Model-based adaptive controller for personalized ventilation and thermal comfort in naturally ventilated spaces

Abstract

This work develops a standalone autonomously controlled personalized ventilation (PV) unit in a naturally ventilated (NV) office space to maintain acceptable thermal comfort (TC) under steady and transient indoor conditions and activity levels. The NV-PV proportional integral derivative (PID) controller adjusts the PV supply temperature (TSPV) at the occupant set flow rate (QSPV) based on predicted TC using a regression model. The target TC level that the controller attains at all times is between 0 (neutral) and 1 (slightly comfortable). Process transfer functions were developed and then used to find the adaptive PID tuning coefficients using the Internal Model Control (IMC) method. The controller was tested in a case study at indoor temperature range of 25 to 33 °C with relative humidity range of 55% and 80%. It was shown that the NV-PV controller adjusted TSPV to maintain acceptable TC under transients of indoor conditions and metabolic rates.