New chaotic flower pollination algorithm for unconstrained non-linear optimization functions

Flower pollination algorithm (FPA) is susceptible to local optimum and substandard precision of calculations. Chaotic operator (CO), which is used in local algorithms to optimize the best individuals in the population, can successfully enhance the properties of the flower pollination algorithm. A ne...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of system assurance engineering and management Jg. 9; H. 4; S. 853 - 865
Hauptverfasser: Kaur, Arvinder, Pal, Saibal K., Singh, Amrit Pal
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New Delhi Springer India 01.08.2018
Springer Nature B.V
Schlagworte:
Algorithms
Engineering
Engineering Economics
Functions (mathematics)
Logistics
Marketing
Mathematical analysis
Operators (mathematics)
Optimization
Organization
Original Article
Quality Control
Reliability
Safety and Risk
Chaos theory
Function optimization
Non-linear functions
Flower pollination algorithm
ISSN:0975-6809, 0976-4348
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Flower pollination algorithm (FPA) is susceptible to local optimum and substandard precision of calculations. Chaotic operator (CO), which is used in local algorithms to optimize the best individuals in the population, can successfully enhance the properties of the flower pollination algorithm. A new chaotic flower pollination algorithm (CFPA) has been proposed in this work. Further FPA and its four proposed variants by using different chaotic maps are tested on nine mathematical benchmark functions of high dimensions. Proposed variants of CFPA are CFPA1, CFPA2, CFPA3 and CFPA4. The result of the experiment indicates that the proposed chaotic flower pollination variant CFPA2 could increase the precision of minimization of function value and CPU time to run an algorithm.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0975-6809
0976-4348
DOI:10.1007/s13198-017-0664-y