The fuzzy logic approach provides the control engineer with a control algorithm which does not require a precise knowledge of the system as required by classical controllers. We applied a fuzzy controller to a second-order system with variable dead-time and compared the performance to that of a classical controller with Smith predictor. The experimental results shown In this paper prove some useful properties of the fuzzy-logic approach versus the classical one. The results show that even in the case of a process with a varying dead-time, this algorithm presents acceptable values of the squared-error integral that enable its usage in this sort of process too. This algorithm was tested In a system for controlling the temperature of a heat radiator with variable transport time between heat source and radiator.
展开▼
