"Artificial Intelligence" or "Logical Discussion and Reasonable Solutions" in Technical Diagnostics of Construction Projects

Abstract

The purpose of this work is to show the possibilities and prospects for the use of intelligent technologies in construction. Currently, the most widely used technologies for this purpose are expert system (ES) technology, the theory of fuzzy sets and fuzzy logic, and artificial neural networks. Artificial intelligence systems make it possible to model the accumulated professional knowledge of experienced specialists, relying on modern methods for solving practical problems, using the existing regulatory framework of the construction industry as a ready-made system model, which should be adequately presented in terms formalized by certain structural rules. The computer allows the expert to autonomously use this, suggesting possible options based on logical inference to solve specific problems. The ability to identify and control in an automated way the cause-and-effect relationship between the characteristics and condition of structures has been implemented, providing support for the decision of the survey engineer in determining the category of the technical condition of structures. An integral part of the diagnostics of construction projects is verification calculations of the bearing capacity of elements, taking into account their real condition. For calculations of complex construction projects, the numerical finite element method is usually used. Considering that artificial neural networks are good approximators of various nonlinear dependencies, today they are proposed to be interpreted as a new universal approach to the numerical solution of problems of mathematical physics.