Solovchuk K. Methods and algorithms for control of interconnected memoryless plants under uncertainty

The thesis is devoted to the development and study of methods and algorithms for the discrete-time control of interconnected memoryless plants where the number of output variables can exceed the number of their input variables in the presence of unmeasurable but bounded disturbances under possible uncertainties about the gain matrices and also the bounds on these disturbances. The novelty of main scientific results is that the proposed pseudoinverse model-based method may be used as a universal method to deal with the interconnected memoryless plants to be controlled irrespective of the ranks of their gain matrices. Simulation results concerning the implementation of the pseudoinverse model-based approach to control a distillation column and also the real practical results obtained after the use of this approach to automatic blast distribution among the tuyeres of a blast furnace are presented.