Correlation of thermal mathematical models for thermal control of space vehicles by means of genetic algorithms

The design of the thermal control system of space vehicles, needed to maintain the equipment components into their admissible range of temperatures, is usually developed by means of thermal mathematical models. These thermal mathematical models need to be correlated with the equipment real behavior registered during the thermal test campaign, in order to adapt them to the real state of the vehicle "as built". The correlation of this type of mathematical models is a very complex task, usually based on manual procedures, which requires a big effort in time and cost. For this reason, the development of methodologies able to perform this correlation automatically, would be a key aspect in the improvement of the space vehicles thermal control design and validation. The implementation, study and validation of a genetic algorithm able to perform this type of correlation in an automatized way are presented in this paper. The study and validation of the algorithm have been performed based on a simplified model of a real space instrument. The algorithm is able to correlate thermal mathematical models in steady state and transient analyses, and it is also able to perform the simultaneous correlation of several cases, as for example hot and cold cases.