This article addresses the issues of monitoring and warning flight risks in flight systems, considering four aspects: aircraft safety, overweight landing, factors leading to exceedances, and pilot flying techniques. Based on real-time flight-related data, an aircraft safety automated warning model is developed, and simulation is conducted. In the problem-solving process, various methods such as coefficient of variation analysis, random forest prediction model, and classification model based on GBDT are employed to identify key data items and features affecting flight safety. The proposed model shows promising results in evaluating flight risks and providing timely warnings, offering valuable insights and support to enhance aviation safety levels and is of significant value to flight management and planning.
GBDT-Based Aircraft System Detection and Warning Model
15.11.2023
474356 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch