A hybrid machine learning framework for wind pressure prediction on buildings with constrained sensor networks

Keyphrases

• Machine Learning Framework 100%
• Sensor Networks 100%
• Wind Pressure 100%
• Hybrid Machine Learning 100%
• Wind Pressure Prediction 100%
• Wind Direction 66%
• Limited Sensors 66%
• Pressure Field Reconstruction 66%
• Machine Learning Models 33%
• Proposed Methodology 33%
• Urban Environment 33%
• Prediction Accuracy 33%
• Spatio-temporal Features 33%
• Generalization Ability 33%
• Sensor Data 33%
• Dimensionality Reduction 33%
• Novel Hybrids 33%
• Pressure Coefficient 33%
• Feature Mapping 33%
• Wind Tunnel Data 33%
• Structural Health Monitoring 33%
• High-rise Building 33%
• Dynamic Mapping 33%
• Sensor Configuration 33%
• Sensor Placement 33%
• Long Short-term Memory Network 33%
• Order Reduction 33%
• Digital Twin Application 33%
• Pressure Field 33%
• Superior Accuracy 33%
• Efficient Prediction 33%
• QR Decomposition 33%
• Right-sided 33%
• Structural Risk 33%
• Pressure Reconstruction 33%
• Pressure Estimation 33%
• Building Faade 33%
• Spatio-temporal Feature Extraction 33%
• Wind Pressure Distribution 33%
• Wind Pressure Coefficient 33%
• Key Stage 1 33%
• Robust Capability 33%
• Alternative Machines 33%
• Coefficient Field 33%
• Wind Sensor 33%

Engineering

• Learning System 100%
• Sensor Network 100%
• Pressure Distribution 100%
• Pressure Coefficient 50%
• Time History 25%
• Limitations 25%
• Dimensionality 25%
• Limited Number 25%
• Feature Extraction 25%
• Sensor Data 25%
• Structural Health Monitoring 25%
• Rise Building 25%
• Lstm 25%
• Pressure Reconstruction 25%
• Digital Twin 25%
• Wind Tunnels 25%

Chemical Engineering

• Learning System 100%
• Structural Health Monitoring 33%
• Long Short-Term Memory Network 33%
• Pattern Recognition 33%
• Feature Extraction 33%
• Digital Twin 33%