Prognostic System for Aircraft Fault Prediction and Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional prognostic systems can predict impending faults in aircraft engines but often fail to avoid imminent faults, leading to costly cancellations, diversions, or delays, as they do not provide sufficient time to alter operational sequences without requiring immediate maintenance.
Innovation Solution
A method that uses a random forest classification model to analyze both parametric and non-parametric data from prior and upcoming operational sequences to predict faults, allowing for adjustments to the upcoming sequence, such as reducing aircraft weight, to minimize the likelihood of faults occurring, thereby deferring maintenance and avoiding costly disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional prognostic systems predict faults using fleet statistics or materials-based models, then fault prediction capability is improved, but the system cannot avoid imminent faults and requires immediate maintenance, leading to costly flight cancellations and delays
Solution Approach 1:
The system performs preliminary actions by altering operational sequence characteristics before the fault occurs. By predicting the fault timing and modifying operations in advance (such as adjusting engine parameters or operational conditions), the system prevents the fault from manifesting, thereby avoiding flight cancellations and maintenance delays while maintaining reliable fault prediction capability
Solution Approach 2:
The system changes operational parameters of the upcoming operational sequence based on predicted fault timing. By adjusting parameters such as engine load, temperature, or pressure within safe limits before the predicted fault occurs, the system extends component life and prevents fault occurrence, thus maintaining both prediction reliability and flight availability
2Measurement precision
If materials-based prognostic systems are used to estimate remaining useful life, then prediction precision is improved, but development cost and system complexity increase significantly
Solution Approach 1:
The system introduces an intermediary operational sequence modification layer between the simple fault prediction model and the physical system. Instead of using complex materials-based models directly, the system uses a simpler prediction model to forecast faults and then modifies operational sequences as an intermediary mechanism to prevent faults, achieving high precision predictions with low system complexity
Solution Approach 2:
The system replaces complex materials-based prognostic models with a data-driven prediction approach combined with operational sequence modification. Instead of relying on complex physical models of material degradation, the system uses predictive analytics on operational data and prevents faults by adjusting operations, thereby achieving high precision with reduced complexity
3Ease of manufacture
If data driven prognostic systems are deployed to predict faults, then development cost is reduced and applicability is widened, but the system cannot prevent imminent faults and still requires immediate maintenance
Solution Approach 1:
The system performs preliminary actions by modifying operational sequences before faults occur. Using data-driven prediction to identify upcoming faults, the system adjusts operational parameters in advance to prevent fault manifestation, thereby transforming a passive prediction system into an active fault avoidance system that maintains low development cost while improving reliability
Solution Approach 2:
The system implements feedback by using predicted fault information to adjust upcoming operational sequences. The prediction outcome feeds back into operational decision-making, creating a closed-loop system where operational characteristics are continuously optimized based on predicted fault risks, enabling fault avoidance while maintaining cost-effective data-driven architecture
Data Source
AI summary
A method, apparatus and computer program product are provided to not only predict an impending fault, but also to avoid the occurrence of the fault such that continued operations are permitted with a reduced likelihood of the occurrence of the fault. In this regard, a plurality of features are provided to at least one model, such as a random forest classification model. The plurality of features include features representative of at least one prior operational sequence as well as features representative of at least one upcoming operational sequence. The plurality of features are then processed with at least one model to determine a likelihood of a fault during the upcoming operational sequence. The method also alters the characteristics of the upcoming operational sequence without requiring maintenance of the equipment to thereby permit the upcoming operational sequence to be conducted with a reduced likelihood of the fault.