Fuzzy logic in gain scheduling of multivariable control
Joonas Varso, Heikki N. Koivo · 2008
Controlling nonlinear systems is very difficult. Multivariable case makes control even harder. Combining these characteristics results in a problem, which can be solved explicitly very seldom. In this paper the use of fuzzy logic in gain scheduling of nonlinear and multivariable processes is discussed. The developed method is tested using a simulation model and a real heating process. The proposed approach uses a linear controller to control a nonlinear system. The controller is tuned at chosen operating points, which are selected to cover the nonlinear range of the process. Gain scheduling means that the tuned parameters of the controller are collected into a table. The global controller observes the state of the process and chooses proper parameters from the table. Fuzzy logic fits very well for gain scheduling, because the change of the state of the process is uncertain by nature. In multivariable control one of the most difficult problems appears in interactions, which are the reason why scalar controllers have poor performance. To overcome these interactions a multivariable PI-controller is used and combined with fuzzy gain scheduling to form a powerful control scheme.