scholarly journals Heterogeneous Cooperative Bare-Bones Particle Swarm Optimization with Jump for High-Dimensional Problems

Electronics ◽  
2020 ◽  
Vol 9 (9) ◽  
pp. 1539
Author(s):  
Joonwoo Lee ◽  
Won Kim
Keyword(s):  
Particle Swarm Optimization ◽  
Large Scale ◽  
Particle Swarm ◽  
Optimization Method ◽  
High Dimensional ◽  
Benchmark Problems ◽  
Swarm Optimization ◽  
Local Optima ◽  
One Step ◽  
Bare Bones

This paper proposes a novel Bare-Bones Particle Swarm Optimization (BBPSO) algorithm for solving high-dimensional problems. BBPSO is a variant of Particle Swarm Optimization (PSO) and is based on a Gaussian distribution. The BBPSO algorithm does not consider the selection of controllable parameters for PSO and is a simple but powerful optimization method. This algorithm, however, is vulnerable to high-dimensional problems, i.e., it easily becomes stuck at local optima and is subject to the “two steps forward, one step backward” phenomenon. This study improves its performance for high-dimensional problems by combining heterogeneous cooperation based on the exchange of information between particles to overcome the “two steps forward, one step backward” phenomenon and a jumping strategy to avoid local optima. The CEC 2010 Special Session on Large-Scale Global Optimization (LSGO) identified 20 benchmark problems that provide convenience and flexibility for comparing various optimization algorithms specifically designed for LSGO. Simulations are performed using these benchmark problems to verify the performance of the proposed optimizer by comparing the results of other variants of the PSO algorithm.

scholarly journals Optimization of water distribution networks using a modified particle swarm optimization algorithm

2017 ◽  
Vol 18 (2) ◽  
pp. 660-678 ◽  
Author(s):  
Douglas F. Surco ◽  
Thelma P. B. Vecchi ◽  
Mauro A. S. S. Ravagnani
Keyword(s):  
Particle Swarm Optimization ◽  
Water Distribution ◽  
Large Scale ◽  
Particle Swarm ◽  
Minimum Cost ◽  
Distribution Networks ◽  
Water Distribution Networks ◽  
Benchmark Problems ◽  
Swarm Optimization ◽  
Local Optima

Abstract In the present work, a model is presented for the optimization of water distribution networks (WDN). The developed model can be used to verify node pressures, head losses, and fluid flow rate and velocity in each pipe. The algorithm is based on particle swarm optimization (PSO), considering real and discrete variables and avoiding premature convergence to local optima using objective function penalization. The model yields the minimum cost of the network, the node pressures and the velocities in the pipes. The pressures and velocities are calculated using the hydraulic simulator Epanet. Some benchmark problems are used to test the applicability of the developed model, considering WDN for small-, medium-, and large-scale problems. Obtained results are consistent with those found in the literature.

A High-Dimensional Model Representation Guided PSO Methodology With Application on Compressor Airfoil Shape Optimization

2016 ◽  
Author(s):  
A. Safari ◽  
K. H. Hajikolaei ◽  
H. G. Lemu ◽  
G. G. Wang
Keyword(s):  
Particle Swarm Optimization ◽  
Particle Swarm ◽  
Pso Algorithm ◽  
Population Based ◽  
Superior Performance ◽  
High Dimensional ◽  
Benchmark Problems ◽  
High Dimensional Model Representation ◽  
Swarm Optimization ◽  
Design Variables

Although metaheuristic techniques have recently become popular in optimization, still they are not suitable for computationally expensive real-world problems, specifically when the problems have many input variables. Among these techniques, particle swarm optimization (PSO) is one of the most well-known population-based nature-inspired algorithms which can intelligently search huge spaces of possible arrangements of design variables to solve various complex problems. The candidate solutions and accordingly the required number of evaluated particles, however, dramatically increase with the number of design variables or the dimension of the problem. This study is a major modification to an original PSO for using all previously evaluated points aiming to increase the computational efficiency. For this purpose, a metamodeling methodology appropriate for so-called high-dimensional, expensive, black-box (HEB) problems is used to efficiently generate an approximate function from all particles calculated during the optimization process. Following the metamodel construction, a term named metamodeling acceleration is added to the velocity update formula in the original PSO algorithm using the minimum of the metamodel. The proposed strategy is called the metamodel guided particle swarm optimization (MGPSO) algorithm. The superior performance of the approach is compared with original PSO using several benchmark problems with different numbers of variables. The developed algorithm is then used to optimize the aerodynamic design of a gas turbine compressor blade airfoil as a challenging HEB problem. The simulation results illustrated the MGPSO’s capability to achieve more accurate results with a considerably smaller number of function evaluations.

scholarly journals Variable-Length Particle Swarm Optimization for Feature Selection on High-Dimensional Classification

2021 ◽  
Author(s):  
B Tran ◽  
Bing Xue ◽  
Mengjie Zhang
Keyword(s):  
Feature Selection ◽  
Particle Swarm Optimization ◽  
Computational Cost ◽  
Particle Swarm ◽  
Search Space ◽  
Classification Performance ◽  
Variable Length ◽  
High Dimensional ◽  
Swarm Optimization ◽  
Local Optima

© 1997-2012 IEEE. With a global search mechanism, particle swarm optimization (PSO) has shown promise in feature selection (FS). However, most of the current PSO-based FS methods use a fix-length representation, which is inflexible and limits the performance of PSO for FS. When applying these methods to high-dimensional data, it not only consumes a significant amount of memory but also requires a high computational cost. Overcoming this limitation enables PSO to work on data with much higher dimensionality which has become more and more popular with the advance of data collection technologies. In this paper, we propose the first variable-length PSO representation for FS, enabling particles to have different and shorter lengths, which defines smaller search space and therefore, improves the performance of PSO. By rearranging features in a descending order of their relevance, we facilitate particles with shorter lengths to achieve better classification performance. Furthermore, using the proposed length changing mechanism, PSO can jump out of local optima, further narrow the search space and focus its search on smaller and more fruitful area. These strategies enable PSO to reach better solutions in a shorter time. Results on ten high-dimensional datasets with varying difficulties show that the proposed variable-length PSO can achieve much smaller feature subsets with significantly higher classification performance in much shorter time than the fixed-length PSO methods. The proposed method also outperformed the compared non-PSO FS methods in most cases. © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

scholarly journals Improved Dynamical Particle Swarm Optimization Method for Structural Dynamics

2019 ◽  
Vol 2019 ◽  
pp. 1-11
Author(s):  
W. D. Annicchiarico ◽  
M. Cerrolaza
Keyword(s):  
Particle Swarm Optimization ◽  
Particle Swarm ◽  
Nonlinear Problems ◽  
Optimization Method ◽  
Benchmark Problems ◽  
Structural Behaviour ◽  
Seismic Excitations ◽  
Swarm Optimization ◽  
Building Model ◽  
Optimization Methodology

A methodology to the multiobjective structural design of buildings based on an improved particle swarm optimization algorithm is presented, which has proved to be very efficient and robust in nonlinear problems and when the optimization objectives are in conflict. In particular, the behaviour of the particle swarm optimization (PSO) classical algorithm is improved by dynamically adding autoadaptive mechanisms that enhance the exploration/exploitation trade-off and diversity of the proposed algorithm, avoiding getting trapped in local minima. A novel integrated optimization system was developed, called DI-PSO, to solve this problem which is able to control and even improve the structural behaviour under seismic excitations. In order to demonstrate the effectiveness of the proposed approach, the methodology is tested against some benchmark problems. Then a 3-story-building model is optimized under different objective cases, concluding that the improved multiobjective optimization methodology using DI-PSO is more efficient as compared with those designs obtained using single optimization.

Hybridized Particle Swarm Optimization on Constrained Economic Dispatch Problem

2020 ◽  
Vol 17 (1) ◽  
pp. 322-328 ◽  
Author(s):  
Namarta Chopra ◽  
Y. S. Brar ◽  
J. S. Dhillon
Keyword(s):  
Particle Swarm Optimization ◽  
Large Scale ◽  
Economic Dispatch ◽  
Particle Swarm ◽  
Classical Problem ◽  
Optimization Method ◽  
Small Scale ◽  
Swarm Optimization ◽  
Suggested Approach ◽  
Large Scale System

The hybridization of particle swarm optimization (PSO) with simplex search method (SSM) is presented on the problem of economic dispatch in the thermal plants so as to minimizes the overall operating fuel cost while subjected to various constraints. This hybridization of stochastic with deterministic optimization method helps the global optimum solution to further refine by the local search. It also overcome some of the drawbacks of conventional PSO like premature convergence and stagnation in the solution if the number of iterations are increased. This proposed optimization method is used to get the overall minimum cost of fuel by including transmission line losses and valve point loading effect (VPLE) in the classical problem of economic dispatch, so as to have the more practical impact in the case considered. The validness of the suggested algorithm is tested using small scale and large scale system and the analogy of results obtained are done with existing algorithms cited in the literature, showing improvement of 29.3% in small scale system and 6.4% in large scale system, which proves the robustness of the suggested approach.

scholarly journals Variable-Length Particle Swarm Optimization for Feature Selection on High-Dimensional Classification

2021 ◽  
Author(s):  
B Tran ◽  
Bing Xue ◽  
Mengjie Zhang
Keyword(s):  
Feature Selection ◽  
Particle Swarm Optimization ◽  
Computational Cost ◽  
Particle Swarm ◽  
Search Space ◽  
Classification Performance ◽  
Variable Length ◽  
High Dimensional ◽  
Swarm Optimization ◽  
Local Optima

© 1997-2012 IEEE. With a global search mechanism, particle swarm optimization (PSO) has shown promise in feature selection (FS). However, most of the current PSO-based FS methods use a fix-length representation, which is inflexible and limits the performance of PSO for FS. When applying these methods to high-dimensional data, it not only consumes a significant amount of memory but also requires a high computational cost. Overcoming this limitation enables PSO to work on data with much higher dimensionality which has become more and more popular with the advance of data collection technologies. In this paper, we propose the first variable-length PSO representation for FS, enabling particles to have different and shorter lengths, which defines smaller search space and therefore, improves the performance of PSO. By rearranging features in a descending order of their relevance, we facilitate particles with shorter lengths to achieve better classification performance. Furthermore, using the proposed length changing mechanism, PSO can jump out of local optima, further narrow the search space and focus its search on smaller and more fruitful area. These strategies enable PSO to reach better solutions in a shorter time. Results on ten high-dimensional datasets with varying difficulties show that the proposed variable-length PSO can achieve much smaller feature subsets with significantly higher classification performance in much shorter time than the fixed-length PSO methods. The proposed method also outperformed the compared non-PSO FS methods in most cases. © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Sign in / Sign up

Export Citation Format

Share Document