In this work we propose several algorithms to solve the reconfiguration problem for linear and hybrid particular, we consider the decision about the usage of redundant hardware in order to compensate for faults. While this problem can be translated into a constrained model predictive control framework, the computation' grows very fast, as the number of possible decisions increases. In this work we prohave components that need not to be used in order to meet nominal control specifications, but can be used in extraordinary situations. The redundancy is Often given by subsystems with similar functions, different parameters or alternative locations within the system toPoloW*The main contribution of this work is to propose several solutions for the decision problem about the choice of redundant hardware using mathematical pose schemes, that require low computational effort. We discuss the applicability of the methods considering the reconfiguration of the three tank benchmark system.Keywords: Hybrid Systems, Fault Tolerant Control, tools. We will refer to this decision simply as reconfiguration of the control scheme. T~ this purpose we are using the modeling formalism of Mixed Logic Dynamical (MLD) systems introduced in [5,4]. The equations describing an MLD system are -
Reconfiguration, Model Predictive ControlMany control systems are subject to faults or malfunctions. The failure of components, like sensors or actuators can partially or fully disable the action of the controller. This results in an unsatisfactory performance of the system, decreased availability, emergency shutdowns or even significant damages to the plant and to its environment. For this reason, several research activities recently focused on the safety aspects of control systems [B, 61. One way to classify the different approaches is the distinction between passive and active fault accommodation [16]. In passive fault accommodation the closed loop system is designed, such that it is robust, within some performance range, to fault occurrences. On the other hand, active fault accommodation aims at changing the control operation when a fault is detected. These changes can comprise reconfiguration of the controller scheme, modifications of controller parameters or alternative setpoint trajectories. In this work we will focus on active fault accommodation. The key issue we are considering is the usage of redundant hardware for reconfiguration when some faults occur. The term hardware redundancy is used for systems where a replication of hardware components is present [16]. In other words, such systems leave from: NKK Corporation, Sensing & Control Research Department, Applied Technology Research Center 1-1; Minamiwatarida-cho; Kawasaki-ku Kawasaki-shi, 210-0855 JAPAN; Email: tsudamlab. keihin.nkk. C o . j pl}nc are the continuous and binary-states, U E Etmc x {O,l}mc are the inputs, 9 E RPc x {O,l}pc the outputs, and 6 E {O,l}Q, t E RrC represent auxiliary binary and continuous variables, respectively, 4 E { O , l } f t denote the occurrence of faults....