Masters Degrees (Mechanical and Mechatronic Engineering)
Permanent URI for this collection
Browse
Recent Submissions
- ItemResponse surface modelling and investigation into release kinetics and in vivo toxicity of nanocellulose-based slow-release devices for delivery of quercetin(Stellenbosch : Stellenbosch University, 2022-11) Keirsgieter, Hannah; Chimpango, AFA; Smith, Carine; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH ABSTRACT: The delivery of many anti-inflammatory and anti-cancer drugs is hindered due to their low solubility in water, leading to poor bioavailability and therapeutic efficacy. The high specific surface area and customisable properties of nanoscale materials have established them as innovative solutions in many biomedical applications, such as wound healing, tissue engineering and drug delivery. Nanocellulose (NC) in particular is of special interest as a drug carrier, due to its inherent biocompatibility, biodegradability, and low toxicity. This study focused on cellulose nanocrystals (CNC) and nanofibres (CNF), as potential drug delivery systems (DDSs) for slow-release of the model hydrophobic drug, quercetin. NC is naturally hydrophilic and anionic, and was therefore first modified with the cationic surfactant, cetyltrimethylammonium bromide (CTAB), in order to facilitate effective drug binding through hydrophobic interaction. The effect of surfactant and drug concentration on particle size (Z), polydispersity index (PdI), zeta potential (ζ) and binding efficiency (BE) was investigated by response surface methodology (RSM), an empirical modelling technique in parametric optimisation. A design of experiments (DOE) approach was taken to obtain the experimental data, through a full factorial design (FFD), followed by a central composite design (CCD). The regressed Z and PdI models for both designs reported R2 values < 75%, while the ζ and BE models reported mean R2 values of 78% and 90%, respectively, indicating good model fits. The optimal responses for CNC were reported as Z = 5436 nm, PdI = 0.56, ζ = – 18.3 mV, and BE = 76.9%, at a CTAB and quercetin concentration of 3.3 mM and 4.2 mg/mL, respectively. The optimal responses for CNF were reported as Z = 4183 nm, PdI = 0.56, ζ = – 14.3 mV, and BE = 80.8%, at a CTAB and quercetin concentration of 2.0 mM and 5.1 mg/mL, respectively. Design validation resulted in experimental errors of 18.2% for CNC and 9.9% for CNF. Characterisation of the DDSs was performed by dynamic light scattering (DLS) using a Malvern Zetasizer, and further investigation into particle morphology was carried out by scanning electron microscopy (SEM). The in vitro quercetin release profile of a CNC-CTAB-QT formulation was tested using the dialysis bag method, and best fitted by the Korsmeyer-Peppas model (R2 = 99.9%), with a release exponent n > 1 suggesting super case II (non-Fickian) transport. In the first hour, the DDS exhibited a delayed cumulative release of 29%, compared to the cumulative release of 62% by the free drug. The in vivo safety profile of this formulation was evaluated by performing a toxicity assay on zebrafish larvae, but was constrained by excessive aggregation in the incubation medium.
- ItemA digital twin architecture for a water distribution system.(Stellenbosch : Stellenbosch University, 2022-11) Goosen, Ockert Johann; Basson, AH; Kruger, K; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH ABSTRACT: The incorporation of Industry 4.0 technologies in the water sector is providing water utilities with the necessary tools and insights to combat WDS-related challenges such as water management (e.g. leak detection) and future planning to meet increasing demand. A key enabler in Industry 4.0-related technologies is the concept of a digital twin. Digital twins may provide water utilities with improved insight into their WDSs and enable services that can facilitate data-driven decision-making. This thesis proposes and evaluates a digital twin architecture for a water distribution system (WDS) based on the Six Layer Architecture for Digital Twins with Aggregation (SLADTA). The digital twin architecture aims to facilitate the unique challenges and characteristics present in WDSs. Aspects include the complexity of heterogenous monitoring devices installed on WDSs and potential physical and management-related reconfigurations of WDSs. This thesis evaluates whether digital twin architecture designed according to SLADTA can overcome these challenges and provide the framework to develop a WDS digital twin with the desired characteristics. Several modifications were made in the process of mapping SLADTA (previously implemented in the manufacturing and renewable energy infrastructure domains) to a WDS-focused digital twin architecture. The thesis evaluates the WDS digital twin architecture using two case studies, each with several experiments, to determine whether the proposed architecture can meet the determined WDS digital twin requirements. The case studies also explore how the aggregation philosophy, inherent in SLADTA, can be implemented on WDSs. The first case study implements the WDS digital twin architecture for a laboratory scale WDS. The laboratory scale WDS replicates the real-world complexity of heterogeneous data sources by containing a variety of monitoring devices commonly used in industry. This case study also shows the ability of the WDS digital twin to integrate services, such as anomaly detection. In the second case study, a large-scale, simulated WDS is used to evaluate the WDS digital twin architecture’s ability to accommodate reconfiguration. WDS-related reconfiguration includes changes to a WDS’s monitoring capacity, changes to a WDS management organisations’ boundaries or zone and adding new WDS infrastructure to an existing digital twin. Additionally, the WDS digital twin architecture’s ability to incorporate WDS domain services, such as modelling of the hydraulic behaviour in a WDS, is investigated. The performance of the WDS digital twin architecture presented in this thesis is, therefore, demonstrated in a real-world, but laboratory-scale, case study and in a complex, but simulated, case study. The case study implementations show that the proposed architecture is capable of meeting the functional requirements that were derived from WDS operator needs.
- ItemA digital twin system for the integration of railway infrastructure data.(Stellenbosch : Stellenbosch University, 2022-11) Doubell, Gerhardus Christiaan; Kruger, K; Basson, AH; Conradie, PDF; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH ABSTRACT: The integration of data within large complex systems, such as railway infrastructure, forms a key component of enabling data-led decision making. Data can exist in numerous different data silos, provided by different data sources, with different interfaces and encapsulated data properties. The Passenger Rail Agency of South Africa (PRASA) currently faces this challenge. Research on digital twins (DTs) highlight the integration of data as one of the key advantages of DTs. However, very little research has been published on the implementation of DTs within railway infrastructure and, in particular, for the integration of data. Considering the above, this thesis presents the development of a DT system technology demonstrator for the integration of data for PRASA’s railway infrastructure. This technology demonstrator is evaluated based on a case study implementation for PRASA maintenance management in the Western Cape province. The DT system architecture utilises a layer with various “wrapper” components to provide specific interfaces to each of the data sources to be included in the system. In addition to the wrapper layer, the DT system architecture utilises a DT aggregation hierarchy to store data that represents physical reality, where each DT can also provide services that rely purely in the DT's scope of reality. The DT hierarchy is complemented with a services network to provide the required services that, as dictated by its purpose, integrates information from various DTs and, potentially, external sources. The DT system architecture is intended to be vendor-neutral and can therefore be broadly implemented. The developed DT system promises to be a valuable tool for maintenance management, by providing a holistic view of the railway infrastructure. The view can be reconfigured to adapt to changing infrastructure, data sources and user needs. Having been tested with diverse data sources and interfaces, complex relationships between infrastructural elements, and changing requirements and infrastructure, the DT system technology demonstrator can be considered suitable for PRASA’s railway infrastructure network, but also for other contexts where a similar environment (with regards to data sources and interfaces) is encountered. The integration of data through the DT system provides a means through which the number of data sources that need to be consulted for decision making, can be reduced. At the same time, the DT system provides a platform from which additional services can be offered, both now and in the future.
- ItemA holonic human cyber-physical system in healthcare(Stellenbosch : Stellenbosch University, 2022-11) Defty, TW; Kruger, K; Basson, AH; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH ABSTRACT: The fourth industrial revolution (I4.0) aims to enhance operational performance and promote interconnectivity between system components. Labour-intensive industries which are reliant on humans to perform tasks along with automation and integrated equipment are typically classified as Human Cyber-Physical Systems (HCPSs). Many I4.0 developments within HCPSs lack the appropriate consideration and integration of humans, resulting in performance inefficiencies, quality concerns, and under-utilisation. Healthcare is one such industry, with challenges due to a lack of integration developments which are exacerbated by limited resources and high patient volumes. The objective of this thesis is the development of a holonic HCPS for healthcare. These holonic HCPS developments address challenges in healthcare through the application of Human-System Integration (HSI) principles and requirements. HSI aims to enhance human intelligence and cooperation within digitised environments. HCPSs within this study are defined by human, physical and digital components that interact in a complex manner to achieve overall system goals. An ambulatory clinic case study was selected to study the complex interactions between system components and actors. This thesis presents a Representation-Communication-Interfacing (RCI) framework to better define HSI and guide the development and evaluation of the HSI maturity in HCPSs. A holonic system approach is selected, using the RCI framework, to improve the maturity of HSI developments in healthcare. This thesis presents a structured design process for holonic systems using the Activity-Resource-Type-Instance (ARTI) architecture and the Biography-Attribute-Schedule-Execution (BASE) architecture. The design process is presented in response to a lack of available implementation details expressed in reviewed holonic applications. The developed holonic HCPS is evaluated experimentally, to showcase how the system meets the requirements and enhances the HSI maturity of the ambulatory case study. The evaluation shows that the holonic HCPS effectively integrates humans, resulting in improved operational efficiencies and lower workloads experienced by humans. The RCI framework offers valuable guidance for elevating the human component to the cyber layer, improving the autonomy and cooperability of humans in the system and easing the reconfigurability of the system. The holonic design process guides the system development by standardising the partitioning of system components and interactions into distinct holons. Furthermore, this design process reduces the development complexity and time. The holonic HCPS demonstrates how HSI developments can improve clinic workflow efficiencies, aid decision-making, improve the traceability of activities and reduce the workload for healthcare practitioners.
- ItemIntegration of digital twins and virtual reality for data visualisation.(Stellenbosch : Stellenbosch University, 2022-11) Da Silva, GS; Kruger, K; Basson, AH; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.ENGLISH ABSTRACT: This thesis considers an integration of digital twins (DTs) and virtual reality (VR) for enhancing the data driven decision-making process. An integrated DT and VR system is designed according to an available complex DT system design framework. The Six Layer Architecture for Digital Twins with Aggregation (SLADTA) is used for the internal architecture of the DTs in the system. A custom developed VR application is used to visualise the information using VR equipment. The Facilities Management (FM) Division at Stellenbosch University serves as the chosen case study context for the evaluation. The energy usage information for various facilities of FM is visualised in VR. The DT method of transferring information to VR is compared to a method that does not make use of DTs. Three experiments are used to evaluate the two implementation methods to allow for an adequate comparison of the two methods. The experiments focus on, respectively, latency, computer resource utilisation (in terms of RAM and CPU usage), and reconfigurability when a new feature is to be added to the system. The experiment results indicate that the DT method has lower latencies, the two methods have similar computational resource needs, and the non-DT method is more reconfigurable than the DT method. However, the DT method offers other advantages such as allowing for two VR experiences to visualise the same information, or allowing for a different visualisation tool, other than VR, to be integrated seamlessly into the system. The DT method also allows for a distributed operation functionality that reduces the computational load required from a single hardware device. The Non-DT method does not offer such advantages. The thesis concludes that the integration of DTs and VR for data visualisation is possible and is favourable for a system that will not only use VR as a data visualisation means.