On the use of model-based techniques for achieving multi-mode control architectures

Multi-operation mode processes are often described through a hybrid model using well-known formal descriptions such as hybrid automata. A multi-controller architecture has been proposed in the literature for coping with different dynamics. It is composed by as many control algorithms as operation modes and a supervisor that tracks the system's operation, switching to the appropriate control algorithm when a new operation mode is reached. This paper explores the advantages of model-based techniques to automate the implementation of the supervisor and its integration into the final real-time control system from the simulation dynamic model. In particular, the Model Driven Engineering approach has been followed in order to provide a framework to support the development cycle of the control system for mode switched processes. It uses information from the model of the process and from the control loops designed and using model transformation techniques the supervisor is automatically generated. In order to provide independency from the particular simulation tools used for defining the process model, meta-models are proposed for both, the process and the supervisor, as well as for the necessary transformations. A proof of concept framework has been designed for the target field of automation applications, generating the supervisor in PLCopen XML format for achieving vendor independency.