Robust Structural Control for Hazard Mitigation Using Probable and Possible Models

该项目将研究两种方法来控制结构在环境和小到中等动态载荷下的设计。第一种方法是使用“硬边界”或规范来描述结构的可能模型及其相应的可能预测误差集，其目标是在给定的一组可能激励下最大化控制器的鲁棒性能。在第二种方法中，概率分布放置在可能的模型集和可能的激励上。这种控制设计的概率模型方法在第一种方法的基础上引入了一个额外的维度，它将产生关于控制器可能性能的额外信息，例如实现各种性能级别的概率。实验数据将用于更新和完善控制设计模型，以提高闭环性能。形状记忆合金作动器的使用也将进行研究。主动和被动结构控制的改进可以挽救生命，减少结构破坏和经济损失。* * *

9503370 Balas The project will examine two approaches to control design for structures subjected to environmental as well as small- to-moderate dynamic loads. The first approach is to use "hard-bounds" or norms to describe the possible models for a structure and its corresponding set of possible prediction errors with a goal of maximizing the robust performance of the controller for a given set of possible excitations. In the second approach probability distributions are placed over the sets of possible models as well as over the possible excitations. This probable-model approach to control design introduces an extra dimension to the first approach that will yield additional information about the possible performance of the controller, such as the probability of achieving various levels of performance. Experimental data will be used to update and refine the control design models to improve the close-loop performance. The use of shape memory alloy actuators also will be investigated. Improvements in active and passive structural control can save lives and reduce structural damage and economic losses. ***