breakdown voltage at power frequency – Test method”
- ASTM D971-20, Standard Test Method for Interfacial Tension of Insulating Liquids Against Water by the Ring Method, ASTM International, West Conshohocken, PA, 2020.
- 1ZSC000498-AAA EN, REV. A, 2018-08-17, Dissolved gas analysis in tap-changer oil, http://www.abb.com/transformercomponents
- IEC 60137, International Electrotechnical Commission. “Insulated bushings for alternating voltages above 1000 V.” Geneva: IEC (2008).
- IEEE Standard C57. 104. “Guide For The Interpretation Of Gases Generated In Mineral Oil-Immersed Transformers.” (2019).
- Duval, M., “A Review of Faults Detectable by Gas-in-oil Analysis in Transformers,” IEEE Electrical Insulation Magazine, Vol. 18, No. 3, Pages 8-17, May/June 2002
- N. Moodley and C. T. Gaunt, “Low Energy Degradation Triangle for power transformer health assessment,” in IEEE Transactions on Dielectrics and Electrical Insulation, vol. 24, no. 1, pp. 639-646, Feb. 2017, doi: 10.1109/TDEI.2016.006042.
- IEC 62535, Insulating liquids – Test method for detection of potentially corrosive sulphur in used and unused insulating oil