Hrama M.P. Automated control of the evaporation plant based on intelligent regulators. – Qualifying scientific work on manuscript rights.
Dissertation for obtaining the scientific degree of Doctor of Philosophy in specialty 151 "Automation and computer-integrated technologies". - National University of Food Technologies of the Ministry of Education and Science of Ukraine, Kyiv, 2023.
The dissertation is devoted to increasing the efficiency of the evaporation plant to ensure quality indicators of the evaporation process by improving the system of automated control of the evaporation process by developing and including in its composition intelligent regulators and subsystems of forecasting and decision-making support. The study analyzed the process of moisture evaporation from beet juice as an object of automation. In order to increase the efficiency of the automation system, this paper investigates the use of intelligent regulators in the control system of the evaporation plant, which improves the quality indicators of the regulation process.
A model of the evaporation plant of a sugar factory was developed, in which intelligent control was used. Intelligent regulators are widely used in this work. Among such regulators are fuzzy and neural network regulators.
A prediction model for the system of fuzzy and neural network regulation has been developed. The use of intelligent systems in the automation of the beet juice evaporation process involves the emergence of a large number of options for the development of events, some of which may lead to the emergence of emergency and emergency situations. That is why it is very important to prevent their appearance in advance. For this purpose, this work uses forecasting of the state of this system. The technical condition of the evaporation plant is also predicted for a short time. The task of forecasting the technical condition consists in diagnosing a set of indicators of the system condition. Forecasting of the operation of the evaporation plant by the method of fuzzy local trends is carried out with the help of fuzzy time series models.
Algorithms and structures of the system of fuzzy and neural network regulation have been developed. In order for the user to quickly make decisions based on the predicted data, the human-machine interface was developed in such a way that it could provide objectively correct hints to the user. For this purpose, the decision-making subsystem was used. In this work, a method of calculating and forecasting user priorities was chosen for the development of a decision-making subsystem.
The human-machine interface of the system has been developed. Modern software and hardware were used to implement the system. This project uses a Schneider Electric Modicon M340 PLC.
The obtained results can be used in the design, development and implementation of new or in the improvement of the existing systems of automation of sugar factory evaporation plants.
The results of the work were tested at the sugar factories of JSC "Shamrayivskyi Sugar Factory" and Branch "Zhdanivskyi Sugar Factory" LLC "Tsukoragroprom", which was confirmed by relevant certificates.
Keywords: sugar, evaporation, regulation, PI regulator, neural network, juice, model, evaporation plant, neuro-fuzzy, neuro-fuzzy regulators, management, prediction of behavior, steam pressure, automated control, evaporation, syrup, prognostication, model predictive management, optimal management, energy saving, control system, neurocontroller, Neural Network Predictive Controller, state space model, PID regulator, prediction error, automatic adjustment system, automation, management, modeling, control object, simulation model, productivity, energy costs, automated system, algorithm, intellectual, forecasting methods.
