COMPARISON BETWEEN PARTICLE SWARM OPTIMIZATION AND GENETIC ALGORITHM IN ARTIFICIAL NEURAL NETWORK FOR LIFE PREDICTION OF NC TOOLS

2008 ◽  
Vol 07 (01) ◽  
pp. 1-7 ◽  
Author(s):  
SHILONG WANG ◽  
FEI ZHENG ◽  
LING XU
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Artificial Neural Network ◽  
Particle Swarm Optimization ◽  
Convergence Rate ◽  
Tool Life ◽  
Life Prediction ◽  
Particle Swarm ◽  
Swarm Optimization ◽  
Artificial Neural

Accurate life prediction of NC (Numeric Control) tools is very essential in an advanced manufacturing system. In this paper, tool life prediction in a drilling process was researched. An Artificial Neural Network (ANN) has been established for prediction, with drill diameter, cutting speed and feed rate as input parameters and tool life as an output parameter. To improve the performance of the network, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) were applied independently to train the network instead of standard Backward Propagation (BP) algorithm, which has drawbacks of low convergence rate and weak generalization capacity. And the two methods were compared in terms of algorithm complexity, convergence rate and prediction accuracy, with reference to standard BP method.

scholarly journals Modeling and optimizing an electrochemical oxidation process using artificial neural network, genetic algorithm and particle swarm optimization

2018 ◽  
Vol 83 (3) ◽  
pp. 379-390
Author(s):  
Banghai Liu ◽  
Chunji Jin ◽  
Jiteng Wan ◽  
Pengfang Li ◽  
Huanxi Yan
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Artificial Neural Network ◽  
Particle Swarm Optimization ◽  
Electrochemical Oxidation ◽  
Back Propagation ◽  
Particle Swarm ◽  
Swarm Optimization ◽  
Operation Conditions ◽  
Artificial Neural

This study proposes a novel hybrid of artificial neural network (ANN), genetic algorithm (GA), and particle swarm optimization (PSO) to model and optimize the relevant parameters of an electrochemical oxidation (EO) Acid Black 2 process. The back propagation neural network (BPNN) was used as a modelling tool. To avoid over-fitting, GA was applied to improve the generalized capability of BPNN by optimizing the weights. In addition, an optimization model was developed to assess the performance of the EO process, where total organic carbon (TOC) removal, mineralization current efficiency (MCE), and the energy consumption per unit of TOC (ECTOC) were considered. The operation conditions of EO were further optimized via PSO. The validation results indicted the proposed method to be a promising method to estimate the efficiency and to optimize the parameters of the EO process.

Sign in / Sign up

Export Citation Format

Share Document