Optimal Thicknesses Determination in a Multilayer Structure to Improve the SPP Efficiency for Photovoltaic Devices by an Hybrid FEM – Cascade Neural Network Based Approach

As the global energy needs to grow, there isincreasing interest in the electricity generation by photovoltaics(PVs) devices or solar cells. Analytical and numerical methodsare used in literature to study the propagation of surface plasmonpolaritons (SPP) but the optimal thicknesses in a multilayerstructure can’t be established for an optimal propagation bythese. In this paper a new method based on cascade NeuralNetwork (NN) is used to predict the propagation characteristicsof a multilayer plasmonic structure and coupling FEM analysisof the involved electromagnetic field. The trained NNs are able toprovide the required optimal values of the SPP propagation withgood accuracy at different value of thicknesses in the multilayerstructure.