Enhancing dynamic metabolic rate estimation using calibration models for portable devices

Regarding thermal comfort and safety, metabolic rate is crucial for understanding human responses to environmental heat. Outdoor field experiments were conducted to investigate the variations in M measurement methods across different activities, including five walking speed levels: 1.2 m/s, 1.4 m/s, 1.6 m/s, 1.8 m/s, and 2.0 m/s. with This study presents a significant advancement in the accurate estimation of metabolic rates using portable devices at common walking speeds. By introducing correction formulas, the accuracy of metabolic rate measurements was improved, thereby enhancing thermal comfort research. The key findings included the consistent overestimation of metabolic rates by heart-rate-based methods during sedentary phases by approximately 1.5 met, and the underestimation by ISO 7730 and ASHRAE standards during walking phases by approximately 0.5 met at higher speeds. The conversion formulas between the ECG-based and PPG-based devices demonstrate high compatibility. The developed correction models reduced the deviations in the walking stages to <15% and aligned with the ISO 8996:2004 guidelines. Moreover, the predictive models for activity transitions exhibited high R² values of approximately 1 and RMSE below 0.2. These findings have practical implications for improving thermal comfort assessments, urban planning, and guidelines for outdoor safety and wellbeing. © 2025 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.