Editor’s Note: While this article is primarily focused on using a Health Index to assure Transformer Reliability, we believe that transformer reliability is a fundamental contributor to overall system resilience. While certain system outages may stem from external or uncontrollable events, a resilient power system strategy can begin with a thorough data-driven assessment of transformer health during operation. The healthier the transformer, the lower the likelihood of unplanned outages and operational issues.
INTRODUCTION
This article examines how predictive maintenance can drive operational excellence in power transformers, utilizing a real-world case study as an example.
For large-scale energy and utility operators, improving the reliability of critical power transformers is crucial for minimizing asset downtime and associated costs, as well as enhancing the reliability of power systems.
Many asset owners face similar challenges, including fleets acquired from multiple entities with inconsistent maintenance histories, aging assets, frequent load fluctuations, and industrial operations with large motors that frequently start and stop. These conditions place transformers under constant stress, accelerating wear, degrading performance, and increasing the risk of failures or unplanned outages. Limited centralized data further hampers predictive analytics, forcing reliance on traditional time-based maintenance.
To overcome these obstacles, asset owners need to implement a centralized data repository and a tracking dashboard to consolidate maintenance information across the enterprise into a single index, known as the Static Asset Health Index (SAHI) [1]. SAHI provides a good visibility into asset health through the support of advanced analytics. It is a cost-effective predictive tool that integrates inspection results, test data, and maintenance history into a single, actionable health score. SAHI enables proactive issue detection and prioritization of maintenance, helping prevent critical failures and reduce downtime.
In this article, we will discuss a specific health case, highlighting how SAHI helped the asset owner identify the issue before it led to failure.
PROPOSED SOLUTION:
The proposed solution is a cost-effective Static Asset Health Index (SAHI). SAHI is a quantitative value that represents the overall condition of the transformer. The database behind SAHI provides:
- Comprehensive dashboard based on as-left condition information
- Quantitative value that represents the overall condition of the asset
- Centralized maintenance data that enables insights and provides situational awareness
The Static Asset Health Index (SAHI) integrates high-quality data, including oil analysis results from databases such as Delta-X or SDMyers, with visual inspections, offline tests, and maintenance history. Weighted condition parameters produce a single health score, enabling failure prediction, improving reliability, and prioritized maintenance—without the cost of continuous real-time monitoring.
SAHI takes the following into consideration:
- In-service life, preventive maintenance (PM) test results include the condition of bushings, grounding, cooling system, on-load tap changer, and components such as gaskets, joints, surge arresters, gauges, protection devices, structures, and the control cabinet.
- Upset conditions like alarms, trips, loading data, environment, background history, etc.
- Electrical parameters and historical data, such as loading conditions, oil change dates, etc.
By integrating high-quality data and applying weighted scoring, the Static Asset Health Index (SAHI) generates a single health score, ranging from 0 to 100%, offering a clear snapshot of transformer condition. This score supports early failure prediction, informed maintenance planning, and enhanced reliability in a cost-effective and practical manner. Based on the trends, AI can be utilized to predict failures, including the remaining life of the transformer.
The following items, when available, are considered in developing SAHI for the transformers:
- Dissolved Gas Analysis (DGA) report
- Preventive Maintenance’s Test Results
- Bushings Condition
- Grounding
- Bushings
- Cooling system
- On-Load Tap Changer
- Components such as Gaskets, joints, Surge Arresters, gauges, protection devices, structures, and the Control Cabinet, CTs
- Loading data
- Historical faults
Each category has its own dedicated weight and the health indicator, with the highest weight to the oil analysis (main tank and tap changers), followed by the preventive maintenance test results, including winding resistance, excitation current, insulation resistance, winding power factor test and turns ratio test.
Sample of the Health Indicators and the SAHI Index
The grading scales for SAHI computation are outlined in the following table:
| SAHI (%) | Condition |
| 86-100 | Very Good |
| 70-85.9 | Good |
| 50-69.9 | Fair |
| 30-49.9 | Poor |
| 0-29.9 | Extremely Poor |
SAHI is typically updated every 1–5 years. However, incorporating more frequent data provides a clearer and more detailed view of the asset’s condition.
The health indicators that form the backbone of SAHI and its byproducts come with three statuses: green for normal, yellow for warning, and red for the alert zone.
An example of the data displayed by SAHI dashboard is shown below. The dashboard presents detailed, visualized data for each gas and its rate of change relative to previous samples, aligned with IEEE C57.104-19 limits. It also displays key indicators, such as recent temperature-gauge readings (offline), results from the latest preventive-maintenance tests. The dashboard can be customized for additional complementary data beyond SAHI, such as arc-flash label validity for the substation, and other system information relevant to the transformer.
Why Implement SAHI?
Even with established preventive maintenance programs—including offline tests, online inspections, and oil analysis—gaps remain in consolidating and interpreting data. SAHI addresses these gaps by integrating and analyzing data from various sources, delivering a single number associated with a centralized analytics dashboard for informed decision-making. The SAHI platform can be further integrated with data-based platforms, such as SAP, Oracle, SDMyers, Delta X, or other platforms, to collect and consolidate all data.
Key SAHI benefits include:
CASE STUDY:
Our health case study showcases how SAHI effectively detected a partial discharge in a 7.5MVA 13.8/4.16kV, 12-year-old transformer and prevented a failure.
In Sep. 2020, the gas results appeared normal. However, subsequent analysis in Sep. 2021 revealed a substantial rise in hydrogen concentration, escalating from ethane and methane levels, which also exhibited notable increases, exceeding the limits specified in IEEE C57.104-2019 Table 3 [3]. This increasing trend in gases continued till November 2021, when the hydrogen level reached 10400 ppm, methane reached 798 ppm, and ethane reached 161 ppm. The asset owner requested a SAHI development for the transformer.
During the SAHI development, it was noted that the oil analysis conducted in September 2021 shows a high moisture content of 27.8% relative saturation. Additionally, it was found that the power factor test conducted in May 2021 had a reading of 2.67%, surpassing the recommended maximum value of 2.0% according to ANSI/NETA MTS-2023-Table 100-3 [2]. This indicates increased dielectric losses, which usually means the insulation system is deteriorating or contaminated
For the first step in the diagnosis, Duval’s pentagon analysis [4] was carried out, the results of which are presented below. It suggested the presence of partial discharge in the transformer.
Considering the increasing ethane and methane levels, the possibility of a thermal fault was also considered.
Legend:
PD: Partial Discharge
DT: Partial discharge, leading to a mixture of thermal and electrical faults.
D1: Low energy discharge
D2: High energy discharge
T1: Thermal fault below 300°C
T2: Thermal fault between 300°C and 700°C
T3: Thermal fault above 700°C
S: Stray gassing of mineral oil
The transformer’s SAHI dropped to 56% and then 40%, prompting a health case on November 1, 2021, recommending immediate removal from service. The owner opted for further confirmation, ordering another oil analysis on November 22 and December 10, which showed worsening results and a SAHI decline to 29%.
A third-party specialist performed an additional oil test on December 10, finding even higher gas levels and a further SAHI drop to 27%. After conducting detailed testing and analysis on December 16, they confirmed the critical condition and advised immediate shutdown for offline electrical testing and internal inspection.
The transformer was taken out of service on December 22, 2021, and a full inspection and testing were completed in January 2022, supported by system redundancy to prevent a plant outage.
During the inspection, poor transformer oil quality due to contamination and/or water was identified, which caused partial discharge (PD) in a transformer by lowering the oil’s dielectric strength.
Further findings and corrective actions during the offline inspection:
- The transformer oil was further analyzed, revealing a large amount of sediment and debris in it. Flakes of grey primer were discovered in the oil.
- The cause was identified as contaminants/moisture/ionizable species or particles in the oil, as no internal solid-insulation damage was identified in the component of the transformer.
- The transformer tank and windings were flushed with hot oil to remove suspected external contaminants and moisture.
- New clean transformer oil was used to replace the old oil.
- It was recommended to do the oil sampling of the transformer within 2-4 weeks of energization to monitor combustible gas generation and verify moisture test results, which has been going on for more than two years with increasing intervals, as no deficiencies were found.
- The last dissolved gas analysis (DGA) test in Summer 2025 showed normal dissolved gas concentration in the transformer.
Why did changing the oil of the transformer remove the PD activities?
- There was no physical damage to the internal component of the transformer, and its insulations were intact.
- Fresh oil lowers dissolved moisture, conductivity, and dissolved ionic/organic contaminants.
- Dissolved gases that feed PD are reduced/removed.
This health case highlights the effectiveness of the SAHI framework in integrating diverse data to accurately assess transformer condition and risk. By leveraging SAHI, organizations can prioritize maintenance and make informed decisions that enhance asset performance and reliability.
REFERENCES:
[1] Asset Performance Management, Transformer Monitoring, CIGRE Conference Sep.2023, Vancouver, Canada, By Reza Gol Mohamadi & Khaled Chaabani.
[2] ANSI/NETA, Maintenance Test Standard, ANSI/NETA, 2023.
[3] IEEE C57.104 Guide for the Interpretation of Gases Generated in Mineral Oil-Immersed Transformers, IEEE, 2019.
[4] Duval, M., Lamarre, L., “The Duval Pentagon – A new Complementary Tool for the Interpretation of Dissolved Gas Analysis in Transformers, IEEE Electrical Insulation Magazine 30, no 6 (2014):9-12.
Mr. Reza Gol Mohamadi is an electrical engineer with over 27 years of experience in power-system engineering, reliability, and project management across the energy sector. A recognized expert in power-system reliability, he founded EAMEC Canada Inc., a Calgary-based firm specialized in reliability engineering services, where he leads power systems reliability improvement projects that strengthen the performance and resilience of the power systems. Recently, Reza has focused on protection-system improvement, electrical obsolescence management, and advanced online monitoring of major assets—including power transformers, switchgear, large motors, and HV/MV cables—as well as developing a Static Asset Health Index using advanced analytics.
This article was originally published in the November 2025 issue of the Resilience of the Power System magazine.
View Magazine
