Optimization of cutting parameters in longitudinal turning using genetic algorithms and simulated annealing

Tema diplomskog rada je primjena genetskih algoritama i simuliranog žarenja u optimiranju parametara obrade kod uzdužnog tokarenja. U uvodu je ukratko opisana važnost projektiranja tehnološkog procesa. U drugom poglavlju teoretski su obrađeni parametri tehnološkog procesa i kvaliteta obrađene površine. U trećem poglavlju opisana su osnovna načela projektiranja tehnološkog procesa. Četvrto poglavlje bavi se eksperimentima koji su provedeni u diplomskom radu. U radu je primjenjen centralno kompozitni plan eksperimenata te je detaljnije opisan. Matematički model dobiven je metodom odzivne površine pomoću programa Design Expert. U ovom poglavlju prikazan je postupak dobivanja matematičkog modela, te 3D prikazi hrapavosti obrađene površine u ovisnosti o parametrima obrade. Također, dobivena jednadžba odzivne površine testirana je na 24 dodatna eksperimenta koji su provedeni unutar područja eksperimenta, ali u drugim mjernim točkama. U petom poglavlju izvedeno je optimiranje matematičkog modela iz četvrtog poglavlja. Najprije su teoretski opisane metode optimizacije i njihova podjela. Zatim su detaljnije opisane metode genetskog algoritma i simuliranog žarenja, kao metode kojima će se vršiti optimiranje. Za optimizaciju je korišten programski paket Mathlab. Na kraju su prikazani optimalni parametri obrade za kontinuirane nezavisne varijable, te vrijednost hrapavosti površine za iste parametre.The topic of the master's thesis is optimization of cutting parameters in longitudinal turning using genetic algorithms and simulated annealing. The first chapter briefly describes the importance of technological process design. The second chapter theoretically deals with the parameters of the technological process and the quality of the surface roughness. The third chapter theoretically describes technological process design. The fourth chapter deals with the processing of data obtained experimentally. The experimental plans are described. The central composite design was used as the experimental plan and it is described in more detail. The mathematical model was obtained by the response surface method using the Design Expert program. This chapter presents the steps for obtaining a mathematical model, and 3D representations of the surface roughness depending on the processing parameters. Also, the previously obtained response surface equation was tested on 24 additional experiments conducted within the experimental area but at other measurement points. In the fifth chapter, the optimization of the mathematical model from the fourth chapter is performed. First, optimization methods and their division are theoretically described. Then, the methods of genetic algorithm and simulated annealing are described in more detail, as methods by which optimization will be performed. The Mathlab software package was used for optimization. Finally, the optimal cutting parameters for continuous independent variables and the surface roughness value for the same parameters are presented