AI method combines optimal transport and deep learning for industrial stiction detection

Researchers have developed a new method for detecting control valve stiction in industrial processes by combining optimal transport imaging with deep covariance alignment (Deep CORAL). This approach converts control loop signals into 2D images that capture dynamic behavior. A convolutional neural network is trained to learn domain-invariant features by optimizing both a cross-entropy loss on simulated data and a Deep CORAL loss on unlabeled industrial data, effectively mitigating domain shift. The methodology demonstrated high accuracy, diagnosing 18 out of 20 industrial loops with 100% recall for stiction cases. AI

IMPACT This novel AI approach promises more robust and reliable detection of control valve stiction in real-world industrial settings, improving process efficiency and reducing oscillations.

RANK_REASON The cluster contains an academic paper detailing a new methodology for industrial process control using AI techniques. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• CNN
• convolutional neural network
• Coral
• Deep Covariance Alignment
• Hugging Face
• optimal transport

• paper
• other

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