Adaptive Infrared & Reinforcement Learning based advanced curing

Objectif

AUREL develops an adaptive infrared (IR) curing system for polymer and composite manufacturing that combines multi-zone IR heating with reinforcement learning (RL)–based real-time control. The project will raise the technology from TRL 4 to TRL 6, demonstrating faster, more energy-efficient, and more reliable curing compared to conventional convection or autoclave methods.

Current industrial curing processes rely on static or manually tuned heating profiles, leading to high energy demand, long cycle times, and inconsistent quality. AUREL introduces a data-driven digital twin of the curing process that models coupled thermal and kinetic behaviour and is linked to an RL controller trained to minimise curing time and energy use while preserving material quality. The controller interprets sensor feedback during operation and continuously adjusts IR power distribution in real time, ensuring uniform curing even under process variability.

The 24-month work plan integrates four technical stages: (1) development and calibration of the digital-twin simulation environment; (2) RL algorithm design and hardware-in-the-loop validation; (3) integration and pilot demonstration under industrial-like conditions; and (4) business validation and investor readiness. Parallel business activities will advance BRL 2 → 4, producing a market analysis, ROI model, and investment case for scaling.

Expected impacts include at least 30 % reduction in curing time, 25 % energy savings, and improved reliability of composite manufacturing. The project strengthens the innovation ecosystem of a Widening Country (Greece) through ESA BIC Greece incubation and industrial partnerships across the composites and coatings value chain. By project end, AUREL will deliver a pilot-validated, investment-ready adaptive curing platform positioned for EIC Accelerator or private-sector scale-up.