Research on the inversion method of concrete dam deformation parameters based on DE-ABC-SVM.

Saved in:

Bibliographic Details
Title:	Research on the inversion method of concrete dam deformation parameters based on DE-ABC-SVM.
Authors:	Yong, Xiong^1,2, Yang, Cheng¹, Kui, Wang¹ anhuiwk@163.com, Qiang, Chen¹, Mingjie, Zhao^1,3
Source:	Insight: Non-Destructive Testing & Condition Monitoring. Jul2026, Vol. 68 Issue 7, p469-478. 10p.
Subjects:	Parameter estimation, Support vector machines, Finite element method, Metaheuristic algorithms, Bees algorithm, Deformations (Mechanics), Differential evolution, Concrete dams
Abstract:	In view of the limitations of the traditional inversion method for concrete dam deformation parameters, such as the tendency to fall into local optima, slow convergence rates and low inversion accuracy, this paper proposes an improved support vector machine (SVM)-based inversion method for concrete dam deformation parameters. Firstly, finite element analysis is employed to calculate the amount of deformation under various deformation parameter schemes, thereby constructing a dataset that serves as a solid foundation for subsequent algorithm training. The partial least-squares method is subsequently employed to develop an appropriate statistical model for concrete dam deformation. This approach not only enhances the inversion accuracy and interpretability of the model but also isolates the precise hydraulic component. Then, by analysing the strengths and limitations of the artificial bee colony (ABC) algorithm and the differential evolution (DE) algorithm, the two algorithms are integrated into a hybrid algorithm to enhance performance. This hybrid approach is further applied to improve the SVM, addressing issues such as poor algorithm stability, susceptibility to local optima and low search efficiency. Finally, the proposed deformation parameter inversion method is applied to the Longgang Reservoir. The results show that, under the same modelling framework, the inversion error of the original SVM model exceeds 40%, whereas the optimised differential evolution-artificial bee colony-support vector machine (DE-ABC-SVM) model reduces the relative inversion error to around 10%, indicating a clear improvement in inversion performance. [ABSTRACT FROM AUTHOR]
	Copyright of Insight: Non-Destructive Testing & Condition Monitoring is the property of British Institute of Non-Destructive Testing and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	In view of the limitations of the traditional inversion method for concrete dam deformation parameters, such as the tendency to fall into local optima, slow convergence rates and low inversion accuracy, this paper proposes an improved support vector machine (SVM)-based inversion method for concrete dam deformation parameters. Firstly, finite element analysis is employed to calculate the amount of deformation under various deformation parameter schemes, thereby constructing a dataset that serves as a solid foundation for subsequent algorithm training. The partial least-squares method is subsequently employed to develop an appropriate statistical model for concrete dam deformation. This approach not only enhances the inversion accuracy and interpretability of the model but also isolates the precise hydraulic component. Then, by analysing the strengths and limitations of the artificial bee colony (ABC) algorithm and the differential evolution (DE) algorithm, the two algorithms are integrated into a hybrid algorithm to enhance performance. This hybrid approach is further applied to improve the SVM, addressing issues such as poor algorithm stability, susceptibility to local optima and low search efficiency. Finally, the proposed deformation parameter inversion method is applied to the Longgang Reservoir. The results show that, under the same modelling framework, the inversion error of the original SVM model exceeds 40%, whereas the optimised differential evolution-artificial bee colony-support vector machine (DE-ABC-SVM) model reduces the relative inversion error to around 10%, indicating a clear improvement in inversion performance. [ABSTRACT FROM AUTHOR]
ISSN:	13542575
DOI:	10.1784/insi.2026.68.7.469