Analysis of methods for increasing the level of automation in an aerodynamic product designer’s workplace

Abstract

Increasing the level of automation of production processes has always led to a signifi-cant increase in production efficiency, which in turn has caused the rapid economic develop-ment of society. The field of designing products with certain aerodynamic characteristics has many areas that ensure the production of a wide range of goods. Cars, airplanes, UAVs are the most popular areas that require complex and multi-level calculations to create products with such aerodynamic characteristics that are acceptable for the relevant tasks and pur-poses. In aerodynamics, the main mathematical apparatus is the Navier-Stokes equation, this is a partial differential equation (PDE), in addition, this equation is one of the unsolved "millennium problems". This shows how complex this field of production is from a computational point of view. Although it has already undergone automation at various stages, the phase of designing the aerodynamic shape of the product still requires the time of a narrowly specialized professional, because modeling products with the necessary characteristics still remains mostly a manual process. With the development of machine learning, these complex processes can be largely automated. This article presents a comparative analysis of various developments in the field of automation of aerodynamic product design. The introduction of technologies such as PINNs and quantum computing at different stages of aerodynamic product design, based on the results of the comparative analysis, was identified as the most promising direction for further research. The general architecture of an improved version of the information system of the automated workplace of the aerodynamic product designer is also proposed. In future research, this architecture may undergo appropriate changes and improvements.