Item request has been placed! ×
Item request cannot be made. ×
loading  Processing Request

Utjecaja parametara algoritma diferencijalne evolucije na njegovu učinkovitost ; Impact of the parameters of differential evolution on its efficiency

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
  • معلومة اضافية
    • Contributors:
      Bajer, Dražen
    • بيانات النشر:
      Sveučilište Josipa Jurja Strossmayera u Osijeku. Fakultet elektrotehnike, računarstva i informacijskih tehnologija Osijek. Zavod za programsko inženjerstvo. Katedra za programske jezike i sustave.
      Josip Juraj Strossmayer University of Osijek. Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Department of Software Engineering. Chair of Programming Languages and Systems.
    • الموضوع:
      2025
    • Collection:
      Repository of the University of Osijek
    • نبذة مختصرة :
      Ovaj rad bavi se istraživanjem diferencijalne evolucije, jednog od evolucijskih algoritama koji se pokazao vrlo učinkovit za probleme kontinuirane optimizacije. Rad uključuje analizu učinkovitosti algoritma diferencijalne evolucije pri različitim postavkama parametara. Parametri koje se postavljaju su veličina populacije, faktor skaliranja i stopa križanja. Detaljno su opisani parametri, te su navedene česte postavke istih. Opisuje se ostvareno programsko rješenje, te način rada samog rješenja. Eksperimentalna analiza provedena je na tri unimodalne i tri multimodalne funkcije, te su korištene dvije dimenzionalnosti u svrhu istraživanja ponašanja algoritma u rješavanju različitih problema. Analizirali su se rezultati, s posebnim fokusom na prosječne rezultate i standardne devijacije. Zaključno, rad pokazuje kako promjene u parametrima algoritma mogu utjecati na njegovu učinkovitost, te pruža smjernice o odabira postavki parametara ovisno o problemu. ; This thesis investigates differential evolution, one of the evolutionary algorithms that has proven highly effective for continuous optimization problems. The paper includes an analysis of the efficiency of the differential evolution algorithm under various parameter settings. The parameters considered are population size, scaling factor, and crossover rate. The parameters are described in detail, and common settings are provided. The implemented software solution is described, as well as its functionality. The experimental analysis was conducted on three unimodal and three multimodal functions, with two different dimensionalities, in order to explore the algorithm's behavior in solving various problems. The results were analyzed with a particular focus on average outcomes and standard deviations. In conclusion, the thesis demonstrates how changes in the algorithm's parameters can affect its efficiency, and provides guidelines for selecting parameter settings depending on the problem.
    • File Description:
      application/pdf
    • Relation:
      https://repozitorij.unios.hr/islandora/object/etfos:5508; https://urn.nsk.hr/urn:nbn:hr:200:449388; https://repozitorij.unios.hr/islandora/object/etfos:5508/datastream/PDF
    • الدخول الالكتروني :
      https://repozitorij.unios.hr/islandora/object/etfos:5508
      https://urn.nsk.hr/urn:nbn:hr:200:449388
      https://repozitorij.unios.hr/islandora/object/etfos:5508/datastream/PDF
    • Rights:
      http://rightsstatements.org/vocab/InC/1.0/ ; info:eu-repo/semantics/openAccess
    • الرقم المعرف:
      edsbas.9D7064A1