Sheffield engineers harness weather forecasts for industrial green energy

Researchers at the institution’s Advanced Manufacturing Research Centre (AMRC) are addressing a key economic challenge: how to ensure that sustainable energy use does not come at the expense of productivity and efficiency.

While renewable energy use in the UK has grown considerably, key sources such as wind and solar power fluctuate according to climate conditions, leaving the grid reliant on contributions from more carbon intensive fossil fuels including coal and gas.

The AMRC created a system designed to coincide energy-intensive manufacturing tasks with periods of renewables use when carbon footprint is lowest, thereby reducing manufacturers’ carbon dioxide equivalent.

Previous work done by the centre had confirmed the link between weather behaviour and energy carbon intensity.

From this the AMRC created a machine learning model trained on historical data and energy production trends that was capable of forecasting carbon intensity over a 48 hour period.

Discrete event simulations (DES) using real-world manufacturing process data from AMRC North West were employed in order to validate the system’s effectiveness.

Simulations were filtered in order to gauge results for tasks including melding, cutting and machining processes. They also compared traditional production schedules optimised for speed with those prioritising minimal carbon emissions. 

One exercise demonstrated a significant reduction in overall emissions, of 4.74 kilograms of CO?E. However, this simulation also extended production time by 2.8 hours,

Dr Jon Hall, technical lead for connected factories at the AMRC, said: “The model's ability to accurately predict the maxima and minima in carbon intensity as a function of time is particularly valuable, as it helps identify the most and least sustainable periods for machine operation, and can also calculate the potential carbon dioxide equivalent reduction if sustainable scheduling was prioritised.”

The centre has created a web interface tool to enable manufacturers to use the technology to inform scheduling, offering the chance to assess trade-offs between production timelines and environmental impact.