Metrology for smart metering in gas networks

To meet the goals of the European Green Deal, the EU aims to establish a carbon-neutral energy supply by 2050. In particular, the REpowerEU plan outlines the urgent need to reduce dependence on natural gas imports and increase use of biomethane and hydrogen. However, the introduction of these gases into the grid will bring increased fluctuations due to the more diverse demand and supply. These must be managed carefully to ensure safety, quality and fair billing.

Current gas metering methods and models assume grid measurements are independent of each other. Partnership project Met4H2 showed this can lead to uncertainty being underrated by as much as 35 %. There is also a lack of FAIR (findable, accessible, interoperable and reusable) experimental data for modelling gas dynamics that includes relevant metadata. The use of artificial intelligence (AI) and machine learning (ML) to manage gas grids is being investigated but it is unclear whether such models can provide the required traceability.

This project will create FAIR open-access datasets for gas flow measurements which cover different dynamic scenarios and include relevant metadata. It will develop validated models for assessing correlation in gas metering data and supplement experimental data with synthetic data. This will be used to extend the software framework originally developed during Met4H2. The use of AI/ML and the requirements of “training data” will also be explored.

This work will improve the metering of gas networks, facilitating the uptake of renewable energy gases with improved safety and reliability to meet the targets needed to decarbonise the EU gas grid.