Co-optimizing microgrid asset sizing and dispatch with building automation and load control

The combination of microgrids and load shifting creates stacked economic benefits and, in some cases, may create synergistic value in which “the whole is greater than the sum of its parts.” These behaviors were explored and contrasted for three building types–apartment complex, hospital, school–using the same retail tariff structure for the case study site of Los Angeles, California. Microgrid assets were sized to minimize annualized project costs under a range of load shifting scenarios that varied duration (up to 3 h) and percentage (up to 25% of total load). Results showed stacked value improvements through annualized cost reductions of 6.2%, 8.1% and 21.6% for an apartment complex, hospital and school, respectively, and synergistic value of 0.3%, 2.1%, and −1.2%, respectively, for the mid-point case with 2 h and 15% of total load shifted. The rank order relationship of synergistic value shows a negative correlation with the load factor, suggesting that load profiles with more “peakiness” may not be able to access synergistic value, though the school did have the greatest stacked value. As load shifting percentage increases, the synergistic value for the school trends negative while the synergistic value for the hospital is flat and the apartment trends positive. This underscores results showing that the stacked value and the synergistic value are primarily affected by the load shifting duration rather than the load shifting percentage. Greater synergistic value is created when microgrid asset sizes can be downsized as more load control is added. When looking at the hospital, results indicate that microgrid focuses on demand charge reductions whereas load shifting emphasizes energy charge reductions, thereby promoting more synergistic value when combined because the two approaches address different parts of the utility bill. This trend does not hold in all scenarios, however, and further identifies that the cost optimal solution may not be achieved by simply shifting load to periods of day with greater solar photovoltaics production, a commonly held general assumption.