Abstract
The search for resource-saving technologies implies not only research in the field of new methods of processing raw materials, but also new approaches to managing technological processes.Another important way of resource saving is the ability to quickly assess the technical condition of equipment, which will allow to timely prevent the emergency state of the main and auxiliary technological equipment.The paper deals with the development of intelligent control systems and diagnostics of processes using fuzzy logic, neural networks and hybrid models for creating control algorithms.The subject of the research is control algorithms developed to determine the key variables of the copper melting process in a liquid bath and algorithms for evaluating the technical condition of turbine units.All algorithms are developed on the basis of PFE matrices compiled by experienced technology experts during a "thought" experiment.The resulting models are carefully studied and analyzed for their sensitivity, stability, and single-valuedness.The absolute error of discrepancy between experimental and calculated variables became the criterion of adequacy.Finally, the conducted researches have shown high efficiency of the control algorithms, obtained by using the artificial intelligence methods.In comparison with classical methods of building analytical and statistical models, methods based on the knowledge and experience of human experts allow creating optimal control systems for complex technological processes significantly easier, faster and more efficient.