Adaptive Switching Control Based On the Balance Equation

This paper presents an adaptive switching control strategy based on the Balance Equation. Previous work has described the development of the Balance Equation inspired by the PI/PID design rule in the framework of Internal Model Control. The linear algebraic relationship between the proportional and integral contributions of a PI controller revealed by the Balance Equation is utilized as the driving force for the adaptive switching among a set of candidate controllers. A performance index is calculated through the difference between the integral contribution given by the ideal controller and that given by a candidate controller. By minimizing this performance index, the adaptive switching strategy aims at finding the candidate controller that is closest to the ideal controller. For step changes in the set-point, as long as the candidate controller set covers the ideal controller, the adaptive control strategy can always switch to this controller in a timely fashion. Extensive simulation results have shown that the adaptive switching control strategy can always identify the best controller from the candidate controller set and achieve good set-point tracking performance even in the presence of significant variation in the process time constant. This adaptive control switching strategy serves as a good demonstration for the value of the Balance Equation, which can enable more new insight, better control performance, and stronger adaptation ability in wider applications.