Arivukkarasan Raja, PhD

The Short-Term Load Forecasting System for Smart Grid (SG) Environment is a predictive modelling technique developed for anticipating the energy consumption over a comparably small time horizon within the Smart Grid framework. Leveraging advanced predicting methods, this system integrates different data sources like weather conditions, real-time grid data, and historical consumption patterns for generating exact prediction of short-term load requirement. The system empower grid operators and… Show more

The Short-Term Load Forecasting System for Smart Grid (SG) Environment is a predictive modelling technique developed for anticipating the energy consumption over a comparably small time horizon within the Smart Grid framework. Leveraging advanced predicting methods, this system integrates different data sources like weather conditions, real-time grid data, and historical consumption patterns for generating exact prediction of short-term load requirement. The system empower grid operators and utilities to enhance resource allocation, improve grid stability, and proactively manage energy distribution by using sophisticated techniques, statistical, and machine learning algorithms. This predicting scheme plays a major role in enabling effective energy utilization and planning, which contribute to the overall sustainability and reliability of Smart Grid system. This manuscript offers the design of Gravitational Search Algorithm with Deep Learning Enabled Short-Term Load Forecasting (GSADL-STLF) Scheme for Smart Grid Environment. The GSADL-STLF technique make use of deep belief network (DBN) for modelling and forecasting the short-term electricity utilization pattern. Through the assimilation of different and adaptive data sources like historical load profile, weather condition, and grid operational parameter, the DBN model properly predicts the load in the SG environment. For fine tuning the parameters related to the DBN model, the GSA has been applied. By extensive performance analysis of the GSADL-STLF technique, the performance is validated on real world dataset. The experimental values inferred that the GSADL-STLF technique exhibit significant performance to predict the short term load demand. Show less

Composite materials are substituted by traditional materials owing to their unique properties and the inclusion of reinforcement materials, which are creating a new trend in the material world. This research paper relates to an investigation carried out using Al6061base alloy, as the matrix changes the weight percentage (5%, 10%, 15%) of SiC, B4C (3%) and keeps the weight percentage (2%) of Talc constant, which acted as the reinforcing material. The addition of SiC, B4C, and talc to the… Show more

Composite materials are substituted by traditional materials owing to their unique properties and the inclusion of reinforcement materials, which are creating a new trend in the material world. This research paper relates to an investigation carried out using Al6061base alloy, as the matrix changes the weight percentage (5%, 10%, 15%) of SiC, B4C (3%) and keeps the weight percentage (2%) of Talc constant, which acted as the reinforcing material. The addition of SiC, B4C, and talc to the composite improved its mechanical qualities. EDX and a scanning electron microscope were both employed in this work to investigate the specimens for microstructural analyses. The wear characteristics of composites were investigated using a pin-on-disc machine at sliding velocities of (1, 1.5, 2) m/s, a sliding distance of (1000, 1200, 1500) m, 15% reinforcement weight (5, 10, 15), and loads of 9.81, 19.62, and 29.43 N. The experiment was conducted on the orthogonal array of L27 using Taguchi's method. ANOVA and Grey relational equations were used to determine the optimal value and coefficient of friction based on the influence of input factors. These characteristics allowed us to calculate the ideal wear rate and friction coefficient. It is claimed in the present publication that the results of the experiments have been confirmed by the confirmation tests. Show less

In Each nation across the globe has established a mechanism for reporting instances of violence perpetrated against women. The statistical data pertaining to law enforcement agencies in Asian nations indicates a notable incidence of criminal offences perpetrated against the female gender. During the late 1990s, the National Crime Records Bureau released a report which forecasted that the escalation in the incidence of crimes perpetrated against women could potentially surpass the growth rate by… Show more

In Each nation across the globe has established a mechanism for reporting instances of violence perpetrated against women. The statistical data pertaining to law enforcement agencies in Asian nations indicates a notable incidence of criminal offences perpetrated against the female gender. During the late 1990s, the National Crime Records Bureau released a report which forecasted that the escalation in the incidence of crimes perpetrated against women could potentially surpass the growth rate by the year 2020. Historically, a significant number of offences perpetrated against women were not brought to the attention of law enforcement officials owing to the negative connotations attached to such actions. Based on the data gathered by the government, there has been a significant increase in the incidence of various forms of criminal activities perpetrated against women. The principal objective of this study is to address the deficiencies in the current security surveillance systems with the aim of enhancing the safety of female constituents within the community. The extant surveillance technology currently available in the market is deemed insufficient to address the contemporary apprehensions regarding the safety of women. Consider the scenario wherein a male individual gains unauthorized access to a women's hostel facility. In the case of a nocturnal emergency resulting from a fire outbreak, during which the occupants of the hostel are in a state of slumber, the developed system will effectively identify the fire and promptly alert the designated fire department personnel through a pre-programmed message. The proposed system is expected to incorporate supplementary features, including realtime video streaming, email and text message notifications, SMS alerts for unauthorized access, and the capacity to transmit videos of anomalous movements detected within the monitored region. Show less

Emotion detection has emerged as a critical area of study that exposes many relevant inputs. Emotion can be expressed in numerous forms such as written text, facial and speech expressions, and gestures. Emotion analysis represents the task of identifying the attitude against a target or topic. This might be an emotional (positive or negative) or polarity state including anger, joy, or sadness. Emotion detection in a textual document is basically a content-based classification issue such as… Show more

Emotion detection has emerged as a critical area of study that exposes many relevant inputs. Emotion can be expressed in numerous forms such as written text, facial and speech expressions, and gestures. Emotion analysis represents the task of identifying the attitude against a target or topic. This might be an emotional (positive or negative) or polarity state including anger, joy, or sadness. Emotion detection in a textual document is basically a content-based classification issue such as notion from deep learning (DL) and natural language processing (NLP) sectors. In this aspect, this article develops Deep learning with Natural Language Processing for Emotion Detection and Classification on Social Networking (DLNLP-EDC) approach. The proposed DLNLP-EDC algorithm carries out proper categorization of various emotions in the social networking data. Primarily, the DLNLP-EDC technique employs data preprocessing and word2vec feature extraction. To identify several kinds of emotions, the DLNLP-EDC technique designs capsule autoencoder (CAE) model. For improving the performance of the CAE model, enhanced cuckoo search optimization (ECOA) is derived. The results of the DLNLP-EDC technique are studied well on benchmark databases. The simulation result signified the improvement of the DLNLP-EDC technique over existing systems. Show less

The post recognition of diseases in fruits and vegetables by farmers in India is a contributing factor in the country's declining crop yield. Farmers everywhere are suffering significant economic setbacks. The majority of the agricultural loss can be attributed to diseases that affect both plants and fruits. Farmers can boost their output by increasing their awareness of the nutritional quality of the fruits and vegetables they grow. Because of this, we are motivated to create and develop a… Show more

The post recognition of diseases in fruits and vegetables by farmers in India is a contributing factor in the country's declining crop yield. Farmers everywhere are suffering significant economic setbacks. The majority of the agricultural loss can be attributed to diseases that affect both plants and fruits. Farmers can boost their output by increasing their awareness of the nutritional quality of the fruits and vegetables they grow. Because of this, we are motivated to create and develop a technology that will assist farmers in detecting diseases in their crops at an early stage. The proposed research utilizes Inception V3 model which is accomplished with the help of convolutional neural networks. The performance of the Inception V3 model is enhanced by tuning its hyperparameters using Particle Swarm Optimization (PSO). The accuracy of the proposed model using Inception V3 and PSO algorithm is 0.987. Show less

Only when crops are handled efficiently are the insights they provide translated into profitable decisions. Current advancements in data management are causing Smart Farming to expand dramatically, since data has become the most important aspect of modern agriculture, assisting farmers in making crucial decisions. With the goal of improving production and sustainability, objective information gathered by sensors yields valuable benefits. This type of data-driven farm management relies on data… Show more

Only when crops are handled efficiently are the insights they provide translated into profitable decisions. Current advancements in data management are causing Smart Farming to expand dramatically, since data has become the most important aspect of modern agriculture, assisting farmers in making crucial decisions. With the goal of improving production and sustainability, objective information gathered by sensors yields valuable benefits. This type of data-driven farm management relies on data that can boost productivity by preventing the waste of resources and the contamination of the environment. Data-driven agriculture, aided by robotic solutions using techniques of artificial intelligence, lays the foundation for the agricultural of the future. The Internet of Things (IoT) and other cloud-based solutions, along with the ever-evolving state of the internet world, present a once-in-a-generation window of opportunity for the creation of automated and robotic systems in the fields of urban agriculture, agriculture, and forestry. The advent of GPS, machine vision, lasers, and mechatronics have all contributed to the creation of cutting-edge robotic systems and intelligent technologies for use in precision farming. This study showcased a flexible Internet of Things (IoT) robotic system for greenhouse cultivation applications, with the goal of realtime illness diagnosis in plants. Instead with ADAM, the small robotic system employs a BRCNN (Backpropagation Recurrent Convolution Neural Network) module trained with MFO (Moth Flame Optimizer) as its Primary Optimizer to identify the presence of disease on a plant and SVM to characterize the type of sickness. Further, the system provides appropriate feedback remedies if and when they are required. There is no longer any need to spend a ton of money to examine plants and conduct thorough chemical research because this method permits a quick and preliminary classification of plant illnesses. Show less

Because of the restricted capabilities in plant pathology and the limited number of reserves that can be retrieved, the automation of procedures has become necessary. Farmers in every region of the world have the challenge of warding off various types of damage caused by bacteria and other types of pathogens like fungus, virus and so forth. Infectious plant diseases have become an increasingly significant element that affects crop productivity and economic efficiency as a result of the… Show more

Because of the restricted capabilities in plant pathology and the limited number of reserves that can be retrieved, the automation of procedures has become necessary. Farmers in every region of the world have the challenge of warding off various types of damage caused by bacteria and other types of pathogens like fungus, virus and so forth. Infectious plant diseases have become an increasingly significant element that affects crop productivity and economic efficiency as a result of the expansion of agriculture. Hence, the safety of plants is significant to meet the food requirements. To solve this, in this paper, we have proposed an Usage of Robots for Phytosanitary Requirements Like Agricultural Plants Health and Safety in Internet of Things (IoT). Here, Convolutional Neural Network (CNN) is employed to classify the images of leaves with disease using improved AlexNet architecture. Initially, the images are resized and augmented using data augmentation to form an even count of images in every class. After that, CNN is performed, and the diseased leaf images are identified with more precision. Results show that the proposed technique improves the performance and accuracy with minimum loss. Show less

One of the primary difficulties created by the expansion in global population is the availability of food for all of the world's inhabitants. These challenges must be overcome by implementing new ways to improve soil capacity and environmental resource security. The availability of real-time critical agricultural metrics such as temperature, moisture, weather, water management, and crop diseases, as well as predictive techniques against parameter changes, could be very helpful in addressing… Show more

One of the primary difficulties created by the expansion in global population is the availability of food for all of the world's inhabitants. These challenges must be overcome by implementing new ways to improve soil capacity and environmental resource security. The availability of real-time critical agricultural metrics such as temperature, moisture, weather, water management, and crop diseases, as well as predictive techniques against parameter changes, could be very helpful in addressing these challenges. The Internet of Things (IoT) is a new technology that has the potential to transform nearly every industry. The Internet of Things is a network of devices that can configure themselves. Intelligent agriculture is gaining traction in developed countries. It makes precision agriculture easier and changes the agricultural production landscape. As a result, it decreases resource waste such as water, fertilizers, and operating costs. Smart miniaturized sensors, processors, and communication technologies are now available, enabling IoT-based smart farming. This study will examine the most recent papers on smart farming from 2017 to 2022. Recent works are highlighted in terms of scope and technique. Future researchers will be able to build a system with a single standard expert and a fully autonomous help system on top of this foundation. Show less

A smart city is a well-planned, self-sustainable urban area which is capable of managing its own resource and assets intelligently and efficiently with the help of the data derived out of various sensing elements. A hazard can be defined as an external or internal agent, which has some unrealized ability to cause harm. Some of the identified potential hazards in industrial and residential environment include, but not limited to, air quality, air temperature, humidity, noise levels, vibration… Show more

A smart city is a well-planned, self-sustainable urban area which is capable of managing its own resource and assets intelligently and efficiently with the help of the data derived out of various sensing elements. A hazard can be defined as an external or internal agent, which has some unrealized ability to cause harm. Some of the identified potential hazards in industrial and residential environment include, but not limited to, air quality, air temperature, humidity, noise levels, vibration, radiation, fire, electric short circuit, water leakage, etc. In the context of smart cities, the management of these potential hazards is very important in both industrial and residential environments to assure superior quality of living, reduce the cost of health care and for improving the morale, productivity, and job satisfaction of the employees. The proposed risk and hazard management system includes multiple sensors, which are programmed to receive data like carbon monoxide, formaldehyde, lead, nitrogen dioxide, butane, LPG, and radon levels in real-time. The end-users can visualize the data through Internet of things (IoT) open-source platform. The services for device integrations are done through Embedded C, loaded in Teensy 3.2. Similarly, the services for data validations, intelligent actions, and the data storage, display using Python, loaded in Raspberry Pi3 Model B+. The remedial or control actions like automated door, exhaust fan, and solenoid valve operations are carried out based on real-time data analysis using artificial intelligence. Show less

Indoor and outdoor contaminants in both residential and industrial properties can be a significant environmental health problem. Various health issues have been linked with occupant exposure to numerous toxic and hazardous substances. Some of the identified potential hazards in industrial and residential environment include but not limited to air quality, air temperature, humidity, noise levels, vibration, radiation, fire, electric short circuit, water leakage etc., The management of these… Show more

Indoor and outdoor contaminants in both residential and industrial properties can be a significant environmental health problem. Various health issues have been linked with occupant exposure to numerous toxic and hazardous substances. Some of the identified potential hazards in industrial and residential environment include but not limited to air quality, air temperature, humidity, noise levels, vibration, radiation, fire, electric short circuit, water leakage etc., The management of these potential hazards are very important in both industrial and residential environments to assure quality health care at reduced insurance cost, improving the morale of the employees, increased productivity and job satisfaction with fewer injuries. The proposed risk and hazard management system includes multiple sensors, which are programmed to receive data like Carbon Monoxide, Nitrogen Dioxide, Temperature, Humidity, Butane, LPG and Ammonia levels in real-time. The end users visualize or measure the collected data through an internet of things (IoT) Open source platform. The system further includes a processor such as Raspberry Pi in which the services for visualizations are loaded. It also includes a MySQL database in which historical data for risk analysis are stored. The remedial or control actions are carried out based on the analyzed data using artificial intelligence. The remedial or control actions include, if the concentration of Sulphur dioxide exceeds 80 mg/m3, door, windows will be closed through a stepper motor interface, and the exhaust fan will be switched on. If the concentration of carbon monoxide exceeds 4 mg/m3, ammonia exceeds 400 mg/m3, there is a substantial amount of gas leakage (butane & propane), the gas connection will be turned off through a smart solenoid valve, door and Windows will be opened through a stepper motor interface and the exhaust fan will be switched on. Show less