Design of imitating model of titanium sponge production

Ufimtseva Valentina N.

The relevance and practical significance of improving the efficiency of production of titanium sponge in Russia and in the world are substantiated. Description of the areas of magnesium electrolysis and the reduction of titanium sponge as a technological object and control object are given. The problem of joint management of reduction and electrolysis stages in the production of titanium sponge is formulated. A system analysis of the processes in the areas of magnesium electrolysis and the recovery of titanium sponge is carried out. As a result objects of the system and the connections between them are identified. Assumptions about deviations from planned processes that may entail a decrease in efficiency are formulated to verification them by means of a computational experiment. A model of joint work of the plots was built with the previously developed tool for modeling multi-agent systems and queuing systems. As steps of this, the constructors of the base classes have been changed, additional private variables of derived classes have been introduced, special states and conditions for transitions between them have been specified, methods have been written to react to the model time for each object, and other necessary actions have been taken. A problem of achieving the stationary state of the model before the start of computational experiments is solved. The optimal initial conditions are determined to minimize auto-oscillations occur in the system. Software methods for suppressing self-oscillations that are absent on a real object are developed, including the use of configuration files that are different for different numbers of bottom and vacuum buckets in the model. An error of the model is determined based on experimental data which not used in the construction of the model, its error was estimated. It is shown that the model with the error of 7-8% can be applied to solving production control problems.

Keywords:titanium sponge, production, control, imitation modeling