Description of the Thesis

This thesis aims to evaluate various anomaly detection methods, namely Machine Learning, Deep Learning, Statistical, and Hybrid approaches, applied to sensor data streams. The focus is on an Electrohydraulic Actuator (EHA) test bench.

The objective is to scientifically compare these methods, providing valuable insights into the most effective approach for our EHA test bench. The evaluation will consider factors such as detection accuracy, computational efficiency, and robustness against noise.

The ultimate goal is to guide the selection of the most suitable anomaly detection technique for our specific application, enhancing the reliability and efficiency of our sensor-based systems.

Task:

The student’s primary the tasks include identifying and evaluating various anomaly detection methodologies applied to sensor data adapted to an Electrohydraulic Actuator (EHA) test bench. Student will design and validate these methodologies, with the ultimate goal of selecting the most suitable one for enhancing the reliability and efficiency of the EHA test bench.

Profile:

• You are studying Mechatronics Engineering, Computer Science, Data Science, or a related field,
• With an interest in anomaly detection and sensor data analysis.
• The candidate should have an initial understanding of machine learning, deep learning, and statistical methods or eager to learn.
• You work purposefully and independently.
• You are interested in the design.

If you are interested, I look forward to hearing from you.