Incorporating thermoreceptor responses in a local sensation model to account for setpoint adaptation during cold-to-warm transition

Abstract

The Berkeley comfort models are well-suited for addressing nonuniform, transient, and cold conditions owing to their comprehensible model structures. Integrating these models with thermoregulation models can aid in formulating energy-efficient local warming and heating, ventilation, and air conditioning (HVAC) operational strategies for occupant-centric winter conditioning in battery electric vehicles (BEVs)—a critical step toward their widespread adoption. However, the Berkeley local sensation (LS) model requires accurate consideration of setpoint and setpoint adaptation to ensure reliable predictions. Since the dynamic responses and excitatory–inhibitory interactions between the cold- and warm-sensitive thermoreceptors may inherently entail setpoint adaptation, expressing the LS model in terms of the receptor responses offers a promising alternative. This study evaluates a thermoreceptor-response-based LS model by incorporatingnet cold- and warm-sensitive receptor responses into the existing framework. Model coefficients were regressed and tested on two independent datasets from experiments simulating routine cabin environments during outdoor winter conditions. The results showed a reasonable fit for the development dataset A, with root mean-squared error (RMSE) values in the range of 0.29–0.59 and coefficient of determination (R²) values of 0.65–0.92. Validation on the test dataset B yielded RMSE values of 0.5–0.92 and moderate-to-strong R² values of 0.53–0.78. Compared to the original Berkeley LS model, the proposed receptor-response-based model demonstrated improved performances across all body segments. Moreover, this new framework has the potential to eliminate the need for explicit setpoint and setpoint adaptation definitions while offering a viable solution for optimizing local warmer and HVAC operating strategies in BEVs operating under outdoor winter conditions.