Aircraft Sensor Fault Detection Using Residual Failure Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aircraft pitot tubes are prone to measurement failures due to icing blockage, heavy water ingestion, and volcanic ash blockage, leading to common mode pneumatic events that corrupt airspeed displays and cause deviations from optimal flight paths, necessitating an improved sensor fault detection and identification technology.
Innovation Solution
A method and system using residual failure pattern recognition, which involves sensing data from multiple sensors, performing majority voting, generating estimated values, calculating residuals, and analyzing patterns to detect and identify faulty sensors, while removing known corruption effects and utilizing statistical filters like the extended Kalman filter to generate alert signals and synthetic data signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple pitot tubes are used for airspeed measurement, then measurement reliability is improved, but the risk of common mode failure increases when blockage occurs
Solution Approach 1:
The system segments the airspeed measurement function across multiple independent sensors (pitot tubes, static ports, accelerometers, GPS) and processes their data separately through individual statistical filters before integration. This segmentation allows the system to identify and isolate failed sensors while maintaining measurement capability through remaining functional sensors.
Solution Approach 2:
The system changes the parameter representation by transforming raw sensor measurements into statistical filter outputs and residuals. By monitoring parameter changes in the residuals and their patterns across different sensors, the system detects faults and distinguishes between random failures and common mode events affecting multiple sensors simultaneously.
2Reliability
If sensor fault detection is implemented, then flight safety is improved, but system complexity increases due to additional processing requirements
Solution Approach 1:
The system implements self-service through autonomous fault detection and identification using statistical filters that automatically monitor sensor data, generate residuals, and identify failures without pilot intervention. The system self-diagnoses sensor health status and can switch to alternative measurement sources or generate synthetic data to maintain flight control accuracy.
Solution Approach 2:
The system employs feedback mechanisms where statistical filters continuously compare expected sensor behavior with actual measurements, generating residuals that feed back into the fault detection logic. This feedback loop enables real-time monitoring and automatic identification of sensor failures, allowing the system to adapt and maintain safety without increasing operational complexity for the flight crew.
3Loss of time
If real-time fault identification is performed, then response time is improved, but computational load increases
Solution Approach 1:
The system applies partial action by using statistical filters that process only the essential features of sensor data needed for fault detection. Rather than performing exhaustive analysis of all sensor parameters, the system focuses computational resources on calculating residuals and identifying patterns indicative of failures, achieving real-time detection with reduced computational burden.
Data Source
AI summary
Systems, methods, and apparatus for sensor fault detection and identification using residual failure pattern recognition are disclosed. In one or more embodiments, a method for sensor fault detection and identification for a vehicle comprises sensing, with sensors located on the vehicle, data. The method further comprises performing majority voting on the data for each of the types of data to generate a single voted value for each of the types of data. Also, the method comprises generating, for each of the types of data, estimated values by using some of the voted values. In addition, the method comprises generating residuals by comparing the estimated values to the voted values. Further, the method comprises analyzing a pattern of the residuals to determine which of the types of the data is erroneous to detect and identify a fault experienced by at least one of the sensors on the vehicle.