AI project aims to supercharge hydrogen production in Scotland

AI project aims to supercharge hydrogen production in Scotland

A new project aims to use artificial intelligence to supercharge hydrogen production in Scotland, helping the country meet its net zero targets and powering thousands more homes and businesses each year.

Computer scientists at the University of Aberdeen and Aberdeen-based software company Intelligent Plant will use explainable AI (XAI) to develop a Decision Support System (DSS) to tackle shortfalls in production and help Scotland meet its target of 5GW of installed hydrogen production (equivalent to a sixth of the country’s energy needs) by 2030.

They are working in partnership with the European Marine Energy Centre (EMEC) on the project which has been funded through the Scottish Government’s Emerging Energy Technologies Fund.

There are complex logistical challenges involved in producing green hydrogen, which is generated from renewable sources and reliant on the vagaries of wind speed or tidal power.

Decisions aimed at optimising production are usually made by experts in the field based on experience, utilising traditional so-called ‘black box’ decision support systems that are unable to provide clear reasoning and are not fully trusted by users.

In order to overcome these shortcomings and the impact on production, researchers will use explainable Artificial Intelligence (XAI) that will allow operators to ask the system questions, receive feedback, and modify their approach if necessary.

Crucially, the use of XAI will enable trust by ensuring that decisions are explained clearly.

Professor Nir Oren, from the University of Aberdeen, commented: “A hydrogen production facility must balance myriad demands, particularly when operating using intermittent renewable energy, and consideration must be given to current and future forecasts for storage, consumption, energy availability and cost.

“An AI-based decision support system aims to take these multiple factors into account to optimise hydrogen production, but the system is only as good as the data it receives – so it is critical that decisions made by the system are explainable, that it can justify its decisions, and that the factors leading to the decisions can be understood and modified.

“In this project we build on ideas from the area of Explainable AI and more particularly formal argumentation theory, to enable users to interrogate the system and understand why it suggested specific courses of action.

“By taking this approach, the DSS will build trust amongst users that we hope will ultimately lead to an increase in the production of green hydrogen – an important factor in helping Scotland meet its net zero ambitions.”

The system will be trialled by operators at the Orkney-based European Marine Energy Centre (EMEC) using Intelligent Plant’s Industrial App Store, which will provide an easy and accessible interface for operators in the field.

Paul Gowans from Intelligent Plant will work alongside Professor Oren as part of the project. He commented: “The use of Intelligent Plant's Industrial App Store as an enabler for XAI will allow operators to better understand its system. It will allow for live connectivity to EMEC's sites and will enable the team to demonstrate how the AI system can be integrated with real systems and data to optimise energy management in a practical and scalable way.

“Our end goal is to create a DSS which can be used to make recommendations around hydrogen logistics, and whose recommendations can be queried and corrected as circumstances change.

“In the longer term we could look to extend this technology to benefit other renewable sources such as wind and solar, further increasing the impact of the project which has the potential to go some way to reaching Scotland’s renewable energy targets."

Search News

Browse by Month

2024

  1. Jan
  2. Feb
  3. Mar
  4. Apr
  5. May
  6. Jun
  7. Jul There are no items to show for July 2024
  8. Aug There are no items to show for August 2024
  9. Sep There are no items to show for September 2024
  10. Oct There are no items to show for October 2024
  11. Nov There are no items to show for November 2024
  12. Dec There are no items to show for December 2024

2023

  1. Jan
  2. Feb
  3. Mar
  4. Apr
  5. May
  6. Jun
  7. Jul
  8. Aug
  9. Sep
  10. Oct There are no items to show for October 2023
  11. Nov
  12. Dec

2013

  1. Jan There are no items to show for January 2013
  2. Feb There are no items to show for February 2013
  3. Mar There are no items to show for March 2013
  4. Apr There are no items to show for April 2013
  5. May There are no items to show for May 2013
  6. Jun
  7. Jul
  8. Aug
  9. Sep
  10. Oct
  11. Nov
  12. Dec