ABSTRACT: Cyber Physical System is showing a very important part in the arena of accuracy agriculture and it is predictable to progress productivity in order to provide food to all and prevent hunger. To accelerate the consciousness of CPS in the arena of accuracy agriculture it is essential to grow some devices, software tools and hardware components depend upon interdisciplinary methods, along with authentication of the ideologies through prototype. In this framework this paper offerings a precision agricultural supervision model through drone based on CPS design technology. Accuracy agriculture means to include the cleverness in the production of the harvest, by real-time identifying technique, optimization for improving health of soil-crop, as well as most efficient cyber-occupied tools for computerization and adeptness. Combined analysis of agricultural and ecosystem of soil is in promising stages, but increasing progressively with enhancements in sensing skills and data-influenced decision making. Our target to grow an adaptive Sensor based -Drone-Satellite system for endorsing agricultural operations and sustainability through matching often-conflicting purposes (e.g. price ecological or environmental and commercial). This integrated system is a theoretical prototype constructed on a cyber-physical interface to enable real multi-scale manual decision making by coupling innumerable existing and yet-to-be implemented data resources with AI techniques (e.g. artificial neural networks). The planned flexible system search for best solutions that can efficiently support fundamental, improve agriculture reliability and finally food safety.

ABSTRACT: In the present world, online social networks are rich in multimodal data sources with various objects, URLs, and comments. These are the real-time dynamic sources for the analysis which will lead to the discovery of facts and hidden relationships among the closed community groups in networks. Finding a closed community in online social networks is a challenging task for various purposes of applications. A closed community is formed with a group of similar-minded people and may be related to political, ethnic, or religious. The governance of such groups consciously applies limitations on the network links with outside communities. Broadly, two concepts of algorithms viz., clustering and network partitioning are used for the detection of such groups. These algorithms are based on dynamic networks with humans as key players and the other one is the graph structure similar to the topological structure. However, these algorithms suffer from limitations such as these communities provide no knowledge of groups in advance, the requirement of an extensive analysis of all possible partitions, etc. This article aims to overcome the said limitations by using the fast greedy approach by fusing with a Density-based clustering technique called DBSCAN for the detection of such communities. Detection and deletion of these noisy nodes in the communities lead to the development of quality. The comparison of the experimental results proved that the removal of noisy nodes will impact the quality of the community detection.

ABSTRACT: In recent times, the population of Blind people has been increasing by around 250 million in the world, and an attempt to improve the quality of their lives. However, despite the promising research outcomes, the existing wearable aids for blind or visually impaired people have numerous weaknesses in terms of weight, feature limitations, and cost. In this manuscript, a novel invention of lightweight design of wearable aid for visually impaired and blind people. The proposed design of a wearable aid will help them to walk and detect the environment around them with ease and make them independent in their life. The proposed system uses the fusion of sensor and vision-based technologies. It includes Google Vision API services, Lidar Technology, Arduino Nano, and Raspberry Pi 4. Based on the appearance of this prototype, this invention is named Blind's Bib. This assures the proposed aid is designed with lightweight, easy to use, and with a minimum number of instructions for operation. All the necessary security and frequently needed features are included in this system. Experimental results are demonstrated with blindfolded subjects and visually impaired participants.

ABSTRACT: This paper reviews the various control strategies used for Dc-Dc boost converter applied with renewable energy systems, like photovoltaic array module and wind energy system. Boost converters are in prime position for lifting up the lower voltage level collected from renewable energy sources (RES), such as fuel cells and solar panel, to cope up the required level of the grid voltage. Useful control techniques are required for assuring circuit efficiency, stability and reliability under different condition. This paper classifies control methods into conventional method and advanced methods. Each method’s benefits, drawbacks and suitability for dynamic and unbalanced renewable energy supply. Distinct concentration is provided for real-time adaptability, fast response and minimized power losses vital for improving renewable energy system applications.

ABSTRACT: Characterizing respiratory reactions to external stimuli is essential for understanding the complicated respiratory system, especially in clinical diagnosis and treatment. This research uses model-based system identification to correctly record and assess the respiratory system's reaction to mechanical ventilation, environmental changes, and pharmaceutical drugs. We use advanced system identification methods to create a dynamic model that accurately captures the respiratory system's nonlinear and time-varying behavior. Data-driven modelling and physiological insights are used to determine respiratory function parameters in the proposed strategy. The model is validated using controlled laboratory and clinical trial data. The results show that the model-based method accurately predicts respiratory responses, improving respiratory dynamics comprehension and management in healthy people and respiratory problem patients. This study advances respiratory physiology and biomedical engineering by improving respiratory function monitoring, prediction, and optimization in response to external stimuli. The system identification framework may improve patient-specific treatment techniques and individualized respiratory medicines.

ABSTRACT: Internet of Things (IoT) is an Industry 4.0 Technology that is rapidly getting implemented in the industries and is replacing the conventional and obsolete parts of the machines. In a transformer, the rise in temperature of the winding and the transformer oil, which is the most common coolant, beyond a certain limit signifies that the insulation may be damaged. Conventionally, the protection scheme implied is the Oil Temperature Indicator (OTI) and the Winding Temperature Indicator (WTI). This work uses IoT to transmit and deploy alarms based on wireless and cloud technologies which significantly increased the efficiency of the protection scheme. The hardware model developed attempts to incorporate the benefits of the IoT and deploy the alarm message and alarm sound in significantly lesser amount of time and enable the required safety measures to be taken as soon as possible.

ABSTRACT: The Battery Management System (BMS) is crucial for the functioning of Electric Vehicles (EVs) and Hybrid Electric Vehicles (HEVs), ensuring the safe and reliable functioning of the battery. The primary objective of the BMS is to monitor and regulate the state of the battery, ensuring optimal performance and longevity. The main functions of the BMS include monitoring and assessing the state of the battery, controlling charging, and balancing the cells. These features are essential for maintaining battery safety and efficiency. Rechargeable batteries supply power to the motor and auxiliary systems in electric vehicles. Over the past decade, battery technology has made significant advancements, leading to the development of high-performance batteries. This paper focuses on the critical tasks handled by the BMS, including monitoring the State of Charge (SoC), State of Health (SoH), State of Life (SoL), and maximum capacity of the battery. By examining the various methodologies used to assess these parameters, the paper identifies future challenges and potential solutions to enhance battery management systems.

ABSTRACT: Statistical clustering technique can be used by crime analysts to the generate suspected list of unsolved crimes, locate crime clusters which have committed by the same person or group of persons, forecasting future events, & develop offender profiles. In this paper, the log-bayes-factor and maximum posterior probability has been used as a similarity measures for solving an unsolved case. Since the offender is known only for a fraction of crimes, in this article the proposed approach is partially semi-supervised. It employs the crime attributes, along with spatial and temporal locations, to describe the offender. It is possible to link and compare crimes using a single link, average link, and a complete link strategy. It employs the agglomerative hierarchal-based clustering method for making crime clusters, based on log Bayes factor and Bayesian clustering model which uses the maximum posterior probability as a similarity measure, for unsolved crime identification. Naive Bayes model calculates the Log-Bayes factor, which helps investigators uncover unsolved murders that are linked to one another. The Naïve Bayes classifier outperforms than the Agglomerative Hierarchical Clustering(AHC) as it uses the log Bayes factor.

ABSTRACT: An innovative method for improving auditory and sensory experiences is the incorporation of vibration-based music therapy into the rehabilitation process for those with hearing impairments. This study uses machine intelligence and advanced digital signal processing (DSP) techniques to assess the efficacy of this therapy. Through the analysis and processing of vibration-based music signals, the research endeavors to measure the influence of these signals on hearing-impaired individuals, emphasizing enhancements in sensory perception and general well-being. As part of the process, a DSP framework is created to precisely record and adjust vibrations that the hard of hearing experience. To provide individualized therapeutic experiences, machine intelligence algorithms are used to understand and modify these vibrations based on each person's unique sensory profile. The three main criteria that are measured are cognitive engagement, emotional well-being, and sensory responsiveness. According to preliminary findings, vibration-based music therapy has potential advantages in enhancing sensory integration and emotional reactions in hearing-impaired individuals when linked with advanced DSP and machine learning methods. This study adds to our knowledge of complementary and alternative therapies and emphasizes how digital technologies might support conventional rehabilitation techniques. The main goals of future study will be to improve these methods and apply them to larger groups of people.

ABSTRACT: Today’s electronic devices market depends on wide band semiconductors with the members of chalcogenide compounds. They are chemicals which contain group 16 elements led by oxygen and followed by anion elements, like selenium, sulfur, selenium, tellurium, and others. On the other hand, second components include cations, like, zinc, beryllium, copper, and others. At present, chalcogenide compounds are used for infrared optical windows with strong refractive index. They are used in different optical sensors, especially in photodiode, photo sensor, optical transmission. Obtainable in crystal-like, and nano-crystal-like forms, chalcogenides are noted for their superior electronic, optical, and semiconducting properties. We have discussed present utilization and prospects of this semiconductor. It has been observed that chalcogenides with Cu as another cation component are good to use due to their less toxicity. Even though it is in nanocrystal form. Other chalcogenides are comparatively toxic, for example, cadmium and lead. But they are effectively used in biomedical applications. On the other hand, some applications in photovoltaics are supplemented with copper chalcogenides. They are efficient in clean energy transformation as they efficiently work on photocatalytic activity. In this paper we have reviewed evolutionary applications of chalcogenide compounds in different domains.

ABSTRACT: With the advancement in automation and smart systems, many problems that occur due to human error and inefficiency has been reduced prominently in different fields. In the field of power system, smart metering is a revolution to overcome problems of human inefficiency and to create awareness for consumers about their energy usage. It is possible due to Internet of Things technology under industrial revolution 4.0. This paper addresses the various aspects of different wireless communication technologies utilized in the IoT integration of smart meters and analysed the performance of these technologies on the parameters which is essential for smart metering application. In the country like India, where most of the populations are living in the rural areas and the network connectivity is still a big challenge. The installation of smart meters in various household is still in progress and various communication technologies have been used for the integration in smart meters but still we are facing many issues which are discussed in this paper. Although there are numerous review papers on wireless communication technologies, none of them have specifically addressed how to incorporate these technologies into applications for smart meters. Based on the issues and challenges associated with existing smart meters, a detailed analysis of wireless communication technologies on multiple aspects which are essential for the integration of communication technologies are discussed in this paper which can be helpful for the researchers and companies working in this area. Since not a single technology is perfect in all parameters, we found LoRaWAN- a viable option for smart metering application.

ABSTRACT: Efficient use of renewable energy sources without limiting power consumption is the problem in demand-side energy management. This paper presents the design and implementation of an IoT-driven hybrid power management system that seamlessly integrates solar and AC power sources. Leveraging the ESP32 microcontroller, this system is engineered to prioritize solar energy as the primary power source, switching to AC only when solar output is insufficient or an overload condition occurs. This dynamic switching mechanism not only ensures a consistent power supply but also optimizes energy usage, contributing to both cost-efficiency and environmental sustainability. Proposed design allows users to track power levels, receive notifications on power source switching, and manage overload conditions remotely.
Keywords: Hybrid Power, IoT, ESP32, Solar Energy, Mobile Application, Renewable Energy .

ABSTRACT: By combining cutting-edge natural language processing methods with the AI-SHAP framework, this study offers a novel approach for identifying and evaluating toxic language in digital communication. The research not only improves the interpretability of hazardous text categorization models by utilising AI-SHAP, but it also explores the complex linguistic subtleties and contextual complexities that underlie toxic behaviour in online contexts. This method clarifies the fundamental causes of toxic communication through thorough assessments on a variety of datasets, opening the door for the creation of practical tactics to promote safer and more welcoming online communities. It makes use of a novel approach AI-SHAP framework to examine and identify harmful language used in online communications. It also uses advances natural language processing techniques for reliable text analysis.
Keywords: Toxic Language, Natural Language Processing, AI-SHAP Framework, Logistic Regression

ABSTRACT: India's agriculture industry has a difficult time cutting expenses without sacrificing crop production. This research suggests uses of Internet of Things (IoT) devices to propose a framework for cost optimization in agriculture. We introduced a multifaceted strategy that incorporates: Using cameras to monitor fields remotely and monitor crops in real time, Automated systems for discouraging animals with remote-controlled buzzers ,utilizing intelligent sensors and actuators to manage irrigation precisely With our IoT-based system, farmers can keep an eye on their fields from a distance, identify possible problems, and take preventative action to avoid damaging their crops. The technique minimizes agricultural losses from animal encroachment, maximizes crop utilization, and lowers labour expenses. Our pilot study's findings show a notable decrease in the expenses related to: Labour, Use of water and Crop loss. Our research indicates that the suggested IoT-enabled framework can significantly optimize agricultural expenses, improving their profitability and sustainability. By advancing smart agriculture techniques, this research opens a new door for upcoming advancements in resource management and cost optimization in farming.
Keywords-Smart-farming, Animal deterrence, Renewable energy, Precision Irrigation and Cost Optimization

ABSTRACT: Voltage sags, characterized by temporary drops in voltage levels, are among the most prevalent and disruptive power quality disturbances in modern electrical systems. These events, caused by factors such as short circuits, sudden load changes, and network faults, pose significant risks to sensitive equipment and industrial processes, leading to operational disruptions and financial losses. Ensuring stable voltage levels is critical for maintaining high power quality standards and uninterrupted operations in today's interconnected and digital environments. Dynamic Voltage Restorers (DVRs) have emerged as effective solutions for mitigating voltage sags by injecting compensatory voltage to stabilize the load-side supply. However, conventional DVR controllers, like proportional-integral (PI) and proportional-integral-derivative (PID) controllers, often face challenges in adapting to dynamic and complex sag conditions. To address these limitations, this study proposes a DVR system enhanced by a fuzzy logic controller (FLC). Fuzzy logic offers a rule-based, adaptive approach, enabling the DVR to respond effectively to varying sag scenarios and improve voltage regulation. Presented FLC is working in different voltage sag ranges in an effective manner. Proposed FLC is adequate, effective and efficient one as it provide required level of voltages by operating in 3 modes (Low compensation, medium compensation and high compensation). This ensure the electric power saving too and hence more efficient than other controller. The research investigates the performance of the FLC-enhanced DVR in mitigating voltage sags under diverse fault conditions.
Keywords: Power quality; Voltage sag; Short circuit fault; Dynamic voltage restorer; Sag mitigation; Fuzzy logic controller.

ABSTRACT: The increasing frequency of negative online behaviour has drawn a lot of interest to predictive analysis of cyberbullying on Twitter data. This research suggests a unique method for predicting incidents of cyberbullying on Twitter by using a multi-model supervised strategy. The proposed approach aims to enhance efficacy and enhance the precision of cyberbullying detection through the integration of textual, social, and network attributes. The models are trained and assessed using Twitter data sets that include both cyberbullying and non- cyberbullying events. Sentiment analysis, bag-of-words, and semantic similarity are examples of textual features; follower count and account age are examples of social features. Analysing the user's interaction patterns and network structure is part of network features. The models are created and assessed using a variety of machine learning algorithms, including support vector machines (SVM), random forests (RF), and neural networks (NN). The outcomes of the experiments show that the combined strategy outperforms the individual models in terms of predictive performance. The significance of feature selection in enhancing model accuracy is further emphasised by the study. This research helps establish practical tactics and countermeasures to lessen the negative impacts of cyberbullying by precisely detecting incidences of cyberbullying on Twitter.
Keywords: Support Vector Machines, Random Forests, neural networks, textual features, social features, network features, cyberbullying prediction, Twitter data, multi- model supervised approach, and feature selection.