Summary
The intense computational demand of artificial intelligence (AI) and machine learning workload in modern datacenters is increasing energy consumption into the gigawatt (GW) range and requires significant upscaling of energy distribution networks and related components. Distribution transformers are critical in these networks and need to be upgraded to meet the challenges related to installation and operating conditions in this environment.
Read more Read lessNext to transient voltage resilience due to vacuum circuit breaker (VCB) operation on the medium voltage (MV) side, enclosure design for installation on a SKID-platform, thermal management of the transformer enclosure and a dedicated interface for efficient and time saving connection to low voltage busways are amongst others major requirements to be fulfilled. In this work a dry-type transformer design solution is presented, that is resilient to transient surges based on the shifting of natural transformer series resonance frequencies above the range of the expected frequencies common to operation of VCB and is upgraded and optimized to meet the mentioned requirements inherent to hyperscale datacenter environment.
Additional informations
| Publication type | Session Materials |
|---|---|
| Reference | A2_12374_2026 |
| Publication year | |
| Publisher | CIGRE |
| Country | Czech Republic |
| Study committees | |
| File size | 817 KB |
| Price for non member | 30 € |
| Price for member | 30 € |
Authors
SVOBODA Michal - SGB Group, s.r.o.; SPECKNER Lukas - SGB Germany; BROCILO Drazen - META
Keywords
AI-DATACENTER, DISTRIBUTION TRANSFORMER, SKID- MOUNTED SUBSTATION
