Advancements in Predictive Modeling and Machine Learning

The field of predictive modeling and machine learning is moving towards increased integration of numerical simulations, experimental validations, and machine learning techniques. Researchers are exploring the application of digital twins, reduced-order models, and large language models to advance thermal management, time series forecasting, and structural health monitoring. A key trend is the development of novel frameworks that combine the strengths of different approaches, such as the integration of physics-based models with supervised machine learning. Noteworthy papers include: Predictive Digital Twins for Thermal Management Using Machine Learning and Reduced-Order Models, which presents a scalable and interpretable framework for thermal management in automotive systems. ChronoSteer: Bridging Large Language Model and Time Series Foundation Model via Synthetic Data, which introduces a multimodal time series forecasting model that leverages both temporal and textual information for future inference.

Sources

Fault Diagnosis of 3D-Printed Scaled Wind Turbine Blades

ONERA's CRM WBPN database for machine learning activities, related regression challenge and first results

Predictive Digital Twins for Thermal Management Using Machine Learning and Reduced-Order Models

ChronoSteer: Bridging Large Language Model and Time Series Foundation Model via Synthetic Data

Informed Forecasting: Leveraging Auxiliary Knowledge to Boost LLM Performance on Time Series Forecasting

Built with on top of