Development of a neural network for controlling the chlorination process of recycled dichloroethane using a virtual analyzer

This article highlights the development of an innovative control system for the chlorination of recycled dichloroethane, based on neural networks and a virtual analyzer. The main purpose of the research is to increase the efficiency and safety of a chemical process critical to the production of a wide range of organochlorine products. Chlorination of dichloroethane, a key step in the production of substances such as chlorinated hydrocarbons, requires careful control of multiple parameters, including temperature, pressure, and chemical composition, which poses significant challenges for traditional control methods. The article discusses in detail the principles of operation and advantages of using neural networks, which enable you to adapt to changing production conditions and optimize parameters in real time. The virtual analyzer developed as part of the research provides accurate measurements of the quality of raw materials and intermediates, an important step towards automating the control and management of chlorination. The use of these technologies contributes not only to increased productivity and product quality but also significantly reduces risks to production personnel and the environment. The research covers the development and testing of a neural network model and the integration of a virtual analyzer with a control system. Both theoretical aspects and practical results of the application of the developed solutions are described