Conclusions

Abstract

In this book, a methodology for the design of predictive control strategies for nonlinear dynamic hybrid systems was developed, including discrete and continuous variables. The methodology is designed for real applications, particularly the study of dynamic transport systems, considering operational and service policies, as well as cost reductions. The modeling structure is based on the appropriate definition of the state-space equations, a flexible objective function that is able to capture the predictive behavior of the key system variables and their evolution in the future and efficient algorithms, which mainly come from computational intelligence techniques, to optimize performance indices for real-time applications. The framework of the proposed predictive control methodology enables the dynamic solving of nonlinear mixed-integer optimization problems, which are known to be NP-hard. The framework is generic, which broadens its applicability to other industrial processes. In this chapter, the major contributions of this book, as well as a number of promising future research directions for these topics, are highlighted.