Artificial Neural Networks and Genetic Algorithms for the Modeling, Simulation, and Performance Prediction of Solar Energy Systems

Abstract

In this chapter, two of the most important artificial intelligence techniques are presented together with a variety of applications in solar energy systems. Artificial neural network (ANN) models represent a new method in system modeling and prediction. An ANN mimics mathematically the function of a human brain. They learn the relationship between the input parameters, usually collected from experiments, and the controlled and uncontrolled variables by studying previously recorded data. A genetic algorithm (GA) is a model of machine learning, which derives its behavior from a representation of the processes of evolution in nature. GAs can be used for multidimensional optimization problems in which the character string of the chromosome can be used to encode the values for the different parameters being optimized. The chapter outlines an understanding of how ANN and GA operate by way of presenting a number of problems in different solar energy systems applications, which include modeling and simulation of solar systems, prediction of the performance, and optimization of the design or operation of the systems. The systems presented include solar thermal and photovoltaic systems.

A power management system for energy harvesting and wireless sensor networks application based on a novel charge pump circuit

“…Therefore, the super-capacitor is used as a storage device in this work ( Figure 1). For these various considerations, light and thermal resources can be significant energy harvesting sources and can guarantee a tradeoff between different design requirements (Kalogirou, 2013).…”

Predictive modelling of building energy consumption based on a hybrid nature-inspired optimization algorithm

“…Different machine learning and predictive modelling techniques have been included as part of energy management sensing and control frameworks. Regression analysis has conventionally been the most common approach in context of vehicle energy consumption prediction, solar energy prediction, and energy consumption prediction in buildings [8][9][10][11][12]. Artificial neural networks (ANNs) have also been adopted for predicting energy consumption where in [12] an ANN has been trained on data extracted from a simulation to draw a mapping between the input and output for anticipating energy consumption.…”

“…Regression analysis has conventionally been the most common approach in context of vehicle energy consumption prediction, solar energy prediction, and energy consumption prediction in buildings [8][9][10][11][12]. Artificial neural networks (ANNs) have also been adopted for predicting energy consumption where in [12] an ANN has been trained on data extracted from a simulation to draw a mapping between the input and output for anticipating energy consumption. In addition, decision trees have also been used in production systems as an efficient decision support technique to dynamically control changing industrial production processes [13].…”

A spatiotemporal universal model for the prediction of the global solar radiation based on Fourier series and the site altitude

“…More elaborated models proposed and applied in several projects are the ones in a fourth group which use artificial neural networks (ANN) to provide for H(n), in solar energy systems [21][22][23]. Finally, there is a group of empirical models which determine H(n) in a site with parameter the number of the day, n, in the year [24][25][26][27].…”