AI generative design tool for low-cost district heating networks
A project to build building tools which can help automate the design of district heat energy networks, using artificial intelligence and geographic data.
City Science Corporation
£598,394 (across two phases)
Summary: impacts and findings
This project built a tool suite which uses artificial intelligence and geographic data to automate the design of district heat networks (DHNs), so that projects can be specified to the right scale for optimum efficiency and viability. It demonstrated that this approach can deliver faster, cheaper and more accurate plans and so accelerate rollout of DHNs.
Project aims and approach
Installing district heating for domestic and commercial property will play a key role in achieving decarbonisation targets. But heat network design is currently costly and time-consuming, and often lacks sufficiently detailed data to ensure system efficiency. In many cases this leads to over-sizing of planned equipment, which increases capital costs and makes projects unviable.
This project by City Science aimed to build generative design tools that will automate the optimal design of heat networks and so help accelerate the rollout of district heat.
Using an automated, computer-driven process, it sought to radically reduce the up-front costs of modelling and system sizing. The tool suite uses artificial intelligence techniques combined with detailed geographic data to optimise system design, increasing the affordability of heat networks and maximising their viability.
Phase 1 of the project was a feasibility study to establish need and outline specifications. The team worked with a selection of key local authorities, consultancies and other organisations including the Energy Systems Catapult. This confirmed the potential for a new application to improve the DHN design process and help enable more effective decarbonisation of domestic and commercial heating systems.
In Phase 2 the team confirmed detailed user requirements, refined the functional specification, clarified data needs and sources and built and tested the application. Four modular tools were delivered:
- Heat demand model/API
- Energy centre optimisation
- Automated network routing
- District heating network optimisation
You can watch a detailed ‘show and tell’ presentation and demonstration of the project, recorded in June 2022.
The show and tell from Phase 1 of the project, recorded in July 2021, is also available here.
City Science Corporation
April 2021 to May 2022
Achievements and barriers
The team has developed an application for designing district heat networks which delivers seamless data integration between design stages, provides automated data pipelines, allows data to be examined and analysed at each design stage, and enables multiple iterations to find the optimum solution.
The new toolset is expected to speed up DHN feasibility studies by up to 50% and significantly reduce costs, increasing the success rate of projects and supporting more rapid progress towards net zero targets.
City Science is now promoting the toolset and its modular component parts to key markets, for example consultancies working with local authorities. City Science’s will continue to update and improve the tool suite as part of its ongoing R&D programme.