Optimized inversion method for thermal parameters of concrete dam under the insulated condition
Chen Bo,Wang Qingyi,Chen Weinan,Gu Hao
DOI: https://doi.org/10.1016/j.engappai.2023.106898
IF: 8
2023-11-01
Engineering Applications of Artificial Intelligence
Abstract:It is crucial to study how thermal characteristics of concrete dam change over time, especially in cold regions, in order to guarantee long-term safety of the engineering projects. In this manuscript, an inverse analysis method by coupling the numerical simulation model and dam safety monitoring data, is proposed to solve the problem that true thermal parameters of the whole structure cannot be obtained by laboratory experiments or field tests of single points. Moreover, a numerical equivalence method of the dam and its insulation layer is introduced, in order to overcome the problem that the thickness of the insulation layer is much smaller than that of the concrete dam, resulting in low accuracy of the simulation model. Thirdly, the particle swarm optimization is introduced, in order to solve the ill-posed problems occurring during the solution process of multi-parameter inverse analysis. Considering the problem that the original particle swarm algorithm is easy to fall into local optimum and stuck edge, this study improves particles velocity and inertia weights. Benchmark function tests are applied to show the performance compared to several typical optimized algorithms. Numerical simulation and engineering verification are used to demonstrate the rationality and feasibility of the proposed method. The cases show that the residual box diagram of two typical measuring points calculated by inversion value is smaller than the median line and mean point calculated by design value. Moreover, the change law of the temperature field of the finite element calculation using the inversion values is closer to the actual measured value than that of the design value. The results indicate that the multi-parameter inverse method is feasible in insulated dam engineering. In addition, the result of the improved PSO optimization converges faster and has a higher fitness value than that of the original algorithm. It suggests that the improved PSO algorithm for thermal parameters inversion has fitting accuracy and rationality for insulated concrete dams in cold regions.
automation & control systems,computer science, artificial intelligence,engineering, electrical & electronic, multidisciplinary