Perm has developed an expert system for assessing the condition of buildings

Researchers of Perm Polytechnic University (PNRPU, participant of Perm REC) have developed an expert system that determines the technical condition of buildings and structures based on fuzzy databases. It allows you to reduce the human factor and expand the range of data on the condition of buildings.

Researchers of Perm Polytechnic University (PNRPU, participant of Perm REC) have developed an expert system that determines the technical condition of buildings and structures based on fuzzy databases. It allows you to reduce the human factor and expand the range of data on the condition of buildings.

The technical condition of the building, which is determined during the engineering inspection of structures and structures, is the main criterion in deciding on the degree of emergency of the building or the need for repair for further operation. The traditional assessment of the condition of buildings depends on the level of professionalism and experience of the expert, as well as the instrument base used. Therefore, such estimates always have some degree of error, inconsistency and uncertainty. In addition, the lack of formulated evaluation criteria in regulatory documents and the short GOST scale, which includes only four categories: regulatory, operational, limited operational or emergency, also contribute to uncertainty.

Perm Polytechnic scientists have automated the analysis process by describing a variety of objects, concepts and connections between them.

— The key feature of the theory of fuzzy logic, which distinguishes it from probabilistic methods, frames and neural networks, is that it is logic with well—defined semantics, since linguistic variables are used to evaluate parameters, - explains corresponding member of the Russian Academy of Architecture and Building Sciences, Professor, Doctor of Technical Sciences, Head of the Department of Building Structures and computational mechanics PNRPU Galina Kashevarova.

The theory of fuzzy sets allows you to formalize and process the most heterogeneous information contained in the description of signs of the technical condition of buildings, to model reasoning such as "a lot", "little", "often", "rarely", "about...", "approximately ...", "no less ...", "no more ...", "in the range from ... to...". In total, the program includes more than 90 possible controlled parameters. And the knowledge base of the research module of the system contains more than 5000 rules.

— The assessment of the adequacy and reliability of the solutions issued by the expert system was carried out on the basis of modeling scenarios of various damages of reinforced concrete bent structures, data from a number of technical reports based on the results of an engineering survey of real structures and his own experience, — said Yuri Tonkov, Candidate of Technical Sciences, Associate Professor of the Department of Building Structures and Computational Mechanics.