Western Power Distribution (WPD) becomes the latest to employ artificial intelligence (AI) to boost the accuracy of its forecasting.
The forecasting project - conducted by Smarter Grid Solutions as part of WPD’s Electricity Flexibility and Forecasting Systems (EFFS) project - used AI to improve the accuracy of load forecasting to better determine where and when flexibility services are required
Forecasts for 132kV and 33kV networks for supply points, bulk supply points, primary substations, large load customers and renewable generation were produced. The time frames examined ranged from six months ahead to one hour ahead.
The two techniques used to make predictions – long short-term memory and extreme gradient boosting – were found to be more effective than both the current most commonly used forecasting tool and techniques used in recent network innovation projects, SGS said.
Dr Graham Ault, executive director and co-founder of SGS, said: “All DSOs will require forecasting as a core input to flexibility services procurement and dispatch, so we expect the outcomes, techniques and specific tools used in this project to be widely referred and utilised. “
The technology is being made readily available to other DNOs to help facilitate the transition to DSO, with open-source tools and a ‘how to guide’ style of project report.
The results of the project are likely to be of most interest to Scottish Power Energy Networks and Scottish and Southern Electricity Networks, SGS said, due to their FUSION and TRANSITION projects, which are also investigating forecasting methods and flexibility.
National Grid ESO has also been testing AI as a method of improving forecasting, seeing results of a 33% improvement in solar generation output forecasting. And solar irradiance is now being predicted through a new platform from Meniscus.
Jenny Woodruff, EFFS project manager at WPD, said the DNO has plans to scale up the method to create forecasts for a larger number of sites.
“Being able to accurately forecast demand and generation will enable us to identify where and when our network will need flexibility services. The more accurate the forecasting, the more efficiently we can purchase and dispatch these services,” Woodruff continued.