Extending the software tool TimeNET by power consumption estimation of UML MARTE models

This paper presents an extension of the software tool TimeNET, which supports modeling and analysis of stochastic Petri nets. The new extension implements a previously proposed method for power consumption modeling and evaluation based on extended UML models. Two new net classes have been developed to support the necessary operational and application models and edit them in the graphical user interface. Several stereotypes of the UML Profile for Modeling and Analysis of Real-Time and Embedded Systems (MARTE) have been added to extend the information about power consumption of system and application states. The two UML models are then automatically transformed into a stochastic Petri net. Power consumption of the system can be predicted by standard Petri net analysis modules of TimeNET. An example of an industrial control system is provided.