UDC 004.021
FORECASTING THE STATE OF TECHNICAL SYSTEMS USING GENETIC ALGORITHMS
V. M. Kureichik, Dr. in Technical Sciences, Chief Researcher, Professor of the CAD Department of ICTTI SFedU; This email address is being protected from spambots. You need JavaScript enabled to view it.
Ye. S. Sinyutin, Techno Center SFedU, chief of the sector for the development of medical technolodies and methods to process electro-physiological signals; This email address is being protected from spambots. You need JavaScript enabled to view it.
T. G. Kaplunov, post-graduate student of the department of CADD IUTIU SFU; This email address is being protected from spambots. You need JavaScript enabled to view it.
In this article, a genetic algorithm for predicting the states of technical systems is given. An original approach to forecasting the states of technical systems is described. The approach is to find future values by extrapolating current observation results. Prediction can be considered as a diagnostic control at zero extrapolation time, or as a general diagnosis case. The developed genetic algorithm includes a number of steps, such as generation of the first generation, crossing, mutation, selection. The operators of crossing-over and mutation are modified with respect to the classical ones with the aim of increasing the efficiency of work. The crossover operator is based on the dichotomy method. The break point in the mutation operator corresponds to the number of the Fibonacci series. This approach allows us to minimize the number of random events. The paper presents the results of testing the algorithm on the task of predicting the temperature of central computer processo
Key words: Genetic algorithm, prediction, time series, diagnostic control, chromosome, population, prediction accuracy, crossing-over, mutation, dichotomy, Fibonacci numbers, extrapolation, algorithm.