Shut-Off Valve Anomaly Detection Using Unsupervised Sensor Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Diagnosing abnormal operation of aircraft valves is time-consuming and challenging with conventional methods, which are often specific to certain types of valves and require manual intervention.
Innovation Solution
A data-driven, unsupervised algorithm analyzes sensor data using an n-second window to identify abnormal valve operation, generating real-time alerts and facilitating maintenance without requiring specific valve architecture knowledge or manual labeling of training data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional diagnostic methods are used for abnormal valve operation, then specific valve types can be diagnosed, but the process is time-consuming and requires manual intervention
Solution Approach 1:
The patent replaces manual diagnostic methods with an automated data-driven unsupervised algorithm that analyzes sensor data to detect abnormal valve operation. The system automatically processes sensor data, identifies abnormal patterns, and generates alerts without requiring manual intervention or specific knowledge of valve architecture, thereby reducing diagnosis time while maintaining detection accuracy.
Solution Approach 2:
The system enables self-diagnosis of valve abnormalities through automated algorithmic analysis of sensor data. The unsupervised learning algorithm independently identifies abnormal patterns in the data without requiring external expert intervention, allowing the system to autonomously detect and alert about valve issues.
2Reliability
If conventional diagnostic approaches are used, then specific valve types can be targeted, but the system lacks flexibility across different valve architectures
Solution Approach 1:
The patent creates a universal diagnostic system that can detect abnormal operation across different valve types and architectures. The data-driven unsupervised algorithm analyzes sensor data patterns that are common to various valve operations, enabling the same system to reliably detect abnormalities in different valve configurations without requiring type-specific customization.
Solution Approach 2:
The system adapts to different valve types by analyzing changes in sensor data parameters and patterns rather than relying on fixed architectural knowledge. The unsupervised algorithm identifies abnormal patterns through parameter variations in the sensor data, allowing it to accommodate different valve architectures while maintaining detection reliability.
3Loss of information
If manual diagnostic intervention is required, then specific valve issues can be identified, but the process becomes complex and resource-intensive
Solution Approach 1:
The patent replaces complex manual diagnostic processes with an automated algorithmic system that processes sensor data to identify complete diagnostic information. The unsupervised learning algorithm automatically extracts relevant information from sensor data, generating comprehensive diagnostic insights without requiring manual intervention or complex human expertise.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A computer-implemented method, system, and computer program product are provided. A plurality of maintenance messages (MMSGs) are identified. Each MMSG is associated with at least one shut-off valve. A sensor parameter is identified based on an analysis of sensor parameters associated with the shut-off valves of each MMSG. A threshold value for the sensor parameter is identified as being associated with abnormal operation of the respective shut-off valves. A sensor associated with a first shut-off valve captures values for the sensor parameter during a first and second predefined time period, the first and second predefined time periods associated with an opening and a closing of the first shut-off valve. Upon determining that a difference between the maximum values of the sensor values captured during the first and second predefined time periods exceeds the first threshold value, a determination is made that the first shut-off valve is operating abnormally.