Dear Customers and Industry Colleagues,
In early October the Unidentified Gas (UIG) Task Force published an executive summary of their Sprint 2 findings. The team is now pleased to share the most recent findings from Sprint 3.
Since the implementation of Project Nexus in June 2017, gas shippers have experienced much higher than expected absolute levels and volatility of UIG. This is severely affecting their ability to predict demand and commercially manage their businesses from an immediate cash-flow perspective, because UIG is reconciled (corrected) over an extended and unknown future period. In July 2018 Ofgem approved the UNC Modification 0658 to drive a more centralised and focussed approach to the resolution of UIG, mandating Xoserve as the Central Data Service Provider to take on a leadership role on behalf of the industry. I’m pleased to confirm that the third sprint of the UIG Task Force completed earlier this week.
Sprint 3 Findings
The priority focus for Sprint 3 was to carry out more detailed and complex modelling. This allowed us to delve further into the current components linked to historic UIG volatility, as well as introduce new components to the model, with a view to pinpoint the problem areas that are driving volatility.
To this end, applying the use of machine learning to the model confirmed that the Non-Daily Metered (NDM) algorithm correctly accounts for wind speed, day of week and holiday factors. However, overall, the use of machine learning showed that it wasn’t possible to deliver materially better results based on the current inputs. This has indicated that as a next step we need to focus on the inputs to the model, rather than the mechanics of the model itself.
Having now procured additional weather data items (e.g. precipitation, solar radiation), we are using these in Sprint 4 to test whether their inclusion in NDM estimation processes would make a material difference to UIG.
A significant finding is that within the NDM demand sample dataset there are a small number of sites, which have a measured consumption far greater than the Annual Quantity (AQ) recorded on our systems. This can occur where sites have had a usage change and the shipper needs to increase the AQ to reflect the real consumption. There could also be erroneous historic meter readings or incorrect site set up data, which would need to be resolved. These outliers alone have been shown to have a marked impact on UIG levels and, depending on the End User Category (EUC) profile allocated on our systems, this impact would affect either UIG base, volatility or both.
Furthermore, we have identified a number of large consuming sites within Classes 3 or 4 (Non-Daily Metered), which all have AQs above the Class 1 (mandatory Daily Metered) AQ threshold of 58.6m kWh. In the absence of Xoserve receiving daily meter reads for these sites, there is a likelihood that if these sites’ actual usage patterns are different from the NDM allocation profile (assumed to be fairly flat for EUC Band 9), for example a large hospital that uses more gas in the winter and less in the summer, then this would likely contribute to daily UIG volatility. We are drilling down to understand the root causes behind each site, in order to work with customers to suggest correction actions, as well as drawing out longer-term industry recommendations over the coming weeks.
We can confirm that the review of the geographic clustering of the NDM sample sites shows that we have a good representation across the country.
Sprint 4 will continue to probe the NDM algorithm through the use of additional weather data inputs, and through further investigations into the NDM sample data set to determine how representative it is of the whole market. In relation to out-of-date or missing AQs, we are now simulating historic allocation variances caused by the AQ changes of the NDM EUC Band 9 sites, to see if the change in EUC profile or increased AQ level has an impact on UIG. Using the sample data set, we will simulate the AQ calculations using different read frequencies (monthly, bi-annually and annually) to see what the impact could be to both base and volatility UIG. This may support recommendations in regard to read frequencies and/or EUC bandings to help support the reduction of UIG. Our investigation into the accuracy of daily metered nominations has been carried forward from Sprint 3 and will be concluded in this sprint.
Building on the work of the Allocation of Unidentified Gas Expert (AUGE), which is investigating the impacts of the standard volume-to-energy conversion factor (1.02264, as prescribed in the Thermal Energy Regulations) on UIG, we have modelled the difference between actual and standard outside temperature on daily volume conversion, for a single colder than average Local Distribution Zone (LDZ). This analysis supported the Allocation of Unidentified Gas Expert (AUGE)s findings that standard temperature conversion contributes to positive UIG in winter and negative UIG in summer. The analysis also showed that the annual effect is non-zero, i.e. that winter under-recording of actual energy does not fully offset the summer over-recording of actual energy. On this basis, these investigations will continue into Sprint 4 to model the impact on warm and average LDZs, and to assess national impact on AQs and therefore UIG.
Please click here to see more detail on the full scope of Sprint 3.
We will continue to provide monthly updates at the DSC Change Management Committee. The Investigation Log on our website provides further detail on all Task Force activities. If you have any further questions or comments, please contact us email@example.com.
Ranjit Patel – Chief Customer Officer