The ESOs ability to forecast electricity demand has reduced as a consequence of the COVID-19 pandemic. Historically, the underlying demand profile was driven primarily by day of week and time of day. That demand has become harder to forecast partly because working patterns have not returned to pre-pandemic levels and also because we are seeing new technologies such as Electric Vehicles (EVs) come onto the system. This project seeks to utilise mass mobility data (anonymised telematics vehicle monitoring data) to generate new features for electricity demand forecasting models. It will investigate potential features of value supporting electricity demand estimation and will use historical data to correlate and evidence potential predictive value from said features.
Benefits
This project has the potential to bring new insights into normal (and temporally abnormal) behaviour patterns concerning citizens in transit, home, workplace and state transitions across the population. The regionally resolved data will also provide features in regional forecasting that can support constraint management. In addition, the use of telematics derived features will provide complete spatial coverage over all of GB with minimal sampling bias due to its diverse geographic sampling. Though the project focus is not EV usage, it may be able to explore features related to EV charging behaviour and its impact on demand.
Learnings
Outcomes
Phase A of the project demonstrated that integrating key telematics data features into electricity demand forecasting models led to notable improvements in accuracy; overall, daily mean absolute percentage error (MAPE) was reduced by 0.9%. The use of telematics features proved particularly valuable for predicting demand on atypical days: even greater reductions in MAPE were observed when considering only public holidays (14.2%) and school holidays (7.1%).
Lessons Learnt
Interim Project Outcomes
Phase A of the project is now completed. It concluded that the integration of telematics data increases the accuracy of electricity demand predictions over the whole historic period tested. The improvement in performance was more marked when considering only special days, such as public or school holidays. Following these encouraging results, the extension of the project to phase B was approved.
Review of benefits case
A model integrating telematics features showed an overall improvement in predicting electricity demand when compared to the baseline demand forecasting model. If this is implemented and replicated in NESO’s operational demand forecasting model, it will help optimise decision making and therefore reduce costs related to system balancing and constraint management. The performance improvement in demand forecasting was especially noticeable for atypical days such as the Christmas festive period; if replicated operationally, this would have a significant impact as this period currently incurs high balancing costs, due in part to inaccurate demand forecasts.
Next Steps
The project was extended to the second phase. Phase B started in April 2026. It will focus on securing the provision of telematics data forecasts and on their deployment into NESO’s operational national demand forecast model. This will imply designing an API platform to supply historic and forecast telematics data and performing further feature assessment and refinement.
