DC Approximate Models for Modeling Minigrid Systems

Component sizing for renewable generation and battery storage is a key economic driver for small off-grid electrical systems (minigrids). Stochastic, time-series simulation is the go-to tool for sizing studies. For computational efficiency, these simulations typically avoid electrical system models and utilize power-balance methods that capture net energy flows without modeling current and voltage - parameters often utilized for common supervisory control actions. As a result, power-balance models often overestimate performance or fail to simulate critical supervisory control states necessary for control system development. In this work we demonstrate a modeling method that retains computational efficiency while calculating required currents and voltages. Since microgrid components frequently utilize battery voltage as a key control input, the method emphasizes an efficient representation of battery terminal voltage as a function of state-of-charge and battery current. We demonstrate that critical battery parameters can be extracted from simple pulsed discharge tests which could be conducted in field conditions. Application to typical microgrid illustrates that the method will identify control instabilities which would be missed by power-balance methods, while executing at acceptable speeds.