header foto

FIND-GO: research on smart asset management for grid operators

The Reliable Power Supply research group at HAN University of Applied Sciences (HAN) is conducting research on smart asset management for high-voltage and medium-voltage grids. The researchers hope to develop a next-generation health index that can be used to make decisions based on real-time measurements and predictions.

Photo: TenneT

An important part of grid management is asset management. That means managing the components that make up a power grid. As part of this process, equipment is closely monitored and replaced before defects occur. Grid operators work tirelessly to maintain a reliable power supply.

On 1 January 2019, HAN, TenneT, Stedin, GE, IWO and TU Delft launched a research project aimed at improving and future-proofing asset management. This project is called FIND-GO, which stands for ‘framework for intelligent net management and data-driven optimisation’. ‘The goal is to create a resilient power grid that can withstand the energy transition,’ says project leader Joke Westra.

While we don’t know exactly how the power grid will be used in the future, we do know that the energy transition will make the generation of power less consistent and harder to predict. ‘That’s the advantage of being able to use a more dynamic power grid, but one that’s still safe’

Health index

One of the ways grid operators inspect the equipment in the power grid is by using a health index. ‘They look at the age of the component, for example. This is a relatively good predictor of a device’s lifespan,’ says Westra. ‘They also conduct regular measurements, often on location. For example, every year they check the state of certain equipment and components to get a general idea of the overall health.’

The goal of FIND-GO is to further develop the health index. ‘Our vision is to create a second-generation health index that can be used for smart asset management. By smart I mean that it uses measurements and data more effectively. It also allows you to take better measurements, which will help you make better maintenance decisions about certain components.’

Good data not only tells grid operators when a component needs to be serviced or replaced, it also provides an incentive to use components differently. ‘In some cases, you can adjust the usage to the condition of the component in real-time.’

Adjusting component usage

‘One example is cable load: if you know that a cable gets very hot on sunny days, you may want to use it less for a while. If you have several cold cables, you can use those at 130 per cent, which helps you make more efficient use of the grid. This isn’t happening at the moment, but we hope to see it in the future.’

Good data and data analysis can give grid operators a better understanding of what will happen to components in the future. ‘We want to make a predictive health index with a built-in assurance. This means that in addition to seeing the current condition of the component, we’ll also see its predicted condition and the likelihood of that condition coming true. This is extremely helpful for scheduling maintenance.’

Making accurate predictions about power grid components is complex, Westra explains, because sometimes very little information is available. ‘You don’t just let components in a power grid break down, especially high-voltage ones. The bigger the transformers, the less failure data we have. We work on models that can help us make accurate predictions about the expected lifespan of a component, based on a small data set.’

Knowledge sharing

The FIND-GO researchers are sharing their knowledge of asset management with the field. Endowed Professor of Reliable Power Supply Rob Ross, who is closely involved in the project, even published the book Reliability Analysis for Asset Management of Electric Power Grids. Ross is an expert in this field, currently working as an R&D strategist for asset management at TenneT and as professor of Performance of High Voltage Energy Systems at Delft University of Technology.

The FIND-GO partners hope that the second-generation health index will generate industry-wide support in the long term. ‘We first want to see if we can reach a clear definition in the Netherlands and possibly throughout Europe. It would be extremely useful if we were all on the same page.’