Proceedings of Fourth International Conference on Soft Computing for Problem Solving

The 0/1 knapsack problem is attempted to solve using various soft computing methods till date. This paper proposes hybrid TLBO-GA algorithm which is hybrid of teaching learning-based optimization (TLBO) algorithm with genetic algorithm (GA). The 0/1 knapsack problem is a combinatorial optimization problem. The 0/1 knapsack problem aims to maximize the benefit of objects in a knapsack without exceeding its capacity as a constraint. In the literature, it is found that TLBO works for real-coded or real-valued problems. Hybrid TLBO-GA combines evolutionary process of TLBO and binary chromosome representation of GA for solving the knapsack problem (KP). Hybrid TLBO-GA combines advantages of both TLBO and GA. Results are taken on random as well as standard date sets using hybrid TLBO-GA for 0/1 knapsack problem. Hybrid TLBO-GA results are compared with the results obtained using simple genetic algorithm (SGA) on the same data sets. The results obtained using hybrid TLBO-GA are found satisfactory.

WSNs have limited computation power, battery life, and memory resources. In this paper, an approach is introduced to selection cluster head by using swarm intelligence. This proposed approach is based on LEACH clustering algorithm. Modified version of ant colony optimization by using residual energy as a parameter is employed over LEACH algorithm for effective cluster head selection. This approach reduces the amount of energy consumption. The proposed technique work in three stages: Cluster members transmit their data directly to their cluster heads, cluster heads transmit their data to leader, and leader transmits data to the base station. The result shows that LEACH-MA algorithm improves the average energy consumption effectively.

Artificial intelligence (AI) methods exhibit surprising ability to capture nonlinear relationship and interaction effect with great success. Due to complicacy during the welding and lots of interferential factors, especially short-time property of the spot welding process, of late, AI methods are used more frequently by the researchers to estimate output responses of the process. The present study is aimed at investigating failure load of spot-welded B.S. 1050 aluminum sheets using two most commonly used AI methods such as adaptive neuro-fuzzy inference system (ANFIS) and support vector machine (SVM). Data generated from an experimental study is fed into the paradigm of ANFIS and SVR for the formulation of mathematical model between input and output process parameters. Based on the input data, AI models estimate the failure load of welded joint and results are compared in terms of percentage of relative error. The details of experimentation, model development, and comparisons of modeling methods are summarized in this paper which will guide welders about the proper setting of the process parameters for a strong weld joint.

This paper aims to combine fuzzy and technique for order preference by simulation of ideal solution (TOPSIS) to solve the multi-response parameters optimization problem in green manufacturing. From the viewpoint of health and environment, tap water is used as working fluid, since it does not release the harmful gases. This work considers discharge current, pulse width/pulse interval ratio, gap voltage, and lifting height are the input parameters and output parameters have been identified as material removal rate (MRR), electrode wear ratio (EWR), and surface roughness (SR). In this paper, initially, an experiment was performed using Taguchi experimental technique. Thereafter, fuzzy-TOPSIS is used to convert multi-response parameters into a single response parameter. Finally, the ranking of the parameter decides the best experimental setup and optimized the input-process parameters. In this work, weighting factors for the output parameters are determined using triangular fuzzy number which influences correlation coefficient values for finding the finest experimental setup. Additionally, an attempt has been made to compare the proposed methodology with the gray relational analysis (GRA). The numerical result shows that the optimum process parameters are A

1

(4.5 A), B

1

(30:70 μs), C

3

(30 V), and D

4

(6 mm) and using tap water machining Ti-6Al-4V material can produce high MRR, decrease the machining cost, and have no harmful to the operators and environment.

Sustainability of environment requires more concentration on reverse logistics. Keeping this in mind, in this article, we have developed a two-warehouse reverse logistic model with finite rate of production and remanufacturing. A forward and reverse supply chain is considered for deteriorating items. During different cycles, remanufactured and fresh products are stored in owned warehouse (OW) which has limited storage space. For the excess amount, supplier hire rented warehouse (RW) at a high cost of holding than the holding cost of OW. Main objective of this study is to find optimal value of total relevant cost. Numerical illustration and sensitivity analysis is given at the end of this paper.

Logarithmic transformation and histogram equalization (HE) are well-known image enhancement techniques in spatial domain. These techniques are very popular for contrast enhancement because the methods are simple and effective. The basic idea of HE is to remap the gray levels of an image. However, HE tends to introduce some annoying artifacts and unnatural enhancement. Here, we propose an integration of the two techniques at pixel level, doing first the HE and then using logarithmic transformation for mapping purpose. Both the

global HE

and

local HE

are preformed on the input image pixels. The histogram equalization has been performed in the MATLAB environment. We have experimented the proposed techniques over a number of sample images and found to produce much better results compared to image enhancement using the inbuilt MATLAB function

histeq.

The use of genetic algorithm for supply chain management with its ability to evolve solutions, handle uncertainty, and perform optimization remains to be a leading field of study. The growing body of publications over the last two decades means that it can be difficult to keep track of what has been done previously, what has worked, and what really needs to be addressed. Hence, this paper presents a review of existing research activities inspired by the genetic algorithm application in supply chain management (SCM) aimed at presenting key research themes, trends, and directions of future research.

The importance of data mining techniques for market segmentation is becoming indispensable in the field of marketing research. This is the first identified academic literature review of the available data mining techniques related to market segmentation. This research paper provides surveys of the available literature on data mining techniques in market segmentation. A categorization has been provided based on the available data mining techniques used in market segmentation. Eight online journal databases were used for searching, and finally, 103 articles were selected and categorized into 13 groups based on data mining techniques. The utility of data mining techniques and suggestions are also discussed. The findings of this study show that neural networks is the most used method, and kernel-based method is the most promising data mining techniques. Our research work provides a comprehensive understanding of past, present as well as future research trend on data mining techniques in market segmentation. We hope this paper provides reasonable insight and clear understating to both industry as well as academic researchers.

In this paper, a novel hybrid algorithm incorporating the cuckoo search optimization (CSO) technique and an Assorted Individual Enhancement Scheme is proposed to solve the ob shop scheduling problem (JSSP) with an objective to minimize the makespan. The proposed hybrid algorithm is applied to 20 job shop problem instances available in OR-Library and the results are compared with simple CSO. Analysis suggests that the hybrid approach is more effective and robust toward JSSP than simple CSO technique.

In this article, we present an artificial neural network technique to solve some of the two-point nonlinear elliptic boundary value problems arising in science and engineering. A trial solution of the differential equation is written in terms of a feed forward neural network with adjustable parameters weights or biases, and error function is prepared to use in the back-propagation algorithm to update the network parameters with momentum term. Comparison of the results obtained by the present method is done with analytical solution and other existing numerical methods which show the efficiency of Neural network method with high accuracy, fast convergence, and low use of memory for solving nonlinear elliptic boundary value problems. The main advantage of the proposed approach is that once the network is trained, it allows evaluation of the solution at any desired number of points instantaneously with spending negligible computing time.

A method is proposed to discriminate rough and smooth scene images of various classes such as forest and coast, highway and inside city. The concept of fuzzy uncertainty texture spectrum (FUTS) is used. The fuzzy uncertainty parameter measures the uncertainty of the uniform surface in an image. Distribution of membership in a fuzzy image is called FUTS. The roughness or smoothness of a scene structure can be well explained by local texture information in a pixel and its neighborhood. Our proposed method uses the concept of FUTS to classify rough and smooth scene images. Probabilistic neural network (PNN) is used in the final stage to classify 150 images. This technique gives good result with less computational complexity.

In this paper, Parikh matrices over ternary alphabet are investigated. Algorithm is developed to display Parikh matrices of words over ternary alphabet. A set of equations for finding ternary words from the respective Parikh matrix is discussed. A theorem regarding the relations of the entries of the 4 × 4 Parikh matrices is proved. Some other results in this regard are also discussed. Significance of graphical representation of binary amiable words is given. Extension of this notion for ternary amiable words is introduced.

Scour is caused by the erosive action of flowing water. Although, different researchers have proposed various empirical models to predict the equilibrium local scour depth around bridge abutment, these are suitable to a particular abutment condition. In this study, an integrated model that combines genetic algorithms (GA) and multilayer perceptron (MLP) network, a class of artificial neural network (ANN), is developed to estimate the scour depth around vertical bridge abutment. The equilibrium scour depth was modeled as a function of four affecting parameters of scour, abutment length, median grain size, approaching flow depth, and average approach flow velocity, and these parameters are considered as input parameter to the MLP model. The efficiency of the developed models is compared with the empirical equations over a dataset collected from literature. The MLP is found to outperform the empirical equations for the dataset considered in the present study. The performance of the best case MLP is further improved by applying GA for weight initialization. The results indicate that the GA-based MLP is more effective in terms of accuracy of predicted results and is a promising approach compared to MLP as well as the previous empirical approaches in predicting the scour depth at bridge abutments.

This paper presents a new rough fuzzy classification approach for class imbalanced data. Here, interval type-2 fuzzy granulation of input features is formulated, various combinations of rough set extension-based methods are used to perform class imbalance learning, and

K

-nearest neighbor (KNN) classifier is used for data classification. The experimental results on the UCI data sets are reported to demonstrate the effectiveness of the proposed rough fuzzy classification model. Performance evaluation measures viz

F

-measure and geometric mean (

G

-mean) are used for analyzing classifier’s performance and suitability of the developed model for class imbalance learning.

Face recognition has always been an area of interest for the researchers because it seems so fascinating in itself. As the time has passed, the evolution of human brain and social intelligence has enhanced the analysing power of a human brain for the displaying of face to infer the emotion accordingly. Face of a human being is considered to be the most common channel to read and infer the emotion in the case of mind reading. Theory of mind (ToM) is the inbuilt technique provided to every human being “as reported by Premack and Woodruff (Behav Brain Sci 1:515–523, 1991)”. According to ToM, a person has the ability to read someone else’ mind. Mind reading has a great amount of future advancement if worked upon with the different equipment and devices in the high-tech laboratories. The researchers have worked in this direction and have developed softwares. The aim of such software is to apply concept of security in face recognition but still there is a scope for improvement. This paper presents a comparative study between algorithms used for detecting the face as the key component in mind reading. Further, a model is proposed to enhance the security in face recognition. The proposed model also suggests interface that would help users to identify the people watching their pictures posted online.

To handle bulk of power, the AC power transmission is not economical over long distance. High-voltage direct current (HVDC) transmission system is selected as the alternative not only in economic aspects but also in stability point of view. But the operation and control of HVDC links pose a challenge for the designers to choose the proper control strategy under various operating conditions. Traditionally, PI controllers are used for the rectifier current control of the HVDC system, but due to fixed proportional (P) and integral (I) gains, these controllers can perform well only over a limited operating range. However, in controlling a nonlinear plant such as the firing angle of the rectifier side in HVDC system, the model controls such as fuzzy logic controllers show better performance to the dynamic disturbances than traditional PI controllers. The CIGRÉ model as one of the conventional methods has been studied and improves the stability HVDC system.

Routing the data in wireless sensor network (WSN) is a main issue in the guaranteed data transmission. Contextually, many algorithms consider the power and resource limitation, energy efficiency, scalability, attributes and location based with the assumption of homogeneous and heterogeneous distribution of nodes for data transferring. Therefore data dissemination is required to provide the effective data transmission from source node to sink node. This paper introduces the cluster head selection scheme used in SPIN to efficiently disseminate observations gathered by individual sensor nodes in the network. The paper proposes the cluster head selection scheme used in SPIN protocol and evaluated the performance of proposed protocol to prolong the time interval from the start of network operation till the death of the first node which is important reliable issues in wireless sensor networks for many feedback applications. The scheme is implemented and simulated with LEACH in NS2.34. Simulation shows proposed protocol exhibits significant performance gains over the LEACH for lifetime of network and guaranteed data transmission.

This paper presents comparison of the optimum performance characteristics of four different bearing configurations. An attempt has been made to find out the effect of four different bearing configurations of hydrodynamic journal bearing by changing groove locations. Various groove angles that have been considered are 10°, 20°, and 30°. The Reynolds equation is solved numerically in a finite difference grid satisfying the appropriate boundary conditions. Four optimum performance parameters considered viz non-dimensional load carrying capacity, flow coefficient, friction variable, and mass parameter. Optimum configuration of bearings ensures best flow, load and stability, and least friction of the bearings. Genetic algorithm (GA) for multi-objective function has been used for optimum performance parameter comparison of the bearings. Flow coefficient value is found higher for elliptical bearing, and optimum value of non-dimensional load carrying capacity mass parameter found to be the highest for four-lobe bearing.

Maintaining population diversity is a challenge for the success of genetic algorithm. A numerous approaches have been proposed by researchers for adding diversity to the population. Dual-population genetic algorithm (DPGA) is one of them which is an effective optimization algorithm and provides diversity to the main population. Problems in GA such as premature convergence and population diversity is well addressed by DPGA. The aim of writing this review paper is to study how DPGA has been evolved. DPGA is inherently parallelizable, and hence, it can be port to parallel programming architecture for large-scale or large-dimension problems.

In this paper, the speed of a three-phase induction motor is controlled by using modified cascaded five-level inverter and we compared the total harmonic distortion of the modified cascaded five-level inverter with the conventional three-level inverter. To reduce the total harmonic distortion, multicarrier PWM is used. An open-loop speed control has been achieved by using V/f method. The simulation result gives that the modified cascaded five-level inverter effectively controls the motor speed and enhances the drive performance through reduction in total harmonic distortion (THD).

Magic square construction is a complex and hard permutation problem of recreational combinatorics with a long history. The complexity level enhances rapidly when the number of magic squares increases with the order of magic square. This paper proposes two hybrid metaheuristic algorithms, so-called cellular harmony search (CHS) and smallest-small-world cellular harmony search (SSWCHS) for solving magic square problems. The inspiration of the CHS is based on the cellular automata (CA) formation, while the SSWCHS is inspired by the structure of smallest-small-world network (SSWN) and CA using the concept of HS. Numerical optimization results obtained are compared with different optimizers in terms of statistical results and number of found feasible solutions. Computational results show that the proposed hybrid optimizers are computationally effective and highly efficient for tackling magic square problems.

Breast cancer is undoubtedly a dreadful and life-threatening disease. It is fairly common in women and also the second deadliest cancer in the world. It is arguably the most frightening type of cancer because of its well-publicized nature and potential for lethality. If identified and properly treated in its early stage, the chance of cure increases. Different imaging techniques are there which plays a vital role in the detection of breast cancer. In recent days, mammography and thermography are the two main techniques accepted in the medical field to detect breast cancer followed by other screening methods. A literature survey is presented in this paper based on these two techniques followed by the analysis of their affordability, reliability, and outcomes.

Cervical cancer is the most common cancer among the women. Pap smear screening is the most effective test for detecting the cervical precancerous. But this process requires a long time to complete and also may be an erroneous procedure. In this paper, an automated cervical cancer detection method is presented. This method introduces adaptive median filter to remove impulse noises from the Pap smear images and then uses bi-group enhancer to discriminate the nuclei pixels from other object pixels. Then, segmentation methodology is presented to separate the nucleus regions from the cervical smear images. Two clustering-based classifiers, minimum distance and K-nearest neighbor classifiers, have been used in the classification phase for verifying the performance. The technique was evaluated using 158 Pap smear images from DTU/HERLEV Pap smear benchmark database. The accuracy of the detection method is 92.37 and 98.31 % for minimum distance and K-nearest neighbor classifiers, respectively.

Identifying moving objects plays an important role in video-based applications. In this paper, a background subtraction approach for object detection technique is proposed, which is an improvised version of an existing background subtraction algorithm called visual background extractor (ViBe). Here, the performance of the existing technique has been modified by a median filter. This technique is implemented on different existing databases and also on newly created Society of Applied Microwave Electronics Engineering and Research-Tripura University (SAMEER-TU) dataset. The detection accuracy of the technique is also measured, and a comparison is also carried out between existing and proposed technique, and results are reported in experimental results, in terms of detection accuracy for color video sequence.

Internet of Things (IoT) possesses a dynamic network where the network nodes (mobile devices) are added and removed constantly and randomly; hence, the traffic distribution in the network is quite variable and irregular. The basic but very important part in any network is route searching. We have many conventional route searching algorithms such as link-state and distance vector algorithms, but they are restricted to the static point-to-point network topology. In this paper, we proposed a hypothetical but feasible model that uses the ant colony optimization (ACO) algorithm for route searching. ACO is dynamic in nature and has a positive feedback mechanism that conforms to the route searching. In addition, we have embedded the concept of deterministic finite automata (DFA) minimization to minimize the number of iterations done by ACO in finding the optimal path from source to sink. Analysis and proof show that ACO gives the shortest optimal path from the source to the destination node, and DFA minimization reduces the broadcasting storm effectively.

The present work does a simultaneous maximization of air permeability and ultraviolet radiation protection of single jersey cotton knitted fabrics. As these two objectives are conflicting in nature, i.e., not a single combination of knitting parameters does exist which produce concurrent maximum air permeability as well as maximum ultraviolet radiation protection. Therefore, it has several optimal solutions from which a trade-off is needed depending upon the requirement of user. In this work, the optimal solutions are obtained with an elitist multi-objective evolutionary algorithm based on Non-dominated Sorting Genetic Algorithm II (NSGA-II). These optimum solutions may lead to the efficient exploitation of knitting parameters to produce fabrics with optimum protection from ultraviolet radiation and comfort.

In this paper, we have tried to manufacture cotton yarns with requisite strength by choice of suitable raw material and process parameters. In an attempt to achieve a yarn having optimal strength, a constrained optimization problem is formulated with the relation between raw material and yarn properties. Frydrych’s theoretical model of yarn strength is used as objective function of the optimization problem. The simulated annealing (SA) method has been used to solve the optimization problem by searching the best combination of raw material and process parameters that can translate into reality a yarn with the desired strength. The results show that SA is capable of identifying the set of parameters that gives optimum yarn strength.

Poor visibility in the presence of fog is a major problem for many applications of computer vision. Still image and video systems are typically of limited use in poor visibility condition as the degraded images/frames lack visual vividness and offer low visibility of the scene contents. This paper investigates the defogging effects on images and frames by using a fast defogging method on our own newly developed database, namely Society of Applied Microwave Electronics Engineering and Research-Tripura University (SAMEER-TU) database which consists of 5,390 color images and 10 videos captured in foggy as well as in clear condition. The first step of the method ensures contrast enhancement yielding better global visibility, but the images/frames containing very dense fog still suffer from low visibility. In that case, Luminance and chromatic weight map have been used. Finally for verifying the robustness of the method, qualitative assessment evaluation in respect of peak-signal-to-noise ratio (PSNR) and root-mean-square error (RMSE) is introduced as a contributory step in this paper.

The present study focuses on the optimum design based on primary loop of a liquid metal fast breeder reactor (LMFBR). Inside the core annular fuel, rods are used for better heat transfer to the liquid sodium coolant. Value of the outlet temperature of the coolant from the core, surface temperature of the fuel pin, and pumping power at different volume flow rates are studied, and genetic algorithm (GA) is used to minimize pumping power required to maintain flow in the primary loop for an optimal design parameters.

Image segmentation is the process of partitioning an image into multiple segments and it is one of the most important steps for automatic cell analysis, because the result of final classification depends mainly on the correct image segmentation. In this paper, some general segmentation methods have reviewed which is mainly used in biomedical image processing especially in erythrocyte image. The main goal of biomedical image segmentation was to extract the foreground which contains the useful information from complicated background for the medical diagnosis.

India is a developing country. India’s population is over 1.27 billion people (2014) which are approximated as one-sixth of the world’s population. Such population growth may leave behind the world’s most heavily populated nation China by 2025 (US Census Bureau 2011). Solid waste management is an important environmental issue of all developed and developing countries. The growth of solid waste is basically due to population explosion, urbanization, and mismanagement of municipal corporation. Limitations of Indian Municipality Corporation are waste gathering inefficiency, lack of financial funds, poor planning, and lack of technical knowledge on changing complication of waste materials. In this paper, few important optimization models/techniques proposed by different researchers are studied that may be beneficial for ongoing project work in Municipal solid waste management (MSWM) in different metropolitan cities at various states in India. Much work has not been done in this direction. On the basis of extensive study of literature, it is suggested that the Indian municipal corporation must adopt 4R principles involving reduce, reuse, recycle, and recover to minimize solid waste.

Self-organizing migrating algorithm (SOMA) is relatively a new population-based stochastic search technique for solving nonlinear global optimization problems. There has been done very less work on hybridization of SOMA with other methodologies in order to improve its performance. This paper presents hybridization of self-organizing migrating algorithm with quadratic approximation or interpolation (SOMAQI) and non-uniform mutation. This hybridization (M-SOMAQI) uses the quadratic interpolation (QI) and non-uniform mutation for creating a new solution vector in the search space. To validate the efficiency of this algorithm, it is tested on 15 benchmark test problems taken from the literature, and the obtained results are compared with SOMA and the SOMAQI. The numerical and graphical results conclude that the presented algorithm shows better performance in terms of population size, efficiency, reliability, and accuracy.

Single and double jersey knitted fabrics are different in many aspects, but it is difficult to identify them in open eye, and in textile industry, it is essential to identify them automatically. So far, no hands-on state-of-the-art technology has been adopted for identification of single and double jersey fabrics. This novel work endeavors to recognize these two kind of knitted fabrics by means of proximal support vector machine (PSVM) using the features extracted from gray level images of both fabrics. A

k

-fold cross-validation technique has been applied to assess the accuracy. The robustness, speed of execution, proven accuracy coupled with simplicity in algorithm holds the PSVM as a foremost classifier to recognize single and double jersey fabrics.

Induction motors (IMs) have many applications in the industries, because of the low maintenance and robustness. The speed control of IM is more important to achieve maximum torque and efficiency. The rapid development of power electronic devices and converter technologies in the past few decades, however, has made possible efficient speed control by varying the supply frequency and voltage, giving rise to various forms of adjustable-speed IM drives. In about the same period, there were also advances in control methods and artificial intelligent (AI) techniques, including expert system, fuzzy logic, neural networks, and genetic algorithm. Researchers soon realized that the performance of IM drives can be enhanced by adopting artificial intelligent-based methods. This paper presents dynamic modeling and simulation of IM using AI controller. The integrated environment allows users to compare simulation results between classical and AI controllers. The fuzzy logic controller and artificial neural network controllers (NNCs) are also introduced to the system for keeping the motor speed to be constant when the load varies. The performance of fuzzy logic and artificial neural network (ANN)-based controllers is compared with that of the conventional proportional integral controller. The performance of the IM drive has been analyzed for constant, variable loads, and induction generator mode.

This paper concerns design and implementation of PID and fractional-order PID (FOPID) controllers to control position of an electromagnetically suspended ferromagnetic ball in a magnetic levitation (Maglev) system in real time. The Maglev system, manufactured by Feedback Instruments (Model No 33-210) is used as a platform to test the performance of proposed controllers. Parameters of PID and FOPID controllers are tuned by firefly algorithm (FA). FA is a metaheuristic algorithm based on movement of fireflies toward more attractive and brighter ones. PID and FOPID controllers are implemented in MATLAB and SIMULINK environment inside PC using fractional-order modeling and control toolbox. Controller in the SIMULINK environment inside PC is connected to the Maglev system through Advantech card. Effectiveness of proposed controllers is tested by checking the ability of the suspended ball to track a reference signal. Step change, sine wave, and square wave are used as reference signals. Real-time results have revealed satisfactory transient and steady-state responses over the contemporary existing controllers. FOPID controller showed better results compared to PID.

There is a remarkable performance of differential evolution (DE) algorithm on continuous space problem. Mutation plays a very vital role in success of DE but in traditional DE all the vectors are selected in random manner. Sometimes, it gives a random exploration in search space. Here, the distance-based analysis for mutation vector selection is carried out and distance-based criteria for base vector (reference point) selection have proposed. Experimentation is conducted on eight standard uni-model and multi-model functions. Later, the results have compared with standard DE and other variant of DE. Experiments show that the proposed strategy has a very steady and stable exploration of search space.

The inherent drawback of the popular evolutionary algorithm as such genetic algorithm (GA) and also bio-inspired algorithm bacterial foraging optimization (BFO) lies in the fact that they very often suffer from the problem of being trapped into the local optimum. In recent past, various popular hybridized techniques of GA and BFO came out through different thought processes of the researches and have been implemented in the algorithm. Inspired by those ideas, in this paper, a novel approach has been opted for the hybridization of GA with BFO by incorporating chemotactic step as a local search operator at the end of the entire GA cycle; thus, the algorithm is named as chemo-inspired genetic algorithm (CGA) and it has also been extended for constrained optimization, and further it is named as CGAC, where “C” stands for being capable of handling constraints. At the outset, experiments are made to validate the superiority of CGAC over another hybrid method, namely LX-PM-C and H-LX-PM-C taking a set of 8 typical benchmark problems of various difficulty labels from the literature. Later, it has been applied to real-life application problem, where economic load dispatch (ELD) problem having 40 generators has been considered with valve point loading effect.

In this paper, a novel hybrid intelligent algorithm, integrating with differential evolution (DE) and particle swarm optimization (PSO), is proposed. Initially, all individual in the population are divided into three groups (in increasing order of function value): inferior group, mid-group, and superior group. DE is employed in the inferior and superior groups, whereas PSO is used in the mid-group. The proposed method uses DE-PSO-DE, then it is denoted by DPD. At present, many mutation strategies of DE are reported. Every mutation strategy has its own pros and cons, so which one of them should be selected is critical for DE. Therefore, over 8 mutation strategies, the best one is investigated for both DEs used in DPD. Moreover, two strategies, namely

elitism

(to retain the best obtained values so far) and

Non

-

redundant search

(to improve the solution quality), have been employed in DPD cycle. Combination of 8 mutation strategies generated 64 different variants of DPD. Top 4 DPDs are investigated through solving a set of constrained benchmark functions. Based on the ‘performance,’ best DPD is reported and further used in solving engineering design problem.

Predicting the risk of extinction of species is an important aspect in conservation biology. Mathematical models describing the density dependent per capita growth rates play a predominating role in quantifying the risk of extinction. We used population time series data from global population dynamics database (GPDD) to predict the threat status of the species using three commonly used growth models, allowing demographic disturbances. The best fitted model from a set of candidate models is used to evaluate the extinction measures. We observed that there are instances where the intrinsic growth rate is negative, which has not been reported earlier. We show that, in such cases, the extinction probability is high, but the species may adopt some strategy that saves them from extinction. The mathematical implications are described in light of ecological concepts.

Multiple layer hierarchical routing protocols require energy management and power output optimization as the individual nodes in these networks tend to be small low-power devices. The superiority of multilayer hierarchical networks containing three layers over a two-layer implementation in terms of stability period, network lifetime, and throughput has already been demonstrated. In this paper, the improvement in these performance parameters brought about by the addition of a fourth layer has been elaborated.

This paper reviews the literature on control of 2-DOF robot manipulator using fuzzy logic control (FLC). Different schemes of FLC laws are considered here. These are PID control, sliding mode control (SMC), and adaptive control. Importance of each control techniques with its advantages and disadvantages is discussed here. It is highlighted that the robustness of the system has improved considerably by using FLC than classical controller. A total of 65 papers were surveyed in this research area, covering contribution on each control technique for the 2-DOF robot manipulator for the time span of 1983–2014.

Optimal selection of material for engineering design from a set of materials is more complex due to availability of huge amount of material. An improper selection of material makes adverse effect on successful engineering design, productivity, quality, customer satisfaction, etc. Since the selection of material involves multiple criteria, therefore, the material selection problem is a multi-criteria decision making problem. A systematic and efficient approach toward the material selection is necessary in order to select the best material. In this paper, an integrated approach comprising of entropy and multiple objective on the basis of simple ratio analysis (MOOSRA) method is used to solve material selection problem. A case study on exhaust manifold has been taken for the analysis. The result shows that the carburized steel is the nest material while cast alloy is the worst material.

In the present era, of globalization and automation, artificial intelligence (AI) has come out as a better implement to solve many problems that require decision making. In this work, it has been tried to amalgamate the research work done related to application of soft computing techniques particularly in the area of mechanical engineering. A lot of researches have being carried out in this domain because it has been a pool of infinite scope for innovative works. This paper gives a quick view of the different applications of AI techniques, such as neural networks, fuzzy logic, neurofuzzy (NF), simulated annealing (SA), genetic algorithm (GA), genetic programming (GP), and data mining (DM), and provides a conscious view in the different paradigms of mechanical engineering where the soft computing techniques are separately used and even in combination with one another to perform certain tasks which if performed by human becomes a dreary task. Several applications of soft computing have been proposed in the literature to solve the problems related with complicated mechanical systems. It is felt that a review of the application in various mechanical areas would help to compare their main features and their relative advantages or limitations to allow choose the most suitable technique for a particular application and also throw light on aspects that needs further attention.

In this paper, a fuzzy goal programming (FGP) approach for solving fuzzy multiobjective quadratic chance-constrained programming (CCP) problem involving exponentially distributed fuzzy random variables (FRVs) is developed. In the proposed methodology, the problem is first converted into interval-valued quadratic programming problem using CCP technique and

$$ \alpha $$

α

-cut of fuzzy numbers. Then, using fuzzy partial order relations, the problem is converted into its equivalent deterministic form. The individual optimal value of each objective is found in isolation to construct the quadratic fuzzy membership goals of each of the objective. The quadratic membership goals are transferred into linear goals by applying piecewise linear approximation technique. A

minsum

goal programming (GP) method is then applied to both the linearized and quadratic model to achieve the highest membership degree of each of the membership goals in the decision-making context. Finally, a comparison is made on the two different approaches with the help of distance function. An illustrative numerical example is provided to demonstrate the applicability of the proposed methodology.

The concept of partial commutation and traces on strings has been recently extended to rectangular picture arrays and languages, motivated by the two-dimensional patterns which appear in studies concerning parallel computing and image analysis. The closure property of Siromoney matrix languages (SML) under partial commutation is already studied. In this paper, we consider partial commutation on a generalization of SML languages and Siromoney array languages (SAG) and establish some interesting results. Also, we examine the partial commutativity applied on some higher classes of 2D picture languages generated by Regional tile rewriting grammars and Prusa grammars. We extend partial commutation concept to hexagonal pictures also.

The destabilizing effect of high gain in voltage regulators persists in power system. The power oscillations of small magnitude and high frequency, which often persisted in power system, present the limitation to the amount of power transmitted within the system. In this paper, a linearized Heffron–Phillips model of a single machine infinite bus (SMIB) is developed using different controllers like fuzzy logic power system stabilizer (FPSS), PID controller, particle swarm optimization (PSO)-based PID controller for analyzing the stability enhancement in power system. For FPSS, speed deviation and acceleration deviation are taken as inputs. Comparison of the effectiveness (steady-state error, ess, overshoot (Mp), and settling time (ts) for a different controller has been done. The performance of the SMIB system using FPSS has been analyzed when comparing with conventional controllers used in SMIB. Similarly the PSO is done using different iterations on conventional PID controller. The results of the simulation show that for low frequency oscillations, FPSS is more effective in damping compared to conventional controllers, and similarly PSO-based PID controller is more effective than a conventional PID controller.

Ordered weighted aggregation operators were introduced in the year 1988 and since then have been used in a wide variety of applications. This paper is an attempt to use a special variant of OWA operator—ordered weighted geometric averaging operator for regression analysis. Regression analysis is used to formulate a suitable business model based on the past performances of the company/organization and is an integral component of data mining techniques. Various regression algorithms have been proposed in the literature. In this paper, a multi-criteria decision-making problem is being formulated using some of these regression algorithms on a real-time industrial web data followed by the analysis of using one the OWA operators on its results.

In this paper, a Niching co-swarm gravitational search algorithm (CoGSA) is designed for solving multi-modal optimization problems. The collective approach of Gravitational Search Algorithm and differential evolution (DE) is used to solve multi-modal optimization problems. A set of twelve multi-modal problems are taken from a benchmark set of CEC 2013. An experimental study has been performed to evaluate the availability of CoGSA over these twelve problems. The performance is measured in an advanced way. It has been observed that CoGSA provides good solution for multi-modal optimization problems.

Sorting is arranging a collection of elements either in ascending or descending order. There are various applications of sorting algorithm in every field of computer science. Already there exist different sorting algorithms with different complexities. In worst case, the best known complexity is

O

(

n

log

n

). We have an algorithm called RAPID SORT and analyzed in detail and also compared with some of the existing algorithm like the Quick Sort, Merge sort, Bubble sort, Insertion sort, and selection sort. This algorithm is much better for closely related datasets. This algorithm is very efficient to sort the elements in reverse order.

In this paper, we address the problem of credit scoring (CS) as a feature selection problem. More specifically, we use wrapper feature selection methods to identify features that contain the most relevant information to distinguish good loan applicants from bad loan applicants. Wrapper feature selection approaches are widely used to select a small subset of relevant features from a dataset. However, wrappers suffer from the fact that they only use a single classifier in the evaluation process and each classifier is of a different nature and will have its own biases. Hence, this paper investigates the effects of using different classifiers for wrapper feature selection. A new ensemble method for feature selection is then proposed and evaluated on four credit datasets, and results illustrate that combining classifiers improves the performance of scoring models.