Deformation Prediction of Concrete Dam Based on Support Vector Machine Optimized by Monarch Butterfly Optimization
Open Access
- 18 October 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1551-1555
Abstract
The establishing of an accurate concrete dam deformation prediction model is a crucial aspect of dam safety monitoring. To this end, a concrete dam deformation prediction method based on support vector machine (SVM) method and selecting optimal parameters through monarch butterfly optimization (MBO) is proposed. In view of the small sample, high dimensional and nonlinear characteristics of dam monitoring data, the SVM method is introduced for predicting dam deformation. To improve the prediction accuracy and stability of SVM, the MBO is utilized to optimize the penalty parameter C and kernel function parameter σ. The solutions of parameters are updated by migration and adjustment operators. Then, a concrete dam deformation prediction model based on the MBO-SVM is established. Taking the measured data of a double-curvature arch dam as an example, the feasibility of the MBO-SVM model is verified by comparing the prediction results with the SVM model.Keywords
This publication has 3 references indexed in Scilit:
- A deep learning prediction model of DenseNet-LSTM for concrete gravity dam deformation based on feature selectionEngineering Structures, 2023
- Monarch butterfly optimizationNeural Computing and Applications, 2015
- Arch Dam Deformation Prediction Model Based on PSO-SVMApplied Mechanics and Materials, 2013