Browse Items (11808 total)
Sort by:
-
Optimal design of controller for automatic voltage regulator performance enhancement: a survey
For regulating the Synchronous Generator (SG) output voltage, the Automatic Voltage Regulator (AVR) system is a significant device. This work propounds a survey on Optimization Algorithms (OAs) utilized for tuning the controller parameters on the AVR system. A device wielded for adjusting the SGs Terminal Voltage (TV) is named AVR. A Controller is utilized for improving stability and getting a superior response by mitigating maximum Over Shoot (OS), reducing Rise Time (RT), reducing Settling Time (ST), and enhancing Steady State Error (SSE) since output voltage has a slower response and instability. The controllers utilized here are Proportional-Integral-Derivative (PID), Intelligent Controller (IC), along with Fraction Order PID (FOPID). Owing to the occurrence of time delays, nonlinear loads, variable operating points, and others, OAs are wielded for tuning the controller. (a) Particle Swarm Optimization (PSO), (b) Genetic Algorithm (GA), (c) Gray Wolf Optimizer (GWO), (d) Harmony Search Algorithm (HSA), (e) Artificial Bee Colony (ABC), (f) Teaching Learned Based Optimization (TLBO), et cetera are the various sorts of OA. For enhancing the TV response along with stability, various OAs were tried by researchers. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024. -
A Review of Optimization Algorithms Used in Proportional Integral Controllers (PID) for Automatic Voltage Regulators
The voltage in electrical grids is maintained at its nominal value by automatic voltage regulators (AVR). In AVR systems, proportional-integral-derivative (PID) is the most popular controllers due to their robust performance and simplicity. Controlling the parameters of proportional-integral-derivative (PID) controllers, which are used in AVR technology, is a nonlinear optimization problem. Optimization issues are of great importance to both the industrial and scientific worlds. A PID controller's objective function is designed to minimize the settling time, rise time, and overshoot of the step response of the resultant voltage. This paper presents the performance comparison of six optimization algorithms such as Enhanced Crow Search Algorithm (ECSA-PID), Slime Mould algorithm (SMA-PID), Future Search Algorithm (FSA-PID), Whale Optimization Algorithm (WOAPID), Equilibrium Optimizer (EO-PID) and Archimede's Optimization algorithm (AOA-PID) used in recent literatures. The Electrochemical Society -
Heart Disease PredictionA Computational Machine Learning Model Perspective
Relying on medical instruments to predict heart disease is either expensive or inefficient. It is important to detect cardiac diseases early to avoid complications and reduce the death rate. This research aims to compare various machine learning models using supervised learning techniques to find a better model that gives the highest accuracy for heart disease prediction. This research compares standalone and ensemble models for prediction analysis. Six standalone models are logistic regression, Naive Bayes, support vector machine, K-nearest neighbors, artificial neural network, and decision tree. The three ensemble models include random forest, AdaBoost, and XGBoost. Feature engineering is done with principal component analysis (PCA). The experimental process resulted in random forest giving better prediction analysis with 92% accuracy. Random forest can handle both regression and classification tasks. The predictions it generates are accurate and simple to comprehend. It is capable of effectively handling big datasets. Utilizing numerous trees avoids and inhibits overfitting. Instead of searching for the most prominent feature when splitting a node, it seeks out an optimal feature among a randomly selected feature set in order to minimize the variance. Due to all these reasons, it has performed better. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
A novel free space communication system using nonlinear InGaAsP microsystem resonators for enabling power-control toward smart cities
Nowadays, the smart grid has demonstrated a great ability to make life easier and more comfortable given recent advances. This paper studies the above issue from the perspective of two important and very useful smart grid applications, i.e., the advanced metering infrastructure and demand response using the instrumentality of a set of well-known scheduling algorithms, e.g., best-channel quality indicator, log rule, round robin, and exponentialproportional fairness to validate the performance. To increase the data transmission bandwidth, a new concept of optical wireless communication known as free-space optical communication (FSO) system based on microring resonator (MRR) with the ability to deliver up to gigabit (line of sight) transmission per second is proposed for the two studied smart grid applications. The range between 374.7 and 374.79THz frequency band was chosen for the generation of 10 successive-carriers with a free spectral range of 8.87GHz. The ten multi-carriers were produced through drop port of the MRR. The results show up to 10 times bandwidth improvement over the radius as large as 600m and maintain receive power higher than the minimum threshold (? 20dBm) at the controller/users, so the overall system is still able to detect the FSO signal and extract the original data without detection. 2019, Springer Science+Business Media, LLC, part of Springer Nature. -
Multifarious pigment producing fungi of Western Ghats and their potential
Concerns about the negative impacts of synthetic colorants on both con-sumers and the environment have sparked a surge of interest in natural col-orants. This has boosted the global demand for natural colorants in the food, cosmetics and textile industries. Pigments and colorants derived from plants and microorganisms are currently the principal sources used by mod-ern industry. When compared to the hazardous effects of synthetic dyes on human health, natural colors are quickly degradable and have no negative consequences. In fact, fungal pigments have multidimensional bioactivity spectra too. Western Ghats, a biodiversity hotspot has a lot of unique eco-logical niches known to harbor potential endophytic pigment-producing fungi having enumerable industrial and medical applications. Most of the fungi have coevolved with the plants in a geographical niche and hence the endophytic associations can be thought to bring about many mutually ben-eficial traits. The current review aims to highlight the potential of fungal pigments found in the Western ghats of India depicting various methods of isolation and screening, pigment extraction and uses. There is an urgent need for bioprospecting for the identification and characterization of ex-tremophilic endophytic fungi to meet industry demands and attain sustain-ability and balance in nature, especially from geographic hotspots like the Western Ghats. 2022 Horizon e-Publishing Group. All rights reserved. -
Dirichlet Feature Embedding with Adaptive Long Short-Term Memory Model for Intrusion Detection System
Intrusion Detection System is applied in the network to monitor the network activity and detect the intruder to protect the user data. Various existing models have been applied in the intrusion detection system and have the limitations of high False Alarm Rate (FAR), overfitting problem and data imbalance problem. In this research, Dirichlet Feature Embedding based Adaptive Long Short Term Memory (DFE-LSTM) model is proposed to improve the efficiency of the intrusion detection. The Dirichlet Feature Embedding (DFE) method is applied to effectively represent the feature to analysis the multi-variate of the input data. The enhanced Adaptive Long Short Term Memory (ALSTM) model is applied to select the optimal parameter for the LSTM model to improve the learning rate. The proposed DFE-ALSTM model is compared to three datasets such as UNSW-NB15, NSL-KDD and Kyoto 2006+ for evaluate the efficiency. The proposed DFE-ALSTM model has the accuracy of 94.32 % and existing NB-SVM has 93.75 % accuracy in intrusion detection on UNSW-NB15 dataset. 2022, Success Culture Press. All rights reserved. -
Tracking and Localization of Devices - An IoT Review
S everal IoT applications have immediate impacts on daily lives. The notion of "connected life, which includes IoT has been discussed. Apps that rely on localization are also featured. IoT is originally used to determine the precise position of things, animals, and people. The second tracks everyone and everything that's on the move, including pets, kids, and the elderly people. Localization and tracking are integral parts of security and surveillance systems in interconnected homes. This study reviews the state-of-the-art IoT-based localization and tracking approaches and outlines the key technical aspects, and contrast localization initiatives based on Internet of Things (IoT) with those that do not show how they might be used in a variety of contexts. It is now well established that localization and tracking methods based on the Internet of Things (IoT) are more pervasive and accurate than their predecessors. 2023 IEEE. -
WSETO: wild stock exchange trading optimization algorithm enabled routing for NB-IoT tracking system
The Narrowband Internet of Things (NB-IoT) communication plays a significant role in the IoT due to the capability of generating broad exploration with the usage of limited power. Over the past few years, the Low Power Wide Area Networks (LPWAN) have been efficient in the data acquisition and remote monitoring area however they failed to generate high data rates, low latency, and the consumption of low power. To solve these problems, NB-IoT technology has developed in long-term asset tracking and it replaces the Global Positioning System (GPS) with its ubiquitous coverage. In this research, the Wild Stock Exchange Trading Optimization technique (WSETO) is proposed for a routing-based NB-IoT tracking system. The WSETO is the combination of the Wild Geese Algorithm (WGA) and SETO. By employing WSETO, the routing to the relevant target location is established effectively. The existing techniques like Low Power Asset Tracking of NB-IoT (LoPATraN), Monitoring system based on NB-IoT and BeiDou System/GPS (BDS/GPS), and Narrowband Physical Uplink Shared Channel (NPUSCH) are used to compare the WSETO approach. In rounds with a value of 2000, the WSETO demonstrates a superior location error of 0.001 in comparison to existing methods such as LoPATraN, a monitoring system utilizing NB-IoT and BDS/GPS, as well as NPUSCH. Bharati Vidyapeeth's Institute of Computer Applications and Management 2024. -
Deep Learning Based Multi Constraint Hybrid Optimization Algorithm for Transshipment-Based Inventory Routing with Dynamic Demands
The Inventory-Routing Problem (IRP) is considered a major issue in supply chain management as it comprises two areas: vehicle routing and inventory control. The existing techniqueswere unable to incorporate location details for enhancing the decision-making and it failed to consider the uncertainty of the demands. Hence to solve this issue, a Snake Artificial Ecosystem Optimization (SAEO) algorithm is proposed in this paper. The SAEO algorithm is developed to address the transshipment IRP with dynamic demands by combining the AEO model and SO to enhance the optimizer's performance. Further, a penalty strategy is proposed, where Deep Quantum Neural Network (DQNN) is employed for calculating the penalty for verifying the feasibility of the solution generated in case of violations in model constraints. In addition, the efficiency of the proposed SAEO-DQNN technique is examined by considering metrics, like transportation cost, transshipment cost, and total cost, and it achieved improved values of 0.391, 0.518, and 1.012 when compared to existing techniques such as Genetic Algorithm with Deep Reinforcement Learning (GA + Deep RL) and Kernel Search Multi-vehicle IRP (KSMIRP). The Author(s) 2024. -
A Review on Influence of Cutting Fluid on Improving the Machinability of Inconel 718
Nickel-based superalloys are widely used in the production and manufacturing sectors that require processes or applications that endure or operate at very high superheating temperatures. With the properties of high tensile strength, high melting point, and lightweight structural arrangement of molecules within the alloy material composition makes it more suitable for industrial utilization in aerospace industries and marine applications. This review paper discusses the use of various coolant lubricants that improves the machinability of Inconel 718 based on parameters such as surface roughness and tool wear under the influence of cutting speed, feed rate, and depth of cut. The machine used for analysis is CNC milling machine which will be used for experimentation using ceramic inserts as end milling tool. Various cooling techniques such as hybrid cooling, flood emulsion cooling, minimum quantity lubrication, and cryogenic cooling are being summarized in this paper from various experimentations and conclusions of other authors. On the basis of review, the hybrid cooling technique is found to be better than other cooling techniques because of its ability to obtain long tool life and smoother surface finish on the workpiece. With the use of these reviewed data, further research for finding a more compatible and effective cooling lubricant has to be done by experimentation in order to obtain an improved machining process for Inconel 718 material. 2020, Springer Nature Singapore Pte Ltd. -
Inplane Lateral Load Behaviour of Masonry Walls
Masonry is one of the commonly used construction technology both in urban and rural areas. In this paper the in-plane behaviour of masonry walls is analytically studied considering existing closed form equations. Previous studies have proven that the lateral load behaviour mainly depends on the aspect ratios (h/L) as well as the axial loads. From this analysis the governing failure is determined and the lateral load versus lateral deflection curve is plotted for various percentages of axial loads. This graph gives the ductility of the wall. This concept is further applied to a simple masonry structure and the push over curve is plotted. 2020, Springer Nature Switzerland AG. -
Artificial Intelligence in Data-Driven Analytics for the Personalized Healthcare
Among the various developments in progress over the last decade, we have seen the generous growth of information investigation to take care of, plan, and use a lot of information beneficially. Be that as it may, because the analysis of evidence will only operate for authentic information and have findings as predefined by individuals, explicit principle-based calculations have been developed to broaden the investigation of information, 'Which is usually referred to as 'AI'. AI didn't expect PCs to be personalized unambiguously, which is a definite bit of leeway. In order to break down information and construct complicated equations to foresee models, which was called prescient analysis, AI was then joined with information inquiry. A set of laws characterized by persons, known as prescient equations, drive the prescient inquiry, and are used to break down genuine knowledge in order to predict potential outcomes. 2021 IEEE. -
Hybrid cryptography security in public cloud using TwoFish and ECC algorithm
Cloud computing is a structure for rendering service to the user for free or paid basis through internet facility where we can access to a bulk of shared resources which results in saving managing cost and time for large companies, The data which are stored in the data center may incur various security, damage and threat issues which may result in data leakage, insecure interface and inside attacks. This paper will demonstrate the implementation of hybrid cryptography security in public cloud by a combination of Elliptical Curve Cryptography and TwoFish algorithm, which provides an innovative solution to enhance the security features of the cloud so that we can improve the service thus results in increasing the trust overthe technology. 2019 Institute of Advanced Engineering and Science. -
An investigation and analysis on automatic speech recognition systems
A crucial part of a Speech Recognition System (SRS) is working on its most fundamental modules with the latest technology. While the fundamentals provide basic insights into the system, the recent technologies used on it would provide more ways of exploring and exploiting the fundamentals to upgrade the system itself. These upgrades end up in finding more specific ways to enhance the scope of SRS. Algorithms like the Hidden Markov Model (HMM), Artificial Neural Network (ANN), the hybrid versions of HMM and ANN, Recurrent Neural Networks (RNN), and many similar are used in accomplishing high performance in SRS systems. Considering the domain of application of SRS, the algorithm selection criteria play a critical role in enhancing the performance of SRS. The algorithm chosen for SRS should finally work in hand with the language model conformed to the natural language constraints. Each language model follows a variety of methods according to the application domain. Hybrid constraints are considered in the case of geography-specific dialects. 2024 by author(s). -
Ensembled convolutional neural network for multi-class skin cancer detection
A skin cancer diagnosis is critically important in medical image processing. The role of dermoscopy and dermatologists is inevitable in skin cancer diagnosis. But, considering the time constraints on diagnosing patients on time, even medical experts need computer-assisted methods to automate the diagnosis process with a higher accuracy rate and with good performance. Such computer-assisted methods with induced artificial intelligence (AI) algorithms are gaining significance. The challenging task of medical image processing is finding benign/malignant pigmented skin lesions after the input image of patients. To identify this difference, AI-based classification algorithms shall be deployed. During the implementation of such algorithms, several performance aspects are evaluated. Once the best such algorithm is identified and evaluated for its performance attributes, it shall be deployed to assist dermatologists. This book chapter explains such a novel multiclass skin cancer classification algorithm. The proposed algorithm uses the best of the attributes and parameters of a deep convolutional neural network (CNN) to give the best-ever enactment among similar existing algorithms. The result achievement of the developed deep CNN based multi-class skin cancer classification algorithm (DCNN-MSCCA) is demonstrated using the HAM10000 dataset. To establish the significance of the developed algorithm, the performance parameters of the DCNN-MSCCA are compared with a few existing significant algorithms. The maximum accuracy of DCNN-MSCCA in predicting the exact multi-class skin cancer is 95.1%. This book chapter explains the implementation details of DCNN-MSCCA using python and libraries supporting CNN. 2024 River Publishers. -
Assessing Academic Performance Using Ensemble Machine Learning Models
Artificial Intelligence (AI) shall play a vital role in forecasting and predicting the academic performance of students. Societal factors such as family size, education and occupation of parents, and students' health, along with the details of their behavioral absenteeism are used as independent variables for the analysis. To perform this study, a standardized dataset is used with data instances of 1044 entries and a total of 33 unique variables constituting the feature matrix. Machine learning (ML) algorithms such as Support Vector Machine (SVM), Random Forest (RF), Multilayer Perceptron (MLP), LightGBM, and Ensemble Stacking (ES) are used to assess the specified dataset. Finally, an ES model is developed and used for assessment. Comparatively, the ES model outclassed other ML models with a test accuracy of 99.3%. Apart from accuracy, other parameters of metrics are used to evaluate the performance of the algorithms. 2023 IEEE. -
The role of big data in predicting consumer behavior
Consumer behavior prediction is a significant task, and it is a prerequisite for marketing activities. Regardless of the product type/market type, predicting consumer behavior plays a vital role in determining the target market. The activities involved in identifying a target market include the tasks of analyzing the offerings, conducting market research, identifying market segments to create consumer profiles, and assessing the competition. In order to complete all four tasks mentioned above, it needs to have comprehensive and precise data/dataset in hand. It also means that the data/fact is the primary source of predicting consumer behavior. In today's digital world, sources of source (data) are multifold. During the process of data collection, if the repository is accepting data from such sources, then all five "V" (Volume, Velocity, Variety, Veracity, and Value) of data should be considered. The role of big data in predicting consumer behavior is inevitable. Machine learning models shall be deployed to analyze data from big data. In this chapter, benchmark datasets, and machine learning, are used to demonstrate the usage of artificial intelligence in analyzing, forecasting, and predicting consumer behavior. Before concluding the chapter, the performance of algorithms is evaluated and compared to find the most suitable models for predicting consumer behavior. Benchmark datasets are used in this chapter to represent the role of big data in predicting consumer behavior. 2023 Nova Science Publishers, Inc. All rights reserved. -
Development of smart energy monitoring using NB-IOT and cloud
IoT-based applications are growing in popularity nowadays because they offer effective answers to numerous current problems. In this research, With the aim of decreasing human efforts for monitoring the power units and increasing users' knowledge of excessive electricity usage, an IoT-based electric metre surveillance system utilising an Android platform has been developed. With the help of an Arduino Uno and an optical sensor, the electric analyser pulse is captured. To reduce human mistake and the expense of energy usage, a low-cost wireless network of sensors for digital energy metres is implemented alongside a smartphone application that can autonomously read the metre of the unit. In this research, an intelligent power monitoring system with effective communication modules has been developed to make wise use of the electricity. The controller, NB-IoT connection module, and cloud are the three main components of an IOT-based smart energy metre system. The controller is essential for maintaining the functionality of each component. This solution reduces the need for human involvement in electricity maintenance by connecting energy metres to the cloud using an NB-IoT communication module. The IoT-based metre reading system in the proposed work is created to monitor and analyse the metre reading, and the service provider can cut off the source of electricity whenever the customer fails to pay the monthly bill. It also eliminates the need for human intervention, provides accurate metre reading, and guards against billing errors. The proposed SPM improves the overall accuracy ranges of 7.42, 27.83, and 20% better than DR, OREM, and SLN respectively. 2023 -
Assessing Human Stress Through Smartphone Usage
Stress occurs in a human being when they are faced with exigent situations in life. Assessing stress has been always challenging. Smartphones have become a part of everyones day-to-day activity in the present time. Considering humansmartphone interaction, sensing of stress in an individual can be assessed as todays youth spends most of their time with smartphones. Taking this into consideration, a study is carried out in this paper on assessing stress of an individual based on their interaction with the smartphone. In this work, humansmartphone interaction features, like swipe, scroll, and text input, are examined. Text input is incorporated by disabling the autocorrection and spelling checker features of the keyboard. Moreover, sensor data is used by Google activity recognition API to analyze the physical activity of the individual to assess the stress level. 2019, Springer Nature Singapore Pte Ltd. -
Analyzing students academic performance using multilayer perceptron model
Identification of the students behavior in the class room environment is very important. It helps the lecturer to identify the needs of the students. It also aids in identifying the strength and weakness of the individual and guide them to improve on their performance. Observing and supervising the students regularly can improve their performance. The data has been collected from 120 students who took the common the course taught by two different lectures. The students were observed based on the internal assignments and quizzes and the model exam given by the respective lecturers. In this paper the students are categorized into different groups based on their performance using Multilayer Perceptron (MLP) and also different factors which are influencing the performance of the students are identified. BEIESP.