Data Centre Waste Heat Utilisation: A Closed-Loop Thermal Integration with Jacketed CSTR Systems

AI data centres in Sydney use large amounts of electricity for GPU cooling, yet the resulting low-grade heat is released to the surrounding and wasted. The proposed system sized a closed-loop glycol-water recovery system using thermal camera measurements, hierarchical scaling, SOLIDWORKS CFD, and standard conduction and convection heat-transfer equations. For a representative 16,000-GPU hyperscale facility (NVIDIA H100), about 9.25 MW of waste heat is recoverable, with coolant heated from 20 °C to roughly 37 °C per coolant distribution pipe and only approximately 755 W of heat lost over 199 m of insulated stainless-steel pipe. The heat is delivered to the jacket of a continuous stirred-tank reactor (CSTR) in a co-located manufacturing plant. Switching from air to liquid cooling at this scale drops annual cooling cost from about 51 to 30 million AUD and annual CO₂ emissions from 78.4 to 7.84 million kg. Payback is under one year.

Publication Details

Journal
The Sydney Journal of Interdisciplinary Engineering
Published
2026-11-18
DOI
https://doi.org/10.30722/t29tnb69
Primary Topic
Thermodynamic and Exergetic Analyses of Power and Cooling Systems
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Data Centre Waste Heat Utilisation: A Closed-Loop Thermal Integration with Jacketed CSTR Systems

3 links
The Sydney Journal of Interdisciplinary Engineering
Thermodynamic and Exergetic Analyses of Power and Cooling Systems
article

Data Centre Waste Heat Utilisation: A Closed-Loop Thermal Integration with Jacketed CSTR Systems

article en

Abstract

AI data centres in Sydney use large amounts of electricity for GPU cooling, yet the resulting low-grade heat is released to the surrounding and wasted. The proposed system sized a closed-loop glycol-water recovery system using thermal camera measurements, hierarchical scaling, SOLIDWORKS CFD, and standard conduction and convection heat-transfer equations. For a representative 16,000-GPU hyperscale facility (NVIDIA H100), about 9.25 MW of waste heat is recoverable, with coolant heated from 20 °C to roughly 37 °C per coolant distribution pipe and only approximately 755 W of heat lost over 199 m of insulated stainless-steel pipe. The heat is delivered to the jacket of a continuous stirred-tank reactor (CSTR) in a co-located manufacturing plant. Switching from air to liquid cooling at this scale drops annual cooling cost from about 51 to 30 million AUD and annual CO₂ emissions from 78.4 to 7.84 million kg. Payback is under one year.

The Sydney Journal of Interdisciplinary EngineeringVol. 2(1)
Responsible consumption and production, Affordable and clean energy, Climate action, Industry, innovation and infrastructure
Thermodynamic and Exergetic Analyses of Power and Cooling Systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon