Model-Based Direct Control

After a technical system has been modeled, its performance evaluated and possibly improved, the final step is to bring the specified behavior into reality. The model itself needs to be interpreted in order to create an interface to the environment. Only one model should be used throughout the whole design process. It is easily possible to integrate control rules in an SDES model. Afterwards, the influence on the system behavior (liveness, performance) can be analyzed. The behavior specified in the model is directly executed finally. The control interpretation proposed in the following associates control (output) and sensor (input) signals to transitions, which are exchanged between model and “outside world.” In the applications considered so far, this approach turned out to be very natural in the Petri net understanding, and did not require the exchange of state information. It is, however, easy to define state inputs and outputs as well, as it is done in Sect. 11.2 for general SDES models. For a state output, the most general way is similar to the definition of a marking‐dependent boolean expression, the result for which is available to the environment. State inputs can be used in guard functions of transitions.