Mathematical Baseline and Error Detection Techniques for the Analysis of Unaccounted For Gas (UAG).

Completed

Fossil Fuels: Oil Gas and Coal(Oil and Gas, Refining, transport and storage of oil and gas)

Applied Research and Development

PHYSICAL SCIENCES AND MATHEMATICS (Applied Mathematics)
ENGINEERING AND TECHNOLOGY (Mechanical, Aeronautical and Manufacturing Engineering)

Not Cross-cutting

Project Contact
National Grid Gas Transmission

Network Innovation Allowance

Ofgem

01 April 2017

01 April 2020

36 months

£113,000

Technical Consultancy

London

Network Innovation Allowance

Project Contact, National Grid Gas Transmission

To develop a comprehensive set of mathematical models to assist in the management of UAG through:1. The development of a robust suite of UAG baseline (tolerance) models. 2. The development of a suite of mathematical analysis techniques to assist in the on-going management of UAG. It is proposed to evaluate the programme in terms of the following success criteria: Baseline Determination:-The introduction of a baseline metric into the standard UAG management process will greatly assist in the assessment of the effectiveness of the control and monitoring processes being employed. This should enable the UAG management techniques to be refined and targeted in specific areas greatly improving response and holistic management with all stakeholders. Mathematical Analysis Techniques:-The mathematical techniques will build on the baseline analysis to provide a suite of tools and models that will be embedded into National Grid’s UAG management processes.

National Grid is responsible for the management of Gas Transmission Shrinkage of which the major component is the Unaccounted for Gas (UAG) element and contributed £43m to total shrinkage (community) costs in 2014/15. UAG is mainly attributed to meter error as a consequence of measurement uncertainty but it can also be as the consequence data error, whether systematic or individual occurrences. Thus the detection of errors within the measurement arena resulting in UAG is very problematic. While there is considerable work by asset owners and National Grid to maintain high levels of metering management and best practice to improve measurement and minimize data error, UAG is still a feature of NTS Shrinkage. To gain further understanding of the dynamics of UAG, a dedicated mathematical study is to be undertaken to explore the ‘music in the numbers’ by exploring the latest mathematic analysis techniques. These techniques will sit outside the standard data mining activities and will incorporate the use of non-linear dynamical (Chaos) analysis, matrix and Bayesian methodologies. This study will through a set of defined methods, aim to provide a deeper understanding of observed UAG providing analysis techniques which will improve its management and minimization. The programme will be conducted in conjunction with the Mathematical Department of Manchester University and will concentrate on the use of mathematical techniques to provide a deeper understanding of UAG behaviour. The programme will be initially focused on providing an assessed UAG baseline and expected tolerances around that baseline for the NTS. From this baseline analysis, the programme will aim to deliver a range of additional mathematical techniques to further assist in the management of UAG by improving measurement/data error detection. Defining UAG behaviour in terms of a robust baseline will assist the focus of UAG management improving flexibility and capability. Deliverables: Analytical Development of a robust baseline methodology for expected UAG as a consequence of the NTS operation. {C}a. Use basic site flow data and the UAG calculation to determine a robust evaluation of a likely tolerance band for expected NTS UAG volumes. Calibrate baseline analysis against historical behaviour. Provide base models to embed in National Grid’s BAU UAG analysis tools suite. Development of UAG analysis techniques. Provide a range of desktop mathematical techniques to support the on-going analysis of UAG. Calibrate model and techniques against known measurement anomaly data sets. Provide a suite of modelling techniques and methods to embed in the proposed National Grid ‘UAG analysis HUB’. Deployment User analysis of the models. Continuously re-calibrate models against near term and historic data sets. Embedding the techniques within the BAU UAG management processes.Note : Project Documents may be available via the ENA Smarter Networks Portal using the Website link above

12/09/18