Neural Network Algorithms TRIZ Cases
- Device Modeling Engine for Low-Complexity Condition Monitoring
Machine learning matching networks infer device conditions from operation data, reducing sensor complexity while improving abnormality detection.
- Wind Farm Layout Optimization Using Physics-Informed Wind Prediction
Physics-informed neural networks reconstruct 2D wind fields from lidar data to improve ultra-short-term prediction and turbine layout.
- Wind Turbine Energy Estimation Using Infrared Blade Aerodynamics
Infrared images map wind-turbine blade flow regions and aerodynamic polar values to estimate energy production without routine manual inspections.
- Adaptive Wave Energy Control for Reliable Marine Sensor Power
Real-time wave data guides adaptive converter control, combining wave, solar, and battery power for reliable marine sensors.
- Adaptive Wave Energy Converter for Reliable Offshore Sensor Power
Independent arm assemblies convert wave motion into electricity while adapting to changing conditions for remote marine sensors.
- Wind Farm Power Prediction Using LSTM Neural Networks
A trained recurrent neural network correlates historical weather and power data with future forecasts to predict wind farm output.
- Wind Turbine Vertical Wind Profile Determination Using Trained Data Driven Model
A trained data driven model processes wind turbine measurement values to determine the vertical wind speed profile and shear coefficient.
- Wind Turbine Blade Damage Detection via Acoustic Spectrogram Analysis
Acoustic detection system captures wind impingement sounds to generate frequency spectrograms for automated blade damage recognition.