SGER: Reliability-Based Control Algorithms for Semi-Active and Regenerative Systems with Application to Civil Engineering

该项目的最终目标是为土木工程应用开发先进的半主动和再生控制策略。为了使这种控制技术在土木工程从业者中得到更广泛的接受，首先必须证明，与更简单、成本更低的系统相比，与这种设备的设计和安装相关的投资是合理的，因为结构可靠性不断提高。研究结果提供了一种严格的方法来设计控制器，使其在极端地震事件中具有最佳的可靠性。因此，这项工作的基本贡献是一个控制器设计工具，它可以对给定控制设备实现的“最佳可能”可靠性以及实现它的控制律进行明确评估。本文提出的探索性研究是对该领域进行全面研究调查的第一步。本研究开发的控制器设计方法不同于其他现有方法，因为它们将可靠性评估直接纳入控制系统设计优化中。这种新方法代表了理解这些机电系统可用于提高结构可靠性的程度的重要一步。由于这个问题本质上是非线性的，所以数学上的计算是令人生畏的。这项研究也是第一个在统一的分析框架中同时处理半主动和再生系统的研究。基于系统可靠性的半主动和再生控制系统性能评估基础的发展，为该技术在许多新领域的应用打开了大门。这将导致改进的控制策略，为广泛的系统，可靠性是重要的。本研究将为在考虑经济因素的情况下定量评价结构响应控制的效益提供依据。这项研究的结果还将导致土木工程课程中结构分析和控制的理解方式发生根本性变化，从而导致机电一体化更大程度地融入该领域。

Project SummaryThe ultimate goal of this project is the development of advanced semi-active and regenerative control strategies for civil engineering applications. In order for such control technologies to gain wider acceptance by practitioners in civil engineering, it must first be shown that the investment associated with the design and installation of such devices is justified by consistent improvement in structural reliability, in comparison to simpler and less costly systems. The results of this research provide a rigorous method by which controllers may be designed for optimal reliability during extreme seismic events. Thus, the fundamental contribution of this work is a controller design tool which yields a clear assessment of the "best possible" reliability achievable with a given control device, as well as the control law which achieves it. The exploratory research proposed herein constitutes the first step toward a full research investigation in this area. The controller design methodologies developed in this research differ from other existing approaches in that they incorporate reliability assessment directly into the control system design optimization. This new approach represents a significant step toward understanding of the extent to which these mechatronic systems may be used to enhance the reliability of structures. Because of the fundamentally nonlinear nature of this problem, the mathematics are formidable. This research is also the first study to simultaneously treat semi-active and regenerative systems in a unified, analytical framework. The development of a systematic reliability-based foundation for evaluation of the performance of semi-active and regenerative control systems opens the door for application of this technology to many new areas. This will lead to improved control strategies for a wide range of systems for which reliability is of importance. This research will provide a basis for quantitatively evaluating the benefits of use of structural response control taking into account economic considerations. The results of this research will also lead to fundamental changes in the way structural analysis and control is understood taught in civil engineering courses, thereby leading to greater integration of mechatronics into this field.