API integration can make condition data easier to use. Whether it helps prevent a failure depends on the data, the interpretation and the action taken.
APIs let asset-management systems exchange measurements and assessment results. For managers responsible for power transformers, this can reduce manual data handling and make diagnostic information available within existing workflows.
Transformer asset management
Safe, reliable operation of power transformers supports the integrity of the energy supply chain. An oil-filled power transformer contains dielectric liquid (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 produce various gases that are dissolved in oil. Monitoring those gas concentrations is one part of assessing the transformer; it cannot guarantee uninterrupted operation.
Mineral oil, refined from petroleum, is widely used in 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 molecules break to form various gaseous by-products such as hydrogen (H2), methane (CH4), ethane (C2H6) etc. Gas formation involves multiple reactions whose rates depend on the local conditions. Thermal decomposition of cellulose can produce carbon monoxide and carbon dioxide, but oil oxidation also contributes these gases. Carbon oxides alone do not establish the extent of paper damage; furans and other evidence help investigate cellulose involvement.
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 can support early investigation of developing faults1. The results can inform maintenance planning when they are interpreted with the other condition evidence. 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.
Single-gas monitors can be useful for detecting changes in a selected indicator, but they do not provide the information needed for all DGA interpretation methods. Monitor selection should follow the asset’s criticality, known concerns and diagnostic purpose; a low reading from one gas is not a complete assessment of transformer health.
The Duval’s method of DGA interpretation
Michel Duval developed graphical methods for interpreting DGA. For mineral-oil transformers, Triangle 1 uses methane, ethylene and acetylene; Triangles 4 and 5 refine selected initial classifications using different gas combinations. IEC and IEEE describe the scope and limitations of gas interpretation, while the linked CIGRE brochure addresses gas monitors. Before plotting a triangle, check absolute concentrations, analytical quality and gas evolution. IEEE C57.104-2019 permits Triangle 4 after PD, T1 or T2 in Triangle 1, and Triangle 5 after T2 or T3; neither should refine a D1 or D2 result. A zone suggests a fault type, without proving its presence or location.
| 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
Indexing strategies help prioritize assets for attention. A transformer health index can combine selected observations into a ranking, but aggregation may hide a serious individual indicator unless the method addresses that problem. The historical Ronin AI implementation combined health-index estimates with a Triangle 1 view. Together these provide two views of the available data; they do not cover every transformer component or establish an intervention deadline by themselves.
An API can connect those views with laboratory and asset records. It still requires explicit units, sample dates, liquid type, equipment identity and data validation. In this case, the aim was to compare historical maintenance observations with index and DGA outputs. Integration makes that comparison easier to reproduce; it does not remove the work of checking whether the data supports the diagnosis.
Case study
Table 3 reproduces reported gas concentrations for three 66/18 kV transformers operating in a power range of 50–100 MVA. The account reports no overload or communicating tap changers. O2/N2, sample dates, gas-generation trends and laboratory detection limits are not available in the published case. Other oil-quality inputs were reportedly available to RONIN, but are not reproduced. The displayed D1, T2 and T3 entries are historical zone outputs, not confirmed faults. Several concentrations are very low, making analytical uncertainty and the screening step essential before interpreting the zones.
| 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 |
| Reported zone (unconfirmed) | D1 | T2 | T3 |
| ID | HI (%) | Legend | Reported zone (unconfirmed) |
| Trf1 | 89.6 | Very Good | D1 |
| Trf2 | 69.8 | Fair | T2 |
| Trf3 | 88.4 | Very Good | T3 |
The historical output for Trf2 combines a 69.8% index score with a T2 zone. T2 denotes a thermal classification in the 300–700 °C range; it does not by itself establish paper involvement. The reported follow-on Triangle 4 result is also a classification hypothesis, not proof that the paper retains its mechanical strength or that only normal ageing is occurring. IEEE C57.104-2019, clause 5.2.1, cautions against fault identification from low concentrations without confirming accuracy. With no dated gas trend or complete screening context in this case, the published data cannot justify a six-month sampling interval or a statement that no immediate action is needed. A qualified engineer should review the laboratory limits, concentrations, changes and other condition evidence first.
Conclusion
A health index and a DGA plot can help an engineer organize follow-up. They cannot, from these tables alone, select repair, refurbishment or replacement or set a safe deadline. The case illustrates why an integration must retain both the input evidence and the conditions under which an interpretation is valid. If those conditions are not met, the report should identify what remains undetermined.
API integration between RONIN and an in-house DGA calculator can make the assessment process more traceable by bringing measurements, screening results and interpretation together. Seetalabs can help define those data flows and the checks needed before a result reaches an asset manager.
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
