Gain-Scheduling Control Solutions for a Strip Winding System with Variable Moment of Inertia

摘要

This paper presents four design methods for the speed control of a mechatronics application characterized by variable parameters: variable reference, variable load disturbance and variable moment of inertia. The variations of the operating conditions and also the variations of the process parameters require the development of advanced control solutions. In this context the development of advanced control solutions will be influenced and justified significantly by the knowledge of a detailed mathematical model of the process and its parameters. In order to obtain high performance speed control for the electric drive system, referred to as the strip winding system, four proportional-integral (PI) gain-scheduling control solutions are developed and tested: (1) a PI Switching-I Gain-Scheduling version with bump-less switching between three control algorithms (PI-SIGS), (2) a PI Switching-II Gain-Scheduling version with a switching logic based on Euclidean distance metric (PI-SIIGS); (3) a PI Gain-Scheduling version with a switching logic based on a generalization of the monovariable case of the Lagrange interpolating parameter value method (PI-LGS), and (4) a PI Gain-Scheduling version with a switching logic based on a Cauchy kernel distance metric (PI-CGS). The continuous-time speed controllers are tuned by the Modulus Optimum method (MO-m) and are discretized using Tustin's method. The proposed and developed control solutions were embedded in a conventional control structure (CCS) which involves the switching between different digital control algorithms and are validated by means of simulation results. The strip winding system is discussed in this paper due to its applicability as a controlled plant in the field of mechatronics systems.