COMPUTER SCIENCE, COMPUTER ENGINEERING AND MANAGEMENT SYNTHESISING THE MODEL OF THE PREDICTIVE MAINTENANCE OF ATMOSPHERIC PIPE STILL

Abstract

Objectives. The aim of the study is to develop a model of a predictive maintenance system allowing the output parameters of a high-capacity distillation unit to be stabilised.Method. The problem of controlling the collection of oil products of a given fractional composition can be reduced to the task of controlling the temperature regime at draw-off trays. In this case, temperature control tasks are considered as control actions. In order to implement the proposed control principle, a mathematical model was constructed allowing the oil fraction content to be converted to directly controlled temperatures.Results. Based on the analysis of the technological process and design features of the atmospheric distillation column, the vectors of maintaining and disturbing actions are determined along with the optimisation criterion for the maintenance problem. In order to improve the dynamic characteristics of the control object, oil product quality parameters are introduced into the predictive maintenance model along with the supplementation of existing control loops with intermediate parameters.Conclusion. An analysis of the structural features of the atmospheric pipe still and the existing approach to the distillation column maintenance reveals certain disadvantages associated with the lack of operational quality control of the collected oil products, as well as the need to expand the list of stabilised technological parameters in order to optimise the technological process. It is shown that in order to switch to the proposed maintenance approach, it is necessary to develop a static model reflecting the dependency of the potential yield of the required fractions on the raw material composition along with a dynamic model in the form of a multi-parameter optimisation problem.

Keywords

• нефтепереработка
• управление по модели
• статическая и динамическая модель
• ректификация