A DEM Parameter Calibration Method Based on BP Neural Network and Genetic Algorithm

IF 3.6 2区工程技术 Q2 ENGINEERING, GEOLOGICAL

International Journal for Numerical and Analytical Methods in Geomechanics Pub Date : 2025-08-26 DOI:10.1002/nag.70043

Yaodong Ni, Xianlun Leng, Ruirui Wang, Fengmin Xia, Feng Wang, Chengtang Wang

{"title":"A DEM Parameter Calibration Method Based on BP Neural Network and Genetic Algorithm","authors":"Yaodong Ni, Xianlun Leng, Ruirui Wang, Fengmin Xia, Feng Wang, Chengtang Wang","doi":"10.1002/nag.70043","DOIUrl":null,"url":null,"abstract":"<div>\n \n <p>The discrete element method (DEM) represents a crucial numerical simulation approach for investigating the internal damage mechanisms of rocks. However, in order to construct an accurate simulation model, it is essential to set the correct microscopic parameters. Consequently, parameter calibration has emerged as a key area of focus within this field. The existing parameter calibration methods have yielded satisfactory results; however, there is still scope for further improvement and advancement. In this study, a novel intelligent parameter calibration method has been proposed, combining the benefits of the BP neural network and genetic algorithm (GA). The method constructs a parameter relationship model with micro-parameters as inputs and macro-parameters as outputs. Then GA is employed to invert the relationship model to calculate the parameter calibration. The results demonstrate that the method is capable of calculating a set of high-precision micro-parameter solutions in a mere 2 min, with the majority of its errors being within 5%.</p>\n </div>","PeriodicalId":13786,"journal":{"name":"International Journal for Numerical and Analytical Methods in Geomechanics","volume":"49 16","pages":"3897-3916"},"PeriodicalIF":3.6000,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal for Numerical and Analytical Methods in Geomechanics","FirstCategoryId":"5","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/nag.70043","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, GEOLOGICAL","Score":null,"Total":0}

引用次数: 0

Abstract

The discrete element method (DEM) represents a crucial numerical simulation approach for investigating the internal damage mechanisms of rocks. However, in order to construct an accurate simulation model, it is essential to set the correct microscopic parameters. Consequently, parameter calibration has emerged as a key area of focus within this field. The existing parameter calibration methods have yielded satisfactory results; however, there is still scope for further improvement and advancement. In this study, a novel intelligent parameter calibration method has been proposed, combining the benefits of the BP neural network and genetic algorithm (GA). The method constructs a parameter relationship model with micro-parameters as inputs and macro-parameters as outputs. Then GA is employed to invert the relationship model to calculate the parameter calibration. The results demonstrate that the method is capable of calculating a set of high-precision micro-parameter solutions in a mere 2 min, with the majority of its errors being within 5%.

Abstract Image

查看原文本刊更多论文

基于BP神经网络和遗传算法的DEM参数标定方法

离散元法（DEM）是研究岩石内部损伤机制的重要数值模拟方法。然而，为了构建准确的仿真模型，必须设置正确的微观参数。因此，参数校准已成为该领域的重点领域。现有的参数标定方法取得了满意的结果；然而，仍有进一步改进和进步的余地。本研究结合BP神经网络和遗传算法的优点，提出了一种新的智能参数标定方法。该方法构建了以微观参数为输入，宏观参数为输出的参数关系模型。然后利用遗传算法对关系模型进行反演，计算参数定标。结果表明，该方法能够在2分钟内计算出一组高精度的微参数解，其大部分误差在5%以内。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Journal for Numerical and Analytical Methods in Geomechanics 工程技术-材料科学：综合

CiteScore

6.40

自引率

12.50%

发文量

160

审稿时长

9 months

期刊介绍： The journal welcomes manuscripts that substantially contribute to the understanding of the complex mechanical behaviour of geomaterials (soils, rocks, concrete, ice, snow, and powders), through innovative experimental techniques, and/or through the development of novel numerical or hybrid experimental/numerical modelling concepts in geomechanics. Topics of interest include instabilities and localization, interface and surface phenomena, fracture and failure, multi-physics and other time-dependent phenomena, micromechanics and multi-scale methods, and inverse analysis and stochastic methods. Papers related to energy and environmental issues are particularly welcome. The illustration of the proposed methods and techniques to engineering problems is encouraged. However, manuscripts dealing with applications of existing methods, or proposing incremental improvements to existing methods – in particular marginal extensions of existing analytical solutions or numerical methods – will not be considered for review.