首页> 外文会议>International Conference on Advances in Experimental Structural Engineers >UNBALANCED FORCE COMPENSATION IN SUBSTRUCTURE TESTING BASED ON ONLINE SYSTEM IDENTIFICATION
【24h】

UNBALANCED FORCE COMPENSATION IN SUBSTRUCTURE TESTING BASED ON ONLINE SYSTEM IDENTIFICATION

机译:基于在线系统识别的子结构测试中的不平衡力补偿

获取原文

摘要

Concerning algorithms for substructure tests, there are some advanced methods that can deal effectively with interaction between experimental and numerical substructures such as analog feedback (Thewalt et al. 1987), digital feedback (Dorka et al. 1990) or predictor-corrector (Ghaboussi et al. 2005), etc. The main advantage of these methods is that accuracy of implicit schemes can be benefited. In this research, the digital feedback algorithm is used for real-time substructure test. In substructure tests with digital feedback, error of algorithm comes from integration scheme and sub step control. A suitable implicit integration scheme with proper parameters for a certain application should be used in order to ensure stability and critical accuracy. Besides the use of integration scheme, error in sub step control should be treated by choosing an adequately small value for sub step time. The error due to sub step control would theoretically approach null, if the sub step time interval approaches zero. However, a problem is that the sub step time interval cannot be too small because of large numerical effort, digital error, measuring and control limitations. The non-zero error of sub step control is an unbalanced force which should be compensated to improve the accuracy of substructure tests. This paper introduces an unbalanced force compensation method that is based on online system identification in time domain. Since the unbalanced force is very well related to the dynamic response of the system, this force can be estimated and then compensated in every step. Numerical results and tests of some example models with earthquake loads show the applicability of the compensation method. In addition, convergence and other properties of the method are also discussed.
机译:关于子结构测试的算法,有一些高级方法可以有效处理实验和数值子结构之间的相互作用,例如模拟反馈(Thewalt等人,1987),数字反馈(Dorka等,1990)或预测校正器(Ghaboussi eT al。2005)等。这些方法的主要优点是隐式方案的准确性可能受益。在本研究中,数字反馈算法用于实时子结构测试。在具有数字反馈的子结构测试中,算法误差来自集成方案和子步骤控制。应使用具有适当参数的合适的隐式集成方案,以确保稳定性和批判性准确性。除了使用集成方案之外,应通过选择用于子步进时间的充分小值来处理子步骤控制中的误差。如果子步骤时间间隔接近零,则由于子步管控制引起的错误将理论上接近null。然而,问题是由于数值努力,数字误差,测量和控制限制,子步骤时间间隔不能太小。子步骤控制的非零误差是不平衡的力,应补偿以提高子结构测试的准确性。本文介绍了一种不平衡力补偿方法,基于时域中的在线系统识别。由于不平衡力与系统的动态响应非常好,因此可以估计该力,然后在每个步骤中得到补偿。具有地震载荷的一些示例模型的数值结果和测试,显示了补偿方法的适用性。另外,还讨论了该方法的收敛性和其他性质。

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