Column: Condition monitoring
A technical guide to condition monitoring: Where to begin?
When you own a transformer, it is often challenging to decide whether you have the data you need to confidently make critical asset health decisions. The very volume and variety of information at your fingertips – dissolved gases and load data, for example — can be overwhelming. You think you are looking at an important data trend that is going to help you understand the condition of a key asset in your fleet, but in reality, after thorough analysis and thought, you realize it does not really tell you anything useful. A spike in dissolved gas analysis (DGA) may reflect a bad transformer, but it could also indicate a load spike, a bad sensor or a communication glitch. You think you have found something “interesting,” but how do you really know?
Condition monitoring provides near ‘real time’ information which supports both short-term condition decisions and long-term asset replacement plans. Condition monitoring can be a huge benefit – but you have to be prepared when that “interesting” data comes in: both how to deal with it and how to respond.
For successful condition monitoring, you need to be able to gather and analyze data from individual assets or stations. This means creating a scalable program that can grow with you as your needs and conditions change – the unit you are monitoring today may be different from the one you need to pay particular attention to a couple of years down the road, at which point you might need a comprehensive look at all transformers at that particular station.