«Automation and Informatization of the fuel and energy complex»

Study of the efficiency of evolutionary methods when solving the problem of optimal control of petrochemical process

UDC: 519.6:004.4
DOI: -

Authors:

¹ Ufa University of Science and Technology, Ufa, Russia

Keywords: optimal control, petrochemical process, evolutionary calculations, genetic algorithms, differential evolution

Annotation:

The authors of the article comparatively analyze the differential evolution method and genetic algorithm when solving the problem of petrochemical process optimal control. One of the directions being developed in the field of optimization and optimal control of dynamic processes is evolutionary computing. The advantage of evolutionary methods is the ability to overcome the local extremum due to the mechanisms of crossover and mutation. A general formulation of the optimal control problem with restrictions on the control parameter is given, for the solution of which the method of differential evolution and a genetic algorithm with real coding are described. A computational experiment was carried out on a model example of a petrochemical process. Numerical solutions to problems of optimal control of a petrochemical process are obtained. The convergence and computational complexity of the considered evolutionary methods are analyzed. As a result of comparing the obtained solutions to optimal control problems with analytical solutions found on the basis of Pontryagin's maximum principle, their satisfactory agreement with the exact solution of the problem is shown.

Bibliography:

1. Karpenko A.P. Evolyutsionnye operatory populyatsionnykh algoritmov global'noy optimizatsii. Opyt sistematizatsii // Matematika i mat. modelirovanie. – 2018. – № 1. – S. 59–89. – DOI: 10.24108/mathm.0118.0000103
2. Chislennyy algoritm resheniya zadachi optimal'nogo upravleniya s terminal'nymi ogranicheniyami dlya dinamicheskikh sistem / E.V. Antipina, S.I. Mustafina, A.F. Antipin, S.A. Mustafina // Avtometriya. – 2020. – T. 56, № 6. – S. 132–140. – DOI: 10.15372/AUT20200615
3. Antipina E.V., Mustafina S.A., Antipin A.F. Algorithm of Solving a Multiobjective Optimization Problem on the Basis of a Kinetic Chemical Reaction Model // Optoelectronics, Instrumentation and Data Processing. – 2021. – Vol. 57, Issue 6. – P. 668–674. – DOI: 10.3103/S8756699021060029
4. Trokoz D.A. Metod parametricheskoy optimizatsii dlya shirokikh neyronnykh setey s ispol'zovaniem geneticheskikh algoritmov // Izv. Samarskogo nauch. tsentra RAN. – 2021. – T. 23, № 2(100). – S. 51–56. – DOI: 10.37313/1990-5378-2021-23-2-51-56
5. Balancer Genetic Algorithm – A Novel Task Scheduling Optimization Approach in Cloud Computing / R. Gulbaz, A.B. Siddiqui, N. Anjum [et al.] // Applied Sciences. – 2021. – Vol. 11, Issue 14. – P. 6244. – DOI: 10.3390/app11146244
6. Kochueva O.N. Approksimatsiya koeffitsienta szhimaemosti gaza na osnove geneticheskikh algoritmov // Avtomatizatsiya i informatizatsiya TEK. – 2023. – № 11(604). – S. 59–68. – DOI: 10.33285/2782-604X-2023-11(604)-59-68
7. Panteleev A.V., Bortakovskiy A.S. Teoriya upravleniya v primerakh i zadachakh: ucheb. posobie. – M.: Vyssh. shk., 2003. – 583 s.