Topology optimization of neural networks based on a coupled genetic algorithm and particle swarm optimization techniques (c-GA–PSO-NN)

2016 ◽  
Vol 29 (11) ◽  
pp. 1073-1076 ◽  
Author(s):  
Azam Marjani ◽  
Saeed Shirazian ◽  
Mehdi Asadollahzadeh
Keyword(s):  
Genetic Algorithm ◽  
Neural Networks ◽  
Particle Swarm Optimization ◽  
Topology Optimization ◽  
Particle Swarm ◽  
Optimization Techniques ◽  
Swarm Optimization

A New Hybrid Genetic Algorithm and Particle Swarm Optimization

2012 ◽  
Vol 498 ◽  
pp. 115-125 ◽  
Author(s):  
H. Hachimi ◽  
Rachid Ellaia ◽  
A. El Hami
Keyword(s):  
Genetic Algorithm ◽  
Particle Swarm Optimization ◽  
Hybrid Algorithm ◽  
Hybrid Genetic Algorithm ◽  
Particle Swarm ◽  
Optimization Techniques ◽  
Hybrid Technique ◽  
Swarm Optimization ◽  
New Generation ◽  
Crossover And Mutation

In this paper, we present a new hybrid algorithm which is a combination of a hybrid genetic algorithm and particle swarm optimization. We focus in this research on a hybrid method combining two heuristic optimization techniques, genetic algorithms (GA) and particle swarm optimization (PSO) for the global optimization. Denoted asGA-PSO, this hybrid technique incorporates concepts fromGAandPSOand creates individuals in a new generation not only by crossover and mutation operations as found inGAbut also by mechanisms ofPSO. The performance of the two algorithms has been evaluated using several experiments.

Hybridization of Biogeography-Based

2013 ◽  
Vol 2 (3) ◽  
pp. 86-101 ◽  
Author(s):  
Provas Kumar Roy ◽  
Dharmadas Mandal
Keyword(s):  
Genetic Algorithm ◽  
Particle Swarm Optimization ◽  
Differential Evolution ◽  
Optimal Power Flow ◽  
Particle Swarm ◽  
Population Based ◽  
Optimization Techniques ◽  
Mixed Integer ◽  
Optimization Approach ◽  
Swarm Optimization

The aim of this paper is to evaluate a hybrid biogeography-based optimization approach based on the hybridization of biogeography-based optimization with differential evolution to solve the optimal power flow problem. The proposed method combines the exploration of differential evolution with the exploitation of biogeography-based optimization effectively to generate the promising candidate solutions. Simulation experiments are carried on standard 26-bus and IEEE 30-bus systems to illustrate the efficacy of the proposed approach. Results demonstrated that the proposed approach converged to promising solutions in terms of quality and convergence rate when compared with the original biogeography-based optimization and other population based optimization techniques like simple genetic algorithm, mixed integer genetic algorithm, particle swarm optimization and craziness based particle swarm optimization.

A hybrid of genetic algorithm and particle swarm optimization for reducing material waste in extrusion-basedadditive manufacturing

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ruiliang Feng ◽  
Jingchao Jiang ◽  
Zhichao Sun ◽  
Atul Thakur ◽  
Xiangzhi Wei
Keyword(s):  
Genetic Algorithm ◽  
Particle Swarm Optimization ◽  
Topology Optimization ◽  
Particle Swarm ◽  
Critical Parameters ◽  
Large Set ◽  
Support Structures ◽  
Swarm Optimization ◽  
Content Type ◽  
Fused Deposition

PurposeThe purpose of this paper is to report the design of a lightweight tree-shaped support structure for fused deposition modeling (FDM) three-dimensional (3D) printed models when the printing path is considered as a constraint. Design/methodology/approachA hybrid of genetic algorithm (GA) and particle swarm optimization (PSO) is proposed to address the topology optimization of the tree-shaped support structures, where GA optimizes the topologies of the trees and PSO optimizes the geometry of a fixed tree-topology. Creatively, this study transforms each tree into an approximate binary tree such that GA can be applied to evolve its topology efficiently. Unlike FEM-based methods, the growth of tree branches is based on a large set of FDM 3D printing experiments. FindingsThe hybrid of GA and PSO is effective in reducing the volume of the tree supports. It is shown that the results of the proposed method lead to up to 46.71% material savings in comparison with the state-of-the-art approaches. Research limitations/implicationsThe proposed approach requires a large number of printing experiments to determine the function of the yield length of a branch in terms of a set of critical parameters. For brevity, one can print a small set of tree branches (e.g. 30) on a single platform and evaluate the function, which can be used all the time after that. The steps of GA for topology optimization and those of PSO for geometry optimization are presented in detail. Originality/valueThe proposed approach is useful for the designers and manufacturers to save materials and printing time in fabricating complex models using the FDM technique. It can be adapted to the design of support structures for other additive manufacturing techniques such as Stereolithography and selective laser melting.

scholarly journals Genetic Algorithm and Particle Swarm Optimization Techniques for Inverted Pendulum Stabilization

2020 ◽  
Vol 9 (6) ◽  
pp. 657-660
Keyword(s):  
Genetic Algorithm ◽  
Particle Swarm Optimization ◽  
Evolutionary Computation ◽  
Inverted Pendulum ◽  
Biological Evolution ◽  
Particle Swarm ◽  
Continuous Variable ◽  
Optimization Techniques ◽  
Swarm Optimization ◽  
Soft Computing Techniques

Inverted Pendulum is a popular non-linear, unstable control problem where implementation of stabilizing the pole angle deviation, along with cart positioning is done by using novel control strategies. Soft computing techniques are applied for getting optimal results. The evolutionary computation forms the key research area for adaptation and optimization. The approach of finding optimal or near optimal solutions to the problem is based on natural evolution in evolutionary computation. The genetic algorithm is a method based on biological evolution and natural selection for solving both constrained and unconstrained problems. Particle swarm optimization is a stochastic search method inspired by collective behavior of animals like flocking of birds, schooling of fishes, swarming of bees etc. that is suited to continuous variable problems. These methods are applied to the inverted pendulum problem and their performance studied.

A Comparative Study of Genetic and Particle Swarm Optimization Algorithms and Their Hybrid Method in Water Flooding Optimization

2018 ◽  
Vol 140 (10) ◽  
Author(s):  
Majid Siavashi ◽  
Mohsen Yazdani
Keyword(s):  
Genetic Algorithm ◽  
Particle Swarm Optimization ◽  
Newton Method ◽  
Oil Recovery ◽  
Optimization Technique ◽  
Particle Swarm ◽  
Optimal Solution ◽  
Optimization Techniques ◽  
Water Flooding ◽  
Swarm Optimization

Optimization of oil production from petroleum reservoirs is an interesting and complex problem which can be done by optimal control of well parameters such as their flow rates and pressure. Different optimization techniques have been developed yet, and metaheuristic algorithms are commonly employed to enhance oil recovery projects. Among different metaheuristic techniques, the genetic algorithm (GA) and the particle swarm optimization (PSO) have received more attention in engineering problems. These methods require a population and many objective function calls to approach more the global optimal solution. However, for a water flooding project in a reservoir, each function call requires a long time reservoir simulation. Hence, it is necessary to reduce the number of required function evaluations to increase the rate of convergence of optimization techniques. In this study, performance of GA and PSO are compared with each other in an enhanced oil recovery (EOR) project, and Newton method is linked with PSO to improve its convergence speed. Furthermore, hybrid genetic algorithm-particle swarm optimization (GA-PSO) as the third optimization technique is introduced and all of these techniques are implemented to EOR in a water injection project with 13 decision variables. Results indicate that PSO with Newton method (NPSO) is remarkably faster than the standard PSO (SPSO). Also, the hybrid GA-PSO method is more capable of finding the optimal solution with respect to GA and PSO. In addition, GA-PSO, NPSO, and GA-NPSO methods are compared and, respectively, GA-NPSO and NPSO showed excellence over GA-PSO.

Sign in / Sign up

Export Citation Format

Share Document