Harnessing the Flexibility of Thermostatic Loads in Microgrids with Solar Power Generation

Abstract

This paper presents a demand response (DR) framework that intertwines thermodynamic building models with a genetic algorithm (GA)-based optimization method. The framework optimizes heating/cooling schedules of end-users inside a business park microgrid with local distributed generation from renewable energy sources (DG-RES) based on two separate objectives: net load minimization and electricity cost minimization. DG-RES is treated as a curtailable resource in anticipation of future scenarios where the infeed of DG-RES to the regional distribution network could be limited. We test the DR framework with a case study of a refrigerated warehouse and an office building located in a business park with local PV generation. Results show the technical potential of the DR framework in harnessing the flexibility of the thermal masses from end-user sites in order to: (1) reduce the energy exchange at the point of connection; (2) reduce the cost of electricity for the microgrid end-users; and (3) increase the local utilization of DG-RES in cases where DG-RES exports to the grid are restricted. The results of this work can aid end-users and distribution network operators to reduce energy costs and energy consumption.

Optimizing electricity consumption of buildings in a microgrid through demand response

“…The optimization problem (5a) signifies choosing the switching schedules (β(i, t)) and the PV production schedule E RES (t) over the whole time horizon, with the objective of minimizing energy consumption (6) or energy cost (7). The building temperatures (T in (i, t)) must not exceed the critical values required by the end-users (5b).…”

“…where Φ is Φ e (6) in the energy consumption minimization and Φ c (7) in the energy cost minimization problems:…”

“…3 represent the temperature setpoints when prices are low, average and high, respectively, while 2δ , 2δ, and 2δ are the thermostat's respective deadbands. All results will also be compared to those of a binary operation mode from our previous work [7].…”

“…: enhanced distribution system monitoring and control functionalities, the use of (electrical and/or thermal) storage, the design of new electricity markets, and an increased demand-side flexibility through demand response (DR) programs [4]- [6]. For the purposes of this work, we define DR as "actions voluntarily taken by consumers to change their energy usage -either in terms of quantity or timing-in response to an external control signal" [7], such as price or a command from the system operator. Furthermore, we define load flexibility as "the ability of loads to be shifted in time by automated DR actions" [7].…”

“…For the purposes of this work, we define DR as "actions voluntarily taken by consumers to change their energy usage -either in terms of quantity or timing-in response to an external control signal" [7], such as price or a command from the system operator. Furthermore, we define load flexibility as "the ability of loads to be shifted in time by automated DR actions" [7]. In this work, we focus on using DR to harness the flexibility of end-user thermostatically-controlled loads (e.g., heating, ventilation and air conditioning systems, refrigerators and water heaters) in response to price signals, while simultaneously considering local DG-RES availability.…”