Journal of Engineering and Sciences

Landslide Susceptibility Mapping Using Analytical Hierarchy Process in Gandaki Province, Nepal

2023-12-03T16:39:57+00:00

Nepal has a diverse geography ranging from the majestic Himalayas to the fertile plains in the Terai and this varied topography makes it susceptible to different hazards. This study analyzed the most recurring destructive natural hazard, i.e., a landslide in Gandaki Province. The result has been presented using Geographic Information System (GIS) based susceptibility mapping employing Analytical Hierarchy Process (AHP). The susceptibility mapping was performed based on 11 conditioning parameters under four groups, mainly topographic factors (Elevation, Slope, Land Use Land Cover and Profile curvature), hydrological factors (Proximity to stream, Precipitation, Drainage Density and Topographic Wetness Index), geological factors (Geology and Fault lines) and infrastructure factor (Proximity to the road). The final result was classified into five classes: shallow, low, moderate, high, and high susceptibility. The validity and accuracy were tested by calculating the areas under the curve (AUC) value of the receiver operating characteristic (ROC) curve. The AUC value of the landslide was found to be 0.793, indicating the model's good performance. The final map can be used for disaster risk reduction, land use planning and early warning systems.

Comparison of Tunnel Construction Cycle with NTNU Model for the Headrace Tunnel of Seti Khola Hydropower Project

2023-12-03T17:38:03+00:00

Achieving smooth and efficient blasting of tunnel walls is a challenging task. Tunneling performance and advancement can best be performed with proper and effective blasting techniques in drill and blast techniques. In this manuscript, the Seti Khola hydropower project's headrace tunnel's real-time statistics are used to compare the overall performance of tunnel excavation for the widely used blasting approach. The headrace tunnel's 663 meters of observed and recorded length comprises rocks with weak to medium blastibility. According to the Q-system, this tunnel section's rock mass quality class ranges from good to bad. The rock type found has a monotonous sequence of metasandstone and phyllite with quartz partings and discontinuities filled with clay to silt. A variation in the excavation cycle is observed in relation to the presence of different rock mass classes. Ultimately, the NTNU model is compared and analysis is done on the drilling patterns and cycles of tunnel building. It was discovered that using the NTNU model for tunnel excavation presents a significant opportunity to enhance performance.

Rice Yield Estimation Based on SAR and Meteorological Parameters

2023-12-03T23:34:24+00:00

Estimating food production, demand, and distribution allows for timely yield prediction, crucial for managing food security. In-depth research has been published in the literature on applying vegetation indices discovered using optical remote sensing observation for yield estimation. The fundamental limitation of optical remote sensing is that it cannot penetrate cloud cover, which may contaminate the data. Therefore, a novel approach has been studied in this study employing SAR images from Sentinel-1 to construct a regression model combining vegetation index derived from SAR images and climatic variables over a 6-year time series (2017 to 2022). It is evident from the findings that the predictor variables have a non-linear relationship with the yield, which a straightforward linear regression model cannot describe. Other regression models, such as Random Forest, could be more useful in explaining such a complicated and non-linear connection. When the Multiple Linear regression mode was used for testing, it was found that the R²value was to 0.918 and the MSE was 0.513 Mt/ha. When the RF regression mode was used for testing, it was found that the R²value increased to 0.918 and the MSE improved to 0.513 Mt/ha. Furthermore, observation showed a prediction error of around 0.353 Mt/ha when employing the Spatial Error model. Therefore, rice yield estimation has considerably improved when employed in a spatial model.

Shadow Removal from Images Using Conditional GANs

2023-12-04T06:13:00+00:00

Shadow removal has many applications in computer vision and shadow-free images have better visual quality. In recent studies, deep learning-based CNN models have shown better performance than traditional approaches to shadow removal. GAN takes the advantage of two independent neural networks. This study about shadow removal is implemented using GAN. Shadow removal is divided into two tasks: detection and removal. The two sub-networks stacked upon each other are based on conditional GAN. The input shadow image 256*256 is fed to the first generator network to produce a shadow mask, which is input to the second generator network along with a shadow image to obtain a shadow-free image.

Kidney CT Scan Image Classification Using Modified Vision Transformer

2023-12-04T06:35:47+00:00

With the rising number of kidney-related health issues, early and precise diagnosis is crucial. The study aims to create a reliable method for categorizing kidney CT scan images into four groups: Cyst, Normal, Tumor, and stone. Traditional approaches usually rely on typical Machine Learning (ML) and Convolution Neural Networks (CNNs). However, in this research, the potential of a novel model called Vision Transformer (ViT) is explored. ViT was initially designed for Natural Language Processing (NLP) tasks but shows promise for medical image classification. ViT’s capabilities are enhanced by coupling it with Fully Connected Networks (FCN). This combination helps to merge the feature extraction capability of the ViT and the classification ability of the FCN, which ultimately helps to overcome the challenge of detecting kidney-related issues.

A Model Predictive Power Control Scheme for PV Inverter and Battery Energy Storage System for a Microgrid

2023-12-04T07:14:31+00:00

In recent decades, there has been a substantial surge in the adoption of renewable energy systems (RESs), particularly photovoltaic systems (PVs). However, the increasing integration of PV systems into distribution networks has exposed limitations in traditional control methods, such as cascaded linear control with PID controllers. These limitations manifest in dynamic response, power regulation capability, and adaptability to varying operating conditions. Voltage fluctuations arising from intermittent PV output have become a significant concern. The primary aim of this study is to develop a Model Predictive Control (MPC) scheme capable of generating a control signal based on a cost function. The study also seeks to validate the effectiveness of MPC under diverse scenarios. Additionally, an energy storage system, represented by a battery, is incorporated to support the PV system during periods of low power output. The chosen methodology involves using a Dual Active Bridge converter for charging and discharging the batteries. Including an LCL filter in the design significantly reduces THD from 51% to 2.62%. When comparing the response to changing linear loads, MPC demonstrates faster performance than DQ control, with a response time ahead of 0.23 seconds.

Evaluation and stability assessment of road cut slope at Bhalupahad of Syangja District along Siddhartha Highway; Western Nepal

2023-12-04T07:20:09+00:00

A road section at Bhalupahad, Syangja of Siddhartha highway was investigated to characterize the rock mass, identify the potential slope failure modes, and determine the stability condition and the major governing factors. The site comprises steep slopes, faces intense rainfall during monsoon periods and falls in a highly active seismic zone. The attitudes of the hill slope and the major discontinuities were measured at 16 different locations and analyzed using stereographic projection. The data was plotted by using Dips 6.0 software. Plane, wedge, and toppling failure modes are possible in the study area. Besides, the Factor of Safety (FOS) for plane and wedge modes of failure has been calculated using Slide2 software. An end-anchored bolt has been installed to increase the FOS at the unstable slope locations. The FOS has increased to 1.3, 1.23, 1.21, 1.28, and 1.29 after support installation, which was 0.43, 0.71, 0.4, 0.55, and 0.57 before giving support at locations 1, 4, 8, 12 and 14 respectively. Rocks were identified as fair to good quality according to Rock Mass Rating (RMR) and partially stable to stable according to Slope Mass Rating (SMR). The results obtained from RMR and SMR agree well with each other and the real slope conditions. Q-slope suggests a slope angle of 60°-65° with a respective Q-slope value of approximately 1.0. The orientation of discontinuities, steep topography, intense rainfall and human intervention are the main causes of slope failures.

Image Captioning in Nepali Using CNN and Transformer Decoder

2023-12-04T07:33:20+00:00

Image captioning has attracted huge attention from deep learning researchers. This approach combines image and text-based deep learning techniques to create the written descriptions of images automatically. There has been limited research on image captioning using the Nepali language, with most studies focusing on English datasets. Therefore, there are no publicly available datasets in the Nepali language. Most previous works are based on the RNN-CNN approach, which produces inferior results compared to image captioning using the Transformer model. Similarly, using the BLEU score as the only evaluation metric cannot justify the quality of the produced captions. To address this gap, in this research work, the well-known “Flickr8k” English data set is translated into Nepali language and then manually corrected to ensure accurate translations. The conventional Transformer is comprised of encoder and decoder modules. Both modules contain a multi-head attention mechanism. This makes the model complex and computationally expensive. Hence, we propose a noble approach where the encoder module of the Transformer is completely removed and only the decoder part of the Transformer is used, in conjunction with CNN, which acts as a feature extractor. The image features are extracted using the MobileNetV3 Large while the Transformer decoder processes these feature vectors and the input text sequence to generate appropriate captions. The system's effectiveness is measured using metrics to judge the caliber and precision of the generated captions, such as the BLEU and Meteor scores.

Sentiment Analysis of Nepali COVID-19 Tweets using BERT-LSTM

2023-12-04T08:10:38+00:00

The global impact of COVID-19 has significantly reshaped the day-by-day lives of individuals worldwide. COVID-19 is one of the top deadly diseases and has tragically claimed the lives of millions across the globe. The people are affected not only by the physical infection but also mentally. Among the various social media platforms, Twitter is a widely utilized medium, reflecting a substantial surge in discussions about the coronavirus. These discussions encompass a spectrum of positive, negative, and neutral sentiments. The sentiments acknowledged by individuals, encapsulated in their posts and tweets across this platform, offer valuable insights into their emotional states and perspectives. In this investigation, the people's sentiments using Nepali COVID-19-related Twitter datasets are inspected. For this, the approach involves a two-step process. Initially, the multilingual BERT(m-bert) model will utilize whose output is used for subsequent downstream tasks. Secondly, m-BERT's output is connected to the LSTM layer to categorize the people's sentiments. The model was trained and tested using publically available NepCov19Tweets datasets. These tweets were split into three groups (positive, negative, and neutral). The appraisal outcomes for NepCOV19Tweets demonstrate that the proposed model comes up with outstanding performance when compared to the existing model, achieving an average accuracy of 76.04%, 80.03% recall, a precision of 77.12%, and an F1-score of 76%.

Facial Attribute Editing Using Generative Adversarial Network

2023-12-04T08:21:49+00:00

Facial attribute editing tasks have immense applications in today’s digital world, including virtual makeup, generating faces in the animation and gaming industry, social media face image enhancement and improving face recognition systems. This task can be achieved manually or automatically. Manual facial attribute editing, performed with software such as Adobe Photoshop, is a tedious and time-consuming process that requires an expert person. However, Automatic facial attribute editing tasks that can perform facial attribute editing within a few seconds are achievable using encoder-decoder and deep learning-based generative models, such as conditional Generative Adversarial Networks. In our work, we use different attribute vectors as conditional information to generate desired target images, and encoder-decoder structures incorporate feature transfer units to choose and alter encoder-based features. Later, these encoder features are concatenated with the decoder feature to strengthen the attribute editing ability of the model. For this research, we apply reconstruction loss to preserve other details of a face image except target attributes. Adversarial loss is employed for visually realistic editing and attribute manipulation loss is employed to ensure that the generated image possesses the correct attributes. Furthermore, we adopt the WGAN-GP loss function type to improve training stability and reduce the mode collapse problem that often occurs in GAN. Experiments on the Celebi dataset show that this method produces visually realistic facial attribute edited images with PSNR/SSIM 31.7/0.95 and 89.23 % of average attribute editing accuracy for 13 facial attributes including Bangs, Mustache, Bald, Bushy Eyebrows, Blond Hair, Eyeglasses, Black Hair, Brown Hair, Mouth Slightly Open, Male, No Beard, pale Skin and Young.

Suitability analysis of PV solar power plant sites in Gandaki province: Application of GIS and Remote sensing

2023-12-04T08:46:04+00:00

The escalating global demand for energy has triggered a shift towards cleaner alternatives due to mounting environmental concerns. Emerging as a viable substitute, solar power has gained prominence as fossil fuels' adverse impact becomes evident. Historically reliant on hydropower, Nepal is exploring alternative energy sources to mitigate seasonal output variations. Despite abundant renewable resources like biomass, wind, and solar energy, Nepal's energy sector faces funding and technical expertise challenges. Solar energy presents significant promise as a primary renewable source in Nepal, boasting ample sunlight due to its location. However, the current solar capacity remains limited at around 54.6 MW, comprising less than 2.5% of the total installed capacity. The critical factor is the identification of suitable sites for solar power plants. Geographical Information System (GIS) and Multi-Criteria Decision-Making (MCDM) methods have been employed for suitability analysis. This study employs the Analytic Hierarchy Process (AHP), a robust MCDM technique, to assess site suitability. The research is carried out in the Gandaki province, Nepal, encompassing the Himalayan, Hilly, and Terai regions. Criteria like solar radiation, slope, aspect, land use/land cover, proximity to roads, and substations are considered. These criteria are reclassified into suitability categories based on expert opinions and guidelines. The results indicate areas highly suitable for solar power generation, covering 12.40 km2 (5.64% of the study area), followed by regions least suitable, spanning 7681 km2 (34.93% of the province's area). This research contributes to the effective deployment of solar power by identifying optimal locations for solar power plant construction, thus advancing Nepal's renewable energy goals.

Sketch to Image Translation using Generative Adversarial Network

2023-12-04T09:05:59+00:00

Using a Generative Adversarial Network (GAN) has proven its ability to successfully implement realistic images in image translation fields. It has its successful implementation in the sketch-to-image translation, too. Generative adversarial networks are widely used for the purpose of image translation. Most discriminators in generative adversarial networks use encoder or decoder blocks for image segmentation and classification tasks. U-net-based architecture is mostly used in the generator but rarely in the discriminator. If used in the discriminator, it is used for image resolution increment and segmentation tasks. In this research, a U-net-based discriminator is used for image translation tasks. U-net-based discriminator uses local and global differences between the real and fake images, which helps maintain global and local data representation. Resnet-9, used in the generator, uses skip connections, shortcuts, and concatenations, enabling information to flow from earlier to later layers. This helps preserve the original image features and solves the vanishing gradient problems in normal generators. The use of a strong discriminator and effective generator helps in the improvement system's performance. The available dataset was unpaired at the same time. Datasets from various sources were combined and formed a sketch-image pair. The input is a 512x256 human sketch and a corresponding real image pair. The image pair is split into sketch and image with dimensions 256x256. The system's output is the human face image of the corresponding sketch.

A Comprehensive Review on Integrating Climate Science and Machine Learning for Power System Resilience

2023-12-04T09:54:31+00:00

This study focuses on how climate science and machine learning techniques may be used to improve power system resilience in the face of climate change. It emphasizes the significance of resilience and the principles of machine learning application in power systems. Predictive models for climate-related disruptions are among the most recent advances in merging climate research with machine learning. The review assesses the efficacy of various models in improving system resilience and their limitations and problems. Future research prospects, policy consequences, and recommendations for moving climate science and machine learning integration forward for power system resilience are highlighted. Overall, the need to integrate these technologies to address climate change concerns and improve power system resilience is emphasized in this analysis.

Multi-Class Credit Risk Analysis Using Deep Learning

2023-12-04T10:08:38+00:00

Credit risk prediction, reliability, monitoring and effective loan processing are the keys to proper bank decision-making. So, understanding the credit customer during the initial loan processing phase would help the bank prevent future losses. In this regard, this study aims to develop a credit risk evaluation model using deep learning algorithms. The model utilizes a credit risk analysis dataset published in Kaggle. The objective is to build deep learning models for predicting credit risk using real banking datasets published on Kaggle. Firstly, data preprocessing and feature engineering are done. Suitable features such as irrelevant and null valued features are identified and removed with techniques like the Karl Pearson correlation, information values, and weight of evidence. Next, data normalization is performed and target features are separated into three classes: high risk, medium risk and low risk. SMOTE-ENN (Synthetic Minority Oversampling Technique with Edited Nearest Neighbor) was applied to balance the dataset. State-of-the-art deep learning algorithms such as GRU (Gated Recurrent Units) Model and Bidirectional Long Short-Term Memory (Bi-LSTM) are implemented to train and learn from the pre-processed data. GRU and Bi-LSTM models performed well, with F1 scores of 0.92 and 0.93, respectively. The result of this investigation illustrates that deep learning models seem promising for evaluating and predicting multi-class problems.

Modelling and Simulation of Field-Oriented Control of Permanent Magnet Synchronous Motor

2023-12-05T05:27:33+00:00

A permanent Magnet Synchronous Motor is an electric motor driven by permanent magnets, finding widespread use in industrial and robotic applications due to their high efficiency, low inertia and high torque to volume ratio. Various control techniques have been implemented to make drive systems control more precise and efficient. This paper presents a Field Oriented Control (FOC) as a novel method to effectively control motor torque and speed. The primary objective of this research is to develop a detailed model of FOC-PMSM and simulate its dynamic behaviour under various operating conditions. The system is simulated in MATLAB/Simulink to check the validity, reasonability and expected outcomes. Simulation results show that the designed control system can track speed and current references with minimum error. PMSM depicts better dynamic performances with FOC implementation.

Feasibility study on generation and storage of power generated by mechanical footstep power generator

2023-12-04T10:48:16+00:00

With the increasing demand for electrical energy in today’s world, science has discovered many new concepts and methods for energy production. Among those concepts, a mechanical footstep power generator converts mechanical energy to electrical energy. The ultimate purpose of this research is to fabricate the footstep power generator and to produce electrical energy. The project is based on a Rack and Pinion mechanism to convert mechanical energy into electrical energy by utilizing energy during walking. The maximum voltage and current output of 8.22 Volt and 0.12 Ampere were obtained when the maximum load of 70 kg was applied to the system. This research has shown that mechanical footstep power generators have great potential for powering small electrical devices in various public places.

Ranking Road Safety Hazardous Location: A Case Study of Chhorepatan–Machhapuchhre Viewpoint Road Section

2023-12-05T05:07:17+00:00

The importance of road safety, involving pedestrians, cyclists, motorcyclists, and drivers adhering to rules and strategies to prevent accidents and fatalities, is emphasized, especially in Nepal, where road fatalities remain a pressing issue. The study focuses on the Chhorepatan – Machhapuchhre Viewpoint Road Section, which faces increasing traffic density and hazards. It proposes a six-stage methodological framework, incorporating the Analytical Hierarchy Process (AHP) and field surveys, to rank hazardous locations based on safety parameters, resulting in a Safety Hazardous Index (SHI). The research aims to identify and prioritize key safety factors by correlating SHI values with crash records, potentially serving as a model for assessing road sections in similar conditions.