<p>Electric load forecasting is a critical tool for power system planning and the creation of sustainable energy systems. Precise and reliable load forecasting enables power system operators to make informed decisions regarding power generation and transmission, optimize energy efficiency, and reduce operational costs and extra power generation costs, to further reduce environment-related issues. However, achieving desirable forecasting performance remains challenging due to the irregular, nonstationary, nonlinear, and noisy nature of the observed data under unprecedented events. In recent years, deep learning and other artificial intelligence techniques have emerged as promising approaches for load forecasting. These techniques have the ability to capture complex patterns and relationships in the data and adapt to changing conditions, thereby enhancing forecasting accuracy. As such, the use of deep learning and other artificial intelligence techniques in load forecasting has become an increasingly popular research topic in the field of power systems. </p> <p>Although deep learning techniques have advanced load forecasting, the field still requires more accurate and efficient models. One promising approach is federated learning, which allows for distributed data analysis without exchanging data among multiple devices or cen- ters. This method is particularly relevant for load forecasting, where each power station’s data is sensitive and must be protected. In this study, a proposed approach utilizing Federated Deeponet for seven different power stations is introduced, which proposes a Federated Deep Operator Networks and a Lagevin Dynamics-based Federated Deep Operator Networks using Stochastic Gradient Langevin Dynamics as the optimizer for training the data daily for one-day and predicting for one-day frequencies by frequencies. The data evaluation methods include mean absolute percentage error and the percentage coverage under confidence interval. The findings demonstrate the potential of federated learning for secure and precise load forecasting, while also highlighting the challenges and opportunities of implementing this approach in real-world scenarios. </p>

<p>Breast cancer affects 12.5\% of women over their life time and tissue expansion (TE) is the most common technique for breast reconstruction after mastectomy. However, the rate of complications with TE can be as high as 15\%. Even though the first documented case of TE happened in 1957, there has yet to be a standardized procedure established due to the variations among patients and the TE protocols are currently designed based on surgeon's experience. There are several studies of computational and theoretical framework modeling skin growth in TE but these tools are not used in the clinical setting. This dissertation focuses on bridging the gap between the already existing skin growth modeling efforts and it's potential application in the clinical setting.</p> <p><br></p> <p>We started with calibrating a skin growth model based on porcine skin expansions data. We built a predictive finite element model of tissue expansion. Two types of model were tested, isotropic and anisotropic models. Calibration was done in a probabilistic framework, allowing us to capture the inherent biological uncertainty of living tissue. We hypothesized that the skin growth rate was proportional to stretch. Indeed, the Bayesian calibration process confirmed that this conceptual model best explained the data. </p> <p><br></p> <p>Although the initial model described the macroscale response, it did not consider any activity on the cellular level. To account for the underlying cellular mechanisms at the microscopic scale, we have established a new system of differential equations that describe the dynamics of key mechanosensing pathways that we observed to be activated in the porcine model. We calibrated the parameters of the new model based on porcine skin data. The refined model is still able to reproduce the observed macroscale changes in tissue growth, but now based on mechanistic knowledge of the cell mechanobiology.  </p> <p><br></p> <p>Lastly, we demonstrated how our skin growth model can be used in a clinical setting. We created TE simulations matching the protocol used in human patients and compared the results with clinical data with good agreement. Then we established a personalized model built from 3D scans of a patient unique geometry. We verified our model by comparing the skin growth area with the area of the skin harvested in the procedure, again with good agreement.</p> <p><br></p> <p>Our work shows that skin growth modeling can be a powerful tool to aid surgeons design TE procedures before they are actually performed. The simulations can help with optimizing the protocol to guarantee the correct amount of skin is growth in the shortest time possible without subjecting the skin to deformations that can compromise the procedure.</p>

<p>The document describes the computational framework to optimize spring-driven Autoinjectors. It involves Bayesian Optimization for efficient and cost-effective design of Autoinjectors.</p>

<p dir="ltr">The ability to change the shape of aerodynamic surfaces is necessary for modern aircraft, both to provide control while performing maneuvers and to meet the conflicting requirements of various flight conditions such as takeoff/landing and level cruise. These shape changes have traditionally been accomplished through the use of various mechanical devices actuating discrete aerodynamic surfaces, for example ailerons and flaps. Such control surfaces and high-lift devices are generally limited to their specific functionality and create surface discontinuities which increase drag and aircraft noise. Broadly speaking, the design and study of morphing wings typically seeks to improve the performance of aircraft by completing one or more of the following objectives: reducing the drag from discontinuities in the aerodynamic surface of the wing by closing hinge gaps and creating smooth transitions, reducing weight and/or mechanical complexity by integrating mechanism functionality into compliant structures that can bear aerodynamic load and maintain shape adaptability, and providing unique or optimal functionality to the aircraft by allowing it to adjust its aerodynamic shape to meet the needs of various flight conditions with conflicting objectives and constraints.</p><p dir="ltr">The concepts proposed in this work represent potential methods for addressing these objectives. In each case, a compliant structure with multiple stable states is embedded into the wing. Exploiting elastic structural instabilities in this way provides the advantage that a structure can be made relatively stiff while still allowing for large deformations. In the first case, the development of a 3D-printable rib with an embedded bistable element creates a truss-like 2D structure that allows for modification of the airfoil. Switching states of the elements changes their local stiffness, and therefore the global stiffness of the system. By optimizing the topology of the airfoil, a passive deflection of the trailing edge can be leveraged to change the camber to leverage different lift characteristics for varying operating conditions. Primary work on this concept has included the construction of multiple experimental demonstrators for validating the concept through static structural and wind tunnel testing. In the second case, a cellular material has been investigated incorporating a bistable unit cell with a sinusoidal arch. This provides a metamaterial that can exhibit large, reversible deformations with as many stable configurations as there are rows in the honeycomb. This metamaterial is incorporated into a beam-like structure which can serve as a spar for a spanwise morphing wing, providing sufficient bending and torsional stiffness, particularly when utilized at the wing tip. Extending and retracting the wing by switching the states of the honeycomb rows provides a significant change to the wing’s induced drag and wing loading, making it ideal for optimal flight in both loitering and cruising conditions. Contributions to this concept have been the development and characterization of the bistable unit cell and honeycomb, as well as the design and analysis of the metabeam and morphing wing concept.</p>

<p dir="ltr">Precise measurement of the flowrate is crucial for both process control and energy consumption evaluation. The main aim of this work is to develop a methodology to calibrate mechanical flowmeters, designed to measure high viscosity fluids, in water. In order to accomplish this, a series of computational fluid dynamics (CFD) analysis are carried out to determine how the motion of the mechanical component varies with different flow rates of water and high viscosity fluids. This data is recorded and analyzed to develop calibration curves that relate the motion of the mechanical component the flow rates. From the calibration curves, it can be determined the required water flow rate to achieve the equivalent motion of the mechanical component in a specified viscosity. This method provides an efficient and cost-effective calibration process because it eliminates the need for calibrating using heated engine oil to achieve the fluid viscosity of the flow meter is designed. Flowmeter sensitivity analysis was also performed and it was observed that the motion of the mechanical component curves converges as the size of the flowmeter increases suggesting that the effect of viscosity on flowmeter sensitivity decreases as the size of the flowmeter is increased, likely due to reduced resistance to flow and smaller pressure drops. </p><p dir="ltr">The Kanbara Reactor ladle is a commonly used method in the steelmaking industry for hot-metal desulfurization pre-treatment. The impeller's configuration is pivotal to the reactor's performance, yet its precise function remains partially understood. This study introduces a 3-dimensional Volume-of-Fluid (VOF) model integrated with the sliding mesh technique, investigating the influence of five different impeller speeds. After Validating the model through experimental data, this numerical model is applied to investigate the typical developmental phenomena and the consequences of impeller speed variations on fluid flow characteristics, interface profile, and vortex core depth. The findings reveal that the rotational impeller induces a double-recirculation flow pattern in the axial direction due to the centrifugal discharging flow. With increasing impeller rotation speed, the vortex core depth also rises, emphasizing the substantial impact of impeller speed on vortex core depth.</p>

<p dir="ltr">Nanomaterials exhibit many unique properties compared to traditional bulk materials, interfaces play a more important role in nanoscale systems by significantly influencing the mechanical performance. In this thesis, we focus on an intricate exploration of various interfaces, ranging from simple GBs in bicrystal models to intricate GB networks within polycrystalline structures and interfaces within nanocomposite materials. Various computational methodologies, including MD, DFT, and advanced machine learning algorithms<del>,</del> were employed to simulate and predict the mechanical properties of interfaces with microstructural complexity.</p><p dir="ltr">Firstly, utilizing MC/MD simulations, we established a distinct correlation between GB motion in the Cantor alloy and the Cr concentration within the GBs. A formulation is calculated to link the GB mobility with the Cr concentration. Subsequently, DFT simulations highlight that vacancies in Tungsten GBs prefer to appear in the layer adjacent to the GB plane rather than the GB plane itself. These vacancies, as the findings suggest, cause the strength to decrease under tensile loading. Then, to expedite the prediction of interfacial properties, a cGAN model was developed to predict GB network evolution in polycrystalline samples based on the training data of MD simulation results. Finally, two modified deep learning models are introduced including the CNN-Prob and FNN-Prob, to predict the yield stress of a composite material, Cu-Cu/Zr. These models encompassed dual components for predicting both mean values and associated standard deviations.</p>

<p>Thermal barrier coatings (TBCs) made of low-thermal-conductivity ceramic topcoat, metallic bond coat and metallic substrate, have been extensively used in gas turbine engines for thermal protection. Recently, additive manufacturing (AM) or 3D printing techniques have emerged as promising manufacturing techniques to fabricate engine components. The motivation of the thesis is that currently, application of TBCs on AM’ed metallic substrate is still in its infancy, which hinders the realization of its full potential.</p> <p>The goal of this thesis is to understand the processing-structure-property relationship in thermal barrier coating deposited on AM’ed superalloys.</p> <p>The APS method is used to deposit 7YSZ as the topcoat and NiCrAlY as the bond coat on TruForm 718 substrates fabricated using the direct metal laser sintering (DMLS) method. For comparison, another TBC system with the same topcoat and bond coat is deposited using APS on wrought 718 substrates. For thermomechanical property characterizations, thermal cycling, thermal shock (TS) and jet engine thermal shock (JETS) tests are performed for both TBC systems to evaluate thermal durability. Microhardness and elastic modulus at each layer and respective interfaces are also evaluated for both systems. Additionally, the microstructure and elemental composition are thoroughly studied to understand the cause for better performance of one system over the other.</p> <p>Both TBC systems showed similar performance during the thermal cycling and JETS test but TBC systems with AM substrates showed enhanced thermal durability especially in the case of the more aggressive thermal shock test. The TBC sample with AM substrate failed after 105 thermal shock cycles whereas the one with wrought substrate endured a maximum of 85 cycles after which it suffered topcoat delamination. The AM substrates also demonstrated an overall higher microhardness and elastic modulus except for post thermal cycling condition where it slightly underperformed. This study successfully demonstrated the use of AM built substrates for an improved TBC system and validated the enhanced thermal durability and mechanical properties of such a system.</p> <p>A modified YSZ TBC architecture with an intermediate Ti3C2 MXene layer is proposed to improve the interfacial adhesion at the topcoat/bond coat interface to improve the thermal durability of YSZ</p> <p>12</p> <p>TBC systems. First principles calculations are conducted to study the interfacial adhesion energy in the modified and conventional YSZ TBC systems. The results show enhanced adhesion at the bond coat/MXene interface. At the topcoat/MXene interface, the adhesion energy is similar to the adhesion energy between the topcoat and bond coat in a conventional YSZ TBC system.</p> <p>An alternative route is proposed for the fabrication of YSZ TBC on nickel base superalloy substrates by using the SPS technology. SPS offers a one-step fabrication process with faster production time and reduced production cost since all the layers of the TBC system are fabricated simultaneously. Two different TBC systems are processed using the same heating protocol. The first system is a conventional TBC system with 8YSZ topcoat, NiCoCrAlY bond coat and nickel base superalloy substrate. The second system is similar to the first but with an addition of Ti3C2 MXene layer between the topcoat and the bond coat. Based on the first principles study, addition of Ti3C2 layer enhances the adhesion strength of the topcoat/bond coat interface, an area which is highly susceptible to spallation. Further tests such as thermal cycling and thermal shock along with the evaluation of mechanical properties would be carried out for these samples in future studies to support our hypothesis.</p>

<p>Granular systems are ensembles of inelastic particles which dissipate energy during collisions. Granular systems serve as excellent models for a wide variety of materials such as sand, soils, corn, and powder. A rather remarkable property of granular systems is when excited, whether due to an interstitial fluid or via the boundaries, the granular particlesdisplay fluid-like behavior. As a result, there has been decades of granular research with the overarching goal of formulating a general granular hydrodynamic theory.</p> <p>However, the granular hydrodynamic theory is limited, and the underlying transport coefficients often require modifications which are based on empirical observations, and assuch, are system-specific. It is ideally better to devise a general theory which minimizes the information needed about the systema priori. The main thrust of the work undertaken shown here strives to develop such a model by using kinetic theory as the basis. More specifically, I investigate granular gases via the direct simulation Monte Carlo (DSMC) methodand modify the governing equations. In this thesis, two idealized cases of granular gases areconsidered: the homogeneous cooling state and a boundary-heated gas (or the pure conduc-tion case). In the former, the effects of polydispersity are probed. In the latter, the evolutionof the local hydrodynamics due to strong rarefaction effects are divulged. Additionally, amodified, more generalized constitutive relation for the heat flux is proposed and comparedwith DSMC results. Extensions of the DSMC method for dense granular gases and granulargases composed of non-spherical particles are also discussed.</p>

<p>The electric arc furnace (EAF) is a critical steelmaking facility that melts the scrap by the heat produced from electrodes and burners. The migration to EAF steelmaking has accelerated in the steel industry over the past decade owing to the consistent growth of the scrap market and the goal of "green" steel production. The EAF production already hit a new high in 2018, contributing to 67% of total short tons of U.S. crude steel produced. The EAF steelmaking process involves dynamic complex multi-physics, in which electric arc plasma and coherent jets coexist resulting in an environment with local high temperature and velocity. Different heat transfer mechanisms are closely coupled and the phase change caused by melting and re-solidification is accompanied by in-bath chemical reactions and freeboard post-combustion, which further creates a complicated gas-liquid-solid three-phase system in the furnace. Therefore, not all conditions and phenomena within the EAF are well-understood. The traditional experimental approach to study the EAF is expensive, dangerous, and labor-intense. Most of the time, direct measurements and observations are impossible due to the high temperature within the furnace. To this fact, the numerical model has aroused great interest worldwide, which can help to gain fundamental insights and improve product quality and production efficiency, greatly benefiting the steel industry. However, due to the complexity of the entire EAF steelmaking process, the relevant computational fluid dynamics (CFD) modeling and investigations of the whole process have not been reported so far. </p> <p><br></p> <p>The present study was undertaken with the aim of developing the modeling methodologies and the corresponding comprehensive EAF CFD models to simulate the entire EAF steelmaking process. Two state-of-the-art comprehensive EAF CFD models have been established and validated for both the lab-scale direct current (DC) EAF and the industry-scale alternating current (AC) EAF, which were utilized to understand the physical principles, improve the furnace design, optimize the process, and perform the trouble-shootings.</p> <p><br></p> <p>For the lab-scale DC EAF, a direct-coupling methodology was developed for its comprehensive EAF CFD model which includes the solid steel melting model based on the enthalpy-porosity method and the electric arc model (for lab-scale DC arc) based on the Magneto Hydrodynamics (MHD) theory, so that the dynamic simulation of the steel ingot melting by DC arc in the lab-scale furnace can be achieved, which considered the continuous phase changing of solid steel, the ingot surface deformation, and the phase-to-phase interaction. Both stationary DC arc and the arc-solid steel interface heat transfer and force interaction were validated respectively against the experimental data in published literature. For the given lab-scale furnace, the DC arc behavioral characteristics with varying arc lengths generated by the moving electrode were analyzed, and the effects of both the initial arc length and the dynamic electrode movement on the steel ingot melting efficiency were revealed.</p> <p><br></p> <p>For the industry-scale AC EAF, an innovative integration methodology was proposed for its comprehensive EAF CFD model, which relies on the stage-by-stage approach to simulate the entire steelmaking process. Six simulators were developed for simulating sub-processes in the industry-scale AC EAF, and five models were developed for the above four simulators, including the scrap melting model, the electric arc model (for industry-scale AC arc), the coherent jet model, the oxidation model, and the slag foaming model, which can be partially integrated according to the mass, energy, and momentum balance. Specifically, the dual-cell approach and the stack approach were proposed for the scrap melting model to treat the scrap pile as the porous medium and simulate the scrap melting together with its dynamic collapse process. The statistical sampling method, the CFD-compatible Monte Carlo method, and the electrode regulation algorithm were proposed for the electric arc model to estimate the total AC arc power delivery, the arc radiative heat dissipation, and the instantaneous electrode movement. The energetic approach was proposed to determine the penetration of the top-blown jet in the molten bath based on the results from the coherent jet model. The source term approach was proposed in the oxidation model to simulate the in-bath decarburization process, where the oxidation of carbon, iron, and manganese as well as the effect of those exothermic reactions on bath temperature rising was considered. Moreover, corresponding experiments were performed in the industry-scale EAF to validate the proposed simulators. The quantitative investigations and analyses were conducted afterward to explore and understand the coherent jet performance, the AC arc heat dissipation, the burner preheating characteristics, the scrap melting behavior, the in-bath decarburization efficiency, and the freeboard post-combustion status.</p> <p><br></p>

<p>This study is focused on synthesizing Nitinol by additive manufacturing that can provide desirable mechanical properties for orthopedic implants and adding functionally gradient coating that can enhance both safety and biocompatibility for orthopedic implant applications.</p> <p><br></p> <p>The characteristics of additively manufactured Nitinol, by using the direct energy deposition (DED) technique, were experimentally studied. Because of a unique layer-by-layer manufacturing scheme, the microstructure and associated properties (mechanical and thermo-mechanical properties) of the DED Nitinol is different compared to conventionally produced Nitinol. Both the feasibility of manufacturing defect-free microstructure and the precise control of chemical composition were demonstrated. Effects of chemical compositions and post heat-treatment conditions on the phase transformation temperatures of the DED Nitinol were systematically analyzed and compared with those of conventional Nitinol. More precise control of phase transformation temperature from DED Nitinol was possible due to incoherent precipitate formation during aging heat treatment. In a similar way, the mechanical properties of the DED Nitinol were less sensitive to its chemical compositions and post heat-treatment conditions. The feasibility of the precise control of both mechanical and thermo-mechanical properties of the DED Nitinol was demonstrated which can broaden its applications. </p> <p><br></p> <p>The bulk polycrystalline properties of the NiTi phase were studied via molecular dynamics (MD) simulations. Thermo-mechanical properties that are highly sensitive to chemical composition were not precisely predicted from previous reports and studies. In this study, realistic boundary conditions were applied to calculate bulk polycrystalline properties. Thermally driven phase transitions of NiTi between martensite and austenite are simulated with external stresses in both normal and shear directions. It is shown that phase transformation temperatures are affected by applied external stresses, and realistic values compared to experimental data are correctly predicted only when external stresses in both normal and shear directions are similar to the experimentally observed values of 0.05 – 0.1 GPa. The experimentally observed grain orientation and grain boundary thickness were applied to simulation domains for the prediction of the elastic moduli. The elastic moduli of polycrystalline NiTi structure was calculated as 52 GPa which is close to the experimentally reported value of 20-40 GPa while other studies predicted over 85 GPa. </p> <p><br></p> <p>Lastly, pure titanium gradient layers were coated on the Nitinol surface for orthopedic implant applications to eliminate potentially toxic Ni ion release. Using the DED technique, both the core Nitinol and titanium gradient layers were manufactured with high purity and without microstructural defects. An additional biomedical coating of Hydroxyapatite (HA) was deposited on the outer surface using the cold spray technique. The resultant bonding strength was determined to be 26 MPa which exceeded the requirement of the ISO-13779 standard (15 MPa). The <em>in vitro</em> test of the Ni release rate from the entire gradient Nitinol structure was very low, which was comparable to drinking water.</p>