PROGNOSIS ANALYSIS OF ARCH DAM BEHAVIOR BY HYBRID MODEL
Date Issued
2022
Author(s)
Frosina Panovska, Stevcho Mitovski, Ljupcho Petkovski
Abstract
The assessment of the structural stability and behavior of the dam during construction, at full reservoir and during the service period is of paramount importance for such structures. Each dam, in dependence of the type and dimensions, has installed system for technical monitoring that enables tracking of the dam behavior and assessment of the dam state
throughout registration and interpretation of various data such as displacements, stresses, seepage etc. The case study is double curvature concrete arch dam, with asymmetric shape due to the valley topography. In the paper are systemized acknowledgments for the comparison of the recorded data from technical monitoring of the dam and output results from spatial (3D) numerical model, created by application by application of SOFiSTiK code, based on Finite Element Method. In addition, a Neural Network model is created by recorded data from the technical monitoring. The both models create the so called hybrid model. For both models are used input parameters such as variation of water level in the reservoir and air temperature, and the output results are compared with recorded values for water level in piezometers at specified nodes of the dam. The aim of the task is to compare and calibrate the output results from the both models and recorded values from the technical monitoring of the dam and to carry out prognosis modeling for the future behavior of the dam.
throughout registration and interpretation of various data such as displacements, stresses, seepage etc. The case study is double curvature concrete arch dam, with asymmetric shape due to the valley topography. In the paper are systemized acknowledgments for the comparison of the recorded data from technical monitoring of the dam and output results from spatial (3D) numerical model, created by application by application of SOFiSTiK code, based on Finite Element Method. In addition, a Neural Network model is created by recorded data from the technical monitoring. The both models create the so called hybrid model. For both models are used input parameters such as variation of water level in the reservoir and air temperature, and the output results are compared with recorded values for water level in piezometers at specified nodes of the dam. The aim of the task is to compare and calibrate the output results from the both models and recorded values from the technical monitoring of the dam and to carry out prognosis modeling for the future behavior of the dam.
Subjects
File(s)![Thumbnail Image]()
Loading...
Name
15 wmhe2022.pdf
Size
2.88 MB
Format
Adobe PDF
Checksum
(MD5):086e1a1a5cbac95713b4b948d3194508