Simulating Observer in Supervisory Control- A Domain-based Method

An Observer in the supervisory control observes responses of a discrete system to events of its environment and reports an unsafe/ critical situation if the response is undesired. An undesired response from the system indicates the system response does not adhere to users’ requirements of the system. Therefore, events and conditions of the system environment and user’s requirements of the system are basic elements to observer in determining correctness of the system response. However, the noteworthy matter is that the events, conditions, and requirements should be defined based on data of problem domain because discrete data are primary ingredients of the environment in discrete systems and they are used by system users as a gauge to express their requirements playing a vital role in safety-critical systems, such as medical and avionic ones. A large quantity of methods has already been proposed to model and simulate supervisory control of discrete systems however, a systematic method relying on data of problem domain is missing. Having extracted events, conditions, and user’s requirements from data of problem domain, a Petri-Net automaton is constructed for identifying violation of user’s requirements. The net constitutes the core of the observer and it is used to identify undesired responses of the system. In the third step, run-time simulation of the observer is suggested using multithreading mechanism and Task Parallel Library (TPL) technology of Microsoft. Finally, a case study of a discrete concurrent system is proposed, the method applied and simulation results are analyzed based on the system implementation on a multi-core computer.