An Enhanced Smith-Predictor-based Control System for delayed MIMO Processes and its Use on a CSTR with Multiple Time-delays

Abstract

This paper aims to enhance Smith-predictor-based control systems (SPCSs) for multi-input multi-output (MIMO) time delay processes. Conventional SPCSs for MIMO processes have an array of classical feedback controller(s). In practice, these controllers receive error signal(s), calculated with deducting a predicted output by a Smith predictor from a reference signal at the time of operation. Investigations on underperformance of conventional SPCSs identified two major shortcomings: (i) design of classical feedback controllers is based on trade-off, and their use may lead to winding phenomenon, these adversely influence SPCSs performance, (ii) a predicted output by a Smith predictor belongs to a time in the future and does not concurrent with the reference at the time of operation. That is, in conventional SPCSs, the control error is generated with use of two asynchronous signals. This paper proposes an enhanced SPCS design method for MIMO time-delay systems based on two enhancements to tackle the aforementioned dual shortcomings. The proposed control system evidently outperforms a conventional SPCS with internal model control (IMC) proportional-integral-derivative (PID) feedback controllers. The case study is a catalytic stirred tank reactor (CSTR) with three inputs (feed and water flow rates and auxiliary temperature), two outputs (output flow concentration and temperature) and three time delays. The presented model of the CSTR is more comprehensive than any CSTR model found in the literature.