Polynomial neural network modeling for sequential processes of silicon solar cell fabrication

Abstract

Since the neural network was introduced, significant progress has been made on data handling and learning algorithms. Currently, the most popular learning algorithm in neural network training is feed forward error back-propagation (FFEBP) algorithm. Aside from the success of the FFEBP algorithm, a polynomial neural networks (PNN) learning has been proposed as a new learning method. The PNN learning is a self-organizing process designed to determine an appropriate set of Ivakhnenko polynomials that allow the activation of many neurons to achieve a desired state of activation that mimics a given set of sampled patterns. In this paper, the PNN models have been developed using the sequential processes of high efficiency silicon solar cells experimental data. The output characteristics considered are reflectance, sheet resistance, diffusion depth, and cell efficiency. The influence of each parameters affected to efficiency is investigated through the main effect and interaction analysis. Sequential PNN process models are constructed to characterize the entire 3-step process. To characterize sequential processes using PNN, experiments were processed under varying conditions using central composite design (CCD) with three center points. It was shown that the output of the PNN model follows the real experimental measurement data very well. © 2012 Springer-Verlag GmbH Berlin Heidelberg.