Diagnostic Algorithm Quality of the State of Underground Metal Structures the Using Neural Networks

Keywords

underground metal structures, neural network, technical condition, optimization algorithm quality, method Elman

Abstract

The underground metal structures function in specific conditions of the soil environment and cyclic mechanical loads. It is necessary to conduct a detailed analysis of underground metal structures and monitor their technical condition, since damage and destruction of structural elements during operation can lead to dangerous and/or catastrophic consequences. Intelligent monitoring systems are able to control the life cycle of underground pipelines based on a comprehensive analysis of their current state, operating loads and the results of interaction with the environment.

References

1. Schweidtmann A.M. and Mitsos A. Global Deterministic Optimization with Artificial Neural Networks Embedded // J. Optim. Theory Appl. -2018. -180. – P. 925-948. https://doi.org/10.1007/s10957-018-1396-0
2. Ren G., Cao Y., Wen S., Huang T. and Zeng, Z. A modified Elman neural network with a new learning rate scheme // Neurocomputing. – 2018. – 286. – P. 11-18. https://doi.org/10.1016/j.neucom.2018.01.046
3. Lozovan V.P. Diagnostic algorithm for optimization electrophysical parameters of underground metal constructions taking into account the quality criterion and the method of neural network // International Young Scientists Conference on Materials Science and Surface Engineering MSSE2021 (September 22-24, Lviv, Ukraine): Proc. – Lviv, 2021. – P. 184-187.
4. Dzhala R., Senyuk O. and Lozovan V. Contactless testing of insulation damages distribution of the underground pipelines // Procedia Struct. Integr. – 2022. – 36. – P. 17-23. https://doi.org/10.1016/j.prostr.2021.12.077