LLM-based multi-scale imputation of missing sensor data in structural health monitoring

Key Points

• Improved accuracy in imputation of missing sensor data, which is crucial for monitoring structural health.
• A key improvement was observed, achieving 20% better data recovery compared to traditional methods across various scales.
• Analysis of sensor data utilizing machine learning techniques showcased the potential benefits of multi-scale imputation.
• Highlights the need for further validation in real-world applications to ensure reliable structural health assessments.