Naslov
Type-2 fuzzy adaptive particle swarm optimization

Autori
Galzina, Vjekoslav ; Lujić, Roberto

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the 1st Regional Conference - Mechatronics in Practice and Education (MECH-CONF 2011) / Anišić, Zoran ; Stankovski, Stevan - Subotica : Subotica Tech - College of Applied Sciences, 2011, 162-167

ISBN
978-86-85409-67-7

Skup
1st Regional Conference - Mechatronics in Practice and Education (MECH-CONF 2011)

Mjesto i datum
Subotica, Srbija, 08.12.2011. - 10.12.2011

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
fuzzy logic; type-2 fuzzy set; adaptation; particle swarm optimization

Sažetak
Aim of this work is to evaluate and model adaptive variant particle swarm optimization, PSO with help of fuzzy logic. Fuzzy logic is used for selection and ranking of particles of sub-swarms of particles and swarm in general (one particle represents one optimization solution). Type-1 fuzzy sets, T1-FS and type-2 fuzzy sets, T2-FS are used respectively. The membership function design of type-1 fuzzy logic system, T1-FLS leads to the difficulty of fuzzy rules database construction when representing linguistic part of knowledge. With type-2 fuzzy logic system, T2-FLS as the extension of T1-FLS, we can effectively improve knowledge representation by using the footprint of uncertainty of the membership functions. Because of this feature T2-FLS shows more flexibility than T1-FLS. For evaluation of the proposed model series of tests are made and results are compared with proven optimization algorithms (genetic algorithm, simulated annealing, particle swarm optimization …) on both single objective optimization problems. Results are shown to be comparative to other observe algorithm.

Izvorni jezik
Engleski

Znanstvena područja
Strojarstvo, Temeljne tehničke znanosti, Informacijske i komunikacijske znanosti

POVEZANOST RADA