Thermal energy storage (TES) and distributed generation technologies, such as combined heat and power (CHP) or photovoltaics (PV), can be used to reduce energy costs and decrease CO_{2} emissions from buildings by shifting energy consumption to times with less emissions and/or lower energy prices. To determine the feasibility of investing in TES in combination with other distributed energy resources (DER), mixed integer linear programming (MILP) can be used. Such a MILP model is the well-established Distributed Energy Resources Customer Adoption Model (DER-CAM); however, it currently uses only a simplified TES model to guarantee linearity and short run-times. Loss calculations are based only on the energy contained in the storage. This paper presents a new DER-CAM TES model that allows improved tracking of losses based on ambient and storage temperatures, and compares results with the previous version. A multi-layer TES model is introduced that retains linearity and avoids creating an endogenous optimization problem. The improved model increases the accuracy of the estimated storage losses and enables use of heat pumps for low temperature storage charging. Results indicate that the previous model overestimates the attractiveness of TES investments for cases without possibility to invest in heat pumps and underestimates it for some locations when heat pumps are allowed. Despite a variation in optimal technology selection between the two models, the objective function value stays quite stable, illustrating the complexity of optimal DER sizing problems in buildings and microgrids.

This paper describes the introduction of stochastic linear programming into Operations DER-CAM, a tool used to obtain optimal operating schedules for a given microgrid under local economic and environmental conditions. This application follows previous work on optimal scheduling of a lithium-iron-phosphate battery given the output uncertainty of a 1 MW molten carbonate fuel cell. Both are in the Santa Rita Jail microgrid, located in Dublin, California. This fuel cell has proven unreliable, partially justifying the consideration of storage options. Several stochastic DER-CAM runs are executed to compare different scenarios to values obtained by a deterministic approach. Results indicate that using a stochastic approach provides a conservative yet more lucrative battery schedule. Lower expected energy bills result, given fuel cell outages, in potential savings exceeding 6%.

1 aCardoso, Gonçalo1 aStadler, Michael1 aSiddiqui, Afzal, S.1 aMarnay, Chris1 aDeForest, Nicholas1 aBarbosa-Póvoa, Ana1 aFerrão, Paulo uhttps://gridintegration.lbl.gov/publications/microgrid-reliability-modeling-and00525nas a2200145 4500008004100000020001300041245006000054210005900114260003900173100002300212700002100235700001800256700002200274856008300296 2012 eng d a0-91824900aMicrogrid Dispatch for Macrogrid Peak-Demand Mitigation0 aMicrogrid Dispatch for Macrogrid PeakDemand Mitigation aPacific Grove, Californiac08/20121 aDeForest, Nicholas1 aStadler, Michael1 aMarnay, Chris1 aDonadee, Jonathan uhttps://gridintegration.lbl.gov/publications/microgrid-dispatch-macrogrid-peak