ResStock 2024.2 zone temperature capacitance multiplier and internal mass

savecancel

The zone temperature capacitance multiplier is a knob we adjusted in our residential simulations to better calibrate timeseries indoor temperatures against real data. It is particularly aimed to improve modeling during periods where the HVAC system is not running (e.g., recovery from large thermostat setbacks or power outages). It does not have much of an impact when the HVAC system is running.

The value of 7 was chosen after conducting a literature review several years ago where we found values that ranged from 3-15 for residential buildings:

• Development and validation of an HVAC on/off controller in EnergyPlus for energy simulation of residential and small commercial buildings -- 3.6
• Greenbuilt Retrofit Test House Final Report -- 7
• Residential Mechanical Precooling -- 15
• Measuring and modelling the energy demand reduction potential of using zonal space heating control in a UK home -- 10
• A Plug-and-Play Duct System Evaluation -- 8
• Automated fault detection of residential air-conditioning systems using thermostat drive cycles -- 3

It's hard to say what physical phenomena result in the need for this knob. Temperatures near heat loss surfaces are different from HVAC setpoints due to the presence of furniture/cabinets adjacent to walls? Temperatures near the thermostat are not representative of the entire home due to inadequate air distribution or placement of supply registers or airflow rates that do not achieve sufficient mixing? The time it takes for heated/cooled air to reach the rooms once the HVAC thermostat calls for operation? It's probably a combination of these things and more.

Indeed we've also found that adjusting the internal mass in EnergyPlus does not have a significant effect. Which either means that there's an issue with the way thermal mass is being modeled/handled, or thermal mass is not a significant cause of differences between modeled timeseries temperatures and real world data. Or some of both. There is some interesting discussion on this topic in Integrating physics-based models with sensor data: An inverse modeling approach.