Optimisation of System Dynamics Models Using a Real-Coded Genetic Algorithm with Fuzzy Control

Cybernetics and Information Technologies

Beklaryan Gayane L.¹  Khachatryan Nerses K.¹  Akopov Andranik S.¹ 

[1] Plekhanov Russian University of Economics, Moscow, Russian Federation;

关键词: real-coded genetic algorithm;    fuzzy control;    system-dynamics;    optimisation methods;   

DOI  :  10.2478/cait-2019-0017

来源: DOAJ

【 摘 要 】

This paper presents a new real-coded genetic algorithm with Fuzzy control for the Real-Coded Genetic Algorithm (F-RCGA) aggregated with System Dynamics models (SD-models). The main feature of the genetic algorithm presented herein is the application of fuzzy control to its parameters, such as the probability of a mutation, type of crossover operator, size of the parent population, etc. The control rules for the Real-Coded Genetic Algorithm (RCGA) were suggested based on the estimation of the values of the performance metrics, such as rate of convergence, processing time and remoteness from a potential extremum. Results of optimisation experiments demonstrate the greater time-efficiency of F-RCGA in comparison with other RCGAs, as well as the Monte-Carlo method. F-RCGA was validated by using well-known test instances and applied for the optimisation of characteristics of some system dynamics models.

【 授权许可】

Unknown