MATHEMATICAL MODELING OF THE GAS PIPELINE OF A GAS CLEANING SYSTEM IN STEEL PRODUCTION

Abstract

A gas cleaning system (GCS) is a technological complex boiler-utilizer-gas cleaning-smoke exhauster consisting of a number of interconnected subsystems containing numerous controls. Flue gas purification before its emission by a smoke exhauster into the atmosphere is a complex technological process [1]. Heat separation in gas purification systems is a key task. In this regard, strict requirements are imposed on its operation, which are recognized by the quality of the gas to be cleaned and the performance of the separation unit. The task of the gas purification shop is to remove dust from contaminated gas while ensuring stable operation of the equipment. The stable oper- ation of the entire system affects the quality of the gas being cleaned, the economic efficiency of the installed equipment, repair and maintenance costs, and the cost of air emissions [7]. For optimal system operation, it is necessary to ensure smooth process control. For optimal system operation, it is necessary to ensure smooth process control. As a result of the experiment on removing the temporary characteristic, a disturbing effect was applied to the gas pipeline - a stepwise change in the recycled water flow rate relative to the nominal one by 8 %, from 170 m3/h to 185 m3/h. To determine these values, an experimental curve of the object acceleration through the channel “circulating water flow - temperature of contaminated gas at the inlet to the venturi pipes” was obtained. Different smoothing methods are used to extract the actual transient response. For smoothing of values in this case the method of moving averaging is used [8] Approximation - replacing the graph with mathematical expressions. Dynamic proper-ties of the control object are characterized by differential expressions, transition and transfer functions, frequency characteristics, between which there is an unambiguous dependence. When calculating automatic control systems, it is convenient to represent the mathematical model as a transfer characteristic. It can be obtained as a result of approximation of the time characteristic. A large number of methods have been developed to analyze the transient response in order to obtain the transfer function of a linear control object [3]. The essence of the methods is to determine the coefficients of the transfer function of a pre-selected form, the basis of which is to obtain the calculated characteristic that best matches the experimental one. There are several approximation methods: graphical and logarithmic, area method, method of solving differential equations, etc. The calculation is carried out using a computer. The initial data for the calculation are the experimental transient response of the object, given in the form of equidistant ordinates in time, and the input signal value. To approximate the transient response of this object, we use the Simoy method [6, 9]. The Simoy method is a universal approximation method that allows obtaining approximating expressions of any order. This method is very convenient for computer processing, it is easily algorithmized and has great accuracy. As a result of the approximation, the transfer function of the object, i.e. its mathematical model, is obtained.