From electrical generation to consumption, power transformers continue to play the important role of stepping down energy from higher to lower voltages. Whether located in a transmission or distribution level substation, a single transformer can serve thousands of customers’ needs or be dedicated to a large industrial plant, data center, hospital, or other critical operation.
Our dependence on transformers means that any failure can be highly disruptive to businesses and homes, as well as affecting safety. Not only can it represent a major power outage, like the one affecting 10,000 customers in Kansas City recently, but transformer failures can also cause fires, like the one in Windsor, Ontario, last year. And a fire is often a threat to human safety. A fire that resulted from a recent transformer explosion in Bangkok, Thailand, caused two deaths and many injuries. The fact that the explosion occurred 30 minutes after the local utility had visited the site shows that conventional inspections are not enough to prevent disaster.
Why Is this a Growing Problem?
Transformers are long-lived assets with a failure rate that increases significantly after about 30 years. The average age of distribution transformers in the USA is 42 years, while 70% of transmission line and power transformers are 25 years or older.
There is currently minimal monitoring of transformer assets in electricity networks on all but the largest assets due to the current cost and complexity of monitoring systems. This leads to a lack of visibility into the state of health and robustness of assets which, in turn, drives conservative asset management. Without timely, effective condition analysis, conservative management and heavy investment are needed to avoid catastrophic failure.
This conservative behavior includes retiring network assets early for the peace of mind that failures will be avoided. This strategy encourages overspending due to early replacement of transformer assets that may have years of useful life left.
However, there is now intense commercial and political pressure to drive down costs and improve infrastructure utilization without impacting business continuity and customer services. It is time to consider a more proactive management approach that will save capital expenditure by extending useful life.
As discussed in our report “The Digital Grid Unleashed,” utilities are now recognizing the potential benefits of smarter substations and other assets, with many already making such investments to improve network operations. IDC predicts that, by 2020, one in four utilities will integrate new sensor data and cognitive capabilities to boost their assets’ efficiency and reduce maintenance costs.
As part of this smart substation strategy, monitoring and maintaining the health of transformers has never been more important. However, in the past, accurately assessing risk factors has been a complex problem. Let’s have a look at why that has been the case.
What Are the Biggest Risks to Transformer Reliability?
The short answer is heat and moisture. These factors accelerate transformer aging. Specifically, high temperatures can cause the paper insulation inside transformers to break down. When this happens, the transformer is less able to withstand normal operating stress.
Moisture will exacerbate this aging effect. Perfectly dry Kraft paper operating at 80ºF should last over 20 years. But add 2% moisture, and that lifespan shrinks to only three years.
The Challenge with Measuring Transformer Health
Measuring transformer temperature is relatively easy, using sensors in the transformer windings or measuring top oil temperature. Measuring moisture is harder. Measuring the moisture content of the paper – where 95% of the moisture will be – is impossible while a transformer is in service, and measuring the moisture content of the oil can be prone to error.
There is a complex relationship between the moisture content of the paper and the oil, which you can read more about in our article here. Suffice it to say that when the two are in ‘equilibrium,’ there is a direct relationship between them. However, equilibrium rarely occurs in an active transformer. Measuring moisture is further complicated by factors such as the nature of the oil (age, pollutants, etc.), paper qualities, how heat affects paper differently than oil, moisture movement through layers of paper, and more.
Due to all of this dynamic activity, trying to understand the state of a transformer from a single oil test is like trying to understand the dynamic state of an entire power system by measuring only a single point – it’s simply not comprehensive or accurate enough.
The Key To Accuracy: Online Transformer Monitoring
Continuous monitoring can help average out the inaccuracies caused by the above dynamic conditions. A system that can continuously measure top oil temperature as well as “relative saturation or water activity” (see our article for explanation) can estimate the paper’s water content much more accurately and, consequentially, calculate paper age and life left.
However, due to the moisture-holding characteristics of aging paper, a fully accurate assessment requires performing age profiling at the top, bottom, and hotspot of the transformer. An advanced algorithm is required to perform these complex computations.
Using Smart Analytics to Forecast Transformer Lifespan
EcoStruxure Transformer Expert from Schneider Electric automates the complex analysis for the water content in the transformer paper, taking into account the requirements noted above. The web-based software uses data from the EcoStruxure Transformer Sensor, a low-cost device that measures multiple temperatures, water activity, vibration, and RF signals that indicate partial discharge.
The accurate analysis enables forecasting for paper aging and the remaining life based on current load profiles. The dashboard then allows the user to model other forecasts for aging based on chosen load profiles.
The ease-of-use and ability of EcoStruxure Transformer Expert to simplify these complex analyses can help asset managers:
- assess and forecast health and lifespan across entire fleets of transformers
- make data-driven decisions regarding maintenance and replacement based on transformer risk rankings rather than the costly, conservative approach of retiring assets based on their age.
For more information, download the EcoStruxure Transformer Expert brochure.
Add a comment