Transactions of the Institute of Systems, Control and Information Engineers

Papers

Control System Synthesis to Prevent Undesirable Transient Responses after a Failure for Safety

Koichi Suyama, Noboru Sebe

2010 Volume 23 Issue 4 Pages 65-73
Published: 2010
Released on J-STAGE: April 22, 2011

DOIhttps://doi.org/10.5687/iscie.23.65

JOURNAL FREE ACCESS

Show abstractHide abstract

This paper presents control system synthesis to prevent undesirable transient responses after a device failure for safety against deviation from control range. It is based on multiobjective design for possible contexts and switchings by a failure to optimize the normal-case performance.

View full abstract

Download PDF (167K)
Identification of Crack Profiles using Database and Greedy Search Inverse Analysis Arising in Eddy Current Testing

Thanh-Duong Nguyen, Fumio Kojima

2010 Volume 23 Issue 4 Pages 74-82
Published: 2010
Released on J-STAGE: April 22, 2011

DOIhttps://doi.org/10.5687/iscie.23.74

JOURNAL FREE ACCESS

Show abstractHide abstract

This paper is concerned with an optimization method to recover crack profiles from measurement signals obtained by eddy current testing. This inversion technique uses Greedy search algorithm to find the optimum of least-square error between measurement signals and simulation data. In creating simulation data, a database called ECT database is used. This database is constructed as a set of parameters-to-output mapping data, by using hybrid FEM-BEM simulator derived from Maxwell’s equations. Numerical experimental results show the effectiveness and feasibility of this method.

View full abstract

Download PDF (446K)
Reinforcement Learning of Optimal Supervisor based on the Worst-Case Behavior

Kouji Kajiwara, Tatsushi Yamasaki

2010 Volume 23 Issue 4 Pages 83-89
Published: 2010
Released on J-STAGE: April 22, 2011

DOIhttps://doi.org/10.5687/iscie.23.83

JOURNAL FREE ACCESS

Show abstractHide abstract

The supervisory control initiated by Ramadge and Wonham is a framework for logical control of discrete event systems. In the original supervisory control, the costs for occurrence and disabling of events have not been considered. Then, the optimal supervisory control based on quatitative measures has also been studied. This paper proposes a synthesis method of the optimal supervisor based on the worst-case behavior of discrete event systems. We introduce the new value functions for the assigned control patterns. The new value functions are not based on the expected total rewards, but based on the most undesirable event occurrence in the assigned control pattern. In the proposed method, the supervisor learns how to assign the control pattern based on reinforcement learning so as to maximize the value functions. We show the efficiency of the proposed method by computer simulation.

View full abstract

Download PDF (301K)