Cálculo da Solução Explícita de Controladores MPC por Modelo Takagi-Sugeno Modificado

Abstract

: Multi-parameter quadratic programming is a technique applied to compute optimization solution inherent to Model Predictive Control (MPC) strategies. The present work proposes a modified Takagi-Sugeno model to compute explicit solution, formed by the polyhedral critical regions and their respective affine functions. Advantages of this technique are: (i) capacity to complexity reduction e parallel processing, (ii) dismiss of model's training phase from numerical data; and (iii) synthesis of a single analytical expression for the associated MPC control law. A case study shows that this new method has potential to be more efficient than classical explicit MPC optimization techniques with respect to numbers of parameters and processing time