Sistema de monitoreo apícola mediante el uso de redes neuronales artificiales para identificar la variación de población

Abstract

ABSTRACT The implementation of the beekeeping monitoring system through the use of artificial neural networks to identify population variation allows to predict in the future the physical variables that directly affect the health and population production of the hive. In this project shows the development of a prototype bee monitoring system, the optimal ranges of the physical variables of temperature, humidity, carbon dioxide concentration and weight of the hive are identified to ensure stability within the apiary, as well as why the use of the selected sensors to read the said variables that were chosen and how the information is acquired and sent through WI-FI wireless technology to be stored in a MySql database, the information is processed to the development of the artificial neural network model of short and long-term memory, which allows forecasting in which, given the conditions in which the hive is located in the previous days or hours, it predicts the conditions in the following days or hours, reducing the intervention of man on the hive, avoiding possible damage to its ecosystem and offers information in real time on any changes sig Significant in its environment favoring the user or beekeeper when taking preventive measures to improve the performance of the hive.