Artificial neural network model to estimate investment viability of solar plants for the industry of Jalisco, Mexico

• Published in: 2018 IEEE Latin American Conference on Computational Intelligence (LA-CCI) pp. 1 - 5
• Main Authors: Tzuc, O. May, Bassam, A., Ricalde, L. J., Flota-Banuelos, M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2018
• Subjects: artificial intelligence; Artificial neural networks; Biological neural networks; Computational modeling; Computer architecture; Investment; Mathematical model; net present value; Neurons; solar energy
• DOI: 10.1109/LA-CCI.2018.8625223

The present paper describes the development of a computational model based on artificial neural network (ANN) to estimate the industrial investment viability of solar thermal projects for Jalisco, Mexico. A solar plant, with an auxiliary liquefied petroleum gas heating system, designed for pasteurization process was considered as study case. Net present Value (NPV) was used as the indicator of investment viability. The model was trained considering different plant design scenarios as the independent variables. According to the results, the best ANN architecture was obtained using Levenberg-Marquardt optimization algorithm, the logarithmic sigmoid transfer-function and the linear transfer-function for the hidden and output layer; with 22 neurons at the hidden layer. The developed model presented an estimation capacity of over 99%, indicating that it is an adequate auxiliary tool for making investment decisions. The model described represents an alternative designed to facilitate decision making for the implementation of solar thermal technology in the industrial sector of Jalisco, which can be extrapolated to other climatic regions.