International Journal on Engineering Technology

Editorial Vol.1(1)

Nabin Chandra Sharma — 2023-12-21

No abstract available.

Bottle Recycling Machine Using Convolutional Neural Network

Samir Lohani — 2023-12-21

Every day, billions of plastic bottles and cans are used worldwide. Most of these wastes end up in landfills or as litter. These plastic bottles and aluminium cans can be recycled, to transform into other household products. As plastic bottles are cheaper to produce than to recycle, recycling is not preferred. In the meantime, people are neglecting the adverse effects of plastics and cans. For recycling plastic bottles and cans effective segregation is the major issue which requires high accuracy and precision. Most of the segregation is done manually to separate plastic bottles from other waste, which increases labor costs thus, making recycling unreliable economically. Here we have proposed a system that employs machine vision to distinguish bottles and cans using a CNN. There isn’t any sensor that can accurately identify plastic bottles with high accuracy. So, we have tried to complete this task using a deep learning algorithm. Being an automated system, it doesn't require a full-fledged team to manage which makes the recycling process feasible from monetary perspective. Different compartments are allocated for plastics and cans. When an item is accurately identified by the system, it is placed in the respective compartments. A reward is given to the user as a redeemable code to encourage recycling behavior. The system has an accuracy of 96% in identifying items and proves to be an effective pre-recycling process.

Brain Tumor Detection Using Convolutional Neural Networks: A Comparative Study

Bibhusha Ojha — 2023-12-21

Using Magnetic Resonance Imaging (MRI) images to detect brain tumors by medical practitioners is mundane and prone to errors. Misdiagnosis of brain tumors can be life-threatening, so to lessen misdiagnosis, computational techniques can be used in concert with medical professionals. Deep learning approaches have been gaining popularity in modeling and developing systems for medical image processing that can detect abnormalities quickly. The methods proposed herein are based on Convolutional Neural Networks (CNN) trained on the 'BR35H::Brain Tumor Detection 2020' dataset. A custom CNN architecture was designed, followed by the utilization of transfer learning with four pre-trained models: InceptionV3, ResNet101, VGG19, and DenseNet169 and a comparative analysis of these architectures has been presented in this paper. The experimental results show that the DenseNet169 model outperformed other models with a training accuracy of 99.83 %, test accuracy of 99.66%, precision of 99.67%, and recall of 99.67%. Additionally, ResNet101 has a 95.92% test accuracy, VGG19 has a test accuracy of 97.83%, the custom architecture has a test accuracy of 98.16%, and InceptionV3 has the lowest test accuracy of 91.66%. It has been concluded that DenseNet169 provides better results for the classification of brain tumors than other models.

Comparative Seismic Analysis, Design, and Cost Estimation of a Residential Building

Mahesh Pokhrel — 2023-12-21

The updated National Building Code (NBC 105:2020) for professionally engineered buildings provides guidelines for analyzing and designing earthquake-resistant buildings. Though it has been three years since the update of the NBC, the Indian standard code, IS 1893:2016, still seems to be prevalent; however, it has not been updated after 2016. Therefore, comparing the seismic behavior of buildings analyzed and designed by these two prevailing methods in Nepal is essential. This study aims to compare the seismic parameters, design results, and the total cost of the buildings for NBC and IS for a three-story with a staircase-cover residential building. The response Spectrum Method was used to analyze and compare the response parameters like base shear, story drift, and story displacement. The seismic response parameters are more significant; however, the total cost is not significantly greater (i.e., only 5.5% approx.) for NBC, though the base shear was relatively higher than IS. Thus, this study helps understand the structural performance and emphasizes the use of NBC in all parts of Nepal.

Comparison of CNN Architecture of Image Classification Using CIFAR10 Datasets

Yogesh Pant — 2023-12-21

This paper demonstrates image classification using deep learning. Deep learning has the inherent ability to automatically discover and extract meaningful features for specific applications. Among the popular techniques in deep learning, the convolutional neural network (CNN) stands out. CNN consists of an input layer, hidden layers, and an output layer, where meaningful features are automatically extracted from input images.

This paper presents the performance and identifies the most effective CNN architectures for accurately classifying images in the CIFAR-10 dataset. Five CNN architectures were implemented, namely [Architecture 1], [Architecture 2], [Architecture 3], [Architecture 4], and [Architecture 5] using the CIFAR-10 dataset. The architectures were selected based on the need to explore variations in convolutional filter sizes, dense layers, and batch normalization to assess their impact on CIFAR-10 image classification performance. Each architecture was trained on a standard training set and evaluated on a validation set. We used specific details on data preprocessing and training settings for a consistent and fair comparison. After training and evaluation, we have obtained the following results for each architecture.

Architecture 1 has a training accuracy of 74.7% and validation accuracy of 76.6%, Architecture 2 has 96.09% and 86.08%, Architecture 3 has 77% and 78.1%, Architecture 4 has 67.91% and 69.54%, and Architecture 5 has 94.64% and 87.34% of training accuracy and validation accuracy respectively. After conducting a comparative analysis, we found that Architecture 5 has achieved the highest validation accuracy in classifying images in the CIFAR-10 dataset. These findings suggest that Architecture 5 is a promising choice for image classification tasks involving the CIFAR-10 dataset.

Design Optimization of Fins of a Sounding Rocket for Maximum Lift-to-Drag Ratio and Minimum Radar Cross- Section Area Using ANSYS

Mandeep Prasad Shah — 2023-12-21

The optimization of a sounding rocket's fins is a crucial part of improving its performance. The geometric optimization of a sounding rocket's fins is presented in this paper. The main goal is to optimize the geometry for the minimum radar cross-section area and maximum lift-to-drag ratio. In CATIA, a 3D model was created. The L/D ratio was maximized using the ANSYS adjoint solver, and the radar cross-section area was minimized using ANSYS optimetrics. To determine the total RCS, the RCS of each fin was determined individually and then added together. Root, leading edge, tip, and trailing edge were the four parameters that were defined for the RCS optimization. The L/D ratio was increased by 8.3 times, and the RCS was decreased by 12% after optimization. Additionally, the body surface can be optimized further. The missile industry can benefit from the paper's findings.

Detection of Missing Component in PCB Using YOLO

Shivaji Pandit Chhetri — 2023-12-21

The detection of missing components in printed circuit boards (PCBs) is a critical task in the electronics manufacturing industry. The current practice of manual inspection is time-consuming and prone to human error, which can result in faulty products and increased costs. In this paper, we propose a solution that uses the YOLO (You Only Look Once) object detection algorithm to automatically detect missing electronic components in PCBs. Electronic components detection model is trained using YOLOv3 architecture. Dataset is prepared using high quality printed circuit board images and manual labeling in Label Studio. The model is trained on a dataset of 16 different electronic components commonly found in PCBs including Electrolytic Capacitor, QFP, Toroidal core Inductor, Crystal Oscillator etc. Prepared model recognizes these electronic components with an average map score of 65.8%with IoU 50% and 42.6% with IoU 95%. The results show that the proposed solution can detect the missing components.

Feasibility Study of a Motorcycle Lane: Evaluation of Travel Time and Delay Impacts at New Baneshwor Intersection

Kshitiz Dhungana — 2023-12-21

Like many South Asian countries, Nepal can be identified as a “Motorcycle Dependent Country” as the chief mode of transportation on the urban streets of Nepal compromises of Motorcycles for their daily commute. With inefficient public transportation, haphazard lane discipline, and easy accessibility and mobility of PTW, there is a rising dependence on PTW. The risk of life that comes with this dependence is also increasing in the urban streets of the country. Therefore, this research proposes Motorcycle Lanes as a measure to reduce congestion and create a risk-free riding environment. Three scenarios– a) Considering Dedicated motorcycle lanes b) Excluding right turning traffic into dedicated motorcycle lanes c) Dedicated motorcycle lanes only for two intersection legs Maitighar-Tinkune and Tinkune-Maitighar have been proposed, analyzed and compared with base data to measure the efficiency of motorcycle lanes at the New-Baneshwor intersection. The feasibility of those scenarios was checked based on traffic volume, Q-length, and vehicle delay with base data obtained from simulation in VISSIM. It was observed that difficult turning movements, unpredictable riding behavior and signal phases rendered the lanes unfeasible. A need for a holistic approach to address current and future traffic conditions as PTW are poised to experience a high upswing in foreseeable future is an essence.

Heart Disease Prediction Using Outlier Removal based Max Voting Ensemble Method

Pralhad Chapagain — 2023-12-21

Heart disease has emerged as a serious health concern for many individuals due to its high death rate around the world. The routine clinical data analysis has a significant difficulty in the early diagnosis of cardiac disease. The identification of cardiac disease may benefit from the use of machine learning. To improve machine learning models, several studies have previously been conducted. The suggested study uses the maximum voting ensemble technique of classification to effectively identify heart disease. The suggested classifier is a more reliable and accurate approach. To identify and eliminate outliers, conduct Inter quartile range outlier removal and min-max normalization during preprocessing. Accuracy, Precision, Recall, and F1 Score are calculated and evaluated against various models. For the heart disease dataset collected from the Kaggle, the suggested max voting ensemble classifier has an accuracy of 99.22%.

Identification and Ranking of Road Safety Hazardous Locations: Case Study of Kotre–Aabhukhaireni Section of Prithvi Highway

Srijan Subedi Chowi — 2023-12-21

In Nepal, due to rapid urbanization, there are many emerging towns and settlements near highways, which have raised issues of increased traffic density resulting in frequent road crashes like in Kotre- Aabhukhaireni section of Prithvi Highway. Higher crash rates depict lower safety conditions. To improve safety conditions and reduce the loss of life and property, hazardous conditions along the highways need to be identified and ranked for treatment purpose. This study is based on a methodological framework for identifying hazardous locations based on a Field Survey (condition rating) and Analytical Hierarchy Process (AHP) for ranking road safety hazardous locations of Kotre- Aabhukhaireni road Section of Prithvi Highway by weighing the safety parameters of the road section and calculating the Safety Hazardous Index (SHI). Field condition rating with a focus on pavement surface conditions, structural conditions, shoulder and drainage conditions, pedestrian infrastructure and vehicle restraint system, street lighting, traffic signs, island and road marking conditions was conducted, and risk factors were identified by comparing these variables against norms.

The study identified 15 different locations as safety hazardous locations. The results showed Yampa, Baradi, and Chirrkeni as the most hazardous ones with an SHI index of 0.852, 0.773 and 0.712 respectively, among the study section considered. In contrast, Dhaaptar was considered as the least hazardous one with SHI value of 0.210.

Identifying Product Bundle from Market Basket Analysis

Ashish Neupane — 2023-12-21

Bundling has emerged as an increasingly popular promotional strategy, offering numerous benefits to buyers and sellers and aligning perfectly with the goals of a transaction process. From the consumer’s perspective, bundling enables them to enjoy substantial savings, with an average percentage off, when purchasing a bundle package at a discount price. This significant cost reduction serves as a critical motivator for embracing bundling. Furthermore, bundling allows customers to streamline their purchasing experience by minimizing search costs. They can conveniently find all the desired products and services in a comprehensive package offered by the seller. Additionally, bundles are favorable to some individuals due to their ability to mitigate compatibility risks between various components. From the seller’s view, adopting bundling can lead to increased sales and a broader customer base. Bundling facilitates the attraction of buyers, while simultaneously raising awareness and acceptance of newly released products...The scope of this project, titled "Identifying Product Bundles from Sales Data using Market Basket," a highly performant model has been developed to aid in the identification of product bundles and market basket determination. Recognizing the multitude of prediction challenges during product sales, this project utilizes historical sales data to predict optimal product bundles. Leveraging a dataset obtained from Instacart, the project incorporates clustering and analysis processes. By thoroughly analyzing the data, the model can predict the most effective product bundles, enabling the selling company to boost its sales potential. This project specifically caters to e-commerce websites seeking to address product bundling and market basket analysis challenges. It provides a valuable platform for applying various techniques to solve problems associated with product bundling, generating comprehensive theoretical and practical resources for research-based studies. Ultimately, this project empowers businesses to make more reliable predictions about the future, enhancing their decision-making processes.

Implementation of Indirect Field Oriented Control Using Space Vector Pulse Width Modulation for the Control of Induction Motor

Sabin Kasula — 2023-12-21

For the efficient speed control of three phase Induction motor, the Vector control or Field oriented control (FOC) technique is used for the Voltage source inverter. In implementing FOC the angular position of the rotor flux vector is to be determined. There are two methods (direct and indirect) of vector control for the evaluation of rotor angle. In an Indirect field-oriented control (IFOC) method rotor angle is estimated from slip frequency and rotor frequency without using a sensor (sensors are used in direct FOC) which are achieved in a synchronously rotating frame. IFOC produces high performance in induction motor drives by decoupling rotor flux and torque, so it can be separately controlled by stator direct-axis current and quadrature-axis current respectively, like in a DC motor. The rotor flux orientation method is used for incorporating PI control system using the Space Vector Pulse Width Modulation (SVPWM) technique. The Induction motor drive control generally involves three different PI controllers for torque, speed, and flux respectively. Through the simulation result, varying the load torque, and reference speed is achieved within a few milliseconds, and the behavior of the separately excited DC motor is observed.

Numerical Model Scenario Analysis of Stilling Basin: A Case Study of Tanahu Hydropower Project (140 MW)

Posaraj Khadka — 2023-12-21

The difficulty of experimental methods to modify the complex model motivated the researchers to explore alternative solutions. Tanahu Hydropower Project is a storage-type hydropower project. It has an installed capacity of 140 MW and 140 m high concrete dam along with chute-type stilling basin followed by a complex topography. Computational fluid dynamic (CFD) model is a numerical approximation of partial differential equations. It has been widely used to simulate the fluid flow. In this study, a fluid flow is simulated through the stilling basin using the numerical model for different return period floods. The model’s predictions for flow parameters are validated with the results taken from the 1:60 scaled physical models for the same project. The results regarding the flow velocities and water surface level are within 30% and 1.92 m accuracy respectively. The validated model is run for the three modification cases: i) by opening only two of the three spillway gates, ii) by decreasing the depth of the stilling basin, and iii) by decreasing the length of the basin, aiming to recommend the best alternative solution for the effective dissipation of the high kinetic energy of flow from the 140 m high dam. The results reveal that the base case model is the best solution compared to these three modified cases to pass the flood effectively. This study concludes that the CFD model is the effective alternative tool to analyze the fluid flow problems even in the complex geometry and is recommended to use for the design and modification processes.

Offline Handwritten Text Extraction and Recognition Using CNN-BLSTM-CTC Network

Ranila Shrestha — 2023-12-21

Offline handwriting recognition is a significant research area that aims at tackling problems encountered with handwritten forms in college application and registration processes. The objective of this study is to address the problems of English language offline handwriting recognition via CNN-BLSTM-CTC neural network applied for an NCE Admission form. The system uses OpenCV for image processing, TensorFlow for neural network training and handwritten text recognition, and trains and tests it on the IAM database using image segmentation-based handwriting recognition. With the help of proper image verification, the system allows the users to upload images of the NCE Admission form provided that they strictly comply with the specified format; it denies access to images not conforming to the set standards. Following the successful delivery of a valid image, the form goes through extensive processing that includes text extraction from specific regions of interest (ROIs). The extracted texts are then passed to text recognition block. The recognized texts are then recorded in a CSV file under respective fields. The text recognition model has a CER of approximately 9.33%. The study performed with 15 NCE Admission forms found that the average Character Error Rate (CER) was approximately 12.2% for scanned images and 19.3% for camera-captured images. The results show that accuracy depends on aspects such as the quality and orientation of the image; thus, scanned images are preferred for better performance.

Passengers Flow Analysis of Public Transit Service in Pokhara

Sushmita Midun — 2023-12-21

The motive of the study is to evaluate the existing public transportation system and propose recommendations for Pokhara public transportation to enhance its efficiency and effectiveness. In urban locations, a well-managed public transportation system is unquestionably required. According to the previous study, passengers faced various issues on their journey, such as delays in reaching their destination, bus congestion, an inappropriate time of departure, lane stoppage, and more. (Duwadi, et al., 2019) The primary and secondary data collected from fieldwork and the Yatayat Bibhag, respectively, were used in the research for the analysis. The volumes of passenger flows on ten different routes were analyzed. Mapping and modeling were done using GIS and CAD. Results show that the service frequency did not match the actual swings in consumer demand, resulting in increased time delays. It is recommended that well-managed loading areas be needed in areas with the highest person capacity of more than 3000 per stop per hour, such as Chipledhunga, Mahendrapool, Srijanachowk, and more. A practical operation schedule and service frequency, including 10% for backup and maintenance, were recommended. Demand for passengers on various routes and the number of years the capacity can fulfill the demand is calculated.

Predicting Subcontractor Performance Using Artificial Intelligence

Ida Pradhan — 2023-12-21

Subcontractors contribute to almost 90% of the overall construction work hence, assessing the performance of their work from commencement till the completion stage is essential. This paper focuses on identifying different factors affecting the subcontracted work performance and developing a predictive model using classification - based algorithm to find the proficient subcontractors.

Data collected from the building construction projects was analyzed and utilized. Expert validation method was carried out to validate the factors that were obtained from literature review and a survey was conducted to assess the subcontractor’s performance level. Different classification algorithms such as Naïve Bayes, Logistic, Multilayer Perceptron, Sequential Minimal Optimization (SMO), KStar, J48 and Random Forest were applied to the collected data. Waikato Environment for Knowledge Analysis (WEKA), an opensource machine learning tool was used to compare the performance of several algorithms. Statics were generated and compared using k-folds cross validation (k=10) method.

Among the seven algorithms/classifiers, Random Forest had the highest accuracy in schedule performance model and Multilayer Perceptron in quality performance model

Transforming Structural Engineering: Examining the Opportunities and Risks of ChatGPT and Other Large Language Models

Rishav Pokhrel — 2023-12-21

The advancements in technology, particularly the development of high-performance computing (HPC) and large language models (LLMs) like ChatGPT, can potentially transform the field of Structural Engineering. Use of LLMs, such as ChatGPT, offers several opportunities in Structural Engineering, including the development of innovative design solutions, use in code-based structural analysis programs by automating repetitive coding tasks, conforming to building code requirements by automating compliance checks, and storing information. The critical concerns arise in LLM’s regarding biases, misinformation, safety, reliability, and lack of domain expertise. This paper explores the opportunities and risks associated with using ChatGPT and LLMs in Structural Engineering, focusing on efficiency, accuracy, and reliability. The main aim of the study is to examine the limitations and potential risks of relying solely on machine-generated information and to provide mitigation strategies to overcome them. Careful management to prevent harmful content, collaboration with human experts for accurate results, establishing guidelines and standards are obligatory measures to address ethical concerns such as bias, privacy, and abuse. Continuous monitoring and updating of LLMs are essential to maintain accuracy and relevance. While ChatGPT and LLMs offer significant benefits in Structural Engineering, responsible usage in combination with human expertise and machine-generated insights are vital to maximizing their potential while mitigating risks and ensuring safe as well as reliable engineering practices.

Unlocking the Potential of CT scans: An Explanation-Driven Deep Learning Model for Predicting Lung Cancer

Yuba Raj Oli — 2023-12-21

This research study aims to evaluate the effectiveness of transfer learning with the ResNet50 model for classifying CT scan images of lungs as having cancer or not. Additionally, it explores the interpretability methods of LIME (Local Interpretable Model-Agnostic Explanations) and SHAP (SHapley Additive exPlanations) to provide explanations for the predictions made by the ResNet50 model on CT scan images. The research objectives include developing a deep learning model based on the ResNet50 architecture, evaluating its performance using various metrics, and explaining the predictions using LIME and SHAP techniques. The dataset consists of a collection of CT scan images of lungs, with labels indicating the presence or absence of cancer. Through k-fold cross-validation, the model achieves high accuracy and low loss, demonstrating its effectiveness in classifying lung cancer. The interpretability methods of LIME and SHAP shed light on the crucial features and regions in the CT scan images that contribute to the model's predictions, enhancing the understanding of the model's decision-making process. The results highlight the potential of transfer learning and interpretability techniques in improving the accuracy and explainability of lung cancer detection models. Future directions may involve applying the developed model to larger datasets, classifying different stages of cancer, and identifying the specific regions within the lungs where cancer cells are detected.

Urbanization-Driven Land Use Shifts, Diminished Traditional Water Ponds, and Their Impacts on Water-Dependent Livelihood: A case of Panchkhal Municipality, Kavre District

Kamal Katwal — 2023-12-21

The escalating trend of urbanization is reshaping land use patterns, impacting the hydrological cycle, and giving rise to issues like intensified floods and prolonged droughts. Concurrently, Traditional Water Ponds, essential for sustaining surface and ground hydrology, flood control, and livelihood water access, are rapidly declining. In the Panchkhal Municipality of Kavre District, the dependency of households on water for vegetable cultivation, and effective land and watershed management is imperative. This prompted a comprehensive study to analyze changing land use, the status of traditional water ponds, and their repercussions on water-dependent livelihoods. Presently, a concerning 73% of traditional water ponds are either completely abandoned or repurposed for construction. Urbanization has surged dramatically, with a 1662.5% growth rate in the past two decades, reshaping the landscape. This transformation directly affects supplementary water sources that support livelihoods. Notably, traditional spring sources (kuwas) have dried up, with 60 % of the respondent experiencing a decrease in water availability in kuwas which has been fulfilled by the construction of wells, tap, and water pumps.