Across the various condition monitoring options available for transformer health, Online Dissolved Gas Analysis (ODGA) has become the go‑to early warning system. Yet too often, alarm logic still relies on static ppm thresholds that ignore a unit’s context, leading to nuisance alarms and operator fatigue. This article presents a practical, standards‑aligned methodology moving from fixed limits to transformer‑specific baselines, population statistics, and noise‑robust Rate‑Of‑Change (ROC) and explains how to cut false positives without compromising sensitivity. It draws on advice from IEC 60599, IEEE C57.104-2019, CIGRE TB 403 & TB 771 & TB 783, and recent industry experiences.
Beyond fixed limits: Baseline driven, trend focused Online DGA alarm strategy – Part II
This is the second part of an article which examines why many ODGA systems generate excessive or unreliable alarms when traditional fixed thresholds are applied to high-resolution data.
Read the Article
To read the article, subscribe and choose the option which suits you best. We offer both free and paid options, and registration takes only a minute.
