Compositionalmodels for power systems

Abstract

The problem of integrating multiple overlappingmodels and data is pervasive in engineering, though often implicit. We consider this issue of model management in the context of the electrical power grid as it transitions towards a modern 'Smart Grid.' We present a methodology for specifying, managing, and reasoning within multiple models of distributed energy resources (DERs), entities which produce, consume, or store power, using categorical databases and symmetric monoidal categories. Considering the problem of distributing power on the grid in the presence of DERs, we show how to connect a generic problem specification with implementation-specific numerical solvers using the paradigm of categorical databases.