Martin Zlatić, Felipe Rocha, Laurent Stainier, Marko Čanađija
{"title":"计算力学的数据驱动方法:基于神经网络的方法与无模型方法的公平比较","authors":"Martin Zlatić, Felipe Rocha, Laurent Stainier, Marko Čanađija","doi":"arxiv-2409.06727","DOIUrl":null,"url":null,"abstract":"We present a comparison between two approaches to modelling hyperelastic\nmaterial behaviour using data. The first approach is a novel approach based on\nData-driven Computational Mechanics (DDCM) that completely bypasses the\ndefinition of a material model by using only data from simulations or real-life\nexperiments to perform computations. The second is a neural network (NN) based\napproach, where a neural network is used as a constitutive model. It is trained\non data to learn the underlying material behaviour and is implemented in the\nsame way as conventional models. The DDCM approach has been extended to include\nstrategies for recovering isotropic behaviour and local smoothing of data.\nThese have proven to be critical in certain cases and increase accuracy in most\ncases. The NN approach contains certain elements to enforce principles such as\nmaterial symmetry, thermodynamic consistency, and convexity. In order to\nprovide a fair comparison between the approaches, they use the same data and\nsolve the same numerical problems with a selection of problems highlighting the\nadvantages and disadvantages of each approach. Both the DDCM and the NNs have\nshown acceptable performance. The DDCM performed better when applied to cases\nsimilar to those from which the data is gathered from, albeit at the expense of\ngenerality, whereas NN models were more advantageous when applied to wider\nrange of applications.","PeriodicalId":501309,"journal":{"name":"arXiv - CS - Computational Engineering, Finance, and Science","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Data-driven methods for computational mechanics: A fair comparison between neural networks based and model-free approaches\",\"authors\":\"Martin Zlatić, Felipe Rocha, Laurent Stainier, Marko Čanađija\",\"doi\":\"arxiv-2409.06727\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We present a comparison between two approaches to modelling hyperelastic\\nmaterial behaviour using data. The first approach is a novel approach based on\\nData-driven Computational Mechanics (DDCM) that completely bypasses the\\ndefinition of a material model by using only data from simulations or real-life\\nexperiments to perform computations. The second is a neural network (NN) based\\napproach, where a neural network is used as a constitutive model. It is trained\\non data to learn the underlying material behaviour and is implemented in the\\nsame way as conventional models. The DDCM approach has been extended to include\\nstrategies for recovering isotropic behaviour and local smoothing of data.\\nThese have proven to be critical in certain cases and increase accuracy in most\\ncases. The NN approach contains certain elements to enforce principles such as\\nmaterial symmetry, thermodynamic consistency, and convexity. In order to\\nprovide a fair comparison between the approaches, they use the same data and\\nsolve the same numerical problems with a selection of problems highlighting the\\nadvantages and disadvantages of each approach. Both the DDCM and the NNs have\\nshown acceptable performance. The DDCM performed better when applied to cases\\nsimilar to those from which the data is gathered from, albeit at the expense of\\ngenerality, whereas NN models were more advantageous when applied to wider\\nrange of applications.\",\"PeriodicalId\":501309,\"journal\":{\"name\":\"arXiv - CS - Computational Engineering, Finance, and Science\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Computational Engineering, Finance, and Science\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.06727\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Computational Engineering, Finance, and Science","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.06727","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
我们对利用数据模拟超弹性材料行为的两种方法进行了比较。第一种方法是一种基于数据驱动计算力学(DDCM)的新方法,它完全绕过了材料模型的定义,只使用模拟或实际实验的数据进行计算。第二种是基于神经网络(NN)的方法,即使用神经网络作为构成模型。神经网络通过数据训练来学习基本的材料行为,其实现方式与传统模型相同。DDCM 方法已扩展到包括恢复各向同性行为和局部平滑数据的策略。NN 方法包含某些执行原则的元素,如材料对称性、热力学一致性和凸性。为了对这两种方法进行公平比较,它们使用相同的数据,解决相同的数值问题,并选择一些问题来突出每种方法的优缺点。DDCM 和 NN 的性能都可以接受。DDCM 在应用于与收集数据的情况相似的情况时表现更好,尽管牺牲了一般性;而 NN 模型在应用于更广泛的情况时更具优势。
Data-driven methods for computational mechanics: A fair comparison between neural networks based and model-free approaches
We present a comparison between two approaches to modelling hyperelastic
material behaviour using data. The first approach is a novel approach based on
Data-driven Computational Mechanics (DDCM) that completely bypasses the
definition of a material model by using only data from simulations or real-life
experiments to perform computations. The second is a neural network (NN) based
approach, where a neural network is used as a constitutive model. It is trained
on data to learn the underlying material behaviour and is implemented in the
same way as conventional models. The DDCM approach has been extended to include
strategies for recovering isotropic behaviour and local smoothing of data.
These have proven to be critical in certain cases and increase accuracy in most
cases. The NN approach contains certain elements to enforce principles such as
material symmetry, thermodynamic consistency, and convexity. In order to
provide a fair comparison between the approaches, they use the same data and
solve the same numerical problems with a selection of problems highlighting the
advantages and disadvantages of each approach. Both the DDCM and the NNs have
shown acceptable performance. The DDCM performed better when applied to cases
similar to those from which the data is gathered from, albeit at the expense of
generality, whereas NN models were more advantageous when applied to wider
range of applications.