A Model-Based Method for Estimating Faults in Atmospheric Data
By employing a model-based fault estimation method and utilizing dual-model filters and Kalman filtering techniques, the problem of fault identification and isolation in atmospheric data systems was solved, achieving accurate fault detection and estimation, and improving system availability and aircraft safety.
Patent Information
- Application Number
- CN202310289718.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-03-23
- Publication Date
- 2026-03-06
- Estimated Expiration
- 2043-03-23
AI Technical Summary
Existing technologies struggle to identify fault types and characteristics in atmospheric data systems, especially in common-mode faults where it is difficult to properly isolate faults, leading to inaccurate flight parameter outputs and impacting aircraft safety.
A model-based fault estimation method is adopted. By constructing a dual-model filter, using unscented Kalman filtering for filter updates, calculating model probabilities, and identifying fault information sources through residual component chi-square detection, the filter is selectively initialized to achieve accurate estimation of airspeed, angle of attack, and sideslip angle.
It effectively reduces hardware redundancy requirements, enables timely detection and identification of fault information sources, and improves the availability of the atmospheric data system and the safety of the aircraft.