Skip to main content

Über dieses Buch

This LNCS volume contains the papers presented at the First Swarm, Evolutionary and Memetic Computing Conference (SEMCCO 2010) held during December 16–– 18, 2010 at SRM University, Chennai, in India. SEMCCO 2010 marked the beginning of a prestigious international conference series that aims at bringing together researchers from academia and industry to report and review the latest progress in the cutting-edge research on swarm, evolutionary, and memetic computing, to explore new application areas, to design new bio-inspired algorithms for solving specific hard optimization problems, and finally to create awareness on these domains to a wider audience of practitioners. SEMCCO 2010 received 225 paper submissions from 20 countries across the globe. After a rigorous peer-review process involving 610 reviews in total, 90 fu- length articles were accepted for oral presentation at the conference. This corresponds to an acceptance rate of 40% and is intended for maintaining the high standards of the conference proceedings. The papers included in this LNCS volume cover a wide range of topics in swarm, evolutionary, and memetic computing algorithms and their real-world applications in problems selected from diverse domains of science and engineering.



Self-adaptive Differential Evolution with Modified Multi-Trajectory Search for CEC’2010 Large Scale Optimization

In order to solve large scale continuous optimization problems, Self-adaptive DE (SaDE) is enhanced by incorporating the JADE mutation strategy and hybridized with modified multi-trajectory search (MMTS) algorithm (SaDE-MMTS). The JADE mutation strategy, the “DE/current-to-pbest” which is a variation of the classic “DE/current-to-best”, is used for generating mutant vectors. After the mutation phase, the binomial (uniform) crossover, the exponential crossover as well as no crossover option are used to generate each pair of target and trial vectors. By utilizing the self-adaptation in SaDE, both trial vector generation strategies and their associated control parameter values are gradually self-adapted by learning from their previous experiences in generating promising solutions. Consequently, suitable offspring generation strategy along with associated parameter settings will be determined adaptively to match different phases of the search process. MMTS is applied frequently to refine several diversely distributed solutions at different search stages satisfying both the global and the local search requirement. The initialization of step sizes is also defined by a self-adaption during every MMTS step. The success rates of both SaDE and the MMTS are determined and compared, consequently, future function evaluations for both search algorithms are assigned proportionally to their recent past performance. The proposed SaDE-MMTS is employed to solve the 20 numerical optimization problems for the CEC’2010 Special Session and Competition on Large Scale Global Optimization and competitive results are presented.

Shi-Zheng Zhao, Ponnuthurai Nagaratnam Suganthan, Swagatam Das

Differential Evolution Based Ascent Phase Trajectory Optimization for a Hypersonic Vehicle

In this paper, a new method for the numerical computation of optimal, or nearly optimal, solutions to aerospace trajectory problems is presented. Differential Evolution (DE), a powerful stochastic real-parameter optimization algorithm is used to optimize the ascent phase of a hypersonic vehicle. The vehicle has to undergo large changes in altitude and associated aerodynamic conditions. As a result, its aerodynamic characteristics, as well as its propulsion parameters, undergo drastic changes. Such trajectory optimization problems can be solved by converting it to a non-linear programming (NLP) problem. One of the issues in the NLP method is that it requires a fairly large number of grid points to arrive at an optimal solution. Differential Evolution based algorithm, proposed in this paper, is shown to perform equally well with lesser number of grid points. This is supported by extensive simulation results.

Ritwik Giri, D. Ghose

Dynamic Grouping Crowding Differential Evolution with Ensemble of Parameters for Multi-modal Optimization

In recent years, multi-modal optimization has become an important area of active research. Many algorithms have been developed in literature to tackle multi-modal optimization problems. In this work, a dynamic grouping crowding differential evolution (DGCDE) with ensemble of parameter is proposed. In this algorithm, the population is dynamically regrouped into 3 equal subpopulations every few generations. Each of the subpopulations is assigned a set of parameters. The algorithms is tested on 12 classical benchmark multi-modal optimization problems and compared with the crowding differential evolution (Crowding DE) in literature. As shown in the experimental results, the proposed algorithm outperforms the Crowding DE with all three different parameter settings on the benchmark problems.

Bo Yang Qu, Pushpan Gouthanan, Ponnuthurai Nagaratnam Suganthan

Empirical Study on Migration Topologies and Migration Policies for Island Based Distributed Differential Evolution Variants

In this paper we present an empirical performance analysis of fourteen variants of Differential Evolution (DE) on a set of unconstrained global optimization problems. The island based distributed differential evolution counterparts of the above said 14 variants have been implemented with mesh and ring migration topologies and their superior performance over the serial implementation has been demonstrated. The competitive performance of ring topology based distributed differential evolution variants on the chosen problem has also been demonstrated. Six different migration policies are experimented for ring topology, and their performances are reported.

G. Jeyakumar, C. Shunmuga Velayutham

Differential Evolution Based Fuzzy Clustering

In this work, two new fuzzy clustering (FC) algorithms based on Differential Evolution (DE) are proposed. Five well-known data sets viz. Iris, Wine, Glass, E. Coli and Olive Oil are used to demonstrate the effectiveness of DEFC-1 and DEFC-2. They are compared with Fuzzy C-Means (FCM) algorithm and Threshold Accepting Based Fuzzy Clustering algorithms proposed by Ravi et al., [1]. Xie-Beni index is used to arrive at the ‘optimal’ number of clusters. Based on the numerical experiments, we infer that, in terms of least objective function value, these variants can be used as viable alternatives to FCM algorithm.

V. Ravi, Nupur Aggarwal, Nikunj Chauhan

A Population Adaptive Differential Evolution Strategy to Light Configuration Optimization of Photometric Stereo

Differential Evolution is an optimization technique that has been successfully employed in various applications. In this paper we propose a novel Population Adaptive Differential Evolution strategy to the problem of generating an optimal light configuration for photometric stereo. For ‘n’ lights, any 2


/n of orthogonal light directions minimizes the uncertainty in scaled normal computation. The assumption is that the camera noise is additive and normally distributed. Uncertainty is defined as the expectation of squared distance of scaled normal to the ground truth. This metric is optimized with respect to the illumination angles at constant slant angle. Superiority of the new method is demonstrated by comparing it with sensitivity analysis and classical DE.

B. Sathyabama, V. Divya, S. Raju, V. Abhaikumar

Solving Multi Objective Stochastic Programming Problems Using Differential Evolution

Stochastic (or probabilistic) programming is an optimization technique in which the constraints and/or the objective function of an optimization problem contains random variables. The mathematical models of these problems may follow any particular probability distribution for model coefficients. The objective here is to determine the proper values for model parameters influenced by random events. In this study, Differential Evolution (DE) and its two recent variants LDE1 and LDE2 are presented for solving multi objective linear stochastic programming (MOSLP) problems, having several conflicting objectives. The numerical results obtained by DE and its variants are compared with the available results from where it is observed that the DE and its variants significantly improve the quality of solution of the given considered problem in comparison with the quoted results in the literature.

Radha Thangaraj, Millie Pant, Pascal Bouvry, Ajith Abraham

Multi Sensor Fusion Using Fitness Adaptive Differential Evolution

The rising popularity of multi-source, multi-sensor networks supports real-life applications calls for an efficient and intelligent approach to information fusion. Traditional optimization techniques often fail to meet the demands. The evolutionary approach provides a valuable alternative due to its inherent parallel nature and its ability to deal with difficult problems. We present a new evolutionary approach based on a modified version of Differential Evolution (DE), called Fitness Adaptive Differential Evolution (FiADE). FiADE treats sensors in the network as distributed intelligent agents with various degrees of autonomy. Existing approaches based on intelligent agents cannot completely answer the question of how their agents could coordinate their decisions in a complex environment. The proposed approach is formulated to produce good result for the problems that are high-dimensional, highly nonlinear, and random. The proposed approach gives better result in case of optimal allocation of sensors. The performance of the proposed approach is compared with an evolutionary algorithm coordination generalized particle model (C-GPM).

Ritwik Giri, Arnob Ghosh, Aritra Chowdhury, Swagatam Das

Differential Evolution Algorithm with Ensemble of Parameters and Mutation and Crossover Strategies

Differential Evolution (DE) has attracted much attention recently as an effective approach for solving numerical optimization problems. However, the performance of DE is sensitive to the choice of the mutation and crossover strategies and their associated control parameters. Thus, to obtain optimal performance, time consuming parameter tuning is necessary. Different mutation and crossover strategies with different parameter settings can be appropriate during different stages of the evolution. In this paper, we propose a DE with an ensemble of mutation and crossover strategies and their associated control parameters known as EPSDE. In EPSDE, a pool of distinct mutation and crossover strategies along with a pool of values for each control parameter coexists throughout the evolution process and competes to produce offspring. The performance of EPSDE is evaluated on a set of 25 bound-constrained problems designed for Conference on Evolutionary Computation (CEC) 2005 and is compared with state-of-the-art algorithm.

Rammohan Mallipeddi, Ponnuthurai Nagaratnam Suganthan

Design of Robust Optimal Fixed Structure Controller Using Self Adaptive Differential Evolution

This paper presents a design of robust optimal fixed structure controller for systems with uncertainties and disturbance using Self Adaptive Differential Evolution (SaDE) algorithm. PID controller and second order polynomial structure are considered for fixed structure controller. The design problem is formulated as minimization of maximum value of real part of the poles subject to the robust stability criteria and load disturbance attenuation criteria. The performance of the proposed method is demonstrated with a test system. SaDE self adapts the trial vector generation strategy and crossover rate (


) value during evolution. Self adaptive Penalty (SP) method is used for constraint handling. The results are compared with constrained PSO and mixed Deterministic/Randomized algorithms. It is shown experimentally that the SaDE adapts automatically to the best strategy and


value. Performance of the SaDE-based controller is superior to other methods in terms of success rate, robust stability, and disturbance attenuation.

S. Miruna Joe Amali, S. Baskar

Electromagnetic Antenna Configuration Optimization Using Fitness Adaptive Differential Evolution

In this article a novel numerical technique, called Fitness Adaptive Differential Evolution (FiADE) for optimizing certain pre-defined antenna configuration is represented. Differential Evolution (DE), inspired by the natural phenomenon of theory of evolution of life on earth, employs the similar computational steps as by any other Evolutionary Algorithm (EA). Scale Factor and Crossover Probability are two very important control parameter of DE since the former regulates the step size taken while mutating a population member in DE. This article describes a very competitive yet very simple form of adaptation technique for tuning the scale factor, on the run, without any user intervention. The adaptation strategy is based on the fitness function value of individuals in DE population. The feasibility, efficiency and effectiveness of the proposed algorithm for optimization of antenna problems are examined by a set of well-known antenna configurations.

Aritra Chowdhury, Arnob Ghosh, Ritwik Giri, Swagatam Das

Analyzing the Explorative Power of Differential Evolution Variants on Different Classes of Problems

This paper is focusing on comparing the performance of Differential Evolution (DE) variants, in the light of analyzing their Explorative power on a set of benchmark function. We have chosen fourteen different variants of DE and fourteen benchmark functions grouped by feature: Unimodal Separable, Unimodal NonSeparable, Multimodal Separable and Multimodal NonSeparable. Fourteen variants of DE were implemented and tested on these fourteen functions for the dimension of 30. The explorative power of the variants is evaluated and analyzed by measuring the evolution of population variance, at each generation. This analysis provides insight about the competitiveness of DE variants in solving the problem at hand.

G. Jeyakumar, C. Shanmugavelayutham

A Self Adaptive Differential Evolution Algorithm for Global Optimization

This paper presents a new Differential Evolution algorithm based on hybridization of adaptive control parameters and trigonometric mutation. First we propose a self adaptive DE named ADE where choice of control parameter F and Cr is not fixed at some constant value but is taken iteratively. The proposed algorithm is further modified by applying trigonometric mutation in it and the corresponding algorithm is named as ATDE. The performance of ATDE is evaluated on the set of 8 benchmark functions and the results are compared with the classical DE algorithm in terms of average fitness function value, number of function evaluations, convergence time and success rate. The numerical result shows the competence of the proposed algorithm.

Pravesh Kumar, Millie Pant

Optimization for Workspace Volume of 3R Robot Manipulator Using Modified Differential Evolution

Robotic manipulators with three-revolute (3R) family of positional configurations are very common in the industrial robots (IRs). The manipulator capability of a robot largely depends on the workspace (WS) of the manipulator apart from other parameters. With the constraints in mind, the optimization of the workspace is of prime importance in designing the manipulator. The workspace of manipulator is formulated as a constrained optimization problem with workspace volume as objective function. It is observed that the previous literature is confined to use of conventional soft computing algorithms only, while a new search modified algorithm is conceptualized and proposed here to improve the computational time. The proposed algorithm gives a good set of geometric parameters of manipulator within the applied constrained limits. The availability of such an algorithm for optimizing the workspace is important, especially for highly constrained environments. The efficiency of the proposed approach to optimize the workspace of 3R manipulators is exhibited through two cases.

Bibhuti Bhusan Biswal, Sumanta Panda, Debadutta Mishra

Adaptive Differential Evolution with p-Best Crossover for Continuous Global Optimization

Differential Evolution (DE) is arguably one of the most powerful stochastic real parameter optimization algorithms in current use. DE operates through the similar computational steps as employed by a standard Evolutionary Algorithm (EA). However, unlike the traditional EAs, the DE-variants perturb the current-generation population members with the scaled differences of randomly selected and distinct population members. Therefore, no separate probability distribution has to be used, which makes the scheme self-organizing in this respect. Its performance, however, is still quite dependent on the setting of control parameters such as the mutation factor and the crossover probability according to both experimental studies and theoretical analyses. Our aim is to design a DE algorithm with control parameters such as the scale factor and the crossover constants adapting themselves to different problem landscapes avoiding any user intervention. Further to improve the convergence performance an innovative crossover mechanism is proposed here.

Sk Minhazul Islam, Saurav Ghosh, Subhrajit Roy, Swagatam Das

A New Particle Swarm Optimization Algorithm for Dynamic Environments

Many real world optimization problems are dynamic in which global optimum and local optima change over time. Particle swarm optimization has performed well to find and track optima in dynamic environments. In this paper, we propose a new particle swarm optimization algorithm for dynamic environments. The proposed algorithm utilizes a parent swarm to explore the search space and some child swarms to exploit promising areas found by the parent swarm. To improve the search performance, when the search areas of two child swarms overlap, the worse child swarms will be removed. Moreover, in order to quickly track the changes in the environment, all particles in a child swarm perform a random local search around the best position found by the child swarm after a change in the environment is detected. Experimental results on different dynamic environments modelled by moving peaks benchmark show that the proposed algorithm outperforms other PSO algorithms, including FMSO, a similar particle swarm algorithm for dynamic environments, for all tested environments.

Masoud Kamosi, Ali B. Hashemi, M. R. Meybodi

Power Mutation Embedded Modified PSO for Global Optimization Problems

In the present study we propose a simple and modified framework for Particle Swarm Optimization (PSO) algorithm by incorporating in it a newly defined operator based on Power Mutation (PM). The resulting PSO variants are named as (Modified Power Mutation PSO) MPMPSO and MPMPSO 1 which differs from each other in the manner of implementation of mutation operator. In MPMPSO, PM is applied stochastically in conjugation with basic position update equation of PSO and in MPMPSO 1, PM is applied on the worst particle of swarm at each iteration. A suite of ten standard benchmark problems is employed to evaluate the performance of the proposed variations. Experimental results show that the proposed MPMPSO outperforms the existing method on most of the test functions in terms of convergence and solution quality.

Pinkey Chauhan, Kusum Deep, Millie Pant

PSO Advances and Application to Inverse Problems

Particle swarm optimization (PSO) is a Swarm Intelligence technique used for optimization motivated by the social behavior of individuals in large groups in nature. The damped mass-spring analogy known as the PSO continuous model allowed us to derive a whole family of particle swarm optimizers with different properties with regard to their exploitation/exploration balance. Using the theory of stochastic differential and difference equations, we fully characterize the stability behavior of these algorithms. PSO and RR-PSO are the most performant algorithms of this family in terms of rate of convergence. Other family members have better exploration capabilities. The so called four point algorithms use more information of previous iterations to update the particles positions and trajectories and seem to be more exploratory than most of the 3 points versions. Finally, based on the done analysis, we can affirm that the PSO optimizers are not heuristic algorithms since there exist mathematical results that can be used to explain their consistency/convergence.

Juan Luis Fernández-Martínez, Esperanza García-Gonzalo

Adaptive and Accelerated Exploration Particle Swarm Optimizer (AAEPSO) for Solving Constrained Multiobjective Optimization Problems

Many science and engineering design problems are modeled as constrained multiobjective optimization problem. The major challenges in solving these problems are (i) conflicting objectives and (ii) non linear constraints. These conflicts are responsible for diverging the solution from true Pareto-front. This paper presents a variation of particle swarm optimization algorithm integrated with accelerated exploration technique that adapts to iteration for solving constrained multiobjective optimization problems. Performance of the proposed algorithm is evaluated on standard constrained multiobjective benchmark functions (CEC 2009) and compared with recently proposed DECMOSA algorithm. The comprehensive experimental results show the effectiveness of the proposed algorithm in terms of generation distance, diversity and convergence metric.

Layak Ali, Samrat L. Sabat, Siba K. Udgata

Expedite Particle Swarm Optimization Algorithm (EPSO) for Optimization of MSA

This paper presents a new designing method of Rectangular patch Microstrip Antenna using an Artificial searches Algorithm with some constraints. It requires two stages for designing. In first stage, bandwidth of MSA is modeled using bench Mark function. In second stage, output of first stage give to modified Artificial search Algorithm which is Particle Swarm Algorithm (PSO) as input and get output in the form of five parameter- dimensions width, frequency range, dielectric loss tangent, length over a ground plane with a substrate thickness and electrical thickness. In PSO Cognition, factor and Social learning Factor give very important effect on balancing the local search and global search in PSO. Basing the modification of cognition factor and social learning factor, this paper presents the strategy that at the starting process cognition-learning factor has more effect then social learning factor. Gradually social learning factor has more impact after learning cognition factor for find out global best. The aim is to find out under above circumstances these modifications in PSO can give better result for optimization of microstrip Antenna (MSA).

Amit Rathi, Ritu Vijay

Covariance Matrix Adapted Evolution Strategy Based Design of Mixed H2/H ∞  PID Controller

This paper discusses the application of the covariance matrix adapted evolution strategy (CMAES) technique to the design of the mixed H




PID controller. The optimal robust PID controller is designed by minimizing the weighted sum of integral squared error (ISE) and balanced robust performance criterion involving robust stability and disturbance attenuation performance subjected to robust stability and disturbance attenuation constraints. In CMAES algorithm, these constraints are effectively handled by penalty parameter-less scheme. In order to test the performance of CMAES algorithm, MIMO distillation column model is considered. For the purpose of comparison, reported intelligent genetic algorithm (IGA) method is used. The statistical performances of combined ISE and balanced robust performance criterion in ten independent simulation runs show that a performance of CMAES is better than IGA method. Robustness test conducted on the system also shows that the robust performance of CMAES designed controller is better than IGA based controller under model uncertainty and external disturbances.

M. Willjuice Iruthayarajan, S. Baskar

Towards a Link between Knee Solutions and Preferred Solution Methodologies

In a bi-criteria optimization problem, often the user is interested in a subset of solutions lying in the knee region. On the other hand in many problem-solving tasks, often one or a few methodologies are commonly used. In this paper, we argue that there is a link between the knee solutions in bi-criteria problems and the preferred methodologies when viewed from a conflicting bi-criterion standpoint. We illustrate our argument with the help of a number of popularly used problem-solving tasks. Each task, when perceived as a bicriteria problem, seems to exhibit a knee or a knee-region and the commonly-used methodology seems to lie within the knee-region. This linking is certainly an interesting finding and may have a long-term implication in the development of efficient solution methodologies for different scientific and other problem-solving tasks.

Kalyanmoy Deb, Shivam Gupta

A Relation-Based Model for Convergence Analysis of Evolutionary Algorithm

There have been many results on convergence of evolutionary algorithm (EA) since it was proposed, but few result focused on convergence analysis based on relation theory. This paper proposed a relation-based model to study the equivalence and ordering of EA in convergence. The equivalence relation named equivalence in status (EIS) can be used to divide a given set of EAs into equivalence classes in which the EAs have the same capacity of convergence. EAs belonging to different EIS classes have different capacities of convergence based on the absorbing Markov chain model, which is described as an ordering relation named superiority in status. The performance of an EA can be improved if it is modified to be superior in status to its original version.

Zhi-Feng Hao, Han Huang, Haozhe Li, Shaohu Ling, Benqing Li

Neural Meta-Memes Framework for Combinatorial Optimization

In this paper, we present a Neural Meta-Memes Framework (NMMF) for combinatorial optimization. NMMF is a framework which models basic optimization algorithms as memes and manages them dynamically when solving combinatorial problems. NMMF encompasses neural networks which serve as the overall planner/coordinator to balance the workload between memes. We show the efficacy of the proposed NMMF through empirical study on a class of combinatorial problem, the quadratic assignment problem (QAP).

Li Qin Song, Meng Hiot Lim, Yew Soon Ong

An Improved Evolutionary Programming with Voting and Elitist Dispersal Scheme

Although initially conceived for evolving finite state machines, Evolutionary Programming (EP), in its present form, is largely used as a powerful real parameter optimizer. For function optimization, EP mainly relies on its mutation operators. Over past few years several mutation operators have been proposed to improve the performance of EP on a wide variety of numerical benchmarks. However, unlike real-coded GAs, there has been no fitness-induced bias in parent selection for mutation in EP. That means the i-th population member is selected deterministically for mutation and creation of the i-th offspring in each generation. In this article we present an improved EP variant called Evolutionary Programming with Voting and Elitist Dispersal (EPVE). The scheme encompasses a voting process which not only gives importance to best solutions but also consider those solutions which are converging fast. By introducing Elitist Dispersal Scheme we maintain the elitism by keeping the potential solutions intact and other solutions are perturbed accordingly, so that those come out of the local minima. By applying these two techniques we can be able to explore those regions which have not been explored so far that may contain optima. Comparison with the recent and best-known versions of EP over 25 benchmark functions from the CEC (Congress on Evolutionary Computation) 2005 test-suite for real parameter optimization reflects the superiority of the new scheme in terms of final accuracy, speed, and robustness.

Sayan Maity, Kumar Gunjan, Swagatam Das

Heuristic Algorithms for the L(2,1)-Labeling Problem



(2, 1)-labeling of a graph


is an assignment


from the vertex set




) to the set of nonnegative integers such that |




) − 




)| ≥ 2 if




are adjacent and |




) − 




)| ≥ 1 if




are at distance two for all








). The span of an




is the maximum value of




) over all vertices




. The


(2,1)-labeling number of a graph


, denoted as




), is the least integer


such that


has an


(2,1)-labeling with span



Since the decision version of the


(2,1)-labeling problem is NP-complete, it is important to investigate heuristic approaches. In this paper, we first implement some heuristic algorithms and then perform an analysis of the obtained results.

B. S. Panda, Preeti Goel

Runtime Analysis of Evolutionary Programming Based on Cauchy Mutation

This paper puts forward a brief runtime analysis of an evolutionary programming (EP) which is one of the most important continuous optimization evolutionary algorithms. A theoretical framework of runtime analysis is proposed by modeling EP as an absorbing Markov process. The framework is used to study the runtime of a classical EP algorithm named as EP with Cauchy mutation (FEP). It is proved that the runtime of FEP can be less than a polynomial of n if the Lebesgue measure of optimal solution set is more than an exponential form of 2. Moreover, the runtime analysis result can be used to explain the performance of EP based on Cauchy mutation.

Han Huang, Zhifeng Hao, Zhaoquan Cai, Yifan Zhu

Best Hiding Capacity Scheme for Variable Length Messages Using Particle Swarm Optimization

Steganography is an art of hiding information in such a way that prevents the detection of hidden messages. Besides security of data, the quantity of data that can be hidden in a single cover medium, is also very important. We present a secure data hiding scheme with high embedding capacity for messages of variable length based on Particle Swarm Optimization. This technique gives the best pixel positions in the cover image, which can be used to hide the secret data. In the proposed scheme, k bits of the secret message are substituted into k least significant bits of the image pixel, where k varies from 1 to 4 depending on the message length. The proposed scheme is tested and results compared with simple LSB substitution, uniform 4-bit LSB hiding (with PSO) for the test images Nature, Baboon, Lena and Kitty. The experimental study confirms that the proposed method achieves high data hiding capacity and maintains imperceptibility and minimizes the distortion between the cover image and the obtained stego image.

Ruchika Bajaj, Punam Bedi, S. K. Pal

Ant Colony Optimization for Markowitz Mean-Variance Portfolio Model

This work presents Ant Colony Optimization (ACO), which was initially developed to be a meta-heuristic for combinatorial optimization, for solving the cardinality constraints Markowitz mean-variance portfolio model (nonlinear mixed quadratic programming problem). To our knowledge, an efficient algorithmic solution for this problem has not been proposed until now. Using heuristic algorithms in this case is imperative. Numerical solutions are obtained for five analyses of weekly price data for the following indices for the period March, 1992 to September, 1997: Hang Seng 31 in Hong Kong, DAX 100 in Germany, FTSE 100 in UK, S&P 100 in USA and Nikkei 225 in Japan. The test results indicate that the ACO is much more robust and effective than Particle swarm optimization (PSO), especially for low-risk investment portfolios.

Guang-Feng Deng, Woo-Tsong Lin

Hybrid PSO Based Integration of Multiple Representations of Thermal Hand Vein Patterns

This paper outlines a novel personal authentication approach by integrating the multiple feature representations of thermal hand vein patterns. In the present work, vein patterns are regarded as comprising textures. Accordingly two types of texture features using Gabor wavelets and fuzzy logic are extracted from the acquired vein images. Since both the approaches have different domains of feature representation, their integration is accomplished at the decision level by incorporating individual decisions using the Euclidean distance based classifiers. The optimal decision parameters comprising individual decision thresholds and one fusion rule out of 16 rules for two features are estimated with the help of hybrid Particle Swarm Optimization (PSO) which can optimize the decisions taken by the individual classifiers. The experimental results carried out on 100 user database are promising thus confirming the usefulness of the proposed authentication system.

Amioy Kumar, Madasu Hanmandlu, H. M. Gupta

Detection and Length Estimation of Linear Scratch on Solid Surfaces Using an Angle Constrained Ant Colony Technique

In many manufacturing areas the detection of surface defects is one of the most important processes in quality control. Currently in order to detect small scratches on solid surfaces most of the industries working on material manufacturing rely on visual inspection primarily. In this article we propose a hybrid computational intelligence technique to automatically detect a linear scratch from a solid surface and estimate its length (in pixel unit) simultaneously. The approach is based on a swarm intelligence algorithm called Ant Colony Optimization (ACO) and image preprocessing with Wiener and Sobel filters as well as the Canny edge detector. The ACO algorithm is mostly used to compensate for the broken parts of the scratch. Our experimental results confirm that the proposed technique can be used for detecting scratches from noisy and degraded images, even when it is very difficult for conventional image processing to distinguish the scratch area from its background.

Siddharth Pal, Aniruddha Basak, Swagatam Das

An Intelligence Model with Max-Min Strategy for Constrained Evolutionary Optimization

An intelligence model (IM) is proposed for constrained optimization in this paper. In this model, two main issues are considered: first, handling feasible and infeasible individuals in population, and second, recognizing the piecewise continuous Pareto front to avoid unnecessary search, it could reduce the amount of calculation and improve the efficiency of search. In addition, max-min strategy is used in selecting optimization. By integrating IM with evolutionary algorithm (EA), a generic constrained optimization evolutionary (IMEA) is derived. The new algorithm is applied to tackle 7 test instances on the CEC2009 MOEA competition, and the performance is assessed by IGD metric, the results suggest that it outperforms or performs similarly to other algorithms in CEC2009 competition.

Xueqiang Li, Zhifeng Hao, Han Huang

Parallel Ant-Miner (PAM) on High Performance Clusters

This study implements parallelization of Ant-Miner for classification rules discovery. Ant-Miner code is parallelized and optimized in a cluster environment by employing master-slave model. The parallelization is achieved in two different operations of Ant-Miner viz. discretization of continuous attributes and rule construction by ants. For rule mining operation, ants are equally distributed into groups and sent across the different cluster nodes. The performance study of Parallel Ant-Miner (PAM) employs different publicly available datasets. The results indicate remarkable improvement in computational time without compromising on the classification accuracy and quality of discovered rules. Dermatology data having 33 features and musk data having 168 features were taken to study performance with respect to timings. Speedup almost equivalent to ideal speedup was obtained on 8 CPUs with increase in number of features and number of ants. Also performance with respect to accuracies was done using lung cancer data.

Janaki Chintalapati, M. Arvind, S. Priyanka, N. Mangala, Jayaraman Valadi

A Software Tool for Data Clustering Using Particle Swarm Optimization

Many universities all over the world have been offering courses on swarm intelligence from 1990s. Particle Swarm Optimization is a swarm intelligence technique. It is relatively young, with a pronounce need for a mature teaching method. This paper presents an educational software tool in MATLAB to aid the teaching of PSO fundamentals and its applications to data clustering. This software offers the advantage of running the classical K-Means clustering algorithm and also provides facility to simulate hybridization of K-Means with PSO to explore better clustering performances. The graphical user interfaces are user-friendly and offer good learning scope to aspiring learners of PSO.

Kalyani Manda, A. Sai Hanuman, Suresh Chandra Satapathy, Vinaykumar Chaganti, A. Vinaya Babu

An ACO Approach to Job Scheduling in Grid Environment

Due to recent advances in the wide-area network technologies and low cost of computing resources, grid computing has become an active research area. The efficiency of a grid environment largely depends on the scheduling method it follows. This paper proposes a framework for grid scheduling using dynamic information and an ant colony optimization algorithm to improve the decision of scheduling. A notion of two types of ants -‘Red Ants’ and ‘Black Ants’ have been introduced. The purpose of red and Black Ants has been explained and algorithms have been developed for optimizing the resource utilization. The proposed method does optimization at two levels and it is found to be more efficient than existing methods.

Ajay Kant, Arnesh Sharma, Sanchit Agarwal, Satish Chandra

Runtime Analysis of (1+1) Evolutionary Algorithm for a TSP Instance

Evolutionary Algorithms (EAs) have been used widely and successfully in solving a famous classical combinatorial optimization problem-the traveling salesman problem (TSP). There are lots of experimental results concerning the TSP. However, relatively few theoretical results on the runtime analysis of EAs on the TSP are available. This paper conducts a runtime analysis of a simple Evolutionary Algorithm called (1+1) EA on a TSP instance. We represent a tour as a string of integer, and randomly choose 2-opt and 3-opt operator as the mutation operator at each iteration. The expected runtime of (1+1) EA on this TSP instance is proved to be





), which is tighter than





 + (1/






) of (1+1) MMAA (Max-Min ant algorithms). It is also shown that the selection of mutation operator is very important in (1+1) EA.

Yu Shan Zhang, Zhi Feng Hao

An Evolutionary Approach to Intelligent Planning

With the explosion of information on WWW, planning and decision making has become a tedious task. The huge volume of distributed and heterogeneous information resources and the complexity involved in their coordination and scheduling leads to difficulties in the conception of optimal plans. This paper presents an intelligent planner which uses modified Genetic Algorithm assisted Case Based Reasoning (CBR) to solve the cold start problem faced by CBR systems and generates novel plans. This approach minimizes the need of populating preliminary cases in the CBR systems. The system is capable of generating synchronized optimal plans within the specified constraints. The effectiveness of the approach is demonstrated with the help of case study on e-Travel Planning. Rigorous experiments were performed to generate synchronized plans with one hop and two hops between train and flight modes of transport. Results proved that GA assisted CBR outperforms the traditional CBR significantly in providing the number of optimized plans and solving cold start problem.

Shikha Mehta, Bhuvan Sachdeva, Rohit Bhargava, Hema Banati

Substitute Domination Relation for High Objective Number Optimization

In this paper, we introduce the average rank dominance relation which substitutes the Pareto domination relation for high objective number optimization. The substitute relation is based on the performances of the solutions in each objective and calculated as the average rank of the solutions on each objective. In addition, the paper studies substituting the Pareto domination relation by the new domination relation in the well known multi-objective algorithms NSGAII and SMPSO which are based respectively on the genetic and particle swarm optimization. The new algorithms are tested on the first four problems of DTLZ family and compared to the original algorithms via new performance indicators.

Sofiene Kachroudi

Discrete Variables Function Optimization Using Accelerated Biogeography-Based Optimization

Biogeography-Based Optimization (BBO) is a bio-inspired and population based optimization algorithm. This is mainly formulated to optimize functions of discrete variables. But the convergence of BBO to the optimum value is slow as it lacks in exploration ability. The proposed Accelerated Biogeography-Based Optimization (ABBO) technique is an improved version of BBO. In this paper, authors accelerated the original BBO to enhance the exploitation and exploration ability by modified mutation operator and clear duplicate operator. This significantly improves the convergence characteristics of the original algorithm. To validate the performance of ABBO, experiments have been conducted on unimodal and multimodal benchmark functions of discrete variables. The results shows excellent performance when compared with other modified BBOs and other optimization techniques like stud genetic algorithm (SGA) and ant colony optimization (ACO). The results are also analyzed by using two paired t- test.

M. R. Lohokare, S. S. Pattnaik, S. Devi, B. K. Panigrahi, S. Das, D. G. Jadhav

A Genetic Algorithm Based Augmented Lagrangian Method for Computationally Fast Constrained Optimization

Among the penalty based approaches for constrained optimization, Augmented Lagrangian (AL) methods are better in at least three ways: (i) they have theoretical convergence properties, (ii) they distort the original objective function minimally to allow a better search behavior, and (iii) they can find the optimal Lagrange multiplier for each constraint as a by-product of optimization. Instead of keeping a constant penalty parameter throughout the optimization process, these algorithms update the parameters adaptively so that the corresponding penalized function dynamically changes its optimum from the unconstrained minimum point to the constrained minimum point with iterations. However, the flip side of these algorithms is that the overall algorithm is a serial implementation of a number of optimization tasks, a process that is usually time-consuming. In this paper, we devise a genetic algorithm based parameter update strategy to a particular AL method. The strategy is self-adaptive in order to make the overall genetic algorithm based augmented Lagrangian (GAAL) method parameter-free. The GAAL method is applied to a number of constrained test problems taken from the EA literature. The function evaluations required by GAAL in many problems is an order or more lower than existing methods.

Soumil Srivastava, Kalyanmoy Deb

Evolutionary Programming Improved by an Individual Random Difference Mutation

Evolutionary programming (EP) is a classical evolutionary algorithm for continuous optimization. There have been several EP algorithms proposed based on different mutations strategies like Gaussian, Cauchy, Levy and other stochastic distributions. However, their convergence speed should be improved. An EP based on individual random difference (EP-IRD) was proposed to attain better solutions in a higher speed. The mutation of EP-IRD uses a random difference of individuals selected randomly to update the variance with which offspring are generated. The IRD-based mutation can make the better offspring according to the current population faster than the mathematical stochastic distribution. The numerical results of solving benchmark problems indicate that EP-IRD performs better than other four EP algorithms based on mathematical stochastic distribution in the items of convergence speed, optimal value on average and standard deviation.

Zhaoquan Cai, Han Huang, Zhifeng Hao, Xueqiang Li

Taguchi Method Based Parametric Study of Generalized Generation Gap Genetic Algorithm Model

In this paper, a parametric study of Generalized Generation Gap (G3) Genetic Algorithm (GA) model with Simplex crossover (SPX) using Taguchi method has been presented. Population size, number of parents and offspring pool size are considered as design factors with five levels. The analysis of mean factor is conducted to find the influence of design factors and their optimal combination for six benchmark functions. The experimental results suggest more experiments on granularity of design factor levels for better performance efficacy.

S. Thangavelu, C. Shunmuga Velayutham

EBFS-ICA: An Efficient Algorithm for CT-MRI Image Fusion

Analyzing the spatial and spectral properties of CT and MRI scan medical images; this article proposes a novel method for CT-MRI image fusion. Independent component analysis is used to analyze images for acquiring independent component. This paper addresses an efficient algorithm for ICA-based image fusion with selection of optimal independent components using


Bacterial Foraging Optimization Technique. Different methods were suggested in the literature to select the largest eigenvalues and their corresponding eigenvectors for ICA based image fusion. But, there is no unified approach for selecting optimal ICA bases to improvise the performance. In this context, we propose a new algorithm called EBFS-ICA which uses a nutrient concentration function (cost function). Here the cost function is maximized through hill climbing via a type of biased random walk. The proposed EBFS-ICA algorithm offers two distinct additional advantages. First, the proposed algorithm can supplement the features of ICA. Second, the random bias incorporated in EBFS guide us to move in the direction of increasingly favorable environment. Finally, we use fusion rules to generate the fused image which contain more integrated accurate detail information of different soft tissue such as muscles and blood vessels. Experimental results presented here show the effectiveness of the proposed EBFS-ICA algorithm. Further, the efficiency of our method is better than FastICA method used in medical image fusion field.

Rutuparna Panda, Sanjay Agrawal

Adaptive Nonlinear Signal Approximation Using Bacterial Foraging Strategy

Uniform approximation of signals has been an area of interest for researchers working in different disciplines of science and engineering. This paper presents an adaptive algorithm based on E. coli bacteria foraging strategy (EBFS) for uniform approximation of signals by linear combinations of shifted nonlinear basis functions. New class of nonlinear basis functions has been derived from a sigmoid function. The weight factor of the newly proposed nonlinear basis functions has been optimized by using the EBFS to minimize the mean square error. Different test signals are considered for validation of the present technique. Results are also compared with Genetic algorithm approach. The proposed technique could also be useful in fractional signal processing applications.

Naik Manoj Kumar, Panda Rutuparna

Swarm Intelligence for Optimizing Hybridized Smoothing Filter in Image Edge Enhancement

In this modern era, image transmission and processing plays a major role. It would be impossible to retrieve information from satellite and medical images without the help of image processing techniques. Edge enhancement is an image processing step that enhances the edge contrast of an image or video in an attempt to improve its acutance. Edges are the representations of the discontinuities of image intensity functions. For processing these discontinuities in an image, a good edge enhancement technique is essential. The proposed work uses a new idea for edge enhancement using hybridized smoothening filters and we introduce a promising technique of obtaining best hybrid filter using swarm algorithms (Artificial Bee Colony (ABC), Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO)) to search for an optimal sequence of filters from among a set of rather simple, representative image processing filters. This paper deals with the analysis of the swarm intelligence techniques through the combination of hybrid filters generated by these algorithms for image edge enhancement.

B. Tirumala Rao, S. Dehuri, M. Dileep, A. Vindhya

A Hybrid GA-Adaptive Particle Swarm Optimization Based Tuning of Unscented Kalman Filter for Harmonic Estimation

This paper proposes Hybrid Genetic Algorithm (GA)-Adaptive Particle Swarm Optimization (APSO) aided Unscented Kalman Filter (UKF) to estimate the harmonic components present in power system voltage/current waveforms. The initial choice of the process and measurement error covariance matrices Q and R (called tuning of the filter) plays a vital role in removal of noise. Hence, hybrid GA-APSO algorithm is used to estimate the error covariance matrices by minimizing the Root Mean Square Error(RMSE) of the UKF. Simulation results are presented to demonstrate the estimation accuracy is significantly improved in comparison with that of conventional UKF.

Ravi Kumar Jatoth, Gogulamudi Anudeep Reddy

Using Social Emotional Optimization Algorithm to Direct Orbits of Chaotic Systems

Social emotional optimization algorithm (SEOA) is a new novel population-based stochastic optimization algorithm. In SEOA, each individual simulates one natural person. All of them are communicated through cooperation and competition to increase social status. The winner with the highest status will be the final solution. In this paper, SEOA is employed to solve the directing orbits of chaotic systems, simulation results show this new variant increases the performance significantly when compared with particle swarm optimization algorithm.

Zhihua Cui, Zhongzhi Shi, Jianchao Zeng

A Hybrid ANN-BFOA Approach for Optimization of FDM Process Parameters

This study proposes an integrated approach for effectively assisting the practitioners in prediction and optimization of process parameters of fused deposition modelling (FDM) process for improving the mechanical strength of fabricated part. The experimental data are used for efficiently training and testing artificial neural network (ANN) model that finely maps the relationship between the input process control factors and output responses. Bayesian regularization is adopted for selection of optimum network architecture because of its ability to fix number of network parameters irrespective of network size. ANN model is trained using Levenberg-Marquardt algorithm and the resulting network has good generalization capability that eliminates the chance of over fitting. Finally, ANN network is combined with bacterial-foraging optimization algorithm (BFOA) to suggest theoretical combination of parameter settings to improve strength related responses of processed parts.

Anoop Kumar Sood, R. K. Ohdar, S. S. Mahapatra

Bio Inspired Swarm Algorithm for Tumor Detection in Digital Mammogram

Microcalcification clusters in mammograms is the significant early sign of breast cancer. Individual clusters are difficult to detect and hence an automatic computer aided mechanism will help the radiologist in detecting the microcalcification clusters in an easy and efficient way. This paper presents a new classification approach for detection of microcalcification in digital mammogram using particle swarm optimization algorithm (PSO) based clustering technique. Fuzzy C-means clustering technique, well defined for clustering data sets are used in combination with the PSO. We adopt the particle swarm optimization to search the cluster center in the arbitrary data set automatically. PSO can search the best solution from the probability option of the Social-only model and Cognition-only model. This method is quite simple and valid, and it can avoid the minimum local value. The proposed classification approach is applied to a database of 322 dense mammographic images, originating from the MIAS database. Results shows that the proposed PSO-FCM approach gives better detection performance compared to conventional approaches.

J. Dheeba, Tamil Selvi

A Hybrid Particle Swarm with Differential Evolution Operator Approach (DEPSO) for Linear Array Synthesis

In recent years particle swarm optimization emerges as one of the most efficient global optimization tools. In this paper, a hybrid particle swarm with differential evolution operator, termed DEPSO, is applied for the synthesis of linear array geometry. Here, the minimum side lobe level and null control, both are obtained by optimizing the spacing between the array elements by this technique. Moreover, a statistical comparison is also provided to establish its performance against the results obtained by Genetic Algorithm (GA), classical Particle Swarm Optimization (PSO), Tabu Search Algorithm (TSA), Differential Evolution (DE) and Memetic Algorithm (MA).

Soham Sarkar, Swagatam Das

Sensor Deployment in 3-D Terrain Using Artificial Bee Colony Algorithm

The ability to determine the optimal deployment location of sensor nodes in a region to satisfy coverage requirement is a key component of establishing an efficient network. Random deployment of sensor nodes fails to be optimal when nodes are deployed where no targets need to be covered, resulting in wastage of energy. The objective of this paper is to place the given number of sensor nodes such that all targets are covered and the required sensing range is minimum. We model the sensor deployment problem as a clustering problem and the optimal locations for sensor deployment are obtained using Artificial Bee Colony (ABC) algorithm. We analyze how the sensing range varies with the number of sensor nodes and also carry out sensitivity analysis test to find the variation in sensing range if the sensor nodes are deployed in a near optimal position.

S. Mini, Siba K. Udgata, Samrat L. Sabat

Novel Particle Swarm Optimization Based Synthesis of Concentric Circular Antenna Array for Broadside Radiation

In many applications it is desirable to have the maximum radiation of an array directed normal to the axis of the array. In this paper, the broadside radiation patterns of three-ring Concentric Circular Antenna Arrays (CCAA) with central element feeding have been reported. For each optimal synthesis, optimal current excitation weights and optimal radii of the rings are determined having the objective of maximum Sidelobe Level (SLL) reduction. The optimization technique adopted is Novel Particle Swarm Optimization (NPSO). Standard Particle Swarm Optimization (SPSO) is also employed for comparative optimization but it proves to be suboptimal. The extensive computational results show that the particular CCAA containing 4, 6 and 8 number of elements in three successive rings along with central element feeding yields grand minimum SLL (-56.58 dB) determined by NPSO.

Durbadal Mandal, Sakti Prasad Ghoshal, Anup Kumar Bhattacharjee

A Particle Swarm Optimization Algorithm for Optimal Operating Parameters of VMI Systems in a Two-Echelon Supply Chain

This paper focuses on the operational issues of a Two-echelon Single-Vendor-Multiple-Buyers Supply chain (TSVMBSC) under vendor managed inventory (VMI) mode of operation. To determine the optimal sales quantity for each buyer in TSVMBC, a mathematical model is formulated. Based on the optimal sales quantity can be obtained and the optimal sales price that will determine the optimal channel profit and contract price between the vendor and buyer. All this parameters depends upon the understanding of the revenue sharing between the vendor and buyers. A Particle Swarm Optimization (PSO) is proposed for this problem. Solutions obtained from PSO is compared with the best known results reported in literature.

Goh Sue-Ann, S. G. Ponnambalam

Enhanced Memetic Algorithm for Task Scheduling

Scheduling tasks onto the processors of a parallel system is a crucial part of program parallelization. Due to the NP-hardness of the task scheduling problem, scheduling algorithms are based on heuristics that try to produce good rather than optimal schedules. This paper proposes a Memetic algorithm with Tabu search and Simulated Annealing as local search for solving Task scheduling problem considering communication contention. This problem consists of finding a schedule for a general task graph to be executed on a cluster of workstations and hence the schedule length can be minimized. Our approach combines local search (by self experience) and global search (by neighboring experience) possessing high search efficiency. The proposed approach is compared with existing list scheduling heuristics. The numerical results clearly indicate that our proposed approach produces solutions which are closer to optimality and/or better quality than the existing list scheduling heuristics.

S. Padmavathi, S. Mercy Shalinie, B. C. Someshwar, T. Sasikumar

Quadratic Approximation PSO for Economic Dispatch Problems with Valve-Point Effects

Quadratic Approximation Particle Swarm Optimization (qPSO) is a variant of Particle Swarm Optimization (PSO) which hybridize Quadratic Approximation Operator (QA) with PSO. qPSO is already proven to be cost effective and reliable for the test problems of continuous optimization. Economic dispatch (ED) problem is one of the fundamental issues in power system operations. The problem of economic dispatch turns out to be a continuous optimization problem which is solved using original PSO and its variant qPSO in expectation of better results. Results are also compared with the earlier published results.

Jagdish Chand Bansal, Kusum Deep

Fuzzified PSO Algorithm for OPF with FACTS Devices in Interconnected Power Systems

This paper presents a new computationally efficient improved stochastic algorithm for solving Optimal Power Flow (OPF) in interconnected power systems with FACTS devices. This proposed technique is based on the combined application of Fuzzy logic strategy incorporated in Particle Swarm Optimization (PSO) algorithm, hence named as Fuzzified PSO (FPSO). The FACTS devices considered here include Static Var Compensator (SVC), Static Synchronous Compensator (STATCOM), Thyristor Controlled Series Capacitor (TCSC) and Unified Power Flow Controller (UPFC). The proposed method is tested on single area IEEE 30-bus system and interconnected two area systems. The optimal solutions obtained using Evolutionary Programming (EP), PSO and FPSO are compared and analyzed. The analysis reveals that the proposed algorithm is relatively simple, efficient and reliable.

N. M. Jothi Swaroopan, P. Somasundaram

Co-ordinated Design of AVR-PSS Using Multi Objective Genetic Algorithm

Automatic Voltage Regulator (AVR) regulates the generator terminal voltage by controlling the amount of current supplied to the generator field winding by the exciter. Power system stabilizer (PSS) is installed with AVR to damp the low frequency oscillations in power system by providing a supplementary signal to the excitation system. Optimal tuning of AVR controller and PSS parameters is necessary for the satisfactory operation of the power system. When applying tuning method to obtain the optimal controller parameters individually, AVR improves the voltage regulation of the system and PSS improves the damping of the system. Simultaneous tuning of AVR and PSS is necessary to obtain better both voltage regulation and oscillation damping in the system. This paper deals with the optimal tuning of AVR controller and PSS parameters in the synchronous machine. The problem of obtaining the optimal controller parameters is formulated as an optimization problem and Multi-Objective Genetic Algorithm (MOGA) is applied to solve the optimization problem. The suitability of the proposed approach has been demonstrated through computer simulation in a Single Machine Infinite Bus (SMIB) system.

B. Selvabala, D. Devaraj

A Genetic Algorithm Approach for the Multi-commodity, Multi-period Distribution Planning in a Supply Chain Network Design

Distribution decisions play an important role in the strategic planning of supply chain management. In order to use the most proper strategic decisions in a supply chain, decision makers should focus on the identification and management of the sources of uncertainties in the supply chain process. In this paper these conditions in a multi-period problem with demands changed over the planning horizon is considered. We develop a non-linear mixed-integer model and propose an efficient heuristic genetic based algorithm which finds the optimal facility locations/allocation, relocation times and the total cost, for the whole supply chain. To explore the viability and efficiency of the proposed model and the solution approach, various computational experiments are performed based on the real size case problems.

G. Reza Nasiri, Hamid Davoudpour, Yaser Movahedi

Particle Swarm Optimization with Watts-Strogatz Model

Particle swarm optimization (PSO) is a popular swarm intelligent methodology by simulating the animal social behaviors. Recent study shows that this type of social behaviors is a complex system, however, for most variants of PSO, all individuals lie in a fixed topology, and conflict this natural phenomenon. Therefore, in this paper, a new variant of PSO combined with Watts-Strogatz small-world topology model, called WSPSO, is proposed. In WSPSO, the topology is changed according to Watts-Strogatz rules within the whole evolutionary process. Simulation results show the proposed algorithm is effective and efficient.

Zhuanghua Zhu

Multi-objective Evolutionary Algorithms to Solve Coverage and Lifetime Optimization Problem in Wireless Sensor Networks

Multi-objective optimization problem formulations reflect pragmatic modeling of several real-life complex optimization problems. In many of them, the considered objectives are competitive with each other and emphasizing only one of them during solution generation and evolution, incurs high probability of producing one sided solution which is unacceptable with respect to other objectives. This paper investigates the concept of boundary search and also explores the application of a special evolutionary operator on a multi-objective optimization problem; Coverage and Lifetime Optimization Problem in Wireless Sensor Network (WSN). The work in this paper explores two competing objectives of WSN;network coverage and network lifetime using two efficient, robust MOEAs. It also digs into the impact of special operators in the multi-objective optimization problems of sensor node’s design topology.

Koyel Chaudhuri, Dipankar Dasgupta

Offline Parameter Estimation of Induction Motor Using a Meta Heuristic Algorithm

An offline parameter estimation problem of an induction motor using a well known, efficient yet simple meta heuristic algorithm DEGL (Differential Evolution with a neighborhood based mutation scheme) has been presented in this article. Two different induction motor models such as approximate and exact models are considered. The parameter estimation methodology describes a method for estimating the steady-state equivalent circuit parameters from the motor performance characteristics, which is normally available from the manufacturer data or from tests. Differential Evolution is not completely free from the problems of slow or premature convergence, that’s why the idea of a much more efficient variant of DE comes. The variant of DE used for solving this problem utilize the concept of the neighborhood of each population member. The feasibility of the proposed method is demonstrated for two different motors and it is compared with the genetic algorithm and the Particle Swarm Optimization algorithm. From the simulation results it is evident that DEGL outperforms both the algorithms (GA and PSO) in the estimation of the parameters of the induction motor.

Ritwik Giri, Aritra Chowdhury, Arnob Ghosh, B. K. Panigrahi, Swagatam Das

Performance Evaluation of Particle Swarm Optimization Based Active Noise Control Algorithm

Active noise control (ANC) has been used to control low-frequency acoustic noise. The ANC uses an adaptive filter algorithm and normally uses least mean square (LMS) algorithm. The gradient based LMS algorithm suffers from local minima problem. In this paper, particle swarm optimization (PSO) algorithm, which is a non-gradient but simple evolutionary computing type algorithm, is proposed for the ANC system. Detailed mathematical treatment is made and systematic computer simulation studies are carried out to evaluate the performance of the PSO based ANC algorithm.

Nirmal Kumar Rout, Debi Prasad Das, Ganapati Panda

Solution to Non-convex Electric Power Dispatch Problem Using Seeker Optimization Algorithm

This paper presents the application of Seeker Optimization Algorithm (SOA) to constrained economic load dispatch problem. Independent simulations were performed over separate systems with different number of generating units having constraints like prohibited operating zones and ramp rate limits. The performance is also compared with other existing similar approaches. The proposed methodology was found to be robust, fast converging and more proficient over other existing techniques.

K. R. Krishnanand, P. K. Rout, B. K. Panigrahi, Ankita Mohapatra

Swarm Intelligence Algorithm for Induction Motor Field Efficiency Evaluation

Determining induction motor field efficiency is imperative in industries for energy conservation and cost savings. The induction motor efficiency is generally tested in a laboratories by certain methods defined in IEEE Standard – 112. But these methods cannot be used for motor efficiency evaluations in the field because it disrupts the production process of the industry. This paper proposes a swarm intelligence algorithm, Particle Swarm Optimization (PSO) for efficiency evaluation of in-service induction motor based on a modified induction motor equivalent circuit model. In this model, stray load losses are considered. The proposed efficiency evaluation method combines the PSO and the equivalent circuit method. First, the equivalent circuit parameters are estimated by minimizing the difference between measured and calculated values of stator current and input power of the motor using the PSO algorithm. Based on these parameters, the efficiency of the motor at various load points are evaluated by using the equivalent circuit method. To exemplify the performance of the PSO based efficiency estimation method, a 5 HP motor has been tested, compared with genetic algorithm (GA), torque gauge method, equivalent circuit method, slip method, current method and segregated loss method and found to be superior. Accordingly, the method will be useful for engineers who implement the energy efficiency programs to the electric motor systems in industries.

V. P. Sakthivel, S. Subramanian

Artificial Bee Colony Algorithm for Transient Performance Augmentation of Grid Connected Distributed Generation

In this paper, a conventional thermal power system equipped with automatic voltage regulator, IEEE type dual input power system stabilizer (PSS) PSS3B and integral controlled automatic generation control loop is considered. A distributed generation (DG) system consisting of aqua electrolyzer, photovoltaic cells, diesel engine generator, and some other energy storage devices like flywheel energy storage system and battery energy storage system is modeled. This hybrid distributed system is connected to the grid. While integrating this DG with the onventional thermal power system, improved transient performance is noticed. Further improvement in the transient performance of this grid connected DG is observed with the usage of superconducting magnetic energy storage device. The different tunable parameters of the proposed hybrid power system model are optimized by artificial bee colony (ABC) algorithm. The optimal solutions offered by the ABC algorithm are compared with those offered by genetic algorithm (GA). It is also revealed that the optimizing performance of the ABC is better than the GA for this specific application.

A. Chatterjee, S. P. Ghoshal, V. Mukherjee

Performance Comparison of Attribute Set Reduction Algorithms in Stock Price Prediction - A Case Study on Indian Stock Data

Stock price prediction and stock trend prediction are the two major research problems of financial time series analysis. In this work, performance comparison of various attribute set reduction algorithms were made for short term stock price prediction. Forward selection, backward elimination, optimized selection, optimized selection based on brute force, weight guided and optimized selection based on the evolutionary principle and strategy was used. Different selection schemes and cross over types were explored. To supplement learning and modeling, support vector machine was also used in combination. The algorithms were applied on a real time Indian stock data namely CNX Nifty. The experimental study was conducted using the open source data mining tool Rapidminer. The performance was compared in terms of root mean squared error, squared error and execution time. The obtained results indicates the superiority of evolutionary algorithms and the optimize selection algorithm based on evolutionary principles outperforms others.

P. Bagavathi Sivakumar, V. P. Mohandas

Dimensionality Reduction and Optimum Feature Selection in Designing Efficient Classifiers

In the course of day-to-day work, huge volumes of data sets constantly grow accumulating a large number of features, but lack completeness and have relatively low information density. Dimensionality reduction and feature selection are the core issues in handling such data sets and more specifically, discovering relationships in data. Dimensionality reduction by reduct generation is an important aspect of classification where reduced attribute set has the same classification power as the entire set of attributes of an information system. In the paper, multiple reducts are generated integrating the concept of rough set theory (RST) and relational algebra operations. As a next step, the attributes of the reducts, which are relatively better associated and have stronger classification power, are selected to generate the single reduct using classical Apriori algorithm. Different classifiers are built using the single reduct and accuracies are compared to measure the effectiveness of the proposed method.

A. K. Das, J. Sil

Social Emotional Optimization Algorithm for Nonlinear Constrained Optimization Problems

Nonlinear programming problem is one important branch in operational research, and has been successfully applied to various real-life problems. In this paper, a new approach called Social emotional optimization algorithm (SEOA) is used to solve this problem which is a new swarm intelligent technique by simulating the human behavior guided by emotion. Simulation results show that the social emotional optimization algorithm proposed in this paper is effective and efficiency for the nonlinear constrained programming problems.

Yuechun Xu, Zhihua Cui, Jianchao Zeng

Multi-Objective Optimal Design of Switch Reluctance Motors Using Adaptive Genetic Algorithm

In this paper a design methodology based on multi objective genetic algorithm (MOGA) is presented to design the switched reluctance motors with multiple conflicting objectives such as efficiency, power factor, full load torque, and full load current, specified dimension, weight of cooper and iron and also manufacturing cost. The optimally designed motor is compared with an industrial motor having the same ratings. Results verify that the proposed method gives better performance for the multi-objective optimization problems. The results of optimal design show the reduction in the specified dimension, weight and manufacturing cost, and the improvement in the power factor, full load torque, and efficiency of the motor.A major advantage of the method is its quite short response time in obtaining the optimal design.

Mehran Rashidi, Farzan Rashidi

Genetic Algorithm Approaches to Solve RWA Problem in WDM Optical Networks

Routing and Wavelength Assignment (RWA) problem is a classical problem in Wavelength Division Multiplexing (WDM) networks. It is reported that RWA problem is a NP-hard problem as the global optimum is not achievable in polynomial time due to the memory limitation of digital computers. We model the RWA problem as an Integer Linear Programming (ILP) problem under wavelength continuity constraint and solve it using Genetic Algorithm (GA) approach to obtain a near optimal solution.

Ravi Sankar Barpanda, Ashok Kumar Turuk, Bibhudatta Sahoo, Banshidhar Majhi

Multi-objective Performance Optimization of Thermo-Electric Coolers Using Dimensional Structural Parameters

Thermo-Electric Coolers (TEC) have promising features as it is better than traditional cooling devices based on thermodynamic cycle in many ways like being noiseless, compact and environment friendly as it is free of CFC responsible for ozone layer depletion. However, TEC have poor performance in terms of Coefficient of Performance (COP) and peak value of rate at which heat is extracted from space to be cooled. Hence, it is obviously of interest to designers, that the above mentioned limitation shall be compensated by optimizing structural parameters such as area and length of thermoelectric elements such that these device operate at near optimal conditions. In present work, this problem is systematically decomposed in two segments, namely single objective optimization and multi-objective optimization. In the end, some useful insights are reported for designers about structural parameters of TEC.

P. K. S. Nain, J. M. Giri, S. Sharma, K. Deb

An Artificial Physics Optimization Algorithm for Multi-Objective Problems Based on Virtual Force Sorting Proceedings

In order to maintain the diversity of non-dominated solutions in multi-objective optimization algorithms efficiently the authors have proposed a multi-objective artificial physics optimization algorithm based on virtual force sorting (VFMOAPO). Adopting quick-sort idea, the individuals in non-dominated solutions set were sorted by the total virtual force exerting on the other individuals. So the non-dominated solution set was pruned and the individual with the maximal sum of virtual force exerting on the other individuals was selected as the global best solution. Some benchmark functions were tested for comparing the performance of VFMOAPO with MOPSO, NSGA and SPEA. The simulation results show the algorithm is feasible and competitive.

Yan Wang, Jian-chao Zeng, Ying Tan

Effective Document Clustering with Particle Swarm Optimization

The paper presents a comparative analysis of K-means and PSO based clustering performances for text datasets. The dimensionality reduction techniques like Stop word removal, Brill’s tagger algorithm and mean Tf-Idf are used while reducing the size of dimension for clustering. The results reveal that PSO based approaches find better solution compared to K-means due to its ability to evaluate many cluster centroids simultaneously in any given time unlike K-means.

Ramanji Killani, K. Srinivasa Rao, Suresh Chandra Satapathy, Gunanidhi Pradhan, K. R. Chandran

A Hybrid Differential Invasive Weed Algorithm for Congestion Management

This work is dedicated to solve the problem of congestion management in restructured power systems. Nowadays we have open access market which pushes the power system operation to their limits for maximum economic benefits but at the same time making the system more susceptible to congestion. In this regard congestion management is absolutely vital. In this paper we try to remove congestion by generation rescheduling where the cost involved in the rescheduling process is minimized. The proposed algorithm is a hybrid of Invasive Weed Optimization (IWO) and Differential Evolution (DE). The resultant hybrid algorithm was applied on standard IEEE 30 bus system and observed to beat existing algorithms like Simple Bacterial foraging (SBF), Genetic Algorithm (GA), Invasive Weed Optimization (IWO), Differential Evolution (DE) and hybrid algorithms like Hybrid Bacterial Foraging and Differential Evolution (HBFDE) and Adaptive Bacterial Foraging with Nelder Mead (ABFNM).

Aniruddha Basak, Siddharth Pal, V. Ravikumar Pandi, B. K. Panigrahi, Swagatam Das

Security Constrained Optimal Power Flow with FACTS Devices Using Modified Particle Swarm Optimization

This paper presents new computationally efficient improved Particle Swarm algorithms for solving Security Constrained Optimal Power Flow (SCOPF) in power systems with the inclusion of FACTS devices. The proposed algorithms are developed based on the combined application of Gaussian and Cauchy Probability distribution functions incorporated in Particle Swarm Optimization (PSO). The power flow algorithm with the presence of Static Var Compensator (SVC) Thyristor Controlled Series Capacitor (TCSC) and Unified Power Flow Controller (UPFC), has been formulated and solved. The proposed algorithms are tested on standard IEEE 30-bus system. The analysis using PSO and modified PSO reveals that the proposed algorithms are relatively simple, efficient, reliable and suitable for real-time applications. And these algorithms can provide accurate solution with fast convergence and have the potential to be applied to other power engineering problems.

P. Somasundaram, N. B. Muthuselvan

Tuning of PID Controller Using Internal Model Control with the Filter Constant Optimized Using Bee Colony Optimization Technique

The present research work presents a novel control scheme for tuning PID controllers using Internal Model control with the filter time constant optimized using Bee colony Optimization technique. PID controllers are used widely in Industrial Processes. Tuning of PID controllers is accomplished using Internal Model control scheme. IMC includes tuning of filter constant


. Compromise is made in selecting the filter constant


since an increased value of


results in a sluggish response whereas decreased value of filter constant leads in an aggressive action. In the present work, an attempt has been made to optimize the value of the


by Bee colony optimization technique. Simulation results show the validity of the proposed scheme for the PID controller tuning.

U. Sabura Banu, G. Uma

An Efficient Estimation of Distribution Algorithm for Job Shop Scheduling Problem

An estimation of distribution algorithm with probability model based on permutation information of neighboring operations for job shop scheduling problem was proposed. The probability model was given using frequency information of pair-wise operations neighboring. Then the structure of optimal individual was marked and the operations of optimal individual were partitioned to some independent sub-blocks. To avoid repeating search in same area and improve search speed, each sub-block was taken as a whole to be adjusted. Also, stochastic adjustment to the operations within each sub-block was introduced to enhance the local search ability. The experimental results show that the proposed algorithm is more robust and efficient.

Xiao-juan He, Jian-chao Zeng, Song-dong Xue, Li-fang Wang

Semantic Web Service Discovery Algorithm Based on Swarm System

Lacking effective web services or a chain of web services in web, a semantic web service discovery algorithm based on swarm system is proposed. Firstly, we assume that there is no central registry for services information and we use a distributed registry for storing services information. Then, in the case no single service can satisfy user request we will chain existing services to create a composite service answering user request. At last, we use VAR-GARCH (Vector Auto Regression Generalized Autoregressive Conditional Heteroskedasticity) model for service composition which incrementally gathers required services information and check them to satisfy a desired chain.

Qiu Jian-ping, Chen Lichao

Stochastic Ranking Particle Swarm Optimization for Constrained Engineering Design Problems

This paper presents a novel hybrid algorithm by integrating particle swarm optimization with stochastic ranking for solving standard constrained engineering design problems. The proposed hybrid algorithm uses domain independent characteristics of stochastic ranking and faster convergence of particle swarm optimization. Performance comparison of the proposed algorithm with other popular techniques through comprehensive experimental investigations establishes the effectiveness and robustness of the proposed algorithm for solving engineering design problems.

Samrat L. Sabat, Layak Ali, Siba K. Udgata

A New Improved Particle Swarm Optimization Technique for Daily Economic Generation Scheduling of Cascaded Hydrothermal Systems

Optimum scheduling of hydrothermal plants is an important task for economic operation of power systems. Many evolutionary techniques such as particle swarm optimization, differential evolution have been applied to solve these problems and found to perform in a better way in comparison with conventional optimization methods. But often these methods converge to a sub-optimal solution prematurely. This paper presents a new improved particle swarm optimization technique called self-organizing hierarchical particle swarm optimization technique with time-varying acceleration coefficients (SOHPSO_TVAC) for solving daily economic generation scheduling of hydrothermal systems to avoid premature convergence. The performance of the proposed method is demonstrated on a sample test system comprising of cascaded reservoirs. The results obtained by the proposed methods are compared with other methods. The results show that the proposed technique is capable of producing comparable results.

K. K. Mandal, B. Tudu, N. Chakraborty

Improved Real Quantum Evolutionary Algorithm for Optimum Economic Load Dispatch with Non-convex Loads

An algorithm based on improved real quantum evolutionary algorithm (IRQEA) was developed to solve the problem of highly non-linear economic load dispatch problem with valve point loading. The performance of the proposed algorithm is evaluated on a test case of 15 units. The performance of the algorithm is compared with floating point genetic algorithm (FPGA) and real quantum evolutionary algorithm (RQEA). Results demonstrate that the performance of the IRQEA algorithm is far better than FPGA and RQEA algorithms in terms of convergence rate and solution quality.

Nidul Sinha, Kaustabh Moni Hazarika, Shantanu Paul, Himanshu Shekhar, Amrita Amrita Karmakar

Linear Array Geometry Synthesis with Minimum Side Lobe Level and Null Control Using Dynamic Multi-Swarm Particle Swarm Optimizer with Local Search

Linear antenna array design is one of the most important electromagnetic optimization problems of current interest. This paper describes the synthesis method of linear array geometry with minimum side lobe level and null control by the Dynamic Multi-Swarm Particle Swarm Optimizer with Local Search (DMSPSO) which optimizes the spacing between the elements of the linear array to produce a radiation pattern with minimum side lobe level and null placement control. The results of the DMSPSO algorithm have been shown to meet or beat the results obtained using other state-of-the-art metaheuristics like the Genetic Algorithm (GA),General Particle Swarm Optimization (PSO), Memetic Algorithms (MA), and Tabu Search (TS) in a statistically meaningful way. Three design examples are presented that illustrate the use of the DMSPSO algorithm, and the optimization goal in each example is easily achieved.

Pradipta Ghosh, Hamim Zafar

Constraint Handling in Transmission Network Expansion Planning

Transmission network expansion planning (TNEP) is a very important and complex problem in power system. Recently, the use of metaheuristic techniques to solve TNEP is gaining more importance due to their effectiveness in handling the inequality constraints and discrete values over the conventional gradient based methods. Evolutionary algorithms (EAs) generally perform unconstrained search and require some additional mechanism to handle constraints. In EA literature, various constraint handling techniques have been proposed. However, to solve TNEP the penalty function approach is commonly used while the other constraint handling methods are untested. In this paper, we evaluate the performance of different constraint handling methods like Superiority of Feasible Solutions (SF), Self adaptive Penalty (SP),

$\mathcal E$

-Constraint (EC), Stochastic Ranking (SR) and the ensemble of constraint handling techniques (ECHT) on TNEP. The potential of different constraint handling methods and their ensemble is evaluated using an IEEE 24 bus system with and without security constraints.

R. Mallipeddi, Ashu Verma, P. N. Suganthan, B. K. Panigrahi, P. R. Bijwe

A Novel Multi-objective Formulation for Hydrothermal Power Scheduling Based on Reservoir End Volume Relaxation

The paper presents a new multi-objective approach to determine the optimal power generation for short term hydrothermal scheduling. Generation cost is considered as one objective. Novelty of the paper lies in choosing the second objective. Instead of introducing a hard constraint on the reservoir end volume we have reasoned that allowing it to relax makes better solutions feasible. The degree of relaxation is kept as the second objective. We have tested our approach on a multi-reservoir cascaded hydrothermal system with four hydro and one thermal plant. We have solved the optimization problem using a decomposition based MOEA called MOEA/D-DE.

Aniruddha Basak, Siddharth Pal, V. Ravikumar Pandi, B. K. Panigrahi, M. K. Mallick, Ankita Mohapatra

Particle Swarm Optimization and Varying Chemotactic Step-Size Bacterial Foraging Optimization Algorithms Based Dynamic Economic Dispatch with Non-smooth Fuel Cost Functions

The Dynamic economic dispatch (DED) problem is an optimization problem with an objective to determine the optimal combination of power outputs for all generating units over a certain period of time in order to minimize the total fuel cost while satisfying dynamic operational constraints and load demand in each interval. Recently social foraging behavior of Escherichia coli bacteria has been explored to develop a novel algorithm for distributed optimization and control. The Bacterial Foraging Optimization Algorithm (BFOA) is currently gaining popularity in the community of researchers, for its effectiveness in solving certain difficult real-world optimization problems. This article comes up with a hybrid approach involving Particle Swarm Optimization (PSO) and BFO algorithms with varying chemo tactic step size for solving the DED problem of generating units considering valve-point effects. The proposed hybrid algorithm has been extensively compared with those methods reported in the literature. The new method is shown to be statistically significantly better on two test systems consisting of five and ten generating units.

P. Praveena, K. Vaisakh, S. Rama Mohana Rao

Hydro-thermal Commitment Scheduling by Tabu Search Method with Cooling-Banking Constraints

This paper presents a new approach for developing an algorithm for solving the Unit Commitment Problem (UCP) in a Hydro-thermal power system. Unit Commitment is a nonlinear optimization problem to determine the minimum cost turn on/off schedule of the generating units in a power system by satisfying both the forecasted load demand and various operating constraints of the generating units. The effectiveness of the proposed hybrid algorithm is proved by the numerical results shown comparing the generation cost solutions and computation time obtained by using Tabu Search Algorithm with other methods like Evolutionary Programming and Dynamic Programming in reaching proper unit commitment.

Nimain Charan Nayak, C. Christober Asir Rajan


Weitere Informationen

Premium Partner