Modeling Solar Combisystems Performances Using an Artificial Neural Network Approach
Abstract
: This paper introduces a global Solar Combisystem model that could estimate system performances for any kind of climate and any kind of building, only from a short experimental data set. The aim of this study is to improve the “Short Cycle System Performance Test” (SCSPT) that is being developed at the French National Solar Energy Institute (INES) and that shows relevant results but its performance prediction is limited to only one environment (climate and building). This improvement would lead to a complete and reliable method to characterize SCS performances. The proposed model is based on standard equations conjugated with Artificial Neural Networks (ANN). It shows results very close to TRNSYS simulations of three detailed SCS models.