A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Enhancing chloride concentration prediction in marine concrete using conjugate gradient-optimized backpropagation neural network
The paper proposes a novel approach for predicting surface chloride penetration in marine concrete using a faster and more efficient Backpropagation Neural Network (BPNN) model trained with the Conjugate Gradient Method (CGM) to optimize the weights and biases of the Artificial Neural Network (ANN). Hyper-parameter tuning has been employed to optimize the number of hidden layers and neurons per hidden layer simultaneously, resulting in improved accuracy and faster convergence compared to conventional Gradient Descent methods. The optimized BPNN model has been tested and validated using real-world data through tenfold cross-validations. The model's performance has been evaluated using various statistical metrics, including mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). The results demonstrate that the proposed model outperforms conventional methods and other state-of-the-art models, achieving an R2 value of 0.91, an MAE of 0.11, and an RMSE of 0.15. In addition, partial dependence analysis has been performed to analyze the influence of the features on the output. The proposed approach can be an effective tool for predicting the service life of marine concrete structures and optimizing their maintenance and repair schedules. In summary, this research paper presents a comprehensive and reliable solution to predict surface chloride penetration in marine concrete structures with improved accuracy and efficiency, while providing insights into the importance of input features.
Enhancing chloride concentration prediction in marine concrete using conjugate gradient-optimized backpropagation neural network
The paper proposes a novel approach for predicting surface chloride penetration in marine concrete using a faster and more efficient Backpropagation Neural Network (BPNN) model trained with the Conjugate Gradient Method (CGM) to optimize the weights and biases of the Artificial Neural Network (ANN). Hyper-parameter tuning has been employed to optimize the number of hidden layers and neurons per hidden layer simultaneously, resulting in improved accuracy and faster convergence compared to conventional Gradient Descent methods. The optimized BPNN model has been tested and validated using real-world data through tenfold cross-validations. The model's performance has been evaluated using various statistical metrics, including mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). The results demonstrate that the proposed model outperforms conventional methods and other state-of-the-art models, achieving an R2 value of 0.91, an MAE of 0.11, and an RMSE of 0.15. In addition, partial dependence analysis has been performed to analyze the influence of the features on the output. The proposed approach can be an effective tool for predicting the service life of marine concrete structures and optimizing their maintenance and repair schedules. In summary, this research paper presents a comprehensive and reliable solution to predict surface chloride penetration in marine concrete structures with improved accuracy and efficiency, while providing insights into the importance of input features.
Enhancing chloride concentration prediction in marine concrete using conjugate gradient-optimized backpropagation neural network
Asian J Civ Eng
Tipu, Rupesh Kumar (author) / Panchal, V. R. (author) / Pandya, K. S. (author)
Asian Journal of Civil Engineering ; 25 ; 637-656
2024-01-01
20 pages
Article (Journal)
Electronic Resource
English
Application of backpropagation neural network for the prediction of slope movement
| British Library Conference Proceedings | 2001
A Pavement Condition Rating Model using Backpropagation Neural Network
| Online Contents | 1995