Mechatronics Systems Engineering - Theses, Dissertations, and other Required Graduate Degree Essays

Receive updates for this collection

4D in situ visualization of chemo-mechanical membrane degradation in fuel cells: Understanding and mitigating edge failures

Author: 
Date created: 
2020-08-17
Abstract: 

Fuel cell is a zero-emission energy conversion device using hydrogen and oxygen to generate power with water as the only by-product. Membrane electrode assembly (MEA) edges are sensitive regions that could influence the overall durability of fuel cells, where membrane degradation at poorly designed edges may lead to premature cell failures. In this work, two MEA edge designs were implemented to study their robustness during combined chemical and mechanical accelerated stress testing. Four-dimensional in situ visualization, enabled by X-ray computed tomography, was performed to understand and mitigate the edge failure issue. Interaction of adhesive-containing polyimide gasket with catalyst coated membrane (CCM) was identified as the key contributor to premature edge failures, which was mitigated by using a non-adhesive inert frame at the CCM interface, thus enabling a robust MEA edge wherein the failures were shifted into the active area. Overall, findings of this research may contribute to robust fuel cell manufacturing and enhanced membrane durability.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Erik Kjeang
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Towards the vision of a social robot in every home: A navigation strategy via enhanced subsumption architecture

Author: 
Date created: 
2020-06-29
Abstract: 

In this thesis, we report the studies undertaken in the design and implementation of a behavioristic navigation system for social robots with limited sensors to be deployed in family homes. The project was completed in four phases. Each phase of the project was independently evaluated in virtual or real-time implementation on the NAO humanoid robot. In the first phase of this research study, we address the problem of indoor room classification via several convolutional neural network (CNN) architectures. The main objective was to recognize different rooms in a family home. We also propose and examine a combination model of CNN and a multi-binary classifier referred to as Error Correcting Output Code (ECOC). In the second phase, we propose a new dataset referred to as SRIN, which stands for Social Robot Indoor Navigation. This dataset consists of 2D colored images for room classification (termed SRIN-Room) and doorway detection (termed SRIN-Doorway). The main feature of the SRIN dataset is that its images have been purposefully captured for short robots (around 0.5-meter tall). The methodology of collecting SRIN was designed in a way that facilitated generating more samples in the future regardless of where the samples have come from. In phase three, we propose a novel algorithm to detect a door and its orientation in indoor settings from the view of a social robot equipped with only a monocular camera. The proposed system is designed through the integration of several modules, each of which serves a special purpose. Finally, we report an end-to-end navigation system for social robots in family homes. The system combines a reactive-based system and a knowledge-based system with learning capabilities in a meaningful manner for social robot applications.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Ahmad Rad
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) Ph.D.

Natural graphite sheet heat sinks for power electronics

Author: 
Date created: 
2020-06-04
Abstract: 

In this thesis, a multi-disciplinary investigation of using natural graphite sheet (NGS) for heat sink applications is presented with focus on thermal performance, electromagnetic performance, reliability, cost, energy efficiency, and environmental impact. NGS heat sinks are a promising alternative for weight-sensitive applications in which the heat sink is protected by a case. Contrary to the conventional metals, NGS is also predicted to be feasible at high temperatures or in corrosive environments. To provide the basis for the heat sink design, the thermal conductivity, thermal diffusivity, electrical conductivity, thermal emissivity, coefficient of thermal expansion, and compression behavior are measured and reported in an easy-to-use form. It is shown experimentally that the the thermal contact resistance at metal-NGS interfaces is comparable to metal-metal ones with thermal interface materials, and that the poor through--plane thermal conductivity can be mitigated by embedding heat pipes in NGS heat sinks. The conducted common-mode electromagnetic emissions cannot be reduced by using NGS heat sinks, but potential to reduce the radiated emission by 12 to 97 % was identified. Complex implications on reliability arising from replacing conventional metal heat sinks with NGS ones are discussed. The cost of NGS heat sinks produced in high volumes is predicted to be a double that of mass-produced conventional aluminum ones. The environmental impact of production, manufacturing, and end-of-life management of NGS is reviewed and compared to the conventional heat sink materials. A case-specific approach to evaluating the feasibility of using NGS heat sinks is recommended and the major steps are outlined. The technology is considered to be ready for a transfer to the industrial research and development stage. An audiovisual summary of the work is available at https://www.youtube.com/playlist?list=PLaX55SIXaD20NQQ2JLP-7abmET7l-6LS4.

Document type: 
Thesis
File(s): 
Video summary: 01 - Introduction
Video summary: 02 - Cooling systems, research motivation and goal
Video summary: 03 - Material properties of natural graphite sheet
Video summary: 04 - Thermal performance of NGS heat sinks
Video Summary: 05 - Electromagnetic performance of NGS heat sinks
Video Summary: 06 - Energy efficiency and environmnental impact
Video summary: 07 - Feasibility of NGS heat sinks
Video summary: 08 - Conclusions
Supervisor(s): 
Majid Bahrami
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) Ph.D.

Design and testing of a novel power-assisted wheelchair system

Author: 
Date created: 
2020-05-25
Abstract: 

A power-assisted wheelchair is a hybrid between a manual and power wheelchair that consists of an electric-assist system that can be easily mounted on a manual wheelchair. These devices have a demonstrated benefit on the health and mobility of wheelchair users. However, current power-assisted wheelchairs are not addressing all user needs, and as a result there is room for improvement. In this thesis, a novel power-assisted wheelchair system was developed using Stanford Design Thinking. Design requirements were developed using ISO 13485. Concept designs were iterated and a prototype was fabricated. The result is the NeuwDrive, a lightweight power-assist system. The NeuwDrive demonstrates novelty through the use of a right-angled geared motor and a hub design that maintains the overall wheelchair width and allows for easy removal of the drive system. The functionality of the NeuwDrive was verified in two ways. First, the performance was tested using an absorption dynamometer to measure torque and speed. The test results were within the specifications of class-leading devices on the market. The weight of the NeuwDrive is 10.2 kg, below any currently available hub-motor products. Second, a focus group with power-assist wheelchair users was conducted to collected end-user feedback. The results were favourable, with participants favouring the low device weight, removable batteries and narrow width of the NeuwDrive. The results of the testing indicate that the NeuwDrive is a novel power-assist system with potential for future development.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Carolyn Sparrey
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Human gait monitoring using wearable fabric-based strain sensors and deep supervised learning

Author: 
Date created: 
2019-12-05
Abstract: 

Continuous lower body monitoring is an important step for real-time feedback training of runners and in-home rehabilitation assessment. Optical motion capture systems are the gold standards for gait analysis, but they are spatially limited to laboratories. Recently, wearable sensors have gained attention as unobtrusive methods to analyze gait metrics and health conditions. In this study, a wearable system capable of estimating lower body joint angles in sagittal, frontal, and transverse planes during gait was developed. A prototype with fiber strain sensors was fabricated. The positions of the sensors on the pelvis were optimized using a genetic algorithm. A cohort of ten people completed 15 minutes of running at 5 different speeds for gait analysis by our prototype device. The joint angles were estimated by a deep convolutional neural network in inter- and intra-participant scenarios. In intra-participant tests, root mean squared error (RMSE) and normalized root mean squared error (NRMSE) of less than 2.2° and 5.3 %, respectively, were obtained for hip, knee, and ankle joints in sagittal, frontal, and transverse planes. The RMSE and NRMSE in inter-participant tests were less than 6.4° and 10%, respectively, in the sagittal plane. The accuracy of this device and methodology could yield potential applications as a soft wearable device for gait monitoring.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Carlo Menon
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Thermal performance of heat and water recovery systems: Role of condensing heat exchanger material

Date created: 
2019-09-11
Abstract: 

There is enormous potential for recovering a significant amount of latent heat at temperatures below 100°C from flue gas of combustion-based heating systems due to the presence of water vapor in their exhaust streams. However, condensation of acids along with water vapor in heat and water recovery systems makes a highly corrosive environment, which is a major challenge and a determining factor in selecting suitable materials for condensing heat exchangers. Despite the low cost and great corrosion-resistant properties of plastics, their relatively low thermal conductivities are not ideal for thermal management systems. it is still uncertain how significantly increasing thermal conductivity of the heat exchanger’s material affects thermal performance of the heat recovery systems. The present study aims to shed light on the effect of the thermal conductivity of a condensing heat exchanger’s material on the thermal performance of the unit. For this purpose, an analytical model is developed to predict the thermal performance of condensing heat exchangers, designed for recovering heat and water from wet flue gas. Further, to validate the model, a custom-designed condensing heat exchanger with replaceable tubes is designed in our lab and tested with 304 stainless-steel tubes and FEP plastic tubes under different inlet conditions. For the range of inlet conditions considered in this study, results show that there is a threshold for the thermal conductivity of the material, at which increasing the conductivity any further does not affect the condensation efficiency notably. It is worthy of note that this threshold, with respect to thermal conductivity of commonly used materials for such heat exchangers, has relatively low magnitude (e.g.~10-15 W"∙" m-1"∙" K-1 for stainless steel). This finding is significantly important as it unlocks the potential of using materials such as plastics and polymers with thermally conductive additives for latent heat recovery from flue gas.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Majid Bahrami
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Textile-based electromagnetic soft strain sensors for fast frequency movement and their application in wearable devices measuring multi-axial hip joint angles during running

Date created: 
2020-03-05
Abstract: 

Wearable in situ multi-axis motion tracking with inductive sensors and machine learning is presented. The production, characterization, and use of a modular and size adjustable inductive sensor for kinematic motion tracking are introduced. The sensor was highly stable and able to track high frequency (>15Hz) and high strain rates (>450%/s). Four sensors were used to fabricate a pair of motion capture shorts. A random forest machine learning algorithm was used to predict the sagittal, transverse, and frontal hip joint angle using the raw signals from the sport shorts strain sensors during running with a cohort of 12 participants against a gold standard optical motion capture system to an accuracy as high as R2 = 0.98 and an RMSE of 2° in all three planes. This present study provides an alternative strain sensor to those typically used (piezoresistive/capacitive) for soft wearable motion capture devices with distinct advantages that could find applications in smart wearable devices, robotics, or direct integration into textiles.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Carlo Menon
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Head impact kinematics and location measurement utilizing multiple inertial sensors

Date created: 
2019-09-24
Abstract: 

Impacts that cause high accelerations of the head are a major cause of mTBIs (mild traumatic brain injuries) or concussions. Repetitive head impacts can also potentially cause long term damage to the brain and cognitive abilities. Recently, due to increasing public awareness, wearable technologies and devices targeted towards measuring head impact kinematics during sport are gaining popularity. However, existing devices come with limitations that prevent on-the-field usage in one way or another. In this work, we devised methods to address the problem of accurately measuring impact kinematics and impact location, while also addressing the limitations of existing devices. We developed novel calibration and impact measurement algorithms that allowed us to design a complete impact measurement device; while also reducing the number of sensors and scale. We also proposed a wearable device prototype that can eventually be developed into a low-cost finished product for on-the-field impact measurement. We tested the accuracy of the device and algorithms by comparing the impact linear and rotational acceleration, rotational velocity, and impact location estimate with an industrial-grade IMU and a Hybrid-III dummy Head. Results showed that the device has great potential for relatively low-cost sports applications and can help in establishing a link between impacts and resulting brain injuries in the future.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Farid Golnaraghi
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Self-tuning electromagnetic vibration systems

Date created: 
2019-05-06
Abstract: 

This thesis presents the efforts made toward making industrial mechanical vibration systems smarter. This objective is accomplished in two steps. The first step is realization of mechanical vibration actuators that can mimic the behavior of mechanical dampers and springs with variable and controllable damping and stiffness. The second step includes the design and implementation of algorithms that can find the optimum damping and stiffness in different operating conditions. First, electromagnetic actuators are selected for force generation. It is shown that creating a parallel RL circuit with variable parameters in the shunt of an electromagnetic actuator results in variable damping and stiffness behavior by the actuator. It is shown that this circuit configuration can be realized using a power electronics converter connected to a power source. Next, automatic control methods are developed for adding a self-tuning loop to the system including an electromagnetic actuator. To this end, the sliding mode extremum seeking controller was utilized to make the system self-tuning in a model-free control architecture. The concept is applied to two major problems in vibration systems: vibration energy harvesting and vibration absorption, which is also known as tuned mass damping. In the former application, single variable and multi-variable sliding mode extremum seeking controllers are used for controlling the damping and stiffness of the actuator to maximize the harvested power. In the latter case, the same controller is used with the objective of minimizing the unwanted oscillations in a host structure. Analytical methods, computer simulations, and experimental results are provided to support the proposed concept and verify the theoretical findings. The results show that it is possible to achieve efficient, variable, and controllable damping and stiffness with an electromagnetic actuator comprised of a brushless DC motor and a mechanical motion conversion mechanism. It was also shown that the proposed extremum seeking controllers successfully tune the variables toward the optimum points.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Mehrdad Moallem
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) Ph.D.

Detection of sleep and wake states based on the combined use of actigraphy and ballistocardiography

Author: 
Date created: 
2019-04-30
Abstract: 

Sleep is a vital part of every humans daily circadian rhythm. People can rest and recover their body and mind and live a more active and alert life with an appropriate amount of sleep. The current gold standard method for sleep analysis is polysomnography, but due to the complexity, it is not convenient to perform it regularly and it disrupts the normal sleep environment of the patient. This thesis presents a method of integrating two alternative measurements of sleep analysis for an improved analysis. Combining the motion detection of actigraphy and the cardiac parameters of ballistocardiography, a novel algorithm was developed to analyze sleep and wake states without interfering with the natural sleep cycle of the participant. Without interfering with the natural sleep environment, this system can be implemented for continuous monitoring and be used to evaluate daily sleep patterns to assess overall sleep quality and health over time. The experimental results demonstrate the effectiveness of the novel proposed algorithm in comparison with each device used separately in improving the sleep classification.

Document type: 
Thesis
File(s): 
Supervisor(s): 
Edward J. Park
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.