Substation technology (ST) selection has emerged as a critical decision not only for an efficient power distribution system but also for meeting the UN sustainable development goals (UNSDGs). Hence, this study aims to recommend a suitable ST by applying seven criteria to BestWorst Method (BWM) to calculate their weights and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) approach for choosing the best-suited technology among the Air-Insulated Substation (AIS), Gas-insulated Substation (GIS) and Hybrid Substation technologies. The analyses reveal that GIS is the most suitable substation technology in Indian contexts considering the factors of cost, flexibility, safety, reliability, energy loss, time, and simplicity. The study also offers avenues for further work
For a given graph G, the dilation problem or the grid bandwidth minimization problem (GBMP) seeks to find an embedding of the host graph G onto a guest grid graph H such that the bandwidth of the embedding over all edges is minimized. The Dilation Problem is NP-hard in general. In this paper, a reduced variable neighbourhood search (RVNS) algorithm is developed for GBMP of an l×m grid embedding. This research study has designed four construction heuristics and the best quality solution among them is used as an initial solution for the RVNS procedure. Two new neighbourhood search operators and a shaking procedure are designed to explore the search space. The test suite used for experiments consists of a subset of Harwell-Boeing graphs, grid graphs and cycles. The computational results show that the algorithm is able to achieve optimal results for grid graphs and cycles. Comparison of RVNS with existing approaches shows improvement in results for most Harwell-Boeing instances.
Metaverse has attained global attention since its announcement, and several discussions and analyses have been actively going on in the current research for the past year. The author discussed the marketing frameworks through Gamification in the Metaverse World Wide Web 3.0. Metaverse attempts to bring human connections to the Virtual World with an immersive experience. The author discussed the emerging presence of Artificial Intelligence (AI) in the Metaverse, which could be changing the environment for user experience. Gamification offers different methods to engage the user types in environments where Marketing tries to attain and attract customers in Metaverse. With the growing expansion since the Information technology revolution, the World wide web has grown into a different version of Web 3.0, which is still evolving. The research discusses the newly imminent area in the Metaverse as Web 3.0 evolves with Artificial Intelligence.
Small IoT devices can only support light protocols since they lack significant processing or storage power. Lightweight encryption standards or techniques will be utilized for encryption and decryption to prevent hackers from accessing the data in order to increase the security of IoT devices' network connections and commands. The encryption has to be robust enough to safeguard and lightweight enough to enable light processing hardware. There aren't many standards for lightweight cryptography that are used to safeguard devices in an existing network. If the network is not secured properly, an intruder can get in and take control of the devices without making a visible appearance. A hybrid lightweight cryptographic methodology have been proposed to address this problem. LWC algorithms are designed to provide strong security while minimizing the computational and memory resources required for encryption and decryption. This is particularly important for IoT devices, which often have limited processing power, storage capacity, and battery life. LWC algorithms offer a range of benefits for IoT devices, including improved security, reduced power consumption, and increased battery life. They are also designed to be easy to implement and integrate with existing IoT platforms and protocols. Proposed system's analytical work implementation will combine Salsa20 and AES algorithms to provide greater security and making it well-suited for real-time cryptography.
Agriculture is the primary economic activity in a significant number of nations. Yet, plant-related illnesses are the most difficult obstacle for farmers to overcome. The leaf is the portion of the plant that is noticed the most, but the process of segmenting an image of a sick lesion on a leaf suffers from challenges related to uneven lighting and a crowded, complicated natural environment. The combination of colour balancing and superpixel is the foundation of the approach that has been suggested as a solution to these issues. To begin, the picture that is being read in will be converted into a color-balanced image so that the effects of uneven lighting may be removed. Second, using the superpixel operation, compact areas are produced from the modified picture. An experimentally calculated threshold is then applied to the Histogram of Gradients (HOG) and colour channels of the superpixel to separate the undesirable backdrop from the image of the leaf. In order to identify a sick or contaminated picture, K-means clustering is employed. Pyramid of HOG (PHOG), an expanded version of HOG, together with Grey Level Co-occurrence Matrix (GLCM) characteristics are used to depict a sick infected portion of the body. In the end, many disease classifiers are evaluated against one another, and Random Forest (RF) is chosen as the best option. An experiment is carried out to verify the effectiveness of the proposed approach’s accuracy using natural photographs obtained from a variety of farms, images downloaded from the internet, and the Plant-Village colour dataset. This is done so that the resilience of the method can be evaluated.
COVID-19 is a global pandemic that has caused a substantial increase in the number of people becoming sick and dying all across the globe. The diagnosis of COVID-19 is essential for preventing the disease from becoming more widespread yet, it may be difficult to make owing to the many clinical presentations of the disease and the symptoms’ resemblance to those of other respiratory infections. This research work, presents a multi-modal fusion technique for the detection and categorization of COVID-19 utilising chest X-ray pictures, CT scans, and RT-PCR testing. In order to do the analysis and determine whether or not the images contain COVID-19, and make use of a deep learning strategy that incorporates a convolutional neural network (CNN). To further enhance the precision of the categorization, the proposed model additionally makes use of the findings from the RT-PCR test as an extra input. The findings of the experiments indicate that the multi-modal fusion strategy that this work presented and delivers a greater level of accuracy than employing any particular modality by itself.
As social media has gotten more and more popular in recent years, there are many websites and social media platforms that claim to report on forth coming events, but whose veracity has been called into doubt. This has thrown news websites and blogs into the spotlight. As a result of the controversy surrounding these websites and news organisations the concept of “fake news” is now a crucial component of the news industry. Today’s media content is considerably affected by the falsified news i.e; from satirical pieces to manufactured news and government publicity schemes. Fake news and a lack of media trust are growing problems with serious repercussions for the general public. Automated technologies for spotting false news, including machine learning models, are now a must. The work here presents the evaluation of four machine learning models’ performance on four real-world datasets. Other languages, like Hindi, which is extensively spoken all over India are the subject of relatively fewer study. But the models which we have built are equally efficient in classifying news proliferated in Hindi also. The method proposed utilises NLP techniques for data preprocessing and TF-IDF is the feature extraction technique considered for this model’s development. Using simple machine learning techniques an accuracy of about 0.99 is achieved.
The process of detecting and recognizing disease in plants are crucial for the early control of pests and diseases that significantly affect plant growth. Although maize is the most productive food crop worldwide, it can suffer from physiological lesions caused by viral and fungal infections primarily visible in the leaves. The automatic identification and classification of plant diseases are essential for sustainable agriculture, but it remains a significant challenge. Existing techniques proposed for this purpose are limited in scope and rely solely on models for deep learning. Using cropped photos, convolutional neural networks have demonstrated to be the most adequate method for identifying and forecasting illnesses. An overview of deep learning methods for identifying plant diseases is given in this article, which also includes data collection sources, deep learning architectures, and image processing methods. The study evaluates existing deep learning models and highlights their results, with a focus on future research to improve system performance and accuracy for detecting crop diseases using better deep learning capabilities. The main objective of this review is to strengthen the quality of maize plant leaf disease detection and inspire further research in this field.
Cereals are a significant and vital source of food for humans. To feed the world’s expanding population, farmers must produce more crops. Plant diseases, however, have an impact on crop production and food quality. Wheat is among the most important crops consumed worldwide. Since Wheat Disease causes production loss, early disease detection and classification are absolutely essential. This research study has conducted a literature survey on studies published from the year 2017 to 2022 and summarized the three main types of Wheat Disease (fungal, bacterial, and insects), Wheat Disease datasets, and the current state of the art from the last six years of research. This research study has examined 24 studies on disease identification and classification by using various Machine Learning (ML) and Deep Learning (DL) algorithms. The proposed research analysis shows that the majority of the literature on Wheat Disease focused on fungal disease, and the majority of the datasets used were self-acquired. Many State-of-the-art models have already produced excellent results, and many more need to be developed.
In day-to-day life, Time and Health are two precious assets to achieve anything by an individual. In sustaining good health, rural communities confront numerous obstacles in receiving healthcare services. According to the statistics, 26.5% of uninsured rural residents delayed receiving necessary healthcare in the past year. At the beginning of 2020, the use of telemedicine services was already booming, but responses to the COVID-19 pandemic sped up this expansion. In rural areas, one must make sure that telehealth services continue to be compensated at the existing pandemic rates in order to provide proper treatment at right moment with patient’s medical history. To solve this issue, the “Smart TeleHealthcare using Blockchain and IPFS ”(STBI) system proposes a secure healthcare solution that merges IoT and Blockchain. The Blockchain network connects the patient and medical team in STBI, and both can communicate with the decentralized storage IPFS. STBI system collects healthcare samples (i.e., which are simulated by sensing temperature using Node MCU board interfacing with LM35 sensor) and displays the health reports (i.e., temperature value) on webpage. The lengthy health reports are kept in IPFS, and the resulting hash value is stored in Blockchain for further reference by doctors. In STBI, the electronic medical data is shared between authorized stakeholders such as patients, doctors and medical team which enhances effective medical treatment collaboratively. The experimental system was incorporated with blockchain and IoT technologies to provide an enhanced secure and transparent healthcare facilities in rural areas.
Increasing commuters, bad traffic signal management, and rider mentality are some of the factors contributing to traffic violations in India. Monitoring large traffic volumes physically and tracking violations at the same time is clearly insufficient with only physical traffic police. The result has been that many violators have gone undetected. By violating the traffic laws, the violators cause more serious mishaps on the road, thereby putting both themselves and others in danger. To avoid manual intervention in detecting and catching violators, Artificial Intelligence (AI)-based techniques should be incorporated. This research study demonstrates a novel technique to discover multiple offences on Indian roads electronically, which include helmet detection, using a smartphone while driving, tri cruising, wheeling, and parking illegally, and ultimately model the issuing of tickets by monitoring the infractions and affiliated car number together in record. Through all the automated AI-based traffic offense and booking system, the software will be extremely beneficial in determining diverse safety-related guidelines, facilitating in the imposing of harsher traffic restrictions, and fostering the development of such a green technology atmosphere.
As the Internet of Health (IoH) era has come into being, health services are progressively moving online and have been creating a sizable amount of health-related data related to patients, physicians, medical infrastructure, and other related topics. The effective integration and analysis of this IoH data have good implications for the provision of medical treatment and scientific disaster diagnosis. IoH data, however, are frequently dispersed throughout several departments and only partially protect user privacy. Therefore, efficiently integrating or mining the critical IoH data, while maintaining user privacy, is sometimes a difficult issue. Thus, this study provides IoH as a solution to this problem, allowing the IoH to submit medications to a specific patient without disclosing patient information and the admin to access the medications upon IoH’s approval. Patients can watch physicians and consult them only once admin and the doctor have given their approval.
Internet consumers are extremely worried about identity theft caused by phishing scams. Phishing tactics are frequently used to deceive customers into visiting bogus websites in order to steal their personal information. A new wave of sophisticated malware phishing attempts, however, may be able to get past the present defenses put in place to resist these kinds of attacks. making use of medication It is practically impossible to adequately defend against Trojans that target particular services. The security architecture that guards against phishing scams that originate from both trustworthy and malicious sources will be developed and put into operation using the information in this article. This approach is built on the concept of compartmentalization, which divides programmers with differing levels of trust, using trusted wallets to store credentials and validate sensitive services. This Technique does not require consumers to exert further effort to locate the relevant website when the wallet has set up but credential disclosure is rigorously restricted. Also provide a prototype of the base platform and give a brief description of its operation. a brief explanation of its operation.
Text-based Question Answering (QA) is an essential task in natural language processing that aims to provide relevant and accurate answers to users’ queries. Traditional approaches to QA relied on rule-based systems and hand-crafted features. However, recent advancements in information retrieval and deep learning have shown promising results in improving the performance of QA systems. The combination of IR and deep learning techniques has led to significant improvements in QA performance. Hybrid models, such as the bi-encoder and tri-encoder architectures, have been proposed to leverage the strengths of both IR and deep learning models. Additionally, pre-training techniques, such as BERT and RoBERTa, have been shown to improve the performance of QA systems by providing pre-trained contextualized word embeddings. This study provides a brief overview of the QA system and its different domains and subdomains. This study also provides review of different proposed models and different datasets available for the task and evaluate their performance using some performance metrics. A comparison between various techniques are employed using the obtained result.
The growing interest in biodiversity and conservation has heightened the need for accurate and efficient bird species recognition. This is because bird populations, their distribution, migration patterns, and environmental impact assessments can benefit from the valuable data these recognition methods provide. To address the complex task of manual bird species classification, this research paper presents a transfer learning approach based on MobileNetV2. The proposed solution uses MobileNetV2 to categorize bird species in photographs, achieving 96.79% accuracy on a defined dataset. This work highlights the potential of transfer learning in solving complex recognition tasks and the efficiency of MobileNetV2 in achieving high accuracy despite its lightweight design. The method has real-world applications, including real-time bird identification systems and bird population monitoring, and will become more valuable with advancements in hardware and deep learning. To improve the overall system’s accuracy even further, using a larger dataset, scheduling regular retraining sessions and further fine-tuning the model is suggested. In this study, the model is trained separately on two different datasets. The first dataset contains around 1800 images with 200 types of bird species. The second dataset contains around 180 images with 20 types of bird species, specifically South Indian bird species. The attained accuracy with the first dataset is 90.91% and the attained accuracy with the second dataset is 96.79%.
Pneumonia is a serious health condition where the patient’s lungs get infected by bacteria, virus or sometimes fungi. This causes burning sensation in the alveoli present in the lungs. In response to this, body forms fluids in the lungs that can cause trouble in breathing, coughing and even chest pain. The increasing demand of medical care for elderly care and children below five years of age requires advanced healthcare technologies. Premature diagnosis is vital for fast recovery. This research study intends to propose a novel technique for detecting pneumonia in chest x-ray images by training a model to classify the patient’s condition as normal or pneumonic. The proposed model applies CNN, a deep learning algorithm to detect the health condition of the patient by analyzing their chest X-ray images. The objective of this research study is to attain the model accuracy as high as possible to provide accurate results to the user.
One of the main issues diabetes poses to the medical profession globally is that its consequences are escalating swiftly. Elevated blood glucose levels cause diabetes, also referred to as diabetes mellitus or simply diabetes. On the basis of physical and chemical exams, a number of standard approaches can be used to diagnose diabetes. But, Doctors face a difficult task in predicting diabetes that affects the kidneys, eyes, heart, nerves, feet and other parts of the body.Early diagnosis analyses illness prognosis and diagnosis using a doctor’s training and expertise, although this might be vulnerable to error. Machine learning and data science approaches have the potential to enrich other scientific disciplines by providing new perspectives on well-known issues. One such initiative is to assist in making predictions based on medical data. ‘Machine learning’ describes the processes by which computers tend to learn from experience, and is a recent area of data science. The aim of this research is to develop a system to accurately diagnose diabetes in patients at an early stage by comparing the results of several machine learning algorithms. Use four different supervised machine learning techniques: Random Forests RF, Logistic LR Regression,Decision DT Trees and (SVM) Support Vector Machines. The main objective of this research project is to develop a prognostic tool for early detection and prediction of diabetes.The model is also deployed into a web application using Python Flask, and the web application is built using HTML and CSS. Anyone can input features into the web application, and the model—which was previously developed using machine learning techniques—will then predict whether or not they will be diagnosed with diabetes.
Batteries, which are collections of electrochemical cells, produce electricity to run electrical equipment. Batteries continuously transform chemical energy into electrical energy, and for optimum performance, they need to be properly maintained using systems of management with particular monitoring capabilities. In order to avoid potential health, safety, and property risks associated with battery use, techniques including charge management systems and temperature regulation are implemented. These systems use merit-based metrics to control battery performance. In this work, neural network is used to keep track of the battery’s health. The proposed system consists of a load cell, a temperature sensor, a voltage transformer, and a current transformer. It features automatic battery cleaning powered by servo motors. Also, alert for different battery fault conditions will be sent.
Artificial Intelligence (AD is the wide application that learns the problem and features by given data and processes the data like the human brain. When a computer program imitates a characteristic of the human brain that is considered “innovator.” Among the methods are statistical methods, methods for artificial intelligence, and traditional order to verify the validity. The expansion of AI is also related to virtually infinite storage and an abundance of data, including exchanges, geospatial information, video files, photos, text messages, and audio files. Machine learning is divided into deep learning and deep learning is primarily divided into numerous layers of neural networks. This pattern gives it the ability to learn a lot of information and attempt to replicate the brain function. Increasing the efficiency by attaching more covert layers can be beneficial. It is used to gather the data and transfer the data. Aquaculture production has grown into a barrier to the growth of fish culture and the counting operation represents one of the problems experienced during the spawning process. Previous studies have primarily relied on the application of manual and automated counting techniques, which has prevented it from producing accurate results. The proposed method offers a promising method for enhancing image detection by combining the IoT techniques. The image data were divided into three categories: low frequency, intermediate density, as well as high frequency. The proposed method has used 8200 images to train and 2500 images for verification. Only the data relevant data sources were used during the train and verification phase in order to find the proper parameters and create a better VGG19 parameter calibration strategy. Consequently, the improved VGG19 model can achieve an accuracy of 98%.
Despite its importance. more than 800 million people still lack access to electricity. Individuals in many rural areas lack access to a consistent supply of energy due to the high cost of grid extension. Off-grid solar systems enable rural residents to connect to the energy 2rid and power their homes’ lights. appliances. and other electronic equipment. This document provides a brief overview of a solar power system that can be used to Dower remote homes. Solar panels. a charge controller. batteries. and an inverter make up the system. Sunlight is converted into direct current by solar panels. which is then stored in batteries. The charge controller regulates the charging and discharging cycles of the battery pack in order to keep the cells healthy and long-lasting. The inverter converts the stored DC electricity into the alternating current (AC) power that appliances and other AC loads require. The proposed method is intended to be simple and low-cost. making it an appropriate option for regions with limited resources and low growth. The system can be customized to meet the exact energy needs of the house by adding additional solar panels or batteries. Furthermore. the structure is made up of discrete components. making it easier to assemble and maintain. Off-grid solar power systems are a cost-effective and efficient wav of bringing electricity to areas that are not connected to the power grid. The proposed system describes the design and installation of a solar-powered. direct current (DC) power source for isolated dwellings. The strategy promotes long-term economic growth and raises the living standards of rural families.
