Development and experimental validation of a new control strategy considering device dynamics for large-scale MR dampers using real-time hybrid simulation

Anthony Joseph Friedman, Purdue University

Abstract

This dissertation focuses on the development, evaluation, and validation of a new semi-active control strategy for use with large-scale magnetorheological dampers in structural control applications through real-time hybrid testing. As MR control devices increase in scale for use in real-world civil engineering applications, their dynamics become increasingly complicated. Control designs that are able to take these characteristics into account will be more effective in achieving good performance. A new control algorithm, ODCOC, which utilizes over- and back-driving current control to increase the efficacy of the control device, is proposed. To validate the performance of the new controller under general earthquake excitation and uncertain conditions, a series of three large-scale validation experiments (using large-scale 200kN MR Dampers and steel frames) are performed, including: (1) a three-story linear structure real-time hybrid simulation (RTHS), with a large-scale steel frame and MR damper as tbe physical substructure; (2) a nine-story linear structure realtime hybrid simulation, with a large-scale steel frame and MR damper(s) as the physical substructure; and (3) a non-linear nine-story structure numerical simulation analysis. The performance of the proposed controller is compared to several establish MR damper control methods in each experiment. The main contributions of this research are twofold: (1) RTHS is validated a viable testing technique for large-scale applications, as comparisons to numerical simulations and repeatability testing demonstrate the reliability of this method, and (2) the controller experiments demonstrate the improved structural performance that results when using the ODCOC algorithm compared to other established methods. The ODCOC is able to exceed or match the performance of the other methods in every evaluation category while utilizing less force to accomplish it. From this work, the merits of real-time hybrid testing are demonstrated using large-scale, significant structural components as the physical substructure. RTHS is validated as an acceptable alternative test method. Based on the results of the control performance validation study, using the ODCOC approach in conjunction large- or fullscale MR devices increases the ability of the device to respond in a timely manner to excitation and yields improved global structural responses under seismic loading. Any projects, including both academic and professional engineering applications, that utilize devices at these scales will benefit from employing this control method.

Degree

Ph.D.

Advisors

Dyke, Purdue University.

Subject Area

Civil engineering

COinS