Normalised Diagnostic Contribution Index (NDCI) Integration to Multi Objective Sensor Optimisation Framework...

Area of Science:

• Aerospace Engineering
• Systems Engineering
• Data Science

Background:

• Effective sensor optimisation is crucial for diagnostics, resource management, and maintenance in aerospace.
• Existing methods may not fully address application-specific needs or diverse stakeholder requirements.

Purpose of the Study:

• To integrate the Normalised Diagnostic Contribution Index (NDCI) into the Multi-Objective Sensor Optimisation Framework (MOSOF).
• To enhance sensor optimisation for aerospace systems, considering multiple stakeholder perspectives (OEM, Airlines, MRO).
• To improve diagnostic efficiency and reduce sensor redundancy through a novel optimisation approach.

Main Methods:

• Utilised a multi-objective genetic algorithm within the MOSOF.
• Derived NDCI from simulation data of a Boeing 737-800 Environmental Control System (ECS) using the SESAC platform.
• Analysed sensor performance across four fault modes and evaluated against the Minimum Redundancy Maximum Relevance (mRMR) method.

Main Results:

• The NDCI-MOSOF identified an optimal set of three sensors, outperforming mRMR's six-sensor solution.
• Demonstrated significant improvements in diagnostic efficiency for all stakeholders.
• Expanded the solution space for sensor configurations, prioritising diagnostic value over mere redundancy.

Conclusions:

• The enhanced NDCI-MOSOF provides a scalable, multi-stakeholder solution for next-generation sensor optimisation.
• This approach is vital for advancing predictive maintenance in complex aerospace systems.
• The integration offers a distinct advantage in identifying optimal sensor sets tailored to specific diagnostic needs.