Neural networks optimize resin transfer moulding flow control

Researchers have developed a novel control strategy for resin transfer moulding that utilizes neural network surrogate models to optimize auxiliary gate pressures. This method aims to ensure complete saturation of fiber preforms by preventing premature resin arrival at vents, thereby reducing dry-spot formation. The approach combines Kalman filtering for race-tracking estimation with neural networks to approximate complex finite element models, demonstrating significant improvements in filling efficiency on a forked geometry. AI

IMPACT This research could lead to more efficient manufacturing processes by improving control over resin flow in composite material production.

RANK_REASON The cluster contains an academic paper detailing a new methodology for optimizing an industrial process using AI. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv stat.ML →

• artificial neural network
• arXiv
• Bayesian approximation error framework
• Kalman filter
• Resin transfer moulding
• Ruanui Nicholson

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →