Modeling of a Solar Cooling Machine by Absorption Using RBF Neural Networks

摘要

In this work, the modeling of a solar absorption cooling machine is presented using Artificial Neural Networks of the Radial Basic Function (RBF) type optimized by multi-objective genetic algorithms. The neural model obtained is compared with the results obtained with the Lansing model in order to validate its efficiency for the characterization of the coefficient of performance (COP) of absorption machines that produce cold with solar energy and the energy efficiency of this type of machine in order to reduce consumption. The optimization of the structure of the neural model and its learning are ensured by the NSGA-Ⅱ genetic algorithms by optimizing two functions which are the learning error and the number of neurons in the hidden layer of the neural model. The obtained model offers the possibility of changing several parameters at the same time and facilitates the calculations and opens up fields of future research more push for this type of machine.