Unaccounted for Gas

What is unaccounted for gas and why does it matter?

Unaccounted for gas refers to the difference between the volume of gas entering the NTS and the volume delivered to end-users. Physical gas leaks in the NTS are rare; UAG is a measurement bias due to metering and measurement inaccuracies, data inconsistencies or telemetry errors, opposed to actual physical loss of gas. The need for managing UAG is internationally recognised. The US Environmental Agency and the UK’s Office of Gas and Electric Markets emphasise that whilst UAG is not a direct proxy for methane emissions, it reflects the integrity of metering and data systems.

Due to the complexity and density of the NTS, it is really difficult to isolate sources of error using traditional telemetry-based methods.

What does the project aim to do?

This project is a proof-of-concept that will develop a predictive tool that identifies and quantifies UAG across the NTS. Analysis of historic data alone is not enough to understand the problem and therefore some amount of predictive modelling is required to understand sources of uncertainty and error in the system. CFMS will create a predictive mathematical model for UAG, demonstrating the model’s ability to isolate and attribute sources of UAG. The tool will leverage historic data to detect anomalies in metering behaviours and localise potential sources of UAG at the site or meter level. Over the course of the project, over one billion data points will be analysed, helping to form the model. The project will also validate the model’s accuracy and functionality.

What is the outcome?

This proof-of-concept will help to enhance operational efficiency, improve data transparency and support long-term goals through better system visibility and control. The project looks to develop a working prototype that proves the viability of the approach, highlights any data or modelling gaps, and lays the foundation for broader deployment.  It will also support the long-term goal of integrating predictive UAG analysis into business-as-usual operations.