王贞, 刘进进, 吴斌. 实时混合试验近完全时滞补偿方法的参数确定[J]. 工程力学, 2014, 31(10): 158-166. DOI: 10.6052/j.issn.1000-4750.2013.05.0386
引用本文: 王贞, 刘进进, 吴斌. 实时混合试验近完全时滞补偿方法的参数确定[J]. 工程力学, 2014, 31(10): 158-166. DOI: 10.6052/j.issn.1000-4750.2013.05.0386
WANG Zhen, LIU Jin-jin, WU Bin. PARAMETER DETERMINATION OF NEARLY-COMPLETE COMPENSATION SCHEME FOR TIME DELAY IN REAL-TIME HYBRID SIMULATION[J]. Engineering Mechanics, 2014, 31(10): 158-166. DOI: 10.6052/j.issn.1000-4750.2013.05.0386
Citation: WANG Zhen, LIU Jin-jin, WU Bin. PARAMETER DETERMINATION OF NEARLY-COMPLETE COMPENSATION SCHEME FOR TIME DELAY IN REAL-TIME HYBRID SIMULATION[J]. Engineering Mechanics, 2014, 31(10): 158-166. DOI: 10.6052/j.issn.1000-4750.2013.05.0386

实时混合试验近完全时滞补偿方法的参数确定

PARAMETER DETERMINATION OF NEARLY-COMPLETE COMPENSATION SCHEME FOR TIME DELAY IN REAL-TIME HYBRID SIMULATION

  • 摘要: 实时混合试验是在拟动力试验方法的基础上发展起来的、能够考虑试件率相关力学性能的新型抗震试验方法,时滞补偿是该方法的重要问题。在前期近完全时滞补偿方法的研究基础上,该文研究了该方法的参数确定原则和该补偿方法的性能。近完全时滞补偿方法通过过预测结构位移响应并提前发送,实现实测位移超前期望位移;接着从实测位移数据中寻找与期望位移最接近者,以及与其对应的实测反力,并作为试件反力反馈到数值积分中。数值模拟表明无法回搜到与期望位移相等的实测位移时,方法可能发散。为了确定回搜跨度,分析了试验中可能出现的情况,针对每种回搜时间跨度区间分析了可能存在的最优反力-期望位移滞回环。进一步的数值模拟和真实试验验证了分析的正确性。试验也表明方法具有较好的精度。

     

    Abstract: Real-time hybrid simulation (RHS) is a novel aseismic testing technique developed from a pseudo-dynamic test. RHS can consider the rate-dependent mechanical properties of the test specimen and a time delay-compensation in these tests is of a great importance. Based on the previous study on a nearly-complete compensation scheme, this paper shows the principle and performance of the parameter determination of the scheme. In this scheme, the displacement of structural response is over-predicted and sent out in advance to ensure a measured displacement to be achieved earlier than a desired displacement; then a measured force corresponding to the measured displacement that is closest to the desired displacement is fed back to the numerical computation. Numerical simulations show that the scheme may be instable when the equality between the measured displacement and the desired displacement cannot be found during the search progress. In order to determine the past time span for search, all possible cases in tests are analyzed. The possible loops between optimal restoring forces and desired displacements for each group of past time spans are classified. The correctness of results has been confirmed through further numerical simulations and actual tests. The effectiveness of the scheme in terms of an error reduction is also concluded from the test results.

     

/

返回文章
返回