|
advertisement |
|
|
|
|
|
|
Industrial Automatic Control Systems and Controllers Annotation << Back
Development of a neural network to control the polymerization of isoprene in an isopentane solution using a virtual analyzer |
Ye.A. Muraveva, V.V. Bitkulov, A.I. Nikolaeva
Importance. The process of polymerization of isoprene in an isopentane solution makes it possible to obtain a semi-product for the production of synthetic rubber. Isoprene rubbers are general purpose rubbers. They are used instead of natural, both independently and in combination with other elastomers in the manufacture of almost all rubber products. Isoprene rubbers containing non-coloring and non-toxic stabilizers are used for the manufacture of medical devices, rubbers in contact with food, and consumer goods. Special types of SKI-3 are used for the production of vacuum rubbers and in the cable industry for the manufacture of electrical insulation. On the basis of SKI-3, isoprene rubber latex is obtained, used for the manufacture of sponge rubbers and various film products. To control the quality of the technological process, the development of a virtual analyzer is required. A neural network controller, a predicted virtual analyzer of the quality indicator of the fl ow rate and temperature of the catalytic complex will increase the adaptability of the system.
Objectives. In the course of this work, it is necessary to develop a neural network for regulating parameters such as the temperature of the catalytic complex and the fl ow rate using a virtual analyzer model. The resulting models will increase the degree of automation and control of the technological process.
Methods. In the process of studying the problem of improving the quality of the product obtained during the polymerization of isoprene in an isopentane solution, methods of studying the mathematical model of the virtual analyzer and designing its working model in the Simulink Matlab environment were used.
Results. As a result of the project, a virtual analyzer was developed that continuously predicts the process quality indicator in the form of a mass fraction of the catalytic complex after the polymerizer and a neural network for regulating the technological parameters of isoprene polymerization in an isopentane solution. Conclusions. It is concluded that the combined work of the virtual analyzer and the neural network allows to improve the quality of the technological process, as well as the degree of its automation.
Keywords: analyzer, network, neural network, polymerization, catalytic complex.
DOI: 10.25791/asu.8.2022.1380
Pp. 21-29. |
|
|
|
Last news:
Выставки по автоматизации и электронике «ПТА-Урал 2018» и «Электроника-Урал 2018» состоятся в Екатеринбурге Открыта электронная регистрация на выставку Дефектоскопия / NDT St. Petersburg Открыта регистрация на 9-ю Международную научно-практическую конференцию «Строительство и ремонт скважин — 2018» ExpoElectronica и ElectronTechExpo 2018: рост площади экспозиции на 19% и новые формы контент-программы Тематика и состав экспозиции РЭП на выставке "ChipEXPO - 2018" |