Ilyunin O. Synthesis of an automated system for intelligent control of the process of continuous pickling of strip steel

The dissertation is devoted to the solution of the scientific and technical problem of developing an automated system of intellectual control of the process of continuous pickling of the strip steel with the use of fuzzy and neural network models. Radial-based neural network models designed to work in the contours of the automated intelligent control system of the process of continuous pickling of strip steel allowed to adequately describe the technological process of strip steel in non-stationary operating conditions in the presence of process uncertainties are proposed. A regulator allowing to adaptively adjust the temperature of the pickling solution taking into account the influence of changes in the composition of the solution to maintain the regimented process speed is proposed. A method for identifying the pickling solution components state, using fuzzy clustering for automated control of the content of multi-hydrates in the pickling solution, is proposed. The identification process of strip-rolled surface defects with the use of fuzzy clustering was improved, which allowed to automate the quality control of the final product and to determine the parameters of the pickling solution supply pressure in the pre-irrigation nozzles. The method of assessing the quality of functioning of the technological line of pickling, which uses clustering of factors of the technological process taking into account their mutual influences is improved. Based on the developed intelligent models and methods, a modified structure of the control system of the continuous pickling unit, which as a result allows improving the quality of pickling of the strip steel surface under the conditions of technological resource costs optimization is proposed.