Operation and control strategies for battery energy storage systems to increase penetration levels of renewable generation on remote microgrids

A critical requirement of any remote microgrid is its capability to control the balance between electric generation and load within the confines of the microgrid itself. The integration of significant amounts of “as available” renewable generation to any electric grid (macro or micro) makes it more difficult to maintain this balance and can result in large frequency deviations on a microgrid. Ancillary services provide the resources required to maintain the instantaneous and ongoing balance between generation and load. Battery energy storage systems (BESS) can provide regulating reserves, a type of ancillary service, by modulating active power for frequency control, referred to as load frequency control (LFC), to reduce frequency deviations caused by sudden changes in renewable generation. Historically, the most common methodology for reducing frequency disturbances exacerbated by wind plants with BESS systems is ramp rate control and more recently lead compensation. This thesis proposed a modified lead compensator for use in microgrid applications. A PSS®E microgrid model, based upon existing validated models, was developed to test the effectiveness of the LFC controllers used to dispatch the BESS as a regulating resource to allow increased wind energy penetration levels on remote microgrids. A model of the remote microgrid of the island of Maui, Hawaii was chosen as the basis for the designs. Daily wind power data from 2012 was classified and indexed on an hourly basis by severity of variation. The worst hour for power variation from the wind plants was identified from this indexing and used as the basis for simulating the LFC controllers. The results compared the effectiveness of droop, ramp rate, lead compensation, and modified lead compensation controllers in reducing the variability in the grid frequency caused by changes in wind power generation. An RMS of variation with respect to an average over different time windows was used as the comparison metric. The combined modified lead compensator with ramp rate control showed the best performance of the overall system behavior.