区间算法在结构健康检查中的应用

INTERVAL ARITHMETIC FOR INSPECTING HEALTH OF AN EXISTING STRUCTURE

在线健康检查中,当应用实测振动数据时,如何处理由实测数据引起的不确定性则成为关键问题之一。文中建议和发展了结构健康检查的区间理论,并用实例介绍了健康检查的具体过程。由测试数据得到健康特征量,并以区间量或区间向量形式给出,是健康特征量的非概率建模。健康特征量的估计过程是区间估计问题,估计过程遵循区间算法,它适合于处理非线性问题。以一栋新建5层楼房的抗震性能评估为实例,介绍了评估过程中区间量的表示方法及区间评估准则。用第一层层间刚度系数为健康特征量,用地脉动的测试数据估计第一阶模态频率和振型的区间上限和下限,由于测点的限制,完整的振型与健康特征量一起被估计,并用层间刚度系数的区间值进行抗震评估。

The vibration measurement data are frequently employed for on -line health inspecting of a structure. How to handle the uncertainty in data is a key problem. The interval expressing and arithmetic for the uncertainty in the health inspecting of structure are proposed and developed. To get the symptoms from measurement data an interval estimation procedure, which is in accordance with the rules of interval arithmetic is presented. The measure for assessment of health degree is given by an interval quantity or interval vector. It does not belong to the statistic modeling of the uncertainty, but is available to deal with the non-linear problems. The seismic evaluation of a 5-stories building is introduced to show the procedure of using the interval arithmetic. The stiffness coefficient of the first floor is taken as the symptom of health inspecting. To forecast this stiffness coefficient, the first order modal shape is simultaneously identified together with the symptom estimation.