Geothermal Heat Exchanger’s Temperature Input Sensor Prediction Based on Deep Learning Modelling Technique
摘要
The goal of this study was to test many models in order to check which one better predicted the heat exchanger of a bioclimatic building. A number of regression models, pre-processing methods, and data analyses are compared in the study to forecast the input collector temperature of the heat pump. Specifically, three different techniques have been considered in this research work: Multilayer Perceptrons, Long Short Term Memory networks and Convolutional Neural Networks. Satisfactory results have been obtained in all cases for predicting the temperature 24 h in advance, implementing an useful tool for enhance energy management.