Hybrid systems are dynamical systems consisting of interacting discrete and continuous components. They arise when modeling the complicated behavior of high-technology artificial systems and modern industry process. This topic is becoming very active and important for a series of project items such as production process automation, process control, automatic scheduling, robot and computer communication, due to the emergence of hierarchical structure in many Complex control systems where the lower is controlled continuous variable system, and the high is discrete event system with the characteristics of operation and scheduling. This thesis deals mainly with modeling theory and control methodology for hybrid systems and its applications in modern process control, on the basis of which positive results from both discrete event systems theory and traditional control systems theory are fully utilized. Furthermore, the contributions of thesis include: (1) Advance background, present situation and all kinds of models of hybrid systems are briefly criticized. A much more distinct definition and the research methodology about systems(HCS & HSS) are presented from the point of structure and behavior respectively. (2) According to the methodology, basic structure, ordinary steps for modeling, model consistency and partition criterion of continuous state space about the class of hybrid control systems are studied, based on the hierarchical model consisting of interacting discrete and continubus components, and deciding controllability and expressing performance indices are analyzed. Meanwhile, DES supervisory theory adapted to the design of hybrid control systems so that its control-law can be synthesized fairly well. (3)According to methodology, the class of hybrid state systems we consider is specified by a hybrid automaton model based on the relational model where discrete and continuous dynamics exist in a single system and interact directly with each other. For verification purpose We provided some teachability results for systemsand to obtain semidecision procedures for various verification problems. In particular, we consider the symbolic model-checking algorithms for a hybrid automaton over an infinite state space combined with temporal logic. From the control theory or dynamics view point, the relations between the reachability and the controllability problems are discussed, and controller synthesis procedure is given. (4) In our discussion of hybrid systems with its information structure, the combined qualitative and quantitative information, integrated optimization framework based on Duration Calculus and Grafcet Chart and the design method of the optimal hybrid control are presented so that its synthesis problems are successfully solved constrained by logic indices and real-time indices. In the light of the need in practice, high-performance control approaches for hybrid systems a
修改评论