This is the Seetalabs technical library on power transformer condition assessment. It explains dissolved gas analysis, the CIGRE TB 761 health index, machine-learning methodology, the standards that govern it, and how real fleets were ranked by risk. Written so a non-specialist can follow the reasoning end to end, then decide for themselves. Everything here is grounded in CIGRE, IEC and IEEE references, never a black box.
No articles match your search. Try a broader term such as DGA, health index or CIGRE.
A live map of the diagnostic domain: the standards that govern it, the methods that interpret it, the gases that carry the signal, and the faults, oil and paper condition they resolve to.
This is the text equivalent of the knowledge graph above: the entities RONIN AI reasons over for power transformer diagnostics, and the relationships between them, grouped by cluster.
The transformer health index is built on CIGRE TB 761 and combines dissolved gas analysis, oil quality, paper DP and moisture into a single score used for fleet risk ranking. Dissolved gas analysis (DGA) is interpreted through IEC 60599, IEEE C57.104-2019, CIGRE TB 771 and ASTM D3612, and diagnosed with the Duval triangle, Duval pentagon, Rogers ratios, key gas and rate-of-change methods, each reading the seven fault gases: H2, CH4, C2H2, C2H4, C2H6, CO and CO2.
The Duval triangle is the core DGA fault diagnosis method: it classifies CH4, C2H2 and C2H4 ratios into partial discharge, thermal faults T1 to T3, low-energy discharge (D1) and high-energy arcing (D2). Thermal fault T3 correlates with high ethylene (C2H4), high-energy arcing (D2) with high acetylene (C2H2), partial discharge with hydrogen (H2), and stray gassing with methane (CH4) at normal operating temperature. The Duval pentagon adds ethane (C2H6) to separate stray gassing from genuine faults. Rogers ratios feed into the IEC 60599 interpretation, and rate-of-change trending follows the IEEE C57.104-2019 guide.
On the insulation side, paper DP is estimated from furans (2-FAL), carbon monoxide and carbon dioxide, and drives the remaining-life estimate. Oil quality follows IEC 60422 and is degraded by moisture. Bushings and the OLTC are tracked as separate failure-risk contributors. Fleet ranking, failure risk and maintenance action are the outcomes: the health index ranks the fleet by risk, and that ranking drives the recommended maintenance action.
Every article here explains the method. RONIN AI runs it on your own DGA and oil data and returns a CIGRE TB 761 health index for every asset, ranked worst first. In seconds. No sensors, no IT rollout.
No sales calls · guided by AI · single asset or full fleet, 500 kVA to 800 MVA
