Investigation of human thermal comfort under highly transient conditions for automative applications Part 2 : thermal sensation modeling

This paper focuses on the mathematical modeling of dynamic human thermal comfort under highly transient conditions for automotive applications. A combined physiological and psychological modeling approach was taken. First, the transient environmental and human activity data, plus the
clothing insulation data, were used as inputs to a human thermal model to determine the physiological responses for the vehicle thermal environmental conditions. Secondly, a series
of high dimensional multi-variable statistical analyses were performed to link the physiological variables with the subjective responses. Last, optimized models for summer and winter conditions were obtained to predict whole body and local thermal sensation. They addressed the psychological effects by dynamic path integration, and allowed the transient and steady-state components to be phased in and phased out automatically based on their influences during the entire transient
and pseudo-steady-state processes. The model robustness was examined by challenging the model with widely varying inputs. The models coefficients and choice of primary variables were
evaluated and adjusted repeatedly until a balance of model accuracy and robustness was achieved. The predicted results are in good agreement with the measured thermal sensation
data.