Collaboration Background
GAC Group faced an urgent need to improve vehicle powertrain reliability and reduce after-sales costs. Traditional diagnostic methods rely heavily on expert experience, making them inefficient, costly, and difficult to apply across complex operating conditions.
Solutions
1. Data-driven intelligent fault diagnosis: Based on vehicle operation data, a fault feature library was constructed and integrated with machine learning techniques to enable accurate early-stage fault identification in powertrain systems.
2. Intelligent transmission load spectrum development: By using AI to simulate real-world driving scenarios, high-precision load spectra can be rapidly generated to support optimization and validation of key components.
Research Objects – Four Vehicle Models
Mature Technical Foundation-User-Facing Front-End Interface
Significant Economic and Social Value
1. Substantially improved diagnostic accuracy: AI enables accurate identification of early-stage faults, reducing the risks of missed and false diagnoses and ensuring stable system operation.
2. Significantly improved fault troubleshooting efficiency: Manual troubleshooting time is shortened, downtime is reduced, and delivery and operation efficiency are improved.
3. Significantly reduced overall operating costs: Through fault early warning and optimization, after-sales compensation and validation cycles are reduced, achieving cost reduction and efficiency improvement.