Aircraft Event Association Networks for Cascading Fault Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for diagnosing faults in aircraft systems are inefficient and inaccurate, particularly when dealing with cascading events across multiple subsystems, leading to extraneous efforts during maintenance procedures.
Innovation Solution
A system that includes a fault diagnostics computing device connected to aircraft subsystems via a communication bus, which generates events from sensor and fault code data, calculates statistical dependence between these events, and constructs a temporal precursor network to represent causal chains, facilitating accurate diagnosis and maintenance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing predictive and prognostic models are used to diagnose aircraft component failures, then some prediction capability is provided, but the diagnostic accuracy and efficiency deteriorate when dealing with cascading events across multiple subsystems
Solution Approach 1:
The patent segments the diagnostic process by separating sensor data processing from fault code analysis, and further segments the causal relationship analysis into discrete event chains. Each subsystem's events are analyzed independently first, then integrated to form complete causal chains, improving both accuracy and efficiency in diagnosing cascading failures
Solution Approach 2:
The patent introduces a temporal dimension to the diagnostic analysis by establishing chronological sequences of events across different subsystems. By analyzing events in temporal order and identifying causal relationships over time, the system can accurately trace cascading failures through multiple subsystems while maintaining diagnostic efficiency
2Loss of information
If comprehensive sensor data and fault code data are collected from all aircraft subsystems, then complete diagnostic information is obtained, but the complexity of analyzing and identifying root causes increases
Solution Approach 1:
The patent extracts only the relevant events from the comprehensive sensor data and fault code data by identifying specific anomaly thresholds and fault code patterns. Instead of analyzing all data, the system extracts and focuses on events that meet predefined criteria, significantly reducing analysis complexity while maintaining information completeness for root cause identification
Solution Approach 2:
The patent performs preliminary processing of sensor data and fault code data by pre-defining anomaly thresholds, event generation rules, and causal relationship criteria before actual diagnostic analysis. This preliminary structuring of data and analysis frameworks reduces the complexity of real-time diagnostic operations while ensuring complete information is captured
3Measurement precision
If maintenance personnel manually analyze cascading event chains across multiple subsystems, then thorough investigation is possible, but the time and cost required for diagnosis increases significantly
Solution Approach 1:
The patent implements automated feedback mechanisms where the system continuously monitors sensor data and fault codes, automatically generates events, identifies causal relationships, and provides real-time diagnostic results. This automated feedback loop eliminates manual analysis time while maintaining high precision in root cause identification through systematic event chain analysis
Solution Approach 2:
The diagnostic system performs self-service by automatically analyzing its own collected data without requiring continuous manual intervention. The system autonomously generates events from sensor and fault code data, identifies causal relationships, and produces diagnostic conclusions, significantly reducing both diagnostic time and cost while maintaining accurate root cause identification
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In an example, a method for identifying associated events in an aircraft is described. The method includes obtaining sensor data, obtaining fault code data, generating a set of events, where each event occurs over a time interval over which either (i) the sensor data indicates an anomalous measurement or (ii) a fault code associated with a particular aircraft subsystem of the aircraft was signaled, calculating a value of statistical dependence between the set, based on the value exceeding a threshold, constructing a network representing the set as a sequence of related events and further representing a temporal order in which the sequence occurred, indexing, in a summary table stored in memory and separate from the sensor data and the fault code data, the sequence and the value, and controlling a display device to display the summary table and a visual representation of the network.