Valve Fault Prediction Using Bayesian Prefault Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fault detection methods for complex systems, such as aircraft, rely on preventative maintenance schedules that do not account for the specific state of the system at any given time, leading to extended unscheduled maintenance periods due to unexpected faults in minor subsystems.
Innovation Solution
A Bayesian framework is used to monitor and analyze parameters like temperature, pressure, and flowrate to generate feature metrics, allowing for the detection of prefault states in subsystems, enabling scheduled maintenance and reducing downtime by identifying potential faults before they occur.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional preventative maintenance schedules are used, then maintenance can be performed regularly, but the system cannot account for specific system state and leads to extended unscheduled maintenance periods
Solution Approach 1:
The patent applies preliminary action by defining and detecting a 'prefault state' that occurs before actual system failure. By monitoring parameters and identifying characteristic patterns in this preliminary phase, the system can schedule maintenance in advance, transforming unscheduled maintenance into planned maintenance and eliminating extended downtime.
Solution Approach 2:
The patent implements feedback by continuously monitoring system parameters, comparing them against learned prefault patterns, and providing early warning signals. This closed-loop feedback mechanism enables the system to adapt to specific system states and predict faults before they occur, resolving the contradiction between regular maintenance and system-specific conditions.
2Measurement precision
If physical models of subsystems are used for fault prediction, then accurate predictions can be made, but computing resource expenditure increases and flexibility in sourcing replacement parts decreases
Solution Approach 1:
The patent applies copying by creating a statistical representation of the prefault state based on historical data from multiple operational legs. Instead of using complex physical models, the system learns characteristic parameter patterns from actual system behavior and uses these statistical copies to predict future faults, significantly reducing computing requirements while maintaining accuracy.
Solution Approach 2:
The patent substitutes mechanical/physical modeling approaches with a data-driven statistical approach. By replacing complex physical models with learned parameter patterns from operational data, the system achieves fault prediction with reduced computational complexity and increased flexibility.
Data Source
AI summary
Systems and methods for fault prediction through a Bayesian framework are provided. Fault prediction for a valve system may be provided by generating a Bayesian framework by collecting a plurality of historical parameters related to opening and closing of a valve across a plurality of operational legs; generating a plurality of historical feature metrics based on the plurality of historical parameters; in response to detecting a fault, defining a prefault state corresponding to the historical feature metrics; monitoring a plurality of operational parameters related to opening and closing of the valve during a given operational phase of an operational leg; generating a plurality of operational feature metrics based on the plurality of operational parameters monitored during the given operational phase; and in response to determining, using the generated Bayesian framework, that the operational feature metrics indicate the prefault state of the subsystem, generating a notification.