Predicting The Onset Of Sweat During Cycling In Simulated Environments

Abstract

E-bikes have become a growing alternative to traditional bikes. E- bikes are often used for transportation to and from the workplace, thus, in the initiative to promote adoption, advanced features such as appropriate assistance to eliminate the onset of sweat would be appealing to most users. PURPOSE: To provide a starting point for the development of a regression equation that can predict sweat onset. METHODS: Ten participants volunteered for this study. Participants committed to 5 experimental cycling sessions that varied by workload and climate. Participants cycled on an indoor bike trainer at 2 power outputs (25W and 75W) and 2 climatic conditions (25°C @ 60% RH and 30°C @ 60% RH) until sweating commenced. Physiological measures included: electromyography, heart rate, skin temperature, core temperature, galvanic skin response, and VO2. RESULTS: The average subjective sweat onset time for the 75W condition was less effected by the climatic condition than the 25W condition. The subjective sweat onset times for the 75W condition was 8.53 ± 2.19 minutes and 5.83 ± 1.44 minutes for the low and high temperatures, respectively. The subjective sweat onset times for the 25W condition was 23.52 ± 7.40 minutes and 12.49 ± 7.08 minutes for the low and high temperatures, respectively. A regression equation was developed and is able to predict subjective sweat onset with 61.5% of the variance explained with two measured variables. Workload alone explained 41.5% of the variance for sweat onset determination. When the regression was designed with workload as the outcome instead of sweat onset time, subjective sweat onset time was able to predict wattage with 40.1% of the variance explained. CONCLUSION: For the conditions simulated in this study, external temperature had less of an influence on sweat onset times than cycling workload. Sweat onset can be predicted with 61.5% of the variance explained using only two input variables. Heart rate was a poor indicator of sweat onset and simply using power output would be a better starting point. Overall, workload proved to be the most influential variable for predicting sweat onset. This project was funded by the Natural Sciences and Engineering Research Council Engage Grant