An Adaptive Discrete Event Model for Cyber-Physical System

Cyber-Physical Systems (CPS) often involve a wide spectrum of events, ranging from lower-level signals to higherlevel abstract events. In order to compose different levels of events, a Concept Lattice-based Event (CLE) model was developed. In the CLE model, the traditional first order logic is used in specifying rules composition. However, it is possible that the rules specified by the first order logic may have inconsistency. Furthermore, unanticipated events that are not considered in the initial design may affect systems’ performance, or even lead to system failure. In order to address these issues, an Adaptive Discrete Event (ADE) model is proposed in this paper. ADE model uses Discrete Event Calculus (DEC) to overcome possible inherent inconsistencies in composition rules that are specified by first order logic. In addition, we define abnormal event rules as an adaptive part in the CPS event model to handle unanticipated events. Finally, a CPS application “iLight“ is developed based on the ADE model, results show that ”iLight” adapts to the new environment successfully.