Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.601216
Title: The adaptive reference-state minimal control synthesis algorithm : with application to the control of shaking-tables
Author: Hatano, Toshiaki
ISNI:       0000 0004 5352 6639
Awarding Body: University of Bristol
Current Institution: University of Bristol
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Abstract:
Lessons learnt from the recent catastrophic earthquakes and tsunami across the world motivate research testing using full -scale earthquake shaking-tables. Through this testing avenue, the effects of an earthquake can be reproduced. The dynamic tests involved play significant roles in ensuring the reliability, durability and functionality of engineering systems. This in turn has an impact on lives of people. The fidelity of the results obtained from shaking-tab le tests are largely influenced by the accuracy of the control system. Currently. conventional linear control schemes are widely used on many shaking-tables, but their control performance can significantly deteriorate in the face of dynamic uncertainties, non-linearities and external disturbances. Generally, a feedback control strategy is employed to compensate such disturbances and improve the closed-loop stability of the system. In practice, however, many shaking-tab le facilities have some controller design restrictions: primarily, performance enhancement with in the feedback-loops is explicitly not allowed due to confidentiality and/or safety reasons. Beside these issues, a shaking-tab le system which can generate a large amplitude acceleration (more than -20m/s2)) earthquake wave is in a high demand, due to recent earthquakes recorded in Japan being at such levels of intensity. Nevertheless, most large shaking-tab les cannot reproduce this level of acceleration due to the mechanical limitations in their actuation systems. This thesis describes how a conventional linear controller can be enhanced by a novel, feed forward adaptive control strategy, called the reference-state minimal control synthesis algorithm and its application to the control of shaking-tab les. Furthermore, the development of a new shaking-table configuration. called a multi-stage shaking-table which is used to generate a high-acceleration table motion, is also presented. Comparative simulation and implementation tests show that significant performance improvements are achieved with the new control scheme, in spite of non- linearities and parameter changes in the controlled system.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.601216  DOI: Not available
Share: