Computational Engineering Commons^™

Joint Learning Of Unknown Safety Constraints And Control Policies In Reinforcement Learning, Lunet Abiye Yifru

Graduate Theses, Dissertations, and Problem Reports (ETD)

Reinforcement learning (RL) has revolutionized decision-making across a wide range of domains over the past few decades. Yet, deploying RL policies in real-world scenarios presents the crucial challenge of ensuring safety. Traditional safe RL approaches have predominantly focused on incorporating predefined safety constraints into the policy learning process. However, this reliance on predefined safety constraints poses limitations in dynamic and unpredictable real-world settings where such constraints may not be available or sufficiently adaptable. Bridging this gap, we propose a novel approach that concurrently learns a safe RL control policy and identifies the unknown safety constraint parameters of a given environment. …

Enhancing Pipeline Simulations Through Artificial Intelligence And Machine Learning: A Smart Proxy Modelling Approach, Afeez Shittu

Graduate Theses, Dissertations, and Problem Reports (ETD)

ABSTRACT

Enhancing pipeline simulations is essential for improving operational efficiencies and effectively managing risks in the oil and gas industry. Traditional pipeline simulators, relying heavily on mathematical modeling assumptions, often face limitations due to their high energy and computational demands. This thesis addresses these challenges by introducing an innovative approach that integrates artificial intelligence (AI) and machine learning (ML) through a smart proxy model, offering a more efficient, cost-effective, and flexible alternative to conventional full-physics models used in pipeline simulation software.

The primary aim of this research is to develop and implement a smart proxy model capable of accurately predicting …

Towards Digital Twins For Optimizing Metrics In Distributed Storage Systems - A Review, May Itani, Layal Abu Daher, Ahmad Hammoud

BAU Journal - Science and Technology

With the exponential data growth, there is a crucial need for highly available, scalable, reliable, and cost-effective Distributed Storage Systems (DSSs). To ensure such efficient and fault tolerant systems, replication and erasure coding techniques are typically used in traditional DSSs. However, these systems are prone to failure and require different failure prevention and recovery algorithms. Failure recovery of DSS and data reconstruction techniques take into consideration different performance metrics optimization in the recovery process. In this paper, DSS performance metrics are introduced. Several recent papers related to adopting erasure coding in DSSs are surveyed together with highlighting related performance metrics …

Smart System For Wheat Diseases Early Detection, Rustam Baratov, Himola Sunnatillayeva, Almardon Mamatovich Mustafoqulov

Chemical Technology, Control and Management

This paper presents a smart system for early detection of wheat plant diseases in the vegetation period. The proposed smart system allows detecting three types of wheat diseases, particularly yellow rust, powdery mildew and septoria at early stage and significantly improves the soil and ecology by locally spraying harmful chemicals just to sickness plants. The proposed diagnostic program is created in the C++ programming language. The basic structure of the smart system consists of Raspberry PI 4 MODULE, Logitech HD Pro Webcam C920, buzzer, HC-SR04 distance sensor, DC motor driver, AC motor, power supply, relay and some digital devices.

Combined Risk Based Inspection And Fault Tree Analysis For Repetitive 3-Phase Line Piping Leakage At West Java Offshore Topside Facility, Dona Yuliati, Akhmad Herman Yuwono, D. Rizal Asral, Donanta Dhaneswara

Journal of Materials Exploration and Findings

Hydrocarbon releases might result in serious consequences in various aspects. In addition to the contribution to environmental pollution, repetitive leakages need high repair costs. This study aim is to minimize potential repetitive leakage for other typical 3-phase piping systems. We conducted the risk assessment by adopting Risk Based Inspection (RBI) API 581 to identify risk level, calculating piping lifetime, recommended inspection plan and mitigations. The most relevant root causes can be obtained through quantitative Fault Tree Analysis (FTA). Observation and investigation was taken from eight 3-phase piping systems that experienced repetitive leakages. It has been found that the risk level …

Nitrogen Gas Quenching Pressure Effect On Bs S155 Alloy Steel In Vacuum Furnace, Agus Mulyadi Hasanudin, Eddy Sumarno Siradj

Journal of Materials Exploration and Findings

The production of metal and alloy products requires the use of heat treatment, when during the heat treatment process, quenching is a crucial step. The quenching medium can be anything from water, a salt bath, oil, air and gas. In a vacuum furnace, pressurized gas, most frequently nitrogen (N₂) gas, serves as one of the quenching mediums. One of the drawbacks of the quenching process is the distortion and dimensional change of the parts. This paper aims to investigate the influence of nitrogen gas quenching pressure on the distortion and dimensional change of aerospace actuator gear planet parts …

Pipeline Risk Analysis Optimization With Monte Carlo Method Using Gamma Distribution, Farhan Rama Digita, Jaka Fajar Fatriansyah, Abdul Rahim Ridzuan, Hanna Ovelia, Imam Abdillah Mas'ud, Irma Hartia Tihara, Baiq Diffa Pakarti Linuwih

Journal of Materials Exploration and Findings

The inspection process of piping components in the oil and gas industry is one of the most crucial things, given the high risk posed by pipeline system failures, which have a huge impact on losses, both from environmental and financial aspects. Risk-based inspection with the Monte Carlo method is one of the efforts that can be made to minimize failures in piping systems, by involving data distribution to calculate the probability of component failure. Although the normal distribution is commonly used for generating random variables, its use in corrosion rate calculation can lead to overestimation due to negative corrosion rate …

Early-Age Strength And Failure Characteristics Of 3d Printable Polymer Concrete: Numerical Modelling And Experimental Testing, Mohammad Amin Dehghani Najvani

Civil Engineering ETDs

The time-dependent rheological and early-age strength parameters of Polymer concrete (PC) were investigated. PC flow decreased, and the static yield stress and thixotropy increased as the specimen aged, while the dynamic yield stress remained unchanged. The study found that the Herschel-Bulkley model can accurately describe the PC’s rheological behavior over time. The change in cohesion strength and internal friction angle over time was observed using uniaxial unconfined compression and direct shear tests of fresh PC. A time-dependent MohrCoulomb failure criterion was established. Temperature analysis was used to determine the gel time and explain the evolution of the time-dependent early-age strength …

An Investigation Into Applications Of Canonical Polyadic Decomposition & Ensemble Learning In Forecasting Thermal Data Streams In Direct Laser Deposition Processes, Jonathan Storey

Theses and Dissertations

Additive manufacturing (AM) is a process of creating objects from 3D model data by adding layers of material. AM technologies present several advantages compared to traditional manufacturing technologies, such as producing less material waste and being capable of producing parts with greater geometric complexity. However, deficiencies in the printing process due to high process uncertainty can affect the microstructural properties of a fabricated part leading to defects. In metal AM, previous studies have linked defects in parts with melt pool temperature fluctuations, with the size of the melt pool and the scan pattern being key factors associated with part defects. …

Timeseries Forecasting Of U.S. Housing Price Index Using Machine Learning And Deep Learning Models, Krishna Chaitanya Nunna

Dissertations

Time series forecasting is a promising technique for various applications which predicts future values or patterns by taking historical data as base. Forecasting future trends is very beneficial for different industries to make valuable decisions and strategies. One such industry is housing market; it has biggest influence on U.S. economy. Housing price index (HPI) is a one of the crucial economic indices published by various government funded and private agency to benefit several industries and individuals for better analysis of future trends of housing market.

Several factors influence the HPI, economical, geographical, and demographic features. Development of traditional time series …

Intelpvt And Opt-Stvit: Advances In Vision Transformers For Object Detection, Classification And Video Recognition, Divya Nimma

Dissertations

Since the advent of Transformers, followed by Vision Transformers (ViTs), great success has been achieved by researchers in the field of computer vision and object detection. The difficult mechanism of splitting images into fixed patches posed a serious challenge in this arena and resulted in the loss of useful information at the time of object detection and classification. We propose an innovative Intelligent-based patching mechanism to overcome the challenges and integrate it seamlessly into the conventional Patch-based ViT framework. The proposed method enables the utilization of patches with flexible sizes to capture and retain essential semantic content from input images, …

Automated Usability Evaluation Utilizing Log Files And Data Mining Techniques., Sima Shafaei

Electronic Theses and Dissertations

Usability evaluation is one of the essential aspects of software production. This evaluation should be done during the entire life cycle of a software application, from pre-production to production and post-production. However, the collection and evaluation of usability data can be a very challenging, time-consuming, and expensive task to be conducted manually, particularly for certain types of products and working conditions. These challenges may include the need to recruit participants fully engage and motivate them during evaluation, and factor in environmental conditions. Other challenges may include collecting data in real-world environments, especially when the users are geographically dispersed, minimizing evaluator …

Convolution And Autoencoders Applied To Nonlinear Differential Equations, Noah Borquaye

Electronic Theses and Dissertations

Autoencoders, a type of artificial neural network, have gained recognition by researchers in various fields, especially machine learning due to their vast applications in data representations from inputs. Recently researchers have explored the possibility to extend the application of autoencoders to solve nonlinear differential equations. Algorithms and methods employed in an autoencoder framework include sparse identification of nonlinear dynamics (SINDy), dynamic mode decomposition (DMD), Koopman operator theory and singular value decomposition (SVD). These approaches use matrix multiplication to represent linear transformation. However, machine learning algorithms often use convolution to represent linear transformations. In our work, we modify these approaches to …

Machine Learning For Kalman Filter Tuning Prediction In Gps/Ins Trajectory Estimation, Peter Wright

Electronic Theses, Projects, and Dissertations

This project is an exploration and implementation of an application using Machine Learning (ML) and Artificial Intelligence (AI) techniques which would be capable of automatically tuning Kalman-Filter parameters used in post-flight trajectory estimation software at Edwards Air Force Base (EAFB), CA. The scope of the work in this paper is to design and develop a skeleton application with modular design, where various AI/ML modules could be developed to plug-in to the application for tuning-switch prediction.

Study On Effect Of 3d Printing Parameters On Surface Roughness And Tensile Strength Using Analysis Of Variance, Faqih Fadillah, Heru Suryanto, Suprayitno Suprayitno

Journal of Mechanical Engineering Science and Technology (JMEST)

Fused deposition modeling of 3D printing is the process of making workpieces or parts by adding filaments to each layer. Some indicators of a high-quality product of 3D printing are the precisions dimensions, the surface roughness, and tensile strength. This research aims to find the parameters most affecting surface roughness and tensile strength. The research design used an experimental method with input parameters: (1) print speed (15-35 mm/s), (2) print temperature (200-210C), (3) layer height (0.1 – 0.3 mm), (4) infill line directions (0-90), and dependent variables were surface roughness and tensile strength. The data distribution used the L9 orthogonal …

Analyzing Transition Metallic Catalytic Converter Impact On Four-Stroke Motorcycle Fuel Consumption, Sudirman Rizki Ariyanto, Ata Syifa’ Nugraha, Wahyu Robby Cahyadi, Yelma Dianastiti, Muhammad Yandi Pratama

Journal of Mechanical Engineering Science and Technology (JMEST)

Increased exhaust emissions from motor vehicles have become a major concern in efforts to reduce air pollution. One developed solution is the use of transition metallic catalytic converter (TMCC) technology in vehicle exhaust systems. This study aims to compare the fuel consumption efficiency of three types of exhaust systems, namely standard exhaust without a catalyst (STD WC), the standard exhaust with Original Equipment Manufacturer catalyst (STD OEM), and an exhaust system equipped with a Copper-Coated Chrome Metallic Catalytic Converter (TMCC CuCr). The data analysis method employed a quantitative approach by collecting fuel consumption data at each rpm and analyzing the …

Vertical Shaft Kinetic Turbine Performance Using A Cup-Shape Blade, Kennie A. Lempoy, Nita C. V. Monintja, Arwanto M. A. Lakat

Journal of Mechanical Engineering Science and Technology (JMEST)

The abundant availability of alternative energy sources forms the fundamental foundation for rural energy development. In particular, renewable energy sources are indispensable in rural areas due to their ecofriendly nature and widespread accessibility, which helps preserve the environment by avoiding pollution. Indonesia, a country blessed with vast water resources, especially its rivers, holds enormous potential for harnessing kinetic energy derived from fast-flowing river waters. Effective utilization of this kinetic energy has the potential to address energy shortages in the country. In this context, there is a pressing need to enhance the efficiency of kinetic turbines designed to function as electricity …

Artificial Neural Network-Based Modeling Of Performance Spark Ignition Engine Fuelled With Bioethanol And Gasoline, Susi Marianingsih, Farhanna Mar’I, Hendry Y. Nanlohy

Journal of Mechanical Engineering Science and Technology (JMEST)

Machine learning technology can distinguish the relationship between engine characteristics and performances. Therefore, the goal of the present work is to predict the performance parameters of a singlecylinder 4-stroke gasoline engine at different ignition timings using a blended mixture of gasoline and bioethanol by an artificial neural network (ANN). Experimental data for training and testing in the proposed ANN was obtained at a dynamic speed and full load condition. An ANN model was developed based on standard Back-Propagation algorithm for the spark ignition engine. Multi-layer perception network (MLP) was used for non-linear mapping between the input and output parameters. An …

The Effect Of Adding Aluminium On The Performance Of Zno Nrs/Pani In Their Application As Photoelectrochemical Water Splitting, Eprilia Trikusuma Sari, Nandang Mufti, Anissa Chairani Alfin Nadhira, Hari Wisodo, Markus Diantoro

Journal of Mechanical Engineering Science and Technology (JMEST)

Photoelectrochemical (PEC) is a new renewable energy technology that converts H2O into hydrogen and oxygen gas with the help of sunlight. A photoelectrochemical cell device consists of three main components, one of which is the photoanode. One of the materials that can be used as a photoanode is ZnO which has good electrical properties and is non-toxic. Nanorods-structured ZnO has the advantage of being able to increase light absorption due to its high surface area. However, the resulting performance is still quite low. So it is necessary to make modifications to the photoanode, one of which is by adding aluminium …

Effect Of The Homogenization Process On Titanium Oxide-Reinforced Nanocellulose Composite Membranes, Taufiq Mahsuli, Aisyah Larasati, Aminnudin Aminnudin, Jibril Maulana

Journal of Mechanical Engineering Science and Technology (JMEST)

Indonesian pineapple production can reach 200 tons per day; however, pineapples generate a significant amount of waste. Pineapple peel waste can be used to make membranes. Composite membranes containing TiO2 have dense properties, low porosity, and increase the mechanical strength of the cellulose sheet. This research uses various ultrasonic homogenizers to homogenize the distribution of nanocellulose and TiO2 (50% and 100% power with 30, 60, and 90 minutes). The casting method is used to shape the membrane. The SEM test shows that the higher the power used and the longer the sonication time, the less agglomeration of about 1.63%/ cm2 …