Abstract
<jats:p>An increase in the number of gas deposits in Russia with a high content of acidic components (hydrogen sulfide, carbon dioxide, etc.), as well as stricter environmental requirements for commercial gas, are of paramount im-portance. Along with the fact that such purification processes make it possible to obtain commercial gas with a high degree of purification, there are a number of problems during their operation. The determination of the mutual influence of the main technological parameters of the purification process (gas consumption, amine concentration, defoamer consumption, pressure, temperature, degree of saturation, etc.), as well as the influence of extraneous factors (mechanical impurities, decomposition of amine, thermostable salts, surfactants, etc.), based on statistical data and using computer technology allowed formulate the main critical parameters. The dependence form of the predictive equation and the determination of the reliability of the obtained relationships were determined on the basis of multifactorial regression analysis using standard functions of the S-PLUS program. The array of initial information is represented by a number of observations, represented by seven variables: the content of impurities in the mixture, the absorption rate, the density of the solution, viscosity at 25 °C, surface tension, and the content of thermostable salts. Based on the analysis, the article provides predictive equations for changes in the main performance indicators of the absorbent due to the content of impurities in it, as well as other interfering components. The ability of the technological installation's maintenance personnel to analyze dependencies and graphical relationships makes it possible to stabilize the process and conduct it without deviating from technological regulations. The information resource “Analysis of absorbent consumption at desulfurization plants” allows you to quickly obtain information about the amount of circulating working solution, as well as analyze consumption and losses at any time interval within the database.</jats:p>