Primenenie iskusstvennogo intellekta dlya analiza protsessa izgotovleniya leynera

Abstract

The paper demonstrates a technique that combines the use of artificial intelligence and the finite element method to analyze the production process of the liner – the seamless part of composite overwrapped pressure vessels obtained by tube forming. Instead of expensive physical experiments, computer simulation was used, which made it possible to quickly obtain "big data", including the main technological parameters of tube forming process, some metal flow features and the tendency for defects to appear. The processing of the data array was carried out in MATLAB using neural networks and a genetic algorithm for the multi-objective optimization problem. As a result, the influence of technological parameters on the formation of the liner geometry was determined, and the optimal variant of their manufacture was proposed, which is of practical interest to companies producing metal-composite cylinders.