News, News
20 Years Connected – What the Grid Has Taught Us
May 2026
Spring and summer have come and gone, and developments in the power grid haven’t slowed down.
Electrification continues, the share of renewable power generation keeps growing, and the demands on a robust and reliable power grid keep rising. At the same time, the digitalization of substations continues, with condition monitoring, AI, and data analysis becoming increasingly important tools for ensuring high operational reliability and efficient management of critical infrastructure.
Over the spring and summer, several international publications have touched on the same area. CIGRE released technical reports during this period — one on condition monitoring and predictive maintenance of HVDC converter stations, and another on digital twins for power transformers. IEEE has published research on how AI can contribute to greater reliability, condition monitoring, and better decision support in power electronics. Meanwhile, ENTSO-E’s Summer Outlook from May points to a power system undergoing rapid change, where the growing share of renewable generation and the increasing need for flexibility are placing new demands on system operation and planning.
For us, this confirms a development that has been clear for several years now.
Predictive maintenance is no longer a future project — it’s becoming a natural part of how modern substations are managed.
1. The big challenge is no longer collecting data — it’s using it
A few years ago, digitalization was mainly about installing sensors and starting to collect measurements. Today, many grid owners already have access to large volumes of data from their facilities.
So the critical question has shifted.
How do we turn information into the right decisions? Which deviations require immediate action? Which assets can keep operating? When should maintenance be scheduled to minimize both risk and cost?
This is where predictive maintenance creates its real value.
2. AI becomes a decision-support tool — not a replacement
AI is also taking on a clearer role as a tool for condition monitoring, lifetime assessment, and decision support in power electronics and energy systems. But this development isn’t about replacing experience — quite the opposite.
The greatest benefit comes when AI is used to identify patterns, catch early-stage deviations, and prioritize the right actions at the right time.
The technical expertise of operations and maintenance staff will continue to be essential. What’s changing is that decisions can now be based on far larger volumes of data and faster analysis than before.
3. Operational reliability matters more than ever
The electrification of industry, transportation, and society means the power grid is taking on an increasingly central role. Every substation becomes part of a larger system, where the consequences of an unplanned outage can be significant.
Catching deviations before they turn into failures is therefore about more than just more effective maintenance. It’s about security of supply.
The coming years will likely be decided not by who collects the most data — but by who uses it best.
At its core, predictive maintenance is about making better decisions. When maintenance can be planned based on actual equipment condition rather than fixed intervals, it creates better conditions for high operational reliability, longer asset lifespans, and a more resilient power grid.
This is a development that has already begun — and one that will continue to shape the industry in the years ahead.
Note: these sources are used as background for the analysis, not as direct summaries of the reports’ conclusions.