Abstract: Photovoltaic (PV) power generation is playing a prominent role in rural power generation systems due to its low operating and maintenance cost. The output properties of solar PV mainly depend on solar irradiation, temperature, and load impedance. Hence, the operating point of solar PV oscillates. Due to the oscillatory behavior of operating point, it is difficult to transform maximum power from the source to load. To maintain the operating point constant at the maximum power point (MPP) without oscillations, a maximum power point tracking (MPPT) technique is used. Under partial shading condition, the nonlinear characteristics of PV comprise of multiple maximum power points (MPPs). As a result, discovering true MPP is difficult. The traditional and neural network MPPT methods are not suitable to track the MPP because of oscillations around MPP and impreciseness in tracking under partial shading (PS) condition. Therefore, in this article, a biological intelligence cuckoo search optimization (CSO) technique is utilized to track and extract the maximum power of the solar PV at two PS patterns. MATLAB/Simulink is used to demonstrate the CSO MPPT operation on SEPIC converter.

Abstract: Obstacle detection and avoidance during navigation of an autonomous vehicle is one of the challenging problems. Different sensors like RGB camera, Radar, and Lidar are presently used to analyze the environment around the vehicle for obstacle detection. Analyzing the environment using supervised learning techniques has proven to be an expensive process due to the training of different obstacle for different scenarios. In order to overcome such difficulty, in this paper Reinforcement Learning (RL) techniques are used to understand the uncertain environment based on sensor information to make the decision. Policy free, model-free Q-learning based RL algorithm with the multilayer perceptron neural network (MLP-NN) is applied and trained to predict optimal vehicle future action based on the current state of the vehicle. Further, the proposed Q-Learning with MLP-NN based approach is compared with the state of the art, namely, Q-learning. A simulated urban area obstacles scenario is considered with the different number of ultrasonic radar sensors in detecting obstacles. The experimental result shows that Q-learning with MLP-NN along with the ultrasonic sensors is proven to be more accurate than conventional Q-learning technique with the ultrasonic sensors. Hence it is demonstrated that combining Q-learning with MLP-NN will improve in predicting obstacles for autonomous vehicle navigation.

Abstract: This study explored different models of PV cell, namely, single diode model and double diode models using MATLAB/Simulink Environment. The output power and current characteristics are analyzed for different solar intensity radiations and temperature variations of PV cell. Simulation results are obtained for different atmospheric and temperature conditions. The simulation results reveal that the double diode model generates maximum power and has a higher efficiency compared to single diode model.

Abstract: Maximum power can be extricated when the turbine keeps running at a consistent and constant speed by using all the vitality present in the wind. The turbine can keep running at a steady speed just when the breeze speed is consistent. The wind vitality being wild in nature, maximum power must be achieved by making the turbine to keep running at the specific breeze speed. To achieve most extreme power, distinctive sorts of maximum power point tracking (MPPT) procedures are utilized. So as to comprehend prudent and proficient power age utilizing wind turbines, modification of fuzzy-based MPPT method is displayed and results are compared with different MPPT techniques for wind energy conversion system have been done and are introduced in subtleties.

Abstract: This study applies the principle of artificial neural networks for modelling the mechanical characteristics of a lightweight self-compacting concrete containing pumice and mineral admixtures. Models for predicting compressive strength, split tensile strength and flexural strengths were developed based on several measures of the materials as obtained from the experimental stage. The input parameters for the model were contents of cement, ground granulated blast furnace slag (GGBS), rice husk ash (RHA), fine aggregates, coarse aggregates, pumice stone, water, super-plasticizers and micro-silica. Three output parameters, including compressive strength, tensile strength and flexural strength were considered. The data were trained, tested and validated using the feedforward backpropagation algorithm. The study established the best model for the tested concrete, based on the minimal error criteria, as 9 (input), 12 (hidden layer) and 3 (output layer). This model is expected to serve as a useful tool for concrete designers and constructors.

Abstract: This study develops predictive models for determination of strength parameters of laterized concrete made with ceramic aggregates, based on the principle of Artificial Neural Networks (ANN). The model development follows the results of the experimental phase (covering compressive and split-tensile strengths), where numerous materials were used in varying proportions: ceramics (fine and coarse fractions), river sand, and granite were substituted between 0 and 100%, laterite between 0 and 30%, and curing ages between 3 and 91 days. The cement proportion was maintained at 100%, and the water–cement ratio was 0.6. The model development was performed in MATLAB based on the Levenberg–Marquardt (LM) principles, where input data were separated in ratio 70%:15%:15% for learning, testing, and validation phases, respectively. After several trials, the selected model architecture, based on satisfactory performance in terms of means square error, contains eight-input layer, ten-hidden layer, and two-output layer neurons.

Abstract: Flower pollination algorithm (FPA) is an evolutionary nature-inspired optimization technique, which mimics the pollinating behavior of flowers. FPA has a simple structure and has been applied to numerous problems in different fields of research. However, it has been found that it has poor exploration and exploitation capabilities. In this paper, to mitigate the problems of original FPA, a modified algorithm namely adaptive FPA (AFPA) has been proposed. In the modified algorithm, a four-fold population division has been followed for both global and local search phases. Moreover, to balance the local and global search, switching probability has been decreased exponentially with respect to iterations. For experimental testing, this algorithm has been further applied to antenna design problems. The aim is to optimize linear antenna array (LAA) in order to achieve minimum SLL in the radiation pattern to avoid antenna radiation in the undesired directions. The results of the proposed algorithm for same problems are compared with the results of popular algorithms such as particle swarm optimization (PSO), tabu search (TS), self-adaptive differential evolution (SADE), Taghchi’s method (TM), cuckoo search (CS), and biogeography-based optimization (BBO). The simulation results clearly indicate the superior performance of AFPA in optimizing LAA.

Abstract: With the recent emphasis on large-scale civil engineering constructions, artificial intelligence in the construction activities has received importance. Compressibility behavior is an important property in fine soils to find out the settlements in foundation designs. However, compression index (Cc) from one-dimensional swell-consolidation test is time consuming and laborious. Many traditional prediction-stimulated models rely on simplified assumptions, leading to inaccurate Cc estimations. This paper explores, by comparison, the application of invasive weed optimization (IWO) algorithm and particle swarm optimization (PSO) to predict Cc via multiple linear regression models. The predicted model equations have been developed, uses four input parameters namely plasticity index, free swell index, rate of heave and swell potential in both methods. The results confirm that the developed models using IWO provides accurate prediction than standard particle swarm optimization (PSO) algorithm.

Abstract: Most of the organizations are stuck in a situation to cloudify due to concerns regarding security of sensitive data. Cloud computing provides many aids to the users and organizations in accordance with expenditure and savings. Other than these benefits, cloud computing has some hurdles that result in restriction of it’s usage. Cloud security is the bigger hurdle which is regularly taken into consideration. This paper will give a broad view of major threats challenges in security which are encountered in cloud computing. Cloud computing entities include cloud user, cloud provider, and data owner. The study is carried out based on selection of open-source cloud offerings. This paper can be an exploration tool for any IT person to gain knowledge into security-related risks and challenges which are concerned with cloud computing.

Abstract: Landslides and associated slope movements are common occurrences in the hilly regions. In particular, Tangni in Uttarakhand state between Pipalkoti and Joshimath has experienced a number of landslides in the recent past. Prior research has used a certain moving average and machine-learning (ML) algorithms to predict slope movements. However, a comparison of these methods for real-world slope movements has been less explored. The primary objective of this paper was to compare a seasonal autoregressive integrated moving average (SARIMA) model, a multilayer perceptron (MLP) model, and a long short-term memory (LSTM) model to predict slope movements recorded at the Tangni landslide in Chamoli, India. Time series data about slope movements from five sensors placed on the Tangni landslide hill were collected daily over a 78-week period from July 2012 to July 2014. Different model parameters were calibrated to the training data (first 62 weeks) and then made to predict the test data (the last 16 weeks). Results revealed that the moving-average models (SARIMA) performed better compared to the ML models (MLP and LSTM) during both training and test. We highlight the implication of using moving-average models for predicting slope movements.

Abstract: From some past recent years, Swarm Intelligence (SI) based optimization algorithms have shown their impact in finding the efficient solutions of real-life application problems that occur in engineering, science, industry, and in various other fields. Gray Wolf Optimizer (GWO) is an efficient and popular optimizer in the area of SI to solve nonlinear complex optimization problems. GWO mimics the dominant leadership characteristic of gray wolves to catch the prey. But, like other stochastic search algorithms, GWO gets trapped in local optimums in some cases. Therefore in the present study, Random Walk Gray Wolf Optimizer (RW-GWO) is applied to determine—(1) the optimal redundancies to optimize the system reliability with constraints on volume, weight, and system cost in series, series–parallel, and complex bridge systems and (2) the optimum cost of two different types of complex systems with constraints imposed on system reliability. The obtained results are compared with classical GWO and some other optimization algorithms that are used to solve reliability problems in the literature. The comparison shows that the RW-GWO is comparatively an efficient algorithm to solve the reliability engineering problems.

Abstract: The concept of green manufacturing has gained cognizance among manufacturers due to regulations imposed by the government and rising environmental consciousness of customers. Acknowledging the fact that green manufacturing can yield long-term economic and environmental gains with significant efforts channelized toward green marketing, firms are reinventing their marketing strategies. Although many researchers have discussed the importance as well as theory building of green marketing, none have analyzed the strategies under multi-criteria environment and with multi-stakeholder perspective, which is the novelty of this study. Here, a real-life case of a manufacturing firm has been considered, who wants to select an appropriate green marketing strategy for promoting its newly introduced green product, from four available strategies, namely (i) Lean Green, (ii) Defensive Green, (iii) Shaded Green, and (iv) Extreme Green. The firm’s objective is to select the most appropriate strategy which is ideal for targeting green consumers. Integrated methodology of Fuzzy Analytical Hierarchy Process (Fuzzy AHP) and Fuzzy Technique for Order of Preference by Similarity to Ideal Solution (Fuzzy TOPSIS) is implemented for strategy selection, based on criteria such as green market size, price parity, and top management’s commitment. It confirms accurate selection even with difference of opinion among stakeholders.

Abstract: Lung cancer is one of the most deadly diseases in the world today, the annual number of deaths more than the next three cancers combined. Even with our advancement in medical science, the problem still persists. It can be addressed effectively at earlier stages, but most cases are detected at stages 3 or 4, where it is too late to be addressed properly. The objective of this paper is to design an effective Computer Aided Diagnosis (CAD) system which can segment of the CT scan of the lung, and help radiologists identify and diagnose this issue at an early stage. A novel 3-dimensional CNN is used to segment the nodules present in the CT scan, which will help classify the nodules with better accuracy. Various optimizations have been carried out to ensure that the convergence is quick and fast, while yielding the best possible accuracy. The proposed architecture achieves a Dice coefficient of 0.9615, on the LUNA16 dataset.

Abstract: This paper presents smart and optimal way of allocating power to the utility using global system for mobile communication module-based remote automatic energy meter reading system. The designed device is installed with the energy meter at consumer premises. A smart communication is established between service provider and consumer using GSM module which is capable of calculating the energy consumed at different tariff and time. An artificial neural network using back-propagation approach is employed to obtain optimal allocation of service provider to meet the objective function. The novel idea of smart energy metering not only reduces the cost of energy consumption but also helps in proper repayments, optimal usage of power based on time of day tariff, and theft control with higher reliability and greater flexibility. A smart real-time prototype of the automatic energy reading system was built to demonstrate the effectiveness and efficiency of automatic meter reading, billing, and notification through the use of global system for mobile communication network.

Abstract: Image assessment measures are commonly employed in different domains to extract the helpful information to take essential decisions. This paper implements a soft-computing approach to examine the Benchmark Crop-Weed (BCW) images of Computer Vision Problems in Plant Phenotyping (CVPPP2014) challenge database. The proposed work executes a hybrid procedure based on Spider Monkey Optimization (SMO) algorithm and Kapur’s multi-thresholding and the Watershed Segmentation (WS) based extraction. After extracting the Crop-Weed regions of BCW pictures, the superiority of the proposed tool is then assessed by implementing a relative study among extracted segment and its related ground-truth. Additionally, the prominence of SMO is validated against the Bat-Algorithm (BA) and Firefly-Algorithm (FA). The outcome of this study authenticates that SMO-based technique is competent in examining the BCW pictures with significant accuracy and precision.

Abstract: California bearing ratio (CBR) is one of the soul parameters for the pavement designs. CBR value can assess the stiffness and strength of the subgrade. Moreover, it was influenced by various properties such as index properties and compaction characteristics of the soils. The paper aims the viability of the swarm assisted particle optimization (PSO) for estimation or prediction of subgrade CBR. CBR estimation model equations by using PSO have been developed by considering index properties and compaction characteristics. The results show that the developed model equations are satisfactory agreement with the test data.

Abstract: Rivers are playing an important role in human life and wildlife but due to pollution the quality of river water is extremely deteriorated. The assessment of water quality is a very indeterminate task and associates a lot of uncertainty and subjectivity in the decision making. To cope with this situation, computational intelligence techniques are found competent to develop models for water quality assessment. One of the computational intelligence techniques, fuzzy logic is used to implement such models. In this manuscript, a fuzzy knowledge-based system is developed to classify the water quality of river Ganga in three groups. The open-access software ‘Guaje’ is used to implement the proposed model. The analysis of the results is presented in terms of interpretability and accuracy which are found satisfactory.

Abstract: Image steganography is a technique of embedding secret information inside a cover image and transmits through a public channel without revealing the presence of a message. So that no one except the intended recipient can recognize secret message within the carrier image. Image steganography based on pixel value differencing (PVD) is one of the most important steganographic methods in spatial domain. To embed secret information, PVD method utilizes the difference value of each pixel pair on cover image. In this paper, we study the basic concepts of image steganography based on PVD method and latest research developments in image steganography using PVD. We also presented a detailed analysis of these techniques on the basis of some performance parameters such as payload capacity, imperceptibility, and robustness.

Abstract: The dynamically changing deregulated electricity market involves different entities and the aim of each entity is to achieve maximum profit while performing electricity price and power bidding. The agent-based modeling of electricity systems was used to model the market entities under whole sale electricity market operation. This paper discusses about the strategic learning ability of generators in an IEEE 30 bus system using Variant Roth-Erev learning algorithm. It also analyzes the variation in the generator commitments through the implemented learning algorithm during the present day schedule and helps the generator to perform smart bidding in the next electricity market operation. The results presented show that the smart generators are able to bid strategically in the electricity market and which will reflect in its net earnings in a market scheduled on a day-ahead basis.

Abstract: Detection of animals from a cluttered scene is not a trivial task. So far, convolutional neural network (CNN) architectures have served this purpose. We introduce stacked convolutional autoencoders (SCAE) for this purpose. It is an unsupervised stratified feature extractor that could be used for high-dimensional input images. We also introduce a hybrid feature extraction technique based on Fisher Vectors (FV) and stacked autoencoders (SAE). SCAE learns significant features utilizing plain stochastic gradient descent and finds a good initialization for CNNs so as to eliminate the various unique local minima of exceptionally non-convex target functions emerging in virtually all deep learning problems. We have proposed a parallel pipeline for both detecting animals in both visible and infrared images. The framework model has achieved 97% accuracy.

Abstract: Semi-linear dynamical systems draw attention in many useful real world problems like population model, epidemic model, etc., they also occur in various applications involving parabolic equations. Now, when the modelling of such applications has inbuilt possibilistic uncertainty, it can be efficiently realized using fuzzy numbers. In this paper, we modify the existing Hukuhara derivative and give the pertaining results for it. We also redefine the Fuzzy Laplace Transform (FLT) and use it to solve such fully fuzzy semi-linear dynamical system.

Abstract: The present investigation highlights an experimental study and optimization of machining outcomes characteristics (such as MRR and Ra) during WEDM process of Inconel 625. The present work examined the effects of wire electrode materials, such as Zn-coated brass electrode (ZCBE) and uncoated brass electrode (UBE) on work material during WEDM process. Based on L16 orthogonal array, the experiment was performed in consideration with four process factor: spark-on time (Son), flushing pressure (Pf), wire-tension (Tw), and discharge current (Dc), within selected experimental domain. The additional objective of present investigation is to develop a multi-response optimization tool for selection of satisfactory process parameter setting during WEDM of Inconel 625. Nonlinear regression model was applied to formulate statistical models for multi-objective optimization using, fuzzy inference system (FIS) combination with TLBO for fulfill this objective. Finally, the satisfactory process parameter obtained by TLBO was compared with the genetic algorithm (GA) individually and found out that, the TLBO algorithm was found to be simpler, effective, and time-saving approach while solving multi-objective problems.

Abstract: Visual Cryptography (VC) is a type of image secret sharing scheme which decrypts an original secret image with Human Visual System (HVS). In this, the original image can be alienated into n shadows or shares and allocated to n participants; stacking any k shares reveals the secret image which ensures the security measures. In this paper, we examined the recent research advances in black and white VCSs. We reviewed the existing techniques and a comparative study of VC for binary images is presented. The study is performed with respect to different parameters and draws the current barriers related to the visual cryptography schemes.

Abstract: Malaria being prevalent disease in urban areas, demands its accurate and fast diagnosis. Due to malaria infection in human being, the erythrocyte features got distorted. To diagnose these, various techniques have been developed, i.e., machine learning-based system, rapid diagnostic test, quantitative buffy coat, etc. In machine learning, the system performance depends on the feature set and classifier model. In this paper, the analysis of the importance of the feature set on malaria-infected erythrocyte classification has been performed. Further, a classifier model based on ANN-GA has been developed to classify the erythrocyte. The process consists of illumination correction, erythrocyte segmentation, feature extraction with or without feature selection techniques, and classification. Erythrocytes segmentation is done using image binarization with marker-controlled watershed segmentation. The six feature sets (morphological feature, texture and intensity feature) have been evaluated using various classifiers such as support vector machine (SVM), k-nearest neighbor (k-NN), and Naive Bayes to choose the better feature set. From the experimental results, it has been observed that the feature set \(\textit{f}_6\) (combination of morphological, texture and intensity feature ranked with ANOVA) outperforms other feature sets. Further, erythrocyte classification has been performed using ANN-GA with \(\textit{f}_6\) feature set. It may also conclude that the various features such as morphological feature, texture and intensity feature are equally important to detect the malaria-infected erythrocyte.

Abstract: The development of the Internet provides opportunities for new types of communications between virtual assistant and human. The technology which is mainly used in the communications is chatbot. A chatbot is a simulated computer program that enabled human conversation by the Internet. The virtual assistant is currently used for a variety of purposes. The chatbot database flow is the important activity for the development of software for the virtual assistant. The process of chatbot testing is based on the well-formalized test cases. The test cases are based on the chatbot trace in the database. Trace path identification during the development of the chatbot software is the challenging process. This paper presents the methodology to identify the test cases for virtual assistant using chatbot database flow-oriented specification. Chatbot database flow is one of the few specification languages supporting for formal description into an applied specification. The database specification divided into several of specification trace using the proposed algorithm. Finally, the chatbot intent trace has provided the path for software test case generation. The experiments show trace path-based test cases that have yielded the effective coverage criteria in the chatbot software development.

Abstract: A text classification system’s learning is substantially dependent on the input features and their process of extraction and selection. The solitary drive encouraging feature selection practice is to lessen the dimensionality of the problem at hand; thus, facilitating the process of classification. Among several problem areas, text categorization is one area where feature selection plays a vital role. It is well-known that text categorization suffers from the curse of dimensionality. This results in the creation of feature space which may have redundant or irrelevant features leading to the creation of a poor classifier. Therefore, to build an intelligent classifier feature, selection is an important process. This paper has a fourfold objective: Firstly, it aims to create a word to vector space using a widely used score. Secondly, it intends to optimize text feature space using a nature-inspired algorithm. Thirdly, it aims at comparing classification performance of three prominently used classifiers, SVM, Naive Bayes, and k-Nearest Neighbors in the area of text classification. Lastly, it targets to compare metrics. Besides accuracy, to understand the consequence of optimizing feature space using nature-inspired algorithm. Standard text classification dataset, Reuters-21578, was used, and the classification accuracies reached 95.07%, 92.23%, 87.37% for SVM, Naive Bayes, and k-Nearest Neighbors, respectively. Besides accuracy, precision, recall, and F-measure were the performance metrics. Considering the encouraging results achieved using the ABC algorithm, this method seems promising for other applications of text classification.

Abstract: Internet of Things (IoT) has been changing the way of operations for multiple segments like Industries, Health Care, and Manufacturing. It also holds a chance to change how Educational Institutions operates and enhance student learning experience. It has enormous opportunities for Educational Segment which will enhance the learning experiences for students, teachers and other stakeholders. The development of IOT Systems, devices, applications, and services are already in the consideration and process by the students and researchers. Therefore, this paper presents a framework to capture validated information of individual Higher Educational Institutions (HEI) through IOT devices to avail the assessment based platform to evaluate and enhance the educational experience. It also describes the processes to automate the survey of Educational Institutions and provide analytical report using IOT components and Machine Learning. To ease the understanding of different methods we provide a prototype with its practical implementations using common processes in a friendly manner.

Abstract: Distributed generation (DG) allocation is the most promising source for reducing network loss and enhancing bus voltage stability in a distribution system. Because of the vast availability and nonpolluting character of renewable energy resource, it is gaining more attention nowadays. The most widely used renewable-based DG (RDG) is wind turbine (WT) and solar photovoltaic (SPV). Power generation patterns of the WT and SPV modules are random and nonlinear because the power output of WT and SPV modules are dependent on wind speed and solar irradiation. These require a probabilistic model to represent the actual power generation. The present paper reflects the potency of WT and SPV modules for reducing system losses and enhancing voltage stability. A new hybrid gray wolf optimizer (HGWO) is proposed to solve the DG allocation problem. The proposed optimization method is tested on IEEE 12- and 15-bus radial distribution system (RDS) and it is found that the proposed HGWO has more potency in terms of loss reduction and voltage stability enhancement compared to the existing techniques.

Abstract: This article demonstrates the hybridization of crow search algorithm with pattern search (HCA-PS) algorithm in automatic generation control of an unequal three-area thermal system. A new secondary controller named as fractional order proportional-integral cascaded with proportional-derivative with filter (FOPI-PDN) is proposed for the system. The controller gains and other parameters are simultaneously optimized by using HCA-PS algorithm. Comparison of system dynamics for FOPI-PDN controller is compared with that of PID, FOPI, FOPID controller reveals better performance of FOPI-PDN controller. It is also seen that when HVDC tie-line is connected in parallel to the AC tie-line, the system dynamic is found better than that with AC tie-line alone. Sensitivity analysis explores that the HCA-PS optimized with FOPI-PDN controller parameters at nominal conditions is found to be robust and tough enough against discrepancy in system inertia and loading conditions. The superiority of HCA-PS algorithm is also observed in terms of dynamic responses and convergence characteristics when compared to that obtained with firefly, cuckoo search and crow algorithms in the system comprising of AC-HVDC tie-lines with FOPI-PDN controller.