A New Technique for Seismic Analyses of Nonlinear System

大多数用于解决工程问题的技术，如系统辨识，对于线性系统都是有效的。然而，在非线性系统的某些工程应用中，必须考虑非线性效应。此外，一些系统的输入载荷，如土壤或建筑、海上平台等，可以被视为随机的，通常难以用传统方法预测或表征。该项目旨在开发一种通用的系统辨识方法来表征和分析具有未知和不可测量的随机输入的非线性系统。所提出的研究将扩展和验证随机减量技术的有效性，以涵盖非平稳情况。这样的应用使得使用该程序来分析非线性系统问题成为可能。此外，在不知道随机输入的情况下，系统的非线性形式和系统特征将从测量的系统响应中识别出来。这一一般方法的有效性将通过实验室模拟进行调查。开发的新技术将提高我们调查和预测土壤和建筑物的地震反应的能力，同时考虑到土壤-建筑系统的非线性效应。

Most of the techniques used in solving engineering problems such as system identification are valid for linear systems. However, in certain engineering applications on nonlinear systems one must consider nonlinear effects. In addition input loading on some systems such as soil or building, offshore platforms etc., can be viewed as random and are usually difficult to predict or characterize by traditional means. This project is to develop a general system identification approach to characterize and analyze nonlinear systems with unknown and unmeasured random input. The proposed studies will extend and verify the effectiveness of the random decrement technique to cover the non-stationary cases. Such applications make it possible to use this procedure to analyze nonlinear system problems. In addition the nonlinear form of the system and system signature will be identified from the measured system responses without knowing the random input. The effectiveness of this general approach will be investigated by laboratory simulation. The new technique developed would improve our ability to investigate and predict the seismic response of soil and building taking into account the nonlinear effect of the soil-building system.