Configuring advanced evolutionary algorithms for multicriteria building spatial design optimisation

Autor: Sjonnie Boonstra, Michael Emmerich, Thomas Bäck, H Herm Hofmeyer, Koen van der Blom
Přispěvatelé: Applied Mechanics and Design
Jazyk: angličtina
Rok vydání: 2017
Předmět:
Zdroj: CEC
2017 IEEE Congress on Evolutionary Computation (CEC) Proceedings, June 5-8, 2017, Donostia-San Sebastián, Spain, 1803-1810
STARTPAGE=1803;ENDPAGE=1810;TITLE=2017 IEEE Congress on Evolutionary Computation (CEC) Proceedings, June 5-8, 2017, Donostia-San Sebastián, Spain
Popis: In this paper solution approaches for solving the building spatial design optimisation problem for structural and energy performance are advanced on multiple fronts. A new initialisation operator is introduced to generate an unbiased initial population for a tailored version of SMS-EMOA with problem specific operators. Improvements to the mutation operator are proposed to eliminate bias and allow mutations consisting of multiple steps. Moreover, landscape analysis is applied in order to explore the landscape of both objectives and investigate the behaviour of the mutation operator. Parameter tuning is applied with the irace package and the Mixed Integer Evolution Strategy to find improved parameter settings and explore tuning with a relatively small number of expensive evaluations. Finally, the performances of the standard and tailored SMS-EMOA algorithms with tuned parameters are compared.
Databáze: OpenAIRE
načítá se...