Autotuning design methods for a Smith predictor control scheme

The authors consider the autotuning method for tuning a PID controller used with a Smith predictor. This technique is used to obtain appropriate parameters for the controller, G(s), assuming matched conditions, that is G(s) = 0, and secondly to do so taking into account modelling errors. A third approach which uses a relay with hysteresis to estimate possible time-delay mismatches is also proposed. Some examples are given to illustrate the application of the proposed design methods and some conclusions are given in the last section.