Date of Award

5-2026

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Aerospace, Physics, and Space Sciences

First Advisor

Camilo A. Riano-Rios

Second Advisor

Madhur Tiwari

Third Advisor

Ryan T. White

Fourth Advisor

Donald Platt

Abstract

Fault detection in aircraft is traditionally handled through redundant hardware and comparison algorithms to detect failures. Alternatives like model-based residual generation and data-driven approaches such as supervised fault classification and unsupervised anomaly detection have been explored, but they suffer from practical limitations; model-based methods require accurate system models, and data-driven methods have large constraints on the data limiting scalability and adaptability. This work presents a purely data-driven neural network architecture featuring a custom first layer designed for real-time fault detection where the weights and biases of this layer are used to detect faults. The network requires zero supervision and complements existing redundant-sensor systems by extracting fault-relevant features directly from sensor data. Statistical analysis shows a strong correlation between network weight changes and failure severity, supporting its effectiveness. A demonstration flight case study further validates the system’s real-time applicability. This work offers a scalable, low-overhead solution that bridges critical gaps in current aircraft fault detection strategies, enhancing reliability and operational safety.

Download

Share

COinS