Levin S. Binary Automaton Model for Simulation of Behavior of Dynamic Processes with Large Quantity of Interacting Objects

The research subject is dynamic processes with large quantity of objects. The research purpose is increase of simulation productivity for dynamic processes with large quantity of interacting objects by developing new event-driven model and effective method of its realization. The investigations are based on the research methods of theory of finite automatons, statistical physics, linear algebra, computational geometry and methods of molecular dynamics. The results of computer simulation are validated with using methods of mathematical statistics and theory of similarity. The productivity of the approach developed is confirmed by comparison of results of computational and real experiments with results of experiments on the model. For the first time developed are the binary automaton model (BA-model) in which objects are united only by binary connections; computational method that realizes the BA-model; the events division into classes of “true” and “false” ones is introduced. Improved are the event-driven method for gas-dynamics flows simulation; method of conversion of real values to model ones. The methods for building the flow lines and level lines while researching the gas-dynamic flows are further developed. The practical importance of obtained results consists in development of the simulation modeling template library for practical realization of the BA-model and the software for gas-dynamic flows simulation. The following practically important problems of gas-dynamic flows simulation are solved: the pre-ignition state of stationary plasma engine SPD-20; the propagation of gas-powder fire-extinguishing mixture in the confined space of tunnel.