Graduation Year


Document Type




Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Electrical Engineering

Major Professor

Wilfrido Moreno, Ph.D.

Co-Major Professor

Chung Seop Jeong, Ph.D.

Committee Member

Andrew Hoff, Ph.D.

Committee Member

Alex Volinsky, Ph.D.

Committee Member

Brig Bjorn, Ph.D.


Ambiguity, Detection, Isolation, Digital Twin, Threshold


Induction machines are the workhorse of many military and commercialtypes of equipment. They are asynchronous machines referred to in this dissertation as induction motors. They are widely used due to their reliability in Naval Unmanned Systems like Remotely Operated Vehicles (ROV), Unmanned Untethered Underwater Vehicles (UUV), Autonomous Undersea Vehicles (AUV), and platforms like Missiles and Satellites. These induction motors are mainly controlled using variable speed drive functions, i.e., thrusters, propellers, and actuators. Though they are very robust, to harness the full technical benefits of induction motors, state-of-the-art drivers or controllers must be used to control such motors.

In model-based fault diagnosis, i.e., fault detection, fault isolation,parameter characteristics monitoring also called Prognostic Health Monitoring (PHM) system, drivers or controllers must be used with induction motors of AUV systems to assess when to abort the mission due to malfunctioning hardware issues, as well as sonar detected obstacles to bring the AUV to the surface for recovery.

Induction motor drivers or controllers and model-based motor faultdiagnostic systems require exact knowledge of information about all the major parameter characteristics of the motor, which are usually not available. These characteristics may be functions of joule heating, skin depth, motor linear or non-linear region of operation, and environmental operating conditions, to name a few.

Novel Model-based Fault Diagnostic Systems are the primary goal of thisresearch. In order to validate the proposed novel approach, an induction motor-based system was selected to implement and test the unified approach developed and used in this doctoral dissertation. Such a unified approach comprises of the following items: a) Proposed Model-based Fault Diagnostic System theoretical foundation b) applicable Control Engineering Techniques required in Fault Diagnostic c) System Identification & Verification d) the Induction Motor-based Validation system. Each of these items is a vast field and each constitutes a field of advanced research. This dissertation addresses only their relevant aspects as applicable to advance the research carried out and presented in this dissertation. The driven motivation of this work, the Model-Based Fault Diagnosis through Induction Motors, is mainly to uphold the Department of Defense (DoD) strenuous effectiveness, safety, and performance system requirements [129].