Successful crop production in greenhouses is determined mainly by not only internal and external climatic variables but also by the characteristics of the greenhouse itself. One of the most important variables is the internal temperature, as it greatly determines whether or not the crop will succeed. The objective of this work is to use machine learning models capable of giving a short-term forecast of the indoor temperature of greenhouses in the future. This model will be able to guide growers in determining the best location for new greenhouses, as they will be able to see what the future indoor temperature will be based on the other predictors. The main challenge of this work has been that we have not used the past of the variable to predict, but only the past of the predictor variables. The results obtained show that the model is capable of predicting indoor temperature with significant precision, suggesting that the proposed methodology may be useful to optimize the selection process of greenhouse locations.
CITATION STYLE
Vega-Márquez, B., Pardo-Martínez, J., Villegas-Oliva, M. del M., & Riquelme, J. C. (2023). Forecasting Greenhouse Temperature Using Machine Learning Models: Optimizing Crop Production in Andalucia. In Lecture Notes in Networks and Systems (Vol. 749 LNNS, pp. 239–248). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-42529-5_23
Mendeley helps you to discover research relevant for your work.