Search

Incremental social learning (ISL) was proposed as a way to improve the scalability of systems composed of multiple learning agents. In this paper, we show that ISL can be very useful to improve the performance of population-based optimization algorithms. Our study focuses on two particle swarm optimization (PSO) algorithms: a) the incremental particle swarm optimizer (IPSO), which is a PSO algorithm with a growing population size in which the initial position of new particles is biased toward the best-so-far solution, and b) the incremental particle swarm optimizer with local search (IPSOLS), in which solutions are further improved through a local search procedure. We first derive analytically the probability density function induced by the proposed initialization rule applied to new particles. Then, we compare the performance of IPSO and IPSOLS on a set of benchmark functions with that of other PSO algorithms (with and without local search) and a random restart local search algorithm. Finally, we measure the benefits of using incremental social learning on PSO algorithms by running IPSO and IPSOLS on problems with different fitness distance correlations. [ABSTRACT FROM AUTHOR]

This paper proposes a variable-dimension optimization approach to address the high dimensionality issues in solving the unit commitment problem. This method introduces the concept of adaptive search space dimension. The proposed approach is implemented in particle swarm optimization algorithm. The optimization process starts with an arbitrary problem dimension, adapts with respect to the swarm progress and finally selects the optimal dimensional space. The efficiency of this method is tested on a ten-unit test system. The results are compared with binary programming and fixed duty cycle approaches. The simulation results show that the proposed method results in considerable reduction of problem dimension, faster convergence and improved quality of the final solution. [ABSTRACT FROM AUTHOR]

The particle swarm optimization (PSO) is a recently developed evolutionary algorithm (EA) based on the swarm behavior in the nature. This paper presents recent advances in applying a versatile PSO engine to real-number, binary, single-objective and multiobjective optimizations for antenna designs, with a randomized Newtonian mechanics model developed to describe the swarm behavior. The design of aperiodic (nonuniform and thinned) antenna arrays is presented as an example for the application of the PSO engine. In particular, in order to achieve an improved peak sidelobe level (SLL), element positions in a nonuniform array are optimized by real-number PSO (RPSO). On the other hand, in a thinned array, the on/off state of each element is determined by binary PSO (BPSO). Optimizations for both nonuniform arrays and thinned arrays are also expanded to multiobjective cases. As a result, nondominated designs on the Pareto front enable one to achieve other design factors than the peak SLL. Optimized antenna arrays are compared with periodic arrays and previously presented aperiodic arrays. Selected designs fabricated and measured to validate the effectiveness of PSO in practical electromagnetic problems [ABSTRACT FROM PUBLISHER]

The coordination of directional overcurrent relays (DOCK) is treated in this paper using particle swarm optimization (PSO), a recently proposed optimizer that utilizes the swarm behavior in searching for an optimum. PSO gained a lot of interest for its simplicity, robustness, and easy implementation. The problem of setting DOCR is a highly constrained optimization problem that has been stated and solved as a linear programming (LP) problem. To deal with such constraints a modification to the standard PSO algorithm is introduced. Three case studies are presented, and the results are compared to those of LP technique to demonstrate the effectiveness of the proposed methodology. [ABSTRACT FROM AUTHOR]

The particle swarm algorithm is a newly introduced method for electromagnetic optimization problems that is based on the observation of swarm intelligence and particle behavior. This paper proposes a novel strategy for the initialization of the agents' position within the multidimensional solution domain. In particular, the domain is initially subdivided into subdomains so to have a more uniform distribution of the agents. At a second stage, the sub-boundaries are removed and the best position information of each group is passed to each agent; the agents are therefore allowed to explore the whole search space. This procedure results to be efficient and to improve the convergence rate. A comparison between the performance of this new implementation and that of the basic particle swarm algorithm is presented for several test cases. Finally, this new procedure is successfully applied to the synthesis of artificial magnetic conductors (AMCs) [ABSTRACT FROM PUBLISHER]

Optimization is an important research field, especially in engineering, physical sciences, and economics. The main purpose of optimization is to maximize the profit and minimize the cost of production as well as the loss of the system. Evolutionary computation algorithms, such as the genetic algorithm and the particle swarm optimization (PSO) algorithm have been successfully employed in solving various optimization problems. Owing to its application potential and promising performance in discovering the optimal solution, the PSO algorithm has been recognized as a powerful optimization technique and attracted an ever-increasing interest in the evolutionary computation community. In this article, a novel $N$ -state Markovian jumping PSO (NS-MJPSO) algorithm is presented where the velocity updating equation is adjusted based on the state evolution governed by a Markov chain. The performance of the proposed NS-MJPSO algorithm is evaluated via some widely used mathematical benchmark functions. The experimental results demonstrate that the developed NS-MJPSO algorithm outperforms some currently popular PSO algorithms on the widely used benchmark functions. [ABSTRACT FROM AUTHOR]

The grey wolf optimizer (GWO) is a newly developed swarm intelligence-based optimization technique that mimics the social hierarchy and group hunting behavior of grey wolves in nature. Here, a detailed introduction of the GWO algorithm is given, after which, three sets of examples are investigated: first, numerical experiments on four benchmark functions are conducted; second, the GWO is applied to the synthesis of linear arrays with the aim of reducing the peak sidelobe level under various constraints; and finally, the performance of the GWO is further verified on the optimization design of two representative antennas, namely, a dual-band E-shaped patch antenna and a wideband magneto-electric dipole antenna. The results show that the GWO is capable of outperforming or providing very competitive results compared with some well-known metaheuristics such as the genetic algorithm, particle swarm optimization, and differential evolution. Thus, it may serve as a promising candidate for handling electromagnetic problems. [ABSTRACT FROM AUTHOR]

Particle swarm optimization is a population-based swarm intelligence algorithm based on the simulation of a social psychological metaphor instead of the survival of the fittest individual paradigm. Inspired by the classical particle swarm method and quantum mechanics theories, this work presents a new quantum-behaved approach using a mutation operator with exponential probability distribution. The simulation results demonstrate good performance of the proposed algorithm in solving a significant benchmark problem in electromagnetics, namely the TEAM workshop benchmark problem 22. [ABSTRACT FROM AUTHOR]

A new effective optimization algorithm suitably developed for electromagnetic applications called genetical swarm optimization (GSO) is presented. This is a hybrid algorithm developed in order to combine in the most effective way the properties of two of the most popular evolutionary optimization approaches now in use for the optimization of electromagnetic structures, the particle swarm optimization (PSO) and genetic algorithms (GAs). The algorithm effectiveness has been tested here with respect to both its "ancestors," GA and PSO, dealing with an electromagnetic application, the optimization of a linear array. The here proposed method shows itself as a general purpose tool able to effectively adapt itself to different electromagnetic optimization problems [ABSTRACT FROM PUBLISHER]