A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Temperature gradient impacts on concrete-encased steel I-girder: an ANN optimization approach
Using structural health monitoring sensors, the authors have previously conducted an extensive experimental investigation to assess the impact of varying ambient temperature effects exerted by solar radiation. The experimental findings were accumulated during a period of exceptionally cold temperatures. The findings of the inquiry unveiled the varying temperature gradients over time and associated lateral and vertical dissemination. In the present study, it was aimed at bridging this gap by developing these explicit expressions through the application of the artificial neural network (ANN) approach. The experimental investigation complied with ANN’s training, validation, and testing datasets. Using an artificial neural network (ANN), the best predictions for vertical and lateral gradients were obtained by constructing 4 input layers, 40 hidden layers, and 1 output layer. The neural network was trained using the same experimental study approach as earlier. The ANN model’s prediction of the temperature gradient was so precise that its R2 value was more than 0.99. The web temperature gradient, top slab temperature gradient, and bottom slab temperature gradient were all accurately predicted by the ANN model, with respective R2 values of 0.99842, 0.99081, and 0.99798.
Temperature gradient impacts on concrete-encased steel I-girder: an ANN optimization approach
Using structural health monitoring sensors, the authors have previously conducted an extensive experimental investigation to assess the impact of varying ambient temperature effects exerted by solar radiation. The experimental findings were accumulated during a period of exceptionally cold temperatures. The findings of the inquiry unveiled the varying temperature gradients over time and associated lateral and vertical dissemination. In the present study, it was aimed at bridging this gap by developing these explicit expressions through the application of the artificial neural network (ANN) approach. The experimental investigation complied with ANN’s training, validation, and testing datasets. Using an artificial neural network (ANN), the best predictions for vertical and lateral gradients were obtained by constructing 4 input layers, 40 hidden layers, and 1 output layer. The neural network was trained using the same experimental study approach as earlier. The ANN model’s prediction of the temperature gradient was so precise that its R2 value was more than 0.99. The web temperature gradient, top slab temperature gradient, and bottom slab temperature gradient were all accurately predicted by the ANN model, with respective R2 values of 0.99842, 0.99081, and 0.99798.
Temperature gradient impacts on concrete-encased steel I-girder: an ANN optimization approach
Asian J Civ Eng
Sabarigirivasan, L. (author) / Umamaheswari, N. (author)
Asian Journal of Civil Engineering ; 24 ; 3145-3154
2023-12-01
10 pages
Article (Journal)
Electronic Resource
English
Effect of encased concrete on section temperature gradient of corrugated steel web box girder
| SAGE Publications | 2021
Steel-encased-ultra-high-performance concrete reinforcing method for box girder
| European Patent Office | 2021
Bending behavior of concrete-encased composite I-girder with corrugated steel web
| Online Contents | 2014