【24h】

The philosophy of modelling versus testing

机译:建模与测试的哲学

获取原文
获取原文并翻译 | 示例

摘要

This paper explores the philosophy of modelling and testing. It is argued herein that the objective of modelling is to obtain a prediction, whereas that of testing is to obtain a validation. Through case histories of model studies, it is demonstrated that the requirements of of a model are not always fully achievable. The important thing is for the modeller to know how the simplification he/she is making will affect his/her results. On the other hand, a test need not model a specific or even an idealised prototype; as long as it is definitive. Centrifuge and numerical modelling may be regarded as two complementary as well as competitive approaches to making predictions for geotechnical problems. Numerical modelling with good laboratory and in-situ testing are probably better in defining in-situ state and conditions of the ground, whereas the remoulding and reconsolidation processes involved in preparing centrifuge model may cause the in-situ state of the model to be rather different from that of the prototype. On the other hand, many problems involving very large deformation, break-up of material zones as well as moving and changing interfaces are likely to be much more readily solvable by a centrifuge modelling rather than a numerical modelling approach.
机译:本文探讨了建模和测试的哲学。本文认为,建模的目的是获得预测,而测试的目的是获得验证。通过模型研究的案例历史,可以证明模型的要求并非总是可以完全实现的。对于建模者而言,重要的是要知道他/她所做的简化将如何影响他/她的结果。另一方面,测试不需要建模特定的甚至是理想化的原型。只要是确定的。离心和数值模拟可以被视为对岩土工程问题进行预测的两种互补和竞争性方法。具有良好实验室和原位测试的数值模型可能更好地定义了原位状态和地面条件,而准备离心机模型所涉及的重塑和固结过程可能会导致模型的原位状态大相径庭从原型的。另一方面,许多问题涉及非常大的变形,材料区域破裂以及移动和变化的界面,这些问题很可能通过离心建模而不是数值建模方法来解决。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号