Master’s Thesis: AI-Powered Sensor Fault Detection at University of Technology

The University of Technology’s Department of Electrical Engineering awarded a master’s degree to Ahmed Mohammed Abd for his thesis titled:
“Pattern Recognition-Based Robust Sensor Fault Detection and Isolation Using Artificial Intelligence.”

The thesis aimed to develop a monitoring system for sensor performance degradation. Using data from SG-10 geophone sensor arrays, the study extracted key features, including resistance, noise, leakage, and slope, to identify sensor faults. The research employed advanced pattern recognition techniques like Support Vector Machines (SVM), Quadratic SVM (QSVM), k-Nearest Neighbors (KNN), Decision Trees (DT), and Artificial Neural Networks (ANN).

Ahmed’s findings demonstrated the highest accuracies with ANN (99.4%), QSVM (97.2%), and KNN (100%). The models were trained and validated using MATLAB 2020a. These results highlight the efficiency of the proposed algorithms in real-world applications.

The examining committee included:

• Dr. Raad Sami Fayadh (Chair)
• Dr. Azad Badr Said (Member)
• Dr. Qusay Salem Tawfiq (Member)
• Dr. Abbas Hussein Issa (Supervisor).