Short answer, yes.
Global enterprises rely on integration between different APIs to unlock the full potential of accessible technical services and knowledge about their asset’s performance. API integration is largely (mis)understood as a IT folk-talk. Here’s why you as a non IT manager with access to a capital asset like transformer would want to know more.
Transformer asset management
The safety and reliability of critical assets such as a power transformer is pivotal to ensure integrity of the energy supply chain. A power transformer contains dielectric oil (mineral or non-mineral) and cellulosic paper for insulation and isolation of windings against the active core. These materials decompose under thermal and electric stresses to produces various gases that are dissolved in oil. Hence, transformer state assessment requires adequate monitoring of such gas concentrations for uninterrupted operation.
Mineral oils extracted from petroleum crude is more common for transformer applications. It contains various hydrocarbons such as paraffin, naphthene, and aromatics. Under thermal and electrical stress, the carbon-carbon and carbon-hydrogen bonds of oil molecule breaks to form various gaseous by-products such as hydrogen (H2), methane (CH4), ethane (C2H6) etc. This resultant chemical reaction occurs in simultaneous multiple steps and the mechanism is quite complex. Additional thermal decomposition of cellulosic paper leads to the formation of carbon monoxide and dioxide and directly indicates its loss.
Dissolved gas analysis (DGA)
Dissolved gas analysis (DGA) is a method for identification of transformer faults by measuring concentration of by-product gases such as hydrogen (H2), methane (CH4), ethane (C2H6), ethylene (C2H4), acetylene (C2H2), carbon monoxide (CO) and carbon dioxide (CO2). It ensures safety and reliability of operating transformers by early detection of incipient faults1. It improves strategic predictive maintenance and optimizes operational costs. Various field, laboratory, and on-site gas analyzers operating on-line or off-line modes are now commercially available for this purpose. While it acts as a tool for basic risk management, DGA alone cannot quantify the asset risk and recommend actions over a defined timeline.
A technical brochure recommends the use of single gas (such as hydrogen) monitors on relatively “healthy” transformers2. However, to determine transformer “healthiness” without comprehensive DGA is challenging even by health indexing strategy.
The Duval’s method of DGA interpretation
In 1974, Dr Micheal Duval proposed a set of triangles for interpretation of DGA results for high voltage equipment, mainly transformers. International committees such as IEC, IEEE and CIGRE endorses all Duval triangles and pentagons for such purposes. There are three triangles (Tr1, Tr4, Tr5) and two pentagons available for DGA interpretation of mineral-oil filled transformers. Duval triangle 1 uses various gas ratios to determine low and high-energy discharge and thermal faults; whereas, Duval triangle 4 and 5 helps in identifying sub-fault types and their potential location. A summary of faults types and gas combinations applied to triangles 1, 4, and 5 is tabulated below.
| Gases | Tr 1 | Tr 4 | Tr 5 |
| H2 | • | ||
| CH4 | • | • | • |
| C2H4 | • | • | |
| C2H6 | • | • | |
| C2H2 | • |
| Fault type / Triangles (Tr) | 1 | 4 | 5 |
| Partial Discharges (PD) | Y | Y | Y |
| Electrical discharge of low energy (D1) | Y | ||
| Electrical discharge of high energy (D2) | Y | ||
| Thermal fault of temperature <300oC (T1) | Y | ||
| Thermal fault of temperature between 300oC and 700oC (T2) | Y | Y | |
| Thermal fault of temperature >700oC (T3) | Y | Y | |
| Mixture of discharge and thermal faults (DT) | Y | ||
| Stray gassing of mineral oil <200oC (S) | Y | Y | |
| Overheating <250oC (O) | Y | Y | |
| Hot spot with carbonization of paper >300oC (C) | Y | Y | |
| N/D | Y | Y |
The health index (HI) approach
The indexing strategies are popular industrial practices for prioritization and ranking of assets such as transformers in a fleet. While the purpose to deploy an indexing approach may vary, the target is almost always to determine suitable actions over a defined timeline. These score-weight matrix of transformer health index can be biased and often mask other pre-existing issues thus limiting its scope. Seetalabs’ Ronin AI is a web-based API with enhanced machine learning (ML) framework to predict the transformer HI score. Furthermore, DGA is one of the main components of Ronin AI’s performance. The existing dashboard gives a 360o view of assets in terms of existing health and potential (primary) faults based on triangle 1.
Moreover, without a data context health index score could be completely useless. While deploying data-driven technical services for the transformer industry, API integration could be a cost-effective solution to an asset manager’s conundrum. They can simultaneously work with segregated data silos without worrying about data pre-processing and/or management. This article brings the research outcomes of combining Ronin AI predictions with Duval’s results to put HI score into a context. The primary object for the client here was to review past maintenance actions and plan future strategies.
Case study
Table 3 summarizes the DGA data of three 66/18 kV transformers operating in a power range of 50-100 MVA. None of these transformers are overloaded or have communicating tap changers. The oxygen-to-nitrogen ratio was unavailable at the time of this study. Other oil quality data (breakdown voltage, moisture in ppm, acidity etc.,) was available to Ronin AI for making the health index (HI) prediction. The table also shows the results as seen on Ronin AI dashboard.
| Trf1 | Trf2 | Trf3 | |
| H2 | 12 | 19 | 1 |
| CH4 | 5 | 52 | 1 |
| C2H4 | 1 | 50 | 20 |
| C2H6 | 1 | 257 | 1 |
| C2H2 | 1 | 1 | 1 |
| CO | 197 | 425 | 165 |
| CO2 | 3966 | 2425 | 1728 |
| Fault | D1 | T2 | T3 |
| ID | HI (%) | Legend | Fault |
| Trf1 | 89.6 | Very Good | D1 |
| Trf2 | 69.8 | Fair | T2 |
| Trf3 | 88.4 | Very Good | T3 |
Experts believe that low-energy arching (D1) and low (T2) or high (T3) temperature hot-spots in paper are among the most dangerous faults in transformers. The above table suggests that Trf2 has a “fair” health with a score of 69.8% and low-temperature thermal fault (T2) as predicted by triangle 1. Further use of triangle 4 classifies the sub-fault type as low temperature (<250oC) overheating (O) fault without carbonization. This means that the paper has potentially not lost the withstand-ability and this may simply be the case of normal ageing. The testing frequency of this asset is recommended to increase from annually to half-yearly while emphasis on catching the gas generation rate. Since the oil quality analysis of this asset was normal, no immediate actions are necessary.
Conclusion
An operational asset could require either maintenance, repair, refurbishment and/or replacement recommendations. A health index approach delivers this over a defined timeline. While it is a ranking strategy after all, transformer health index can be improved significantly with data context. In this research, the requirement was to ascertain which maintenance actions are necessary to which asset, and when. The DTM-HI combined approach delivers the end goals succinctly.
This research also opens new paradigms for API integrations between Ronin AI and your in-house DGA calculators to obtain conclusive and comprehensive reports. Seetalabs’ tech lead will help you choose the best recommendations for such integrations.
If you wish to learn more, just contact us or grab your free trial right away!
Footnotes
- IEEE C57.104-2019 Guide for the interpretation of gases generated in mineral oil immersed transformers
- CIGRE TB 409 Report on gas monitors for oil filled electrical equipment