Search results

1 – 10 of 47
Article
Publication date: 6 February 2020

Sajad Ahmad Rather and P. Shanthi Bala

The purpose of this paper is to investigate the performance of chaotic gravitational search algorithm (CGSA) in solving mechanical engineering design frameworks including welded…

Abstract

Purpose

The purpose of this paper is to investigate the performance of chaotic gravitational search algorithm (CGSA) in solving mechanical engineering design frameworks including welded beam design (WBD), compression spring design (CSD) and pressure vessel design (PVD).

Design/methodology/approach

In this study, ten chaotic maps were combined with gravitational constant to increase the exploitation power of gravitational search algorithm (GSA). Also, CGSA has been used for maintaining the adaptive capability of gravitational constant. Furthermore, chaotic maps were used for overcoming premature convergence and stagnation in local minima problems of standard GSA.

Findings

The chaotic maps have shown efficient performance for WBD and PVD problems. Further, they have depicted competitive results for CSD framework. Moreover, the experimental results indicate that CGSA shows efficient performance in terms of convergence speed, cost function minimization, design variable optimization and successful constraint handling as compared to other participating algorithms.

Research limitations/implications

The use of chaotic maps in standard GSA is a new beginning for research in GSA particularly convergence and time complexity analysis. Moreover, CGSA can be used for solving the infinite impulsive response (IIR) parameter tuning and economic load dispatch problems in electrical sciences.

Originality/value

The hybridization of chaotic maps and evolutionary algorithms for solving practical engineering problems is an emerging topic in metaheuristics. In the literature, it can be seen that researchers have used some chaotic maps such as a logistic map, Gauss map and a sinusoidal map more rigorously than other maps. However, this work uses ten different chaotic maps for engineering design optimization. In addition, non-parametric statistical test, namely, Wilcoxon rank-sum test, was carried out at 5% significance level to statistically validate the simulation results. Besides, 11 state-of-the-art metaheuristic algorithms were used for comparative analysis of the experimental results to further raise the authenticity of the experimental setup.

Article
Publication date: 20 January 2022

Vahid Goodarzimehr, Fereydoon Omidinasab and Nasser Taghizadieh

This paper aims to present a new hybrid algorithm of Particle Swarm Optimization and the Genetic Algorithm (PSOGA) to optimize the space trusses with continuous design variables…

147

Abstract

Purpose

This paper aims to present a new hybrid algorithm of Particle Swarm Optimization and the Genetic Algorithm (PSOGA) to optimize the space trusses with continuous design variables. The PSOGA is an efficient hybridized algorithm to solve optimization problems.

Design/methodology/approach

These algorithms have shown outstanding performance in solving optimization problems with continuous variables. The PSO conceptually models the social behavior of birds, in which individual birds exchange information about their position, velocity and fitness. The behavior of a flock is influencing the probability of migration to other regions with high fitness. The GAs procedure is based on the mechanism of natural selection. The present study uses mutation, random selection and reproduction to reach the best genetic algorithm by the operators of natural genetics. Thus, only identical chromosomes or particles can be converged.

Findings

In this research, using the idea of hybridization PSO and GA algorithms are hybridized and a new meta-heuristic algorithm is developed to minimize the space trusses with continuous design variables. To showing the efficiency and robustness of the new algorithm, several benchmark problems are solved and compared with other researchers.

Originality/value

The results indicate that the hybrid PSO algorithm improved in both exploration and exploitation. The PSO algorithm can be used to minimize the weight of structural problems under stress and displacement constraints.

Details

World Journal of Engineering, vol. 20 no. 3
Type: Research Article
ISSN: 1708-5284

Keywords

Article
Publication date: 30 June 2020

Sajad Ahmad Rather and P. Shanthi Bala

In this paper, a newly proposed hybridization algorithm namely constriction coefficient-based particle swarm optimization and gravitational search algorithm (CPSOGSA) has been…

Abstract

Purpose

In this paper, a newly proposed hybridization algorithm namely constriction coefficient-based particle swarm optimization and gravitational search algorithm (CPSOGSA) has been employed for training MLP to overcome sensitivity to initialization, premature convergence, and stagnation in local optima problems of MLP.

Design/methodology/approach

In this study, the exploration of the search space is carried out by gravitational search algorithm (GSA) and optimization of candidate solutions, i.e. exploitation is performed by particle swarm optimization (PSO). For training the multi-layer perceptron (MLP), CPSOGSA uses sigmoid fitness function for finding the proper combination of connection weights and neural biases to minimize the error. Secondly, a matrix encoding strategy is utilized for providing one to one correspondence between weights and biases of MLP and agents of CPSOGSA.

Findings

The experimental findings convey that CPSOGSA is a better MLP trainer as compared to other stochastic algorithms because it provides superior results in terms of resolving stagnation in local optima and convergence speed problems. Besides, it gives the best results for breast cancer, heart, sine function and sigmoid function datasets as compared to other participating algorithms. Moreover, CPSOGSA also provides very competitive results for other datasets.

Originality/value

The CPSOGSA performed effectively in overcoming stagnation in local optima problem and increasing the overall convergence speed of MLP. Basically, CPSOGSA is a hybrid optimization algorithm which has powerful characteristics of global exploration capability and high local exploitation power. In the research literature, a little work is available where CPSO and GSA have been utilized for training MLP. The only related research paper was given by Mirjalili et al., in 2012. They have used standard PSO and GSA for training simple FNNs. However, the work employed only three datasets and used the MSE performance metric for evaluating the efficiency of the algorithms. In this paper, eight different standard datasets and five performance metrics have been utilized for investigating the efficiency of CPSOGSA in training MLPs. In addition, a non-parametric pair-wise statistical test namely the Wilcoxon rank-sum test has been carried out at a 5% significance level to statistically validate the simulation results. Besides, eight state-of-the-art meta-heuristic algorithms were employed for comparative analysis of the experimental results to further raise the authenticity of the experimental setup.

Details

International Journal of Intelligent Computing and Cybernetics, vol. 13 no. 2
Type: Research Article
ISSN: 1756-378X

Keywords

Article
Publication date: 6 March 2017

Soyinka Olukunle Kolawole and Duan Haibin

Keeping satellite position within close tolerances is key for the utilization of satellite formations for space missions. The presence of perturbation forces makes control…

Abstract

Purpose

Keeping satellite position within close tolerances is key for the utilization of satellite formations for space missions. The presence of perturbation forces makes control inevitable if such mission objective is to be realised. Various approaches have been used to obtain feedback controller parameters for satellites in a formation; this paper aims to approach the problem of estimating the optimal feedback parameter for a leader–follower pair of satellites in a small eccentric orbit using nature-based search algorithms.

Design/methodology/approach

The chaotic artificial bee colony algorithm is a variant of the basic artificial bee colony algorithm. The algorithm mimics the behaviour of bees in their search for food sources. This paper uses the algorithm in optimizing feedback controller parameters for a satellite formation control problem. The problem is formulated to optimize the controller parameters while minimizing a fuel- and state-dependent cost function. The dynamical model of the satellite is based on Gauss variational equations with J2 perturbation. Detailed implementation of the procedure is provided, and experimental results of using the algorithm are also presented to show feasibility of the method.

Findings

The experimental results indicate the feasibility of this approach, clearly showing the effective control of the transients that arise because of J2 perturbation.

Originality/value

This paper applied a swarm intelligence approach to the problem of estimating optimal feedback control parameter for a pair of satellites in a formation.

Details

Aircraft Engineering and Aerospace Technology, vol. 89 no. 2
Type: Research Article
ISSN: 1748-8842

Keywords

Article
Publication date: 16 January 2019

Yuanyang Zou

This paper aims to propose a novel nature-inspired optimization algorithm, called whirlpool algorithm (WA), which imitates the physical phenomenon of whirlpool.

Abstract

Purpose

This paper aims to propose a novel nature-inspired optimization algorithm, called whirlpool algorithm (WA), which imitates the physical phenomenon of whirlpool.

Design/methodology/approach

The idea of this algorithm stems from the fact that the whirlpool has a descent direction and a vertex.

Findings

WA is tested with two types of models: 29 typical mathematical optimization models and three engineering problems (tension/compression spring design, welded-beam design, pressure vessel design).

Originality/value

The results shown that the WA is vying compared to the state-of-art algorithms likewise conservative approaches.

Details

Engineering Computations, vol. 36 no. 2
Type: Research Article
ISSN: 0264-4401

Keywords

Article
Publication date: 3 December 2021

R. Ramaporselvi and G. Geetha

The purpose of this paper is to enhance the line congestion and to minimize power loss. Transmission line congestion is considered the most acute trouble during the operation of…

Abstract

Purpose

The purpose of this paper is to enhance the line congestion and to minimize power loss. Transmission line congestion is considered the most acute trouble during the operation of the power system. Therefore, congestion management acts as an effective tool in using the available power without breaking the system hindrances or limitations.

Design/methodology/approach

Over the past few years, determining the optimal location and size of the devices have pinched a great deal of consideration. Numerous approaches have been established to mitigate the congestion rate, and this paper aims to enhance the line congestion and minimize power loss by determining the compensation rate and optimal location of a thyristor-switched capacitor (TCSC) using adaptive moth swarm optimization (AMSO) algorithm.

Findings

An AMSO algorithm uses the performances of moth flame and the chaotic local search-based shrinking scheme of the bacterial foraging optimization algorithm. The proposed AMSO approach is executed and discussed for the IEEE-30 bus system for determining the optimal location of single TCSC and dual TCSC.

Originality/value

In addition to this, the proposed algorithm is compared with various other existing approaches, and the results thus obtained provide better performances than other techniques.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering , vol. 41 no. 1
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 1 April 2019

Arulraj Rajendran and Kumarappan Narayanan

This paper aims to optimally plan distributed generation (DG) and capacitor in distribution network by optimizing multiple conflicting operational objectives simultaneously so as…

Abstract

Purpose

This paper aims to optimally plan distributed generation (DG) and capacitor in distribution network by optimizing multiple conflicting operational objectives simultaneously so as to achieve enhanced operation of distribution system. The multi-objective optimization problem comprises three important objective functions such as minimization of total active power loss (Plosstotal), reduction of voltage deviation and balancing of current through feeder sections.

Design/methodology/approach

In this study, a hybrid configuration of weight improved particle swarm optimization (WIPSO) and gravitational search algorithm (GSA) called hybrid WIPSO-GSA algorithm is proposed in multi-objective problem domain. To solve multi-objective optimization problem, the proposed hybrid WIPSO-GSA algorithm is integrated with two components. The first component is fixed-sized archive that is responsible for storing a set of non-dominated pareto optimal solutions and the second component is a leader selection strategy that helps to update and identify the best compromised solution from the archive.

Findings

The proposed methodology is tested on standard 33-bus and Indian 85-bus distribution systems. The results attained using proposed multi-objective hybrid WIPSO-GSA algorithm provides potential technical and economic benefits and its best compromised solution outperforms other commonly used multi-objective techniques, thereby making it highly suitable for solving multi-objective problems.

Originality/value

A novel multi-objective hybrid WIPSO-GSA algorithm is proposed for optimal DG and capacitor planning in radial distribution network. The results demonstrate the usefulness of the proposed technique in improved distribution system planning and operation and also in achieving better optimized results than other existing multi-objective optimization techniques.

Details

International Journal of Energy Sector Management, vol. 13 no. 1
Type: Research Article
ISSN: 1750-6220

Keywords

Article
Publication date: 20 November 2020

Sudeepa Das, Tirath Prasad Sahu and Rekh Ram Janghel

The purpose of this paper is to modify the crow search algorithm (CSA) to enhance both exploration and exploitation capability by including two novel approaches. The positions of…

Abstract

Purpose

The purpose of this paper is to modify the crow search algorithm (CSA) to enhance both exploration and exploitation capability by including two novel approaches. The positions of the crows are updated in two approaches based on awareness probability (AP). With AP, the position of a crow is updated by considering its velocity, calculated in a similar fashion to particle swarm optimization (PSO) to enhance the exploiting capability. Without AP, the crows are subdivided into groups by considering their weights, and the crows are updated by conceding leaders of the groups distributed over the search space to enhance the exploring capability. The performance of the proposed PSO-based group-oriented CSA (PGCSA) is realized by exploring the solution of benchmark equations. Further, the proposed PGCSA algorithm is validated over recently published algorithms by solving engineering problems.

Design/methodology/approach

In this paper, two novel approaches are implemented in two phases of CSA (with and without AP), which have been entitled the PGCSA algorithm to solve engineering benchmark problems.

Findings

The proposed algorithm is applied with two types of problems such as eight benchmark equations without constraint and six engineering problems.

Originality/value

The PGCSA algorithm is proposed with superior competence to solve engineering problems. The proposed algorithm is substantiated hypothetically by using a paired t-test.

Details

Engineering Computations, vol. 38 no. 2
Type: Research Article
ISSN: 0264-4401

Keywords

Article
Publication date: 30 June 2020

Hadi Kashefi, Ahmad Sadegheih, Ali Mostafaeipour and Mohammad Mohammadpour Omran

To design, control and evaluate photovoltaic (PV) systems, an accurate model is required. Accuracy of PV models depends on model parameters. This study aims to use a new algorithm

Abstract

Purpose

To design, control and evaluate photovoltaic (PV) systems, an accurate model is required. Accuracy of PV models depends on model parameters. This study aims to use a new algorithm called improved social spider algorithm (ISSA) to detect model parameters.

Design/methodology/approach

To improve performance of social spider algorithm (SSA), an elimination period is added. In addition, at the beginning of each period, a certain number of the worst solutions are replaced by new solutions in the search space. This allows the particles to find new paths to get the best solution.

Findings

In this paper, ISSA is used to estimate parameters of single-diode and double-diode models. In addition, effect of irradiation and temperature on I–V curves of PV modules is studied. For this purpose, two different modules called multi-crystalline (KC200GT) module and polycrystalline (SW255) are used. It should be noted that to challenge the performance of the proposed algorithm, it has been used to identify the parameters of a type of widely used module of fuel cell called proton exchange membrane fuel cell. Finally, comparing and analyzing of ISSA results with other similar methods shows the superiority of the presented method.

Originality/value

Changes in the spider’s movement process in the SSA toward the desired response have improved the algorithm’s performance. Higher accuracy and convergence rate, skipping local minimums, global search ability and search in a limited space can be mentioned as some advantages of this modified method compared to classic SSA.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering , vol. 40 no. 2
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 15 March 2022

Ranjitha K., Sivakumar P. and Monica M.

This study aims to implement an improved version of the Chimp algorithm (IChimp) for load frequency control (LFC) of power system.

Abstract

Purpose

This study aims to implement an improved version of the Chimp algorithm (IChimp) for load frequency control (LFC) of power system.

Design/methodology/approach

This work was adopted by IChimp to optimize proportional integral derivative (PID) controller parameters used for the LFC of a two area interconnected thermal system.

Findings

The supremacy of proposed IChimp tuned PID controller over Chimp optimization, direct synthesis-based PID controller, internal model controller tuned PID controller and recent algorithm based PID controller was demonstrated.

Originality/value

IChimp has good convergence and better search ability. The IChimp optimized PID controller is the proposed controlling method, which ensured better performance in terms of converging behaviour, optimizing controller gains and steady-state response.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering , vol. 41 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

1 – 10 of 47