The work begins with the validation of the hydraulic behavior of SAVA via experiment using both single- and two-degree-of-freedom test setups. The work demonstrates that the behavior of the SAVA remains closely tied to the Reynolds number of flow through a valve orifice. In addition the results indicate that for an exact analysis of SAVA, the frictional force between the piston body and the cylinder body should also be included.

This dissertation is devoted to three objectives: (1) the development of a method to estimate the damping characteristics of a hydraulic semiactive vibration absorber (SAVA), (2) the determination of bi-state control gains for a SAVA controller, and (3) the selection of an effective control scheme (among various control algorithms) for the closed loop operation of a system equipped with a SAVA.

The dissertation next reports the results of an investigation of how best to craft a Lyapunov control for the SAVA system. An analytical way to determine the control gains of a bi-state controller is presented. The stability of the SAVA system is also considered and it shows that, given a bounded disturbance, the SAVA system will remain stable. Finally, the work reports experimental verification of a bi-state controller using a two-degree-of-freedom (TDOF) system.

While significant advances have been made in the study of semiactive hydraulic systems, there is still a need to further explore the issue of how to control a semiactive actuator. One key question that has hereto for remained unanswered is; among the many control rules posed for use with semiactive system, which gives the best performance. The purpose of this dissertation is to provide a comprehensive answer to that question. That task was accomplished here. To the author's knowledge, this is the first time. A comprehensive comparison of competing semiactive bi-state controllers has appeared in the open literature.

Finally, the dissertation reports on a comparative analysis of various control architectures for the SAVA system. The work contrasts the performance of heuristic as well as analytical control algorithms.