Graduate Projects (Mechanical Engineering)

Permanent URI for this collection

https://dspace.library.uvic.ca/handle/1828/8101

Browse

Now showing 1 - 20 of 41

Comparative Analysis of Various National Building Codes and Carbon Payback Periods of Insulation Materials at Different Climate Zones in Canada
(2024-05-31) Mascarenhas, Alastair Alphonse; Valeo, Caterina; Mukhopadhyaya, Phalguni
Single-family dwellings make a significant contribution to carbon emissions in Canada. The National Energy Code for Buildings (NECB) emphasizes reducing the operational carbon consumption of buildings. Using thermal insulation material in constructing building envelopes plays a crucial role in decreasing a building's operational carbon. However, since insulation materials have embodied carbon, therefore, for optimal building performance and design, designers should take into account both the operational and embodied carbon of insulation materials. This paper compares the embodied carbon and operational energy savings resulting from the use of thermal insulation material. It also presents Carbon Payback Period (CPP) values of different thermal insulation materials in various Canadian cities representing different climate zones. A model is created using the Athena Impact Estimator (AIE) tool, based on a three-bedroom single-family home with a wood-frame structure. Three insulation materials, namely Batts Fiberglass, Blown Cellulose and Mineral Wool, are evaluated in three different cities, namely Vancouver, Toronto and Calgary, representing three climate zones (zones 4, 5 and 7a). The HOT2000 energy simulator calculates operational carbon consumption using the energy mix comprising electricity and natural gas. The CPPs for selected materials were calculated using operational and embodied carbon data. A comparison of the Whole Building Life Cycle Analysis (WBLCA) Global Warming Potential (GWP) between the National Building Code (NBC) 1995 and 2020 versions revealed an average 25% decrease in Operational Carbon and an average 6% increase in Embodied Carbon. This compromise showed a shift towards standardizing energy-efficient buildings and selecting sustainable thermal insulation materials for construction. Identifying and using less carbon footprint materials can help reduce embodied carbon. In Calgary, the CPP for Blown Cellulose, Batts Fiberglass and Mineral Wool insulation were calculated to be 0.92, 0.94 and 1.09 years, respectively. In Toronto, the CPP for Blown Cellulose, Batts Fiberglass and Mineral Wool is 1.15, 1.17 and 1.39 years, respectively. Vancouver has longer CPP for Batts Fiberglass, Blown Cellulose and Mineral Wool with 2.66, 2.64, and 2.69 years, respectively. This indicates that as the heating degree days (HDD) increases, the CPP shortens. Graphing the CPP vs HDD can help designers and contractors make more informed decisions regarding the available choices of thermal insulations.
Evaluation of Ride Comfort and Road Holding for Heavy Vehicle Suspension (HVS) through Model Predictive Controller (MPC) based on Hybrid Semi-Active Damping Strategy
(2024) Faronbi, Michael O.; Shi, Yang
The continuous expansion of industrial demand in various countries has increased the need for large commercial vehicles to transport products between locations. As a result, freight transportation has become a key driver of economic growth, contributing significantly to a country's GDP, typically accounting for 6-12% of the total. However, this rapid growth in road transportation has also brought about a rise in traffic congestion and a higher probability of road accidents. According to a European Union assessment, large vehicles are a significant factor in these incidents. Nevertheless, the primary cause of road accidents remains driver negligence. Work-related injuries and disorders caused by whole-body vibration have been extensively studied worldwide. To address this problem, researchers have developed a pitch plane model of a large vehicle using a Lagrangian approach coupled with various hybrid semi-active damping schemes based on the model predictive control (MPC) framework. The MPC-based suspension controller is designed to optimize comfort and handling by minimizing a quadratic cost function. The focus has been on reducing the vertical accelerations experienced by the vehicle due to variations in the vehicle and road profile to improve the vehicle's stability and ride comfort level. Additionally, managing the changes in vertical force encountered by each tire during its interaction with the road has been crucial. The ride comfort of the driver has been evaluated by analyzing the vertical accelerations at the center of gravity of the pitch plane model, both with and without the MPC-based controller, using the guidelines specified in ISO 2631-1/2. Simulation results have demonstrated the impact of the MPC-based controller, with and without its implementation, on the ride comfort level and road-holding capability of the heavy vehicle system. These findings highlight the potential of the MPC-based controller to enhance the overall performance of heavy vehicle systems in terms of ride comfort and road holding.
Literature review of 3D-bio printed hair follicles and the proposal for a permanent hair system on the scalp
(2024) Issac, Mathews; Shi, Yang
Modern hair restoration surgery helps restore hair loss or bald areas, which requires a substantial number of hair follicles from the donor area. However, in some cases, people do not have sufficient donor hair follicles for transplant surgery due to diseases, genetics, aging, other biological and environmental issues, and so on. This problem can be addressed using 3D bioprinter technology to cultivate artificial hair follicles. This project report meticulously reviews six different methods for artificially cultivating hair follicles (HF) using bio cells and a cell-transforming environment created using 3D bioprinting technology. The six methods were 3D-bioprinting of a gelatin-alginate hydrogel for tissue-engineered hair follicle regeneration, tissue engineering of human hair follicles using a biomimetic developmental approach, bead-jet printing enabled sparse mesenchymal stem cell patterning augments skeletal muscle and hair follicle regeneration, robot-assisted in situ bioprinting of gelatin methacrylate hydrogels with stem cells induces hair follicle-inclusive skin regeneration, bioprinting of hair follicle germs for hair regenerative medicine, and using bioprinting, and spheroid culture to create a skin model with sweat glands and hair follicles. The main disadvantages of these experimental methods are their complexity, the significantly low number of hair follicles generated, and the fact that it will take time to get approval for human trials for these new technologies. The report also proposes a procedure to overcome the disadvantages of artificially grown HF by developing a permanent hair system on the scalp.
An integrated temperature control system for a 3D printed droplet generation microfluidic device HC-BAR Chip
(2024) Bhatt, Sheshank; Akbari, Mohsen
A microfluidic chip is a small device which deals with a very small amount of fluid. It has microscale channels. It has been used in different fields of science like engineering, physics, biochemistry, etc. A droplet generator is a microfluidic device which is capable of generating small droplets which is used in different applications like drug delivery, cell trapping and gene analysis. Temperature control is an essential part of droplet generation, and it affects the generation of droplets. This project proposes a microfluidic chip design (HC BAR Chip) with a uniform distribution of temperature with no embedded heating equipment. The chip is capable of creating heating and cooling zones unaffected by each other with a flow-focusing droplet generation method. CAD software like Solidworks is used to design and COMSOL Multiphysics® software for the analysis.
Simulation & Formation Control for Leader- Follower Wheeled Mobile Robots Based on Embedded Control Technique
(2023-12-19) Jayasingha Appuhamilage, Chathusanka
Formation control has garnered significant attention from researchers in recent times. This heightened interest can be attributed to its applicability in a wide range of tasks, including but not limited to search and rescue operations, agricultural coverage jobs, and area patrols. This surge in attention is primarily attributed to its ability to enhance efficiency, reliability, and the capacity to accomplish complex tasks effectively within these domains. Formation control for ground vehicles has particularly been useful for application such as cargo transportation, cooperative manipulation, and surveillance and exploration and many more [2]. Combining sensors on robots in a formation enables them to scan larger areas more quickly by directing sensors in different directions, surpassing the smaller area of coverage of a single robot can achieve during the same time. This enhances the efficiency of the search and exploration process compared to individual robot operations. For this project, a formation control strategy based on the embedded control technique for a leader-follower system was used. The proposed technique divides the formation control problem into two subtasks, which is different from the traditional design philosophy of basing the formation controller design directly on the formation tracking errors [3]. Two subtasks are virtual signal generator and trajectory tracking controller. The virtual signal generator achieves the desired formation control goal and outputs a reference signal to the trajectory tracking controller of the follower [3]. The embedded control technique reduces the complexity of directly implementing formation tracking errors while providing several other benefits as well. A predesigned mobile robot model in CoppeliaSim was used as the leader and follower. The robots were programmed to implement the controller and simulate the system. Finally, a number of tests were performed adjusting gains and observing the performance of the robots to identify the optimal performance gain values. Gains were adjusted with the goal of minimizing the error and reducing the motion jerk of the follower.
High-Performance Brick Mortar Mix to Optimize Moisture Management in Brick Wall
(2023-08-16) XIA, HAI; Valeo, Caterina; Mukhopadhyaya, Phalguni
The durability of exterior building envelopes is significantly impacted by the presence of water, particularly through the capillary rise mechanism that allows liquid water to penetrate building materials. This process affects both the energy efficiency and durability of buildings. To assess the capillary water intake into porous building materials, the water absorption coefficient is used as a characterization parameter. Additionally, the water vapor permeability of a material indicates its ability to allow moisture to diffuse and escape. In this project, two concentrations of zinc stearate (0.5% w/w and 1% w/w) were added to commonly used mortar. Following the ASTM standard test procedure, the liquid water absorption coefficients and water vapor permeability of brick, mortars, and brick mortar joints were determined. These experimental values were utilized as inputs for the hygrothermal performance analysis (numerical modelling) of the brick wall assembly. The experimental findings suggest that the addition of zinc stearate to the mortar can reduce water absorption capacity while simultaneously enhancing water vapor permeability. Numerical modelling results further demonstrate that the use of high-performance brick mortar materials can significantly improve the moisture management capability for brick walls in the marine-warm and humid climate of Vancouver, BC, Canada.
Prediction of Orthogonal Cutting Forces Based on Multi-fidelity Modeling
(2023-04-26) Ahmadi, Keivan
Simplified analytical models of chip formation mechanics (e.g. the well-known Merchant’s model) are widely used to compute the machining forces in orthogonal cutting operations. The accuracy of analytical models, however, diminishes when the cutting edge has a rounded shape, known as edge (or hone) radius, which is common for most cutting tools. Finite element (FE) simulation can be used to obtain more accurate predictions of the forces in the presence of edge radius, but FE is computationally expensive because it should numerically solve a thermo-mechanical contact problem with nonlinear material properties to model the plastic deformation and damage of the workpiece. The high computational cost of FE simulations indeed becomes crucial when the force model is used for process optimization or for online simulations in the digital twin of the machining process. In this research, we present a computationally efficient data-driven model with acceptable accuracy when compared to the FE simulation. The presented model combines the predictions of FE simulations (i.e., high-fidelity dataset) and the predictions of the analytical model (i.e., low-fidelity dataset) and generates a new regression multi-fidelity model. The high-fidelity dataset is generated by an FE simulation in Abaqus and using Johnson-Cook constitutive equation to model the plastic deformation and damage of an aluminum workpiece during chip formation. The low-fidelity dataset is generated by Merchant’s analytical model. In both datasets, the inputs are the tool rake angle and uncut chip thickness, and the outputs are the cutting and feed forces. In total, 440 data points (40 high-fidelity points and 400 low-fidelity points) are generated. Based on this dataset, a multi-fidelity model is trained and tested through the emulator-embedded neural network (E2NN) method. The Root Mean Squared Error (RMSE) is then computed between the predictions of the trained model and the predictions from the FE simulation to quantify the performance of the presented multi-fidelity model. The results show a close agreement between the predictions of the high-fidelity and the multi-fidelity models. The computed RMSE was less than 8.5%. Yet, the accuracy would gradually improve by increasing the high-fidelity samples. Moreover, note that the computational time of a FE simulation is typically a few (~5) minutes while it is less than a second for the presented multi-fidelity model. The presented modelling approach therefore can efficiently replace high-cost FE simulations in process optimization or online simulations.
Structural Health Monitoring of the Marine Structure Using Guided Wave and UAV-based testing
(2023-04-25) Gupta, Yugansh; Shi, Yang
The marine industry encompasses a vast range of structures that can be challenging to access, and thus highly mobile systems can be useful for efficient structural health monitoring. This research aims to investigate the feasibility of using unmanned aerial vehicles (UAVs) for deploying various non-destructive testing (NDT) methods to inspect steel structures. In addition, we are evaluating piezoelectric patches (PZT) based detection systems that can provide early warnings of corrosion and fatigue cracks. The study analyzes the pitch and catches method using time-of-flight (TOF) to detect cracks and defects, the UAV-based hammer percussion test for frequency response, and the use of an infrared-mounted camera on the drone for NDT purposes. The goal is to identify the best approach for monitoring the health of marine structures.
The Boundary and Excitation Effect of Non-Spherical Granular Material
(2023-03-21) YASEEN, GHULAM; Nadler, Ben
Non-spherical grains have been gradually receiving attention from both researchers and the industry because of their behavior. Even though these grains possess complex macroscopic orientations that are associated with different applications, such as the pharmaceutical industry, they sometimes can also cause challenges, like jamming while passing collectively through certain narrowed passages. Most published articles have presented studies about granular materials, based on spherical grains and have mainly examined the grain size but ignored the grain’s shape and orientation, especially concerning the interaction of these grains with their boundaries. Further, literature reported that the mechanical properties of the granular materials are critically affected by the alignment of non-spherical grains as conducted in various simulations and associated experiments. To explore more about the shape and orientational effect of non-spherical grains with respect to boundaries, a detailed initial-level observational study is done with the help of different boundary shapes and grains ranging from elongated rice to long cylindrical grains. The collision of grains with boundaries generates an orientational field that results from their interaction with boundaries and neighboring grains. This research shows that the excitation of grains occurs due to their collision with boundaries, and these boundaries can play an important role in the orientation of non-spherical grains. The study provides ‘thought-provoking’ directions for exploring more about the orientation of non-spherical grains.
Impact on building energy performance by deployment of dynamic insulation in residential buildings in Canada
(2023-03-08) Shukla, Anoopkumar; Valeo, Caterina; Mukhopadhyaya, Phalguni
This report summarizes the results of an analysis evaluating the energy performance of small residential buildings in Canada. Using the HOT2000, an energy simulation modelling program created and maintained by Natural Resources Canada, the goal of this work is to investigate dynamic insulations, run simulations, and assess the possible energy savings brought on by using dynamic insulation materials (DIMs) in exterior walls in place of conventional static insulation. DIMs can alter their thermal properties based on control procedures, unlike conventional static insulations, to accomplish desired goals. In this analysis, exterior walls with DIMs are controlled to minimize heating and cooling thermal loads in residential buildings, located in different climate zones in Canada. In particular, 2-step manual controls are used to switch the R-value of variable insulation between low and high levels based on the thermal interactions between the outside and inside a prototypical one-story home, thereby reducing heating and cooling requirements while maintaining thermal comfort. According to the analysis's findings, dynamic insulations can drastically lower the amount of energy needed to run heating and cooling systems. The use of 2-step control techniques operating DIMs, in particular, can lower yearly energy consumption by up to 44% for space cooling and by up to 33% for space heating, resulting in up to 36% annual energy savings.
Researching City-scale Water Resource Improvement through Rainwater: Green Roof in Private Realm
(2023-03-01) Wang, Junlin Jr; Valeo, Caterina Jr; Mukhopadhyaya, Phalguni Jr
Rainwater management has been challenging for many jurisdictions, including the City of Vancouver, as population growth and climate change strain the drainage and sewer systems leading to implications for water safety. Urban rainwater runoff discharges directly to the sewer and drainage system and contributes to pollutants that are toxic to fish and other aquatic species. The green roof, a well-established green rainwater infrastructure, is an innovative approach to enhancing rainwater management and making the urban landscape more sustainable, environmental, and livable using vegetation. From the literature review, a green roof ensures the quality and quantity of collected rainwater, improves building energy efficiency, absorb air pollutants, reduce urban heat island effect and gas house emission, bring aesthetic benefits, and preserve habitat for displaced creatures. The ongoing green roof performance has restrictions on many factors: substrate layer depth, temperature, moisture condition, weather events intensity and period, and proper operation and maintenance. Overall, green roof retains precipitation effectively even aged, with a higher percentage in a moderate climate. Portland and Toronto prioritized on-site infiltration by green rainwater infrastructure in their rainwater management strategies and policies, although their approaches and requirements may differ. Portland and Toronto both have an independent green roof standard in addition to their rainwater management strategy. Portland focuses on a post-occupancy inspection program to monitor the green roof's ongoing performance, while Toronto established a Green Roof Bylaw to encourage the implementation of green roofs. Both cities have advanced strategies which could provide a valuable example with lessons learned from other jurisdictions, including City of Vancouver. This research aims to analyze the available green roof monitoring program in different cities with their establishing process and provide suggestions to jurisdictions for developing comprehensive monitoring programs in the private realm to ensure the implementation and performance of green roofs and other green rainwater infrastructures.
Adaptive Cruise Controller Design in Vehicular Applications
(2022-12-07) Zhao, Yang; Shi, Yang
As the most popular active safety driving technology, adaptive cruise control is favored by major car producers and their end-users. This report presents the history, and the functional and regulatory requirements of the ACC (adaptive cruise control) systems, and also introduces three popular controllers and their design approaches. Then a full or partial realization of the ACC function is accomplished by modelling of the ACC system in MATLAB/ Simulink (R2021b) and using the three controllers for simulation respectively. Finally, through a set of designed test scenarios and the simulations accordingly, the designs of the ACC systems are validated, followed by a discussion on the advantages and disadvantages of each control methodology.
Graphical performance characterization of membrane modules for RO and PRO processes
(2022-11-05) Moreira, Luiz; Struchtrup, Henning
Environmental issues have been stricken our planet in different areas. Current worldwide problems, for instance, water shortage and the increasing demand for energy can be mitigated by employing technological mechanisms, such as a well-established osmotic process for salt water desalination known as reverse osmosis (RO), and a promising technology for generating power from salinity gradient sources, called pressure retarded osmosis (PRO). This work aims to mathematically model the core component of RO and PRO systems, which is the membrane module, working in different conditions and graphically characterize its efficiency using performance indicators to support researchers and people in industry to design and implement RO and PRO systems in a less complex and more reliable way. To reach this goal, segmented mathematical models of a 5-inch scale Toyobo HP5255SI-H3K hollow fibers membrane module were developed for the RO and PRO processes using the solution-diffusion and friction-concentration polarization transport models, mass balances and pressure drop equations. After validating the models and performing simulations, the performance curves obtained were able to provide the optimum values of inlet parameters for both RO and PRO processes that led to generate the best results in terms of volume flow rate and salinity of permeate, recovery ratio, salt rejection rate, power density and net power output. In addition, some interesting discoveries were acquired from the results such as an unused portion of membrane area in the radial direction and the influence of flow velocities on entropy generation, salt and water fluxes within the membrane module in the RO process, as well as how input parameters as hydraulic pressures and flow rates impact power generation in PRO systems and how to mitigate the reverse salt flux in this process. Finally, the possibility of integrating RO and PRO systems to desalinate salt water and produce power from the resulting permeate and brine solutions is also discussed and arguments on the reasons why such systems would not work with current technology are presented.
Measurement of High-Frequency Milling Forces with Dynamic Compensation
(2022-04-28) Jullien-Corrigan, Alan; Ahmadi, Keivan
Piezoelectric dynamometers are widely used to measure cutting forces during milling operations for diagnostic, process monitoring, and research and development purposes. However, the bandwidth of tooth passing frequencies that can be measured has an upper limit due to the electromechanical dynamics of the measurement device. As a result, high-frequency forces cannot be accurately measured. Even if an effort is made to match the cutting conditions to the specifications of the dynamometer, the higher harmonics of the tooth-passing frequency are still affected so that the resulting measurements are distorted. In this work, two new (for milling applications) methods are presented to reconstruct the machining forces from the distorted measurement signal and compared to an existing method, the Augmented Kalman Filter (AKF). The first method implements a Sliding Mode Observer (SMO) to estimate the machining forces at each time-step from the measured signal. The second method, referred to as Regularized Deconvolution (RD), considers the convolution sum of the input machining force and the impulse response of the system, and then reconstructs the machining force signal by regularizing a related inverse problem. All three methods are implemented in a simulation study that imitates the cutting conditions used in a latter experimental cutting test in which the above methods are again used to recover the true machining forces and their relative performance evaluated and compared. A transfer function model of the electomechanical dynamics of a Kistler dynamometer is identified and incorporated into the simulation study and the experiment. The results of this work find that, while all three methods reconstruct the true machining forces reasonably well, SMO has clear advantages for processes carried out over time in which the system dynamics changes. AKF also performs better than RD, but is not robust against variations in system dynamics. Despite its drawbacks, RD does have the advantage of being the method that only requires one parameter to be tuned, whereas the other methods require the tuning of two or more parameters.
Path Planning using Deep Q-learning Network and Artificial Potential Fields for a Robot Formation
(2022-04-28) Yang, Yang; Constantinescu, Daniela
A common path planning in robotics is the artificial potential field method. The artificial potential field is the superposition of the attractive potential field generated by the target and the repulsive potential field generated by the obstacles. The total force on a robot moving in the artificial potential field is the sum of the attractive force from the attractive field and the repulsive force from the repulsive potential field. The robot then moves in the direction of the total force, whose direction is along the negative gradient of the artificial potential field. If the artificial potential field has a unique minimum at the target, the robot will reach it without hitting obstacles. However, if the artificial potential field has multiple minima, the robot may arrive at a location with locally minimum potential. The total force on the robot is zero at such a local minimum and drives the robot towards it in its neighbourhood. Therefore, the robot becomes stationary and cannot arrive at the target. Deep Q-learning network has been proposed to overcome the local minima problem of robot path planning based on artificial potential field. This project investigates the impact of combining deep Q-learning network with an artificial potential field, as proposed in [1], to achieve path planning for a robot formation. Specifically, it uses a deep Q-learning network to guide the robot formation to the target in an artificial potential field created by an environment with multiple targets and obstacles. Deep Q-learning network is a type of deep reinforcement learning, that is, it combines reinforcement learning and deep learning. As in [2], a black-hole potential field is also added in the artificial potential field. Simulation results show that deep Q- learning network can good results in the fixed artificial potential field. Out of 40 tests, a robot reaches the target without hitting any obstacles in 37 tests. However, the deep Q-learning network does not improve path planning performance in different artificial potential fields. During training in four different artificial potential fields, the robot can not achieve collision-free path planning in two of them. Out of four tests, the robot achieves collision-free path planning in three tests. Similarly, the robot formation successfully finds the target in the fixed artificial potential field. Out of eight tests, all tests are successful.
Investigation of Wave Variable Control on Bilateral Teleoperation System with Constant Time Delays
(2021-12-23) Sun, Xizhe; Constantinescu, Daniela
Teleoperation Systems allow human operators to perform complex tasks in remote and hazardous environments. A wide variety of applications based on teleoperation systems, including space exploration, undersea detection, minimally invasive surgery etc. have made great contributions to our society. Various feedbacks like sound, visual and haptic feedbacks are sent to the user in order to enhance the user experience and improve system performance. With the help of haptic feedbacks, human operators can achieve remote control and interact with an inaccessible environment. However, time delays between the master and the slave may cause instability. To guarantee the stability and improve the transparency, many approaches such as passivity-based control, adaptive control, robust control, and sliding mode control are widely researched. This project studied Wave Variable Control approach in teleoperation systems, one of passivity-based control theories, and discussed its advantages and disadvantages.
Development of a Finite Element Analysis Workflow for Studying Reverse Total Shoulder Arthroplasty
(2021-06-13) Shirzadi, Hooman; Giles, Joshua W
The primary articulation of the shoulder joint is a multi-axis synovial ball and socket joint. By having a loose connection, it provides a wide range of motion; however, this means the joint lacks robustness and is prone to damage most commonly from shoulder dislocations. Rotator cuff tears also cause major problems by limiting the ability to lift the arm into abduction positions. It is common that this insufficiency aggravates arthritis within the shoulder. The study focuses on methods for investigating, describing and quantifying the effects of implant geometric properties on fixation and contact mechanics for a reverse total shoulder arthroplasty implant. The investigation presents the result of finite element analyses under heavy loading condition on a reverse shoulder implant. These finite element results are validated through comparison to experimental data on the same prosthesis. The implant is modelled using MIMICS (Materialise, Leuven, BE) and imported into SolidWorks and then ABAQUS (Simulia, Providence, USA) to analyse the distribution of displacement across the scapula. Details of interaction, boundary conditions, loads and material properties are all obtained from research and applied to the model to portray realistic behavior. The micromotion displacements of the implant were observed in the current study. The models follow the expected trends of the mechanics and what was seen in the experimental data and thus the modeling workflow makes sense overall. This can help to demonstrate the differences between different surgical options (e.g. various reverse implant designs), which may provide a basis from which improved designs can be built and allow more accurate methods to be developed in analyzing shoulder implant effectiveness. However, the method presented here needs further refinement to calibrate the models before it could be utilized in order to answer clinical questions.
Analyzing Thermal Mass Property of Concrete using Design Requirements, Thermal Properties and Energy Modelling Simulations
(2021-04-27) Tripuraneni, Akhil; Valeo, Caterina; Gupta, Rishi
Anisotropic Rheology of Non-spherical Grains
(2021-03-08) Ozyuksel, Barbaros; Nadler, Ben
The flow of granular materials is the subject of various academic research and industrial applications. The rheology of granular materials spans from packing, sorting and transportation of the pharmaceuticals to the study of avalanche dynamics. The rheology of shape isotropic (spherical) materials has been studied extensively and successful constitutive stress response models are present in the literature. In most real-life applications the grains are shape anisotropic (ellipsoidal), and their rheological and mechanical response is more complex than spherical grains. The shape anisotropy of the grains brings the effect of the grain orientation to their response. Isotropic granular rheology models neglect the effect of the grain orientation and shape on the mechanical response of the system. This report proposes a novel continuum stress response model based on isotropic granular rheology and utilizes a kinematic continuum model to capture the effect of the grain orientation. The representation theorem has been used to obtain the full description of the novel isotopic tensor valued function of the tensor variable. Dissipation inequality applied as a guide during the construction of the continuum model. The model predictions showed good agreement with the available experimental results of the simple shear flow at the steady-state. To reveal the complete capabilities of the model, the 2-D simple shear flow was studied. The results indicated that the model well captured the nonmonotonic behaviour of the effective viscosity of the flow caused by the effect of grain shape and the orientation with two material parameters. This study revealed the future application potential of the proposed phenomenological model, and concluded as a successful step forward in understanding and modelling the complex character of the shape anisotropic granular materials.
Comparison of different machine learning algorithms to predict mechanical properties of concrete
(2021-01-19) Koya, Bhanu Prakash; Caterina, Valeo; Rishi, Gupta
Concrete is the most widely used construction material throughout the world. Extensive experiments are conducted every year to measure various physical, mechanical, and chemical properties of concrete involving a hefty amount of money and time. This work focuses on the utilization of Machine Learning (ML) algorithms to predict a wide range of concrete properties and avoiding unnecessary experimentation. In this work, six mechanical properties of concrete namely Modulus of Rupture, Compression strength, Modulus of Elasticity, Poisson's ratio, Splitting tensile strength and Coefficient of thermal expansion were estimated by applying five different ML algorithms viz. Linear Regression, Support Vector Machine, Decision Tree, Random Forest, and Gradient Boosting models on the Wisconsin concrete mixes database. Further, these ML models were evaluated to identify the most suitable model that can reliably predict the mechanical properties of concrete. The approach followed in this research was verified using the 10-fold Cross- Validation technique to get rid of training and testing split bias. The Grid Search Cross Validation method was used to find the best hyperparameters for each algorithm. Root mean squared error (RMSE) and Nash and Suctcliffe Efficiency (NS) results showed that the Support Vector Machine outperformed all other models applied on the datasets. Support Vector Machine predicted the Modulus of Rupture at a curing age of 28 days with an NS score of 0.43 which is 34% and 26% better than the NS scores of Random Forest and Gradient Boosting advanced algorithms, respectively. This suggests that the Support Vector Machine algorithm with its NS score further improved can be used for predicting new data points at least for potentially similar systems.

Browse

Recent Submissions