Hybrid multi-objective differential evolution (H-MODE) for optimisation of polyethylene terephthalate (PET) reactor

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Multi-objective evolutionary algorithms (MOEAs) are used to solve the optimisation problems with more than one objective to be optimised simultaneously having conflict among each other. Due to the limitations of traditional deterministic algorithms to handle complex and nonlinear search space, several EAs are developed in the recent past. The multi-objective differential evolution (MODE) algorithm is already tested and found to be a reliable algorithm due to their ability to handle non-linear problems efficiently. Though MODE is accurate in terms of converging to the global Pareto front, traditional method has their advantage in terms of speed. We combined these two algorithms and developed hybrid strategy of MODE thus, achieving both accuracy and speed. Hybrid MODE (H-MODE) algorithm is applied on multi-objective optimisation of industrial wiped film polyethylene terephthalate reactor. The results of the present study are compared with those obtained using MODE algorithm. Smooth and well diverse Pareto optimal front is observed with a much faster speed using H-MODE.

Original languageEnglish
Pages (from-to)213-221
Number of pages9
JournalInternational Journal of Bio-Inspired Computation
Volume2
Issue number3-4
DOIs
Publication statusPublished - 2010

Fingerprint

Differential Evolution
Polyethylene terephthalates
Reactor
Optimization
Differential Evolution Algorithm
Pareto Front
Multi-objective Evolutionary Algorithm
Deterministic Algorithm
Multiobjective optimization
Multi-objective Optimization
Evolutionary algorithms
Search Space
Nonlinear Problem
Optimization Problem

Keywords

  • Bio-inspired computation
  • EAs
  • Evolutionary algorithms
  • H-MODE
  • Hybrid algorithms
  • Mode
  • Modelling and simulation
  • MOO
  • Multi-objective differential evolution
  • Multi-objective optimisation
  • Pareto front
  • Polyethylene terephthalate reactor

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

@article{3f9c677efe2f4b8ab49f3b1f3a198a12,
title = "Hybrid multi-objective differential evolution (H-MODE) for optimisation of polyethylene terephthalate (PET) reactor",
abstract = "Multi-objective evolutionary algorithms (MOEAs) are used to solve the optimisation problems with more than one objective to be optimised simultaneously having conflict among each other. Due to the limitations of traditional deterministic algorithms to handle complex and nonlinear search space, several EAs are developed in the recent past. The multi-objective differential evolution (MODE) algorithm is already tested and found to be a reliable algorithm due to their ability to handle non-linear problems efficiently. Though MODE is accurate in terms of converging to the global Pareto front, traditional method has their advantage in terms of speed. We combined these two algorithms and developed hybrid strategy of MODE thus, achieving both accuracy and speed. Hybrid MODE (H-MODE) algorithm is applied on multi-objective optimisation of industrial wiped film polyethylene terephthalate reactor. The results of the present study are compared with those obtained using MODE algorithm. Smooth and well diverse Pareto optimal front is observed with a much faster speed using H-MODE.",
keywords = "Bio-inspired computation, EAs, Evolutionary algorithms, H-MODE, Hybrid algorithms, Mode, Modelling and simulation, MOO, Multi-objective differential evolution, Multi-objective optimisation, Pareto front, Polyethylene terephthalate reactor",
author = "Gujarathi, {Ashish M.} and Babu, {B. V.}",
year = "2010",
doi = "10.1504/IJBIC.2010.033089",
language = "English",
volume = "2",
pages = "213--221",
journal = "International Journal of Bio-Inspired Computation",
issn = "1758-0366",
publisher = "Inderscience Publishers",
number = "3-4",

}

TY - JOUR

T1 - Hybrid multi-objective differential evolution (H-MODE) for optimisation of polyethylene terephthalate (PET) reactor

AU - Gujarathi, Ashish M.

AU - Babu, B. V.

PY - 2010

Y1 - 2010

N2 - Multi-objective evolutionary algorithms (MOEAs) are used to solve the optimisation problems with more than one objective to be optimised simultaneously having conflict among each other. Due to the limitations of traditional deterministic algorithms to handle complex and nonlinear search space, several EAs are developed in the recent past. The multi-objective differential evolution (MODE) algorithm is already tested and found to be a reliable algorithm due to their ability to handle non-linear problems efficiently. Though MODE is accurate in terms of converging to the global Pareto front, traditional method has their advantage in terms of speed. We combined these two algorithms and developed hybrid strategy of MODE thus, achieving both accuracy and speed. Hybrid MODE (H-MODE) algorithm is applied on multi-objective optimisation of industrial wiped film polyethylene terephthalate reactor. The results of the present study are compared with those obtained using MODE algorithm. Smooth and well diverse Pareto optimal front is observed with a much faster speed using H-MODE.

AB - Multi-objective evolutionary algorithms (MOEAs) are used to solve the optimisation problems with more than one objective to be optimised simultaneously having conflict among each other. Due to the limitations of traditional deterministic algorithms to handle complex and nonlinear search space, several EAs are developed in the recent past. The multi-objective differential evolution (MODE) algorithm is already tested and found to be a reliable algorithm due to their ability to handle non-linear problems efficiently. Though MODE is accurate in terms of converging to the global Pareto front, traditional method has their advantage in terms of speed. We combined these two algorithms and developed hybrid strategy of MODE thus, achieving both accuracy and speed. Hybrid MODE (H-MODE) algorithm is applied on multi-objective optimisation of industrial wiped film polyethylene terephthalate reactor. The results of the present study are compared with those obtained using MODE algorithm. Smooth and well diverse Pareto optimal front is observed with a much faster speed using H-MODE.

KW - Bio-inspired computation

KW - EAs

KW - Evolutionary algorithms

KW - H-MODE

KW - Hybrid algorithms

KW - Mode

KW - Modelling and simulation

KW - MOO

KW - Multi-objective differential evolution

KW - Multi-objective optimisation

KW - Pareto front

KW - Polyethylene terephthalate reactor

UR - http://www.scopus.com/inward/record.url?scp=84555193788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84555193788&partnerID=8YFLogxK

U2 - 10.1504/IJBIC.2010.033089

DO - 10.1504/IJBIC.2010.033089

M3 - Article

AN - SCOPUS:84555193788

VL - 2

SP - 213

EP - 221

JO - International Journal of Bio-Inspired Computation

JF - International Journal of Bio-Inspired Computation

SN - 1758-0366

IS - 3-4

ER -