Control of liquid cooling garments: technical control of body heat storage

This paper describes a concept of how liquid cooling garments (LCG) can be automatically controlled by the objective physiological state. The technically controlled parameter was mean body temperature which was calculated from the measured rectal and mean skin temperature. This was motivated by the fact that mean body temperature is the basis for estimating body heat storage, a commonly used measure of thermal strain. Here the setpoint of mean body temperature was the individual value taken in a preceeding resting period and it was the task of the technical controller to keep the actual value of mean body temperature as close as possible to the setpoint. The most important tuning parameters of the controller were the weighting coefficients for rectal and mean skin temperature in the calculation for mean body temperature. The ratio of these two coefficients determined the degree of compensation for any rectal temperature shift by changing mean skin temperature. Test experiments were carried out (n = 5) in which the controller was able to clamp mean body temperature to the setpoint thereby preventing heat storage. Although exercise rate (75 W) was the same, sweating and warm discomfort occurred in some cases due to the individual rectal temperature rise. Another source of discomfort were delays or paradoxical time courses of rectal temperature at the start or end of exercise which were responsible for a delayed onset of cooling or heating. To avoid these effects, the oxygen consumption signal, which is very fast and directly correlated to the exercise rate, was added to the control loop. Each increase of this parameter above its resting level lowered suit temperature. As heat storage should not be completely rejected by this new signal pathway, the controller for mean body temperature still remained active. The repetition of the experiments showed that the load error in the control loop was smaller and the comfort level in transient phases higher. For a further improvement of this concept it is recommended that the weighting coefficients be tuned to the individual requirements.