Fine-time-granularity fast demand control of building HVAC facilities for future smart grid

Abstract

Real-time demand control is expected for the future smart grid with intensive renewable energy sources (RESs). The Variable Refrigerant Flow (VRF) building HVAC (Heating Ventilation and Air-Conditioning) facilities are popular in modern buildings in Japan, Europe, and other countries. The VRF HVAC type uses sophisticated refrigerant flow control devices. We propose a fine-time-granularity fast demand control scheme that replaces traditional coarse and slow demand controls, such as on/off adjustment of power consumption. However, the power consumption forecast of the VRF's devices for the entire building is extremely complicated. Therefore, we chose statistical forecast model approach. A 5-minute-interval Auto-Regressive (AR) model system identification was introduced. We conducted a trial simulation of fine-time-granularity fast demand control using our 5-minute-interval AR model as a preliminary evaluation for actual experiments.