Härryda Energi, the distribution system operator east of Gothenburg, has taken the step from reactive to predictive grid operations. Eneryield IntelliView® now runs on all outgoing 10 kV lines at the Solsten substation in Mölnlycke — detecting, locating and predicting faults in real time.
Measurement hardware from Gothenburg-based EcoPhi streams waveform data at 2 MHz resolution — millions of measurements per second, where conventional grid monitoring samples in the kilohertz range. That resolution lets the AI hear the earliest signatures of an evolving fault: micro-transients from degrading insulation, vegetation contact, or components approaching failure.
“We can now not only locate faults but also predict them. We want to work proactively to prevent costly failures,” says Lukas Lukiewicz, grid analyst at Härryda Energi.
What it means for utilities: this is the full-digitalization route in practice — AI fault prediction running on a live distribution grid, with substation digitalization hardware supplied through a partner. For a DSO, the result is grid reliability work that happens on schedule instead of on call-out, and a measurement layer that keeps paying back beyond fault prediction alone.
Read more in our Härryda Energi case study.