Mechanical Engineering Commons^™

A Convolutional Neural Network (Cnn) For Defect Detection Of Additively Manufactured Parts, Musarrat Farzana Rahman

Masters Theses

“Additive manufacturing (AM) is a layer-by-layer deposition process to fabricate parts with complex geometries. The formation of defects within AM components is a major concern for critical structural and cyclic loading applications. Understanding the mechanisms of defect formation and identifying the defects play an important role in improving the product lifecycle. The convolutional neural network (CNN) has been demonstrated to be an effective deep learning tool for automated detection of defects for both conventional and AM processes. A network with optimized parameters including proper data processing and sampling can improve the performance of the architecture. In this study, for the …

Design Of Composite Joints Using Machine Learning Approaches, Natalie Richards

Williams Honors College, Honors Research Projects

Adhesively bonded joints have an advantage in joining dissimilar engineering materials due to their high structural efficiency and being lightweight. These joints are either between two opposite laminates or between a composite laminate and a metal structure. The aerospace and automotive industries have seen an increase in utilizing these adhesive joints in their engineering applications. Joint strength along with the failure mode (adhesive, delamination, etc.) is the most important parameter to evaluate when understanding the capability of the adhesive joint. In this paper, a regression and a classification machine learning (ML) model are utilized to predict the failure load and …

Development Of Machine Learning Algorithm To Identify High-Emitters From On-Road Data For Heavy-Duty (Hd) Vehicles, Filiz Kazan

Graduate Theses, Dissertations, and Problem Reports (ETD)

The process of on-road, heavy-duty engine family certification is regulated by the United States Environmental Protection Agency (US EPA). Currently, the US EPA 2010 emissions standards require the threshold from the Federal Testing Procedure (FTP) engine dynamometer cycle to be at or below a brake-specific NO_x (bs-NO_x) value of 0.20 g/bhp-hr for heavy-duty (HD) engines. The engine manufacturers are also required to conduct in-use portable emission measurement system (PEMS) testing to prove their products' compliance. The selected vehicles are required to satisfy not-to-exceed (NTE) analysis under normal driving conditions in the heavy-duty in-use testing (HDIUT) program. California …

Exploiting Building Demand Flexibility Through Machine Learning For Building-To-Grid Integration, Hannah Charlene Fontenot

Dissertations - ALL

Demand flexibility – the ability to adjust a building's load profile across different timescales – is a key aspect of the ongoing effort to increase interconnectivity between buildings and the power grid. By harnessing their demand flexibility, buildings can provide significant benefits to the grid and bolster grid resilience and reliability. To facilitate the transition toward the "smart grid", new and intelligent control approaches are required that can seamlessly integrate building, occupant, and grid data and effectively control multiple building assets to provide grid services while maintaining occupants' required thermal comfort levels and reducing the building's overall energy consumption and …

Predictive Computational Materials Modeling With Machine Learning: Creating The Next Generation Of Atomistic Potential Using Neural Networks, Mashroor Shafat Nitol

Theses and Dissertations

Machine learning techniques using artificial neural networks (ANNs) have proven to be effective tools to rapidly mimic first principles calculations. These tools are capable of sub meV/atom accuracy while operating with linear scaling with respect to the system size. Here novel interatomic potentials are constructed based on the rapid artificial neural network (RANN) formalism. This approach generates precise force fields for various metals that have historically been difficult to describe at the atomic scale. These force fields can be utilized in molecular dynamics simulations to provide new physical insights. The RANN formalism, which is incorporated into a LAMMPS molecular dynamics …

A Machine Learning Method For The Prediction Of Melt Pool Geometries Created By Laser Powder Bed Fusion, Jonathan Ciaccio

LSU New Orleans Theses and Dissertations

A machine learning model is created to predict melt pool geometries of Ti-6Al-4V alloy created by the laser powder bed fusion process. Data is collected through an extensive literature survey, using results from both experiments and CFD modeling. The model focuses on five key input parameters that influence melt pool geometries: laser power, scanning speed, spot size, powder density, and powder layer thickness. The two outputs of the model are melt pool width and melt pool depth. The model is trained and tested by using the k fold cross validation technique. Multiple regression models are then applied to find the …

Investigation Of The Prevalence Of Faults In The Heating, Ventilation, And Air-Conditioning Systems Of Commercial Buildings, Amir Ebrahimifakhar

Durham School of Architectural Engineering and Construction: Dissertations, Theses, and Student Research

This dissertation describes a large-scale investigation of heating, ventilation, and air-conditioning (HVAC) fault prevalence in commercial buildings in the United States. A multi-year dataset with 36,556 pieces of HVAC equipment including air handling units (AHUs), air terminal units (ATUs), and packaged rooftop units (RTUs) was analyzed to determine values for several HVAC fault prevalence metrics. The primary source of data for this study comes from three commercial fault detection and diagnostics (FDD) providers. Since each FDD provider uses different terms to refer to the same fault in an HVAC system, a mapping function was created for each FDD provider’s dataset, …

Searching Extreme Mechanical Properties Using Active Machine Learning And Density Functional Theory, Joshua Ojih

Theses and Dissertations

Materials with extreme mechanical properties leads to future technological advancements. However, discovery of these materials is non-trivial. The use of machine learning (ML) techniques and density functional theory (DFT) calculation for structure properties prediction has helped to the discovery of novel materials over the past decade. ML techniques are highly efficient, but less accurate and density functional theory (DFT) calculation is highly accurate, but less efficient. We proposed a technique to combine ML methods and DFT calculations in discovering new materials with desired properties. This combination improves the search for materials because it combines the efficiency of ML and the …

Use Of Machine Learning For Automated Convergence Of Numerical Iterative Schemes, Leonardo A. Bueno-Benitez

Doctoral Dissertations and Master's Theses

Convergence of a numerical solution scheme occurs when a sequence of increasingly refined iterative solutions approaches a value consistent with the modeled phenomenon. Approximations using iterative schemes need to satisfy convergence criteria, such as reaching a specific error tolerance or number of iterations. The schemes often bypass the criteria or prematurely converge because of oscillations that may be inherent to the solution. Using a Support Vector Machines (SVM) machine learning approach, an algorithm is designed to use the source data to train a model to predict convergence in the solution process and stop unnecessary iterations. The discretization of the Navier …

Real-Time Material State Assessment Of Composites Using Artificial Intelligence And Its Challenges, Muthu Ram Prabhu Elenchezhian

Mechanical and Aerospace Engineering Dissertations - Archive

Over several decades of careful experimental investigation and exhaustive development of discrete damage analysis methods including integrated computational mechanics methods, our community knows a great deal about how discrete defects such as matrix cracks and defect growth (e.g. delamination) can be predicted in structural composites. For many practical situations controlled by laminated multiaxial composite structures, the loss of performance and “sudden death” end of life is controlled by defect coupling which becomes a precursor to fracture plane development. These interaction sequences are highly dependent on local details of manufacture, design configurations, and loading for a given application material and influenced …

Machine Learning Model Optimization With Hyper Parameter Tuning Approach, Md. Riyad Hossain, Douglas Timmer

Manufacturing & Industrial Engineering Faculty Publications

Hyper-parameters tuning is a key step to find the optimal machine learning parameters. Determining the best hyper-parameters takes a good deal of time, especially when the objective functions are costly to determine, or a large number of parameters are required to be tuned. In contrast to the conventional machine learning algorithms, Neural Network requires tuning hyperparameters more because it has to process a lot of parameters together, and depending on the fine tuning, the accuracy of the model can be varied in between 25%-90%. A few of the most effective techniques for tuning hyper-parameters in the Deep learning methods are: …

A Dynamic Active Noise Control System For Live Music Attenuation, Elliot James Krueger

Graduate Research Theses & Dissertations

This thesis proposes a system design that will be suitable for applying active noise control (ANC) effectively to live musical instruments. The design consists of three parts: a signal separation section, an instrument classification section, and the active noise control section. The signal separation section will split up the music signals. The instrument classification section will identify the signals, and the ANC section will attenuate the music signal based on the previous information from the other sections. The two instruments of focus will be the trombone and tuba for their low frequency and ability to be quite loud in a …

A Theory-Supported Machine Learning Model For The Prediction Of Melt Pool Geometry And Optimal Process Window In Metal Additive Manufacturing, Sina Tayebati

Graduate Research Theses & Dissertations

Direct Energy Deposition (DED) is an additive manufacturing (AM) process capable of producing complicate-shaped or functionally graded components, and it is getting intense attention as a revolutionary technology to satisfy high demand in manufacturing process for the aerospace, automotive, and medical industries. However, the repeatability in geometries and properties of fabricated products is one of the most challenging issues for the DED process to be fully utilized, requiring comprehensive understanding of effect of processing conditions on the properties of fabricated parts, and development of relations among those conditions and properties. That is the motivation of this research. In this study, …

Tool Life Prediction Of Ti [C,N] Mixed Alumina Ceramic Cutting Tool Using Gradient Descent Algorithm On Machining Martensitic Stainless Steel, Joseph Daniel S, Senthil Kumar A

Journal of Mechanical Engineering Science and Technology (JMEST)

In automated manufacturing systems, most of the manufacturing processes, including machining, are automated. Automatic tool change is one of the important parameters for reducing manufacturing lead time. Machining studies on Martensitic Stainless Steel was conducted using Ti[C,N] mixed alumina ceramic cutting tool. Tool life was evaluated using flank wear criterion. The tool life obtained from experimental machining process was taken as training dataset and test dataset for machine learning. Tool life model was developed using Gradient Descent Algorithm. The accuracy of the machine learning model was tested using the test data, and 99.83% accuracy was obtained.

Machine Learning In Manufacturing: Review, Synthesis, And Theoretical Framework, Ajit Sharma, Zhibo Zhang, Rahul Rai

Business Administration Faculty Research Publications

There has been a paradigmatic shift in manufacturing as computing has transitioned from the programmable to the cognitive computing era. In this paper we present a theoretical framework for understanding this paradigmatic shift in manufacturing and the fast evolving role of artificial intelligence. Policy, Strategic and Operational implications are discussed. Implications for the future of strategy and operations in manufacturing are also discussed. Future research directions are presented.

Cnn-Based Estimation Of Sagittal Plane Walking And Running Biomechanics From Measured And Simulated Inertial Sensor Data, Eva Dorschky, Marlies Nitschke, Christine F. Martindale, Antonie J. Van Den Bogert, Anne D. Koelewijn, Bjoern M. Eskofier

Mechanical Engineering Faculty Publications

Machine learning is a promising approach to evaluate human movement based on wearable sensor data. A representative dataset for training data-driven models is crucial to ensure that the model generalizes well to unseen data. However, the acquisition of sufficient data is time-consuming and often infeasible. We present a method to create realistic inertial sensor data with corresponding biomechanical variables by 2D walking and running simulations. We augmented a measured inertial sensor dataset with simulated data for the training of convolutional neural networks to estimate sagittal plane joint angles, joint moments, and ground reaction forces (GRFs) of walking and running. When …

Artificial Intelligence In Plasma Electrolytic Micro-Oxidation For Surface Hardening - Insights From Scholarly Citation Networks And Patents., Priya Jadhav, Dr.Arun Bongale, Dr.Satish Kumar, Dr.Amit Kumar Tiwari

Library Philosophy and Practice (e-journal)

Objective - The purpose of this article is to analyze the top work areas and patent domains in the field of surface hardening by micro-arc oxidation. Also, it is directed on the opportunities of data analysis by different machine learning tools. Material and methods - The www.lens.org database is used to collect articles from Elsevier, Trans tech publications, Springer New York, MDPI, etc. to review the relevant articles as well as patents related to the topic. The result - A total of 1057 articles were published in 60 different journals and 756 patents in the area of research under various …

High Tech, High Risk: Tech Ethics Lessons For The Covid-19 Pandemic Response, Emanuel Moss, Jacob Metcalf

Publications and Research

The COVID-19 pandemic has, in a matter of a few short months, drastically reshaped society around the world. Because of the growing perception of machine learning as a technology capable of addressing large problems at scale, machine learning applications have been seen as desirable interventions in mitigating the risks of the pandemic disease. However, machine learning, like many tools of technocratic governance, is deeply implicated in the social production and distribution of risk and the role of machine learning in the production of risk must be considered as engineers and other technologists develop tools for the current crisis. This paper …

Computational Studies Of Thermal Properties And Desalination Performance Of Low-Dimensional Materials, Yang Hong

Department of Chemistry: Dissertations, Theses, and Student Research

During the last 30 years, microelectronic devices have been continuously designed and developed with smaller size and yet more functionalities. Today, hundreds of millions of transistors and complementary metal-oxide-semiconductor cells can be designed and integrated on a single microchip through 3D packaging and chip stacking technology. A large amount of heat will be generated in a limited space during the operation of microchips. Moreover, there is a high possibility of hot spots due to non-uniform integrated circuit design patterns as some core parts of a microchip work harder than other memory parts. This issue becomes acute as stacked microchips get …

An Atomistic Approach For The Survey Of Dislocation-Grain Boundary Interactions In Fcc Nickel, Devin William Adams

Theses and Dissertations

It is well known that grain boundaries (GBs) have a strong influence on mechanical properties of polycrystalline materials. Not as well-known is how different GBs interact with dislocations to influence dislocation movement. This work presents a molecular dynamics study of 33 different FCC Ni bicrystals subjected to mechanical loading to induce incident dislocation-GB interactions. The resulting simulations are analyzed to determine properties of the interaction that affect the likelihood of transmission of the dislocation through the GB in an effort to better inform mesoscale models of dislocation movement within polycrystals. It is found that the ability to predict the slip …

Detection And Classification Of Vibrating Objects In Sar Images, Francisco German Perez Venegas

Electrical and Computer Engineering ETDs

The vibratory response of buildings and machines contains key information that can be exploited to infer their operating conditions and to diagnose failures. Furthermore, since vibration signatures observed from the exterior surfaces of structures are intrinsically linked to the type of machinery operating inside of them, the ability to monitor vibrations remotely can enable the detection and identification of the machinery.

This dissertation focuses on developing novel techniques for the detection and M-ary classification of vibrating objects in SAR images. The work performed in this dissertation is conducted around three central claims. First, the non-linear transformation that the micro-Doppler return …

Training Set Density Estimation For Trajectory Predictions Using Artificial Neural Networks, Zachary Reinke

Masters Theses

Demand on earth orbiting surveillance systems in increasing as more equipment is put into orbit. These systems rely on predictive techniques to periodically track objects. The demand on these systems may be reduced if object trajectory data to develop scalable training sets used for training artificial neural networks (ANNs) to predict trajectories of a dynamic system. These methods use multi-variable statistics to analyze data energy content to provide the ANN with low density, feature-rich, training data. The developed techniques have been shown to increase ANN prediction accuracy while reducing the size of the training set when applied to a linear …

Applied Deep Learning In Orthopaedics, William Stewart Burton Ii

Electronic Theses and Dissertations

The reemergence of deep learning in recent years has led to its successful application in a wide variety of fields. As a subfield of machine learning, deep learning offers an array of powerful algorithms for data-driven applications. Orthopaedics stands to benefit from the potential of deep learning for advancements in the field. This thesis investigated applications of deep learning for the field of orthopaedics through the development of three distinct projects.

First, algorithms were developed for the automatic segmentation of the structures in the knee from MRI. The resulting algorithms can be used to accurately segment full MRI scans in …

Reliability Analysis For Systems With Outsourced Components, Zhengwei Hu

Doctoral Dissertations

"The current business model for many industrial firms is to function as system integrators, depending on numerous outsourced components from outside component suppliers. This practice has resulted in tremendous cost savings; it makes system reliability analysis, however, more challenging due to the limited component information available to system designers. The component information is often proprietary to component suppliers. Motivated by the need of system reliability prediction with outsourced components, this work aims to explore feasible ways to accurately predict the system reliability during the system design stage. Four methods are proposed. The first method reconstructs component reliability functions using limited …

Computational Support For Predicting Requirement Change Volatility In Complex System Design, Phyo Htet Hein

Theses and Dissertations

Requirements play critical role in the design process as objective statements of stakeholders’ expectations. Design process is iterative, and requirements are also constantly changed and updated to reflect stakeholders’ expectations, design changes, regulations, resource limitations, etc. Managing requirement change is one of the most important requirement management tasks. Since requirements are driving factors in product development from initial concept development stage to final production, mismanaged requirement changes can adversely affect project health leading to monetary and time losses. The ability to assess a requirement change and predict its propagation early in the design process will enable engineers to make informed …

Refining Statistical Magnesium Models Via Machine Learning, Andrew Orme, Dr. David Fullwood

Journal of Undergraduate Research

Magnesium is a potential replacement for steels and aluminum in strength applications. Despite desirable strength and weight properties, magnesium is costly to manufacture. To reduce manufacturing costs, extensive research has been done on is a phenomenon called twinning, where a large group of magnesium atoms collectively reorient from a base orientation to a new orientation. This reorientation caused by twinning has the potential to enable easier material deformation, allowing for less costly manufacturing. Our research group pursued a novel approach to twinning research by using data mining and machine learning algorithms. Data collected from samples of magnesium using a scanning …

Micro-Manipulation Using Learned Model, Matthew A. Lyng, Benjamin V. Johnson, David J. Cappelleri

The Summer Undergraduate Research Fellowship (SURF) Symposium

Microscale devices can be found in applications ranging from sensors to structural components. The dominance of surface forces at the microscale hinders the assembly processes through nonlinear interactions that are difficult to model for automation, limiting designs of microsystems to primarily monolithic structures. Methods for modeling surface forces must be presented for viable manufacturing of devices consisting of multiple microparts. This paper proposes the implementation of supervised machine learning models to aid in automated micromanipulation tasks for advanced manufacturing applications. The developed models use sets of training data to implicitly model surface interactions and predict end-effector placement and paths that …

Inverse Analysis Of Heat Generating Body For Safety Applications, Sandeep Patil

Mechanical and Aerospace Engineering Dissertations - Archive

Inverse thermal analysis and its applications have been applied to numerous fields of science and engineering. Historically during the 1950's and early 1960's, space programs played a significant role in the advancement of solution techniques for Inverse Heat transfer Problems(IHTP). It was applied to measure the surface temperature of thermal shield of a space vehicle during its re-entry into atmosphere. Inverse analysis was also used in the estimation of thermo-physical properties of the shield at high operating temperatures. Besides thermal application, inverse technique was also used in other engineering applications such as estimation of alloy specification, design of a shape …

A Two Stage Event Based Data Driven Controller For Improved Grasping Of An Artificial Hand, Christopher Abrego

Mechanical and Aerospace Engineering Dissertations - Archive

The human hand is one of the greatest (if not the greatest) tool known to mankind for grasping objects. So much so, that researchers have been investigating the development of artificial biomimetic hands in an effort to mimic their functionality and dexterity with the indent to apply the technology to various robotic platforms; ranging from end effectors for industrial pick-and-place robotics, to upper-limb prosthetics, to humanoids. There are certain features that make this endeavor challenging such as the mechanical design, actuation and sensorization, and functionality as well as the interaction from both the view point of interacting with an end …

On The Development And Integration Of Pneumatic Extrusion Module And A Methodology To Identify Process Parameters For Additive Manufacturing Using Machine Learning, Kashish Dhal

Mechanical and Aerospace Engineering Theses - Archive

Commonly used additive manufacturing platforms have a single extrusion module based on Fused Filament Fabrication (FFF) and their processing software generates G-Codes for this FFF module using defined process parameters. These platforms and software do not accommodate different processing modules such as viscous extruders or Direct Ink Writing (DIW). This research is focused on the development of a Pneumatic Extrusion Module (PEM) capable of dispensing viscous materials such as gels or slurries controlled through a digital pneumatic valve. A PEM is developed, integrated and its performance is evaluated on a multi-modality additive manufacturing platform in the MARS Lab. The operation …