Yoshio Tange

State Space Parameterization of Stabilizing Multirate Controllers for MIMO Linear Time - Invariant Plants

In this paper, a parameterization of stabilizing multirate controllers which communicate with given plant via a communication channel is proposed. The communications channel does not have any quantizer and the limitation of the communication rate of the channel is model by a projection operator which is defined with the well-known lifting operator. Explicit forms of stabilizing multirate controllers are derived by solving dual periodically weighted model-matching problems with linear periodically time-varying (LPTV) control. The causality constraint of periodic controllers is automatically satisfied without any additional requirement on direct terms of the controllers. A numerical example with 5-periodic control is also provided.

Time analysis on fast distribution control based on formula manipulation

We have developed a fast distribution control method based on formula manipulation for low-resource environment like PLC (Programmable Logic Controller). The developed method consists of offline analysis and online optimization. In order to clarify the required specifications of implementation of the method, we analyzed the relationship between offline analysis result and online calculation time. We evaluated a problem of optimal load distribution of a multiple pump system. As a result, we have shown that online calculation time of the method is linear to the number of feasible on off combinations of pumps.

Overshoot suppression control based on final tracking error estimation and quantifier elimination

This paper presents a method to establish overshoot suppression control based on estimation on the final tracking error and its feedback compensation. Overshoot is one of important issues in industrial control to avoid unsafe situations such as overflow, crash, and so on. We analyze the overshoot amount caused by control input and show a strategy for its suppression. To apply the strategy, we show a control scheme composed of prediction and feedback compensation with novel feedback gain design method based on quantifier elimination. Illustrative examples are given to show the effectiveness of the proposing method.