Ved Pal Singh

A Simple Diversity Guided Particle Swarm Optimization

In this paper we have proposed a new diversity guided particle swarm optimizer (PSO), namely ATRE-PSO, which is a modification of attractive and repulsive PSO (ARPSO), suggested by Riget and Vesterstorm [1]. Depending on the diversity of the population the ATRE-PSO switches alternately between three phases of attraction, repulsion and a combination of attraction and repulsion, called the phase of positive conflict [2]. The performance of ATRE-PSO is compared with basic PSO (BPSO) and ARPSO. The numerical results show that besides preserving the rapid convergence of the BPSO, ATRE-PSO also maintains a good diversity in the population. Under most of the test cases, simulations show that ATRE-PSO finds a better solution than BPSO as well as ARPSO.

Differential Evolution with Parent Centric Crossover

Differential evolution (DE) has emerged as a powerful tool for solving optimization problems in the last few years. However, the convergence rate of DE still does not meet all the requirements, and attempts to speed up differential evolution are considered necessary. In order to improve the performance of DE, we propose a modified DE algorithm called DEPCX which uses parent centric approach to manipulate the solution vectors. The performance of DEPCX is evaluated on a test bed of five functions. Numerical results are compared with original differential evolution (DE) and with TDE, another recently modified version of DE. Empirical results indicate that this modification enables the algorithm to get a better transaction between the convergence rate and robustness.

Particle Swarm Optimization Using Sobol Mutation

In this paper, we present a new mutation operator called the systematic mutation (SM) operator for enhancing the performance of basic particle swarm optimization (BPSO) algorithm. The SM operator unlike most of its contemporary mutation operators do not use the random probability distribution for perturbing the swarm population, but uses a quasi random Sobol sequence to find new solution vectors in the search domain. The comparison of SM-PSO is made with BPSO and some other variants of PSO. The empirical results show that SM operator significantly improves the performance of PSO.

Similarity Solutions for Strong Shocks in a Non-Ideal Gas

A group theoretic method is used to obtain an entire class of similarity solutions to the problem of shocks propagating through a non-ideal gas and to characterize analytically the state dependent form of the medium ahead for which the problem is invariant and admits similarity solutions. Different cases of possible solutions, known in the literature, with a power law, exponential or logarithmic shock paths are recovered as special cases depending on the arbitrary constants occurring in the expression for the generators of the transformation. Particular case of collapse of imploding cylindrically and spherically symmetric shock in a medium in which initial density obeys power law is worked out in detail. Numerical calculations have been performed to obtain the similarity exponents and the profiles of the flow variables behind the shock, and comparison is made with the known results.

Modified onlooker phase in artificial bee colony algorithm

Artificial bee colony (ABC) algorithm is relatively a new bio-inspired swarm intelligence optimization technique comparative to other population based algorithms. In this study BGA (breeder GA) mutation is embedded into onlooker bee phase to improve the capability of local search. The proposed variant is named B-ABC. The experimental results on 10 constrained benchmark functions demonstrate the performance of the proposed variant against those of state-of-the-art algorithms for a set of constrained test problems. Further the efficiency of the proposed variant is tested on the car side impact problem.

Quantum mechanics inspired Particle Swarm Optimisation for global optimisation

This paper presents a novel variant of quantum mechanics inspired Particle Swarm Optimisation (PSO) algorithm named constrained/unconstrained Quantum Particle Swarm Optimisation (CQPSO). The proposed algorithm has the properties of quantum mechanics embedded in the structure of the PSO along with the presence of a quadratic interpolation recombination operator. The performance of CQPSO is validated on three standard non linear, unconstrained functions, eight constrained benchmark problems and two constrained, real life, electrical design problems. The experimental results show that the presence of quadratic interpolation recombination operator enhances the performance of quantum mechanics inspired PSO.

Differential Evolution Using Interpolated Local Search

In this paper we propose a novel variant of the Differential Evolution (DE) algorithm based on local search. The corresponding algorithm is named as Differential Evolution with Interpolated Local Search (DEILS). In DEILS, the local search operation is applied in an adaptive manner. The adaptive behavior enables the algorithm to search its neighborhood in an effective manner and the interpolation helps in exploiting the solutions. In this way a balance is maintained between the exploration and exploitation factors. The performance of DEILS is investigated and compared with basic differential evolution, modified versions of DE and some other evolutionary algorithms. It is found that the proposed scheme improves the performance of DE in terms of quality of solution without compromising with the convergence rate.