Project Details
Description
• Model based calibration (Matlab/Simulink) of GPF automate in conventional and ePower concept vehicles to verify yearly customer service regeneration for china and European market.
• Data based Machine Learning models (TWC, GPF, CVT) from scratch with NARX and RNN structures to reduce calibration time (by 60%) using tools like ASCMO (ETAS) and Python.
• Emission Platform development using ML in collaboration with HiLs team for significant reduction in cost and time in powertrain calibration.
• Performance Optimization through Dynamic DOE (Design of Experiment) method using ASCMO tool.
• Automation of various calibration tasks and data analysis using Python (open source platform) bypassing conventional software license requirements.
• Data based Machine Learning models (TWC, GPF, CVT) from scratch with NARX and RNN structures to reduce calibration time (by 60%) using tools like ASCMO (ETAS) and Python.
• Emission Platform development using ML in collaboration with HiLs team for significant reduction in cost and time in powertrain calibration.
• Performance Optimization through Dynamic DOE (Design of Experiment) method using ASCMO tool.
• Automation of various calibration tasks and data analysis using Python (open source platform) bypassing conventional software license requirements.
| Status | Finished |
|---|---|
| Effective start/end date | 3/03/20 → 30/12/21 |
Keywords
- data driven models
- Machine learning
- Neural Networks
- Deep Learning
- Vehicle Calibration
- Model In Loop
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.