The Colombian agricultural sector faces significant challenges, one of which is the lack of implementation of technologies that allow for the technological advancement of production processes related to crops such as coffee, as these are currently still managed manually and empirically. The research presented proposes the development of a model called the “Digital Coffee Grower,” which aims to technically streamline and systematize all production processes currently performed manually by coffee growers on their crops. The methodology used consisted of an exploratory study where the implementation of a wireless sensor network (WSN) was carried out, which uses the LoRaWAN communication protocol to communicate the sensor nodes with the base station to collect data on certain agro-environmental variables, subsequently these data are sent from the Gateway or base station to the router; Then the data is uploaded to an Internet of Things (IoT) platform where said data will be stored, purified and transformed to later be analyzed and serve as support for the decision-making process. Among the main results obtained we have: 1) The configuration and implementation of a wireless sensor network 2) The connectivity and data transmission between sensor nodes, base station, router and Internet of Things (IoT) platform 3) Collection and storage of data on the IoT platform. 4) Collection, cleaning, debugging and transformation of data of agro-environmental variables. 5) Development of a web application to monitor and analyze agro-environmental variables in real time through a dashboard. 6) Evaluation of the collected data using a predictive model that uses the algorithm known as Random Forest to obtain a more precise and stable prediction. In conclusion, this research provides important evidence on the technification of agriculture and the possibility of improving the productive processes of crops such as coffee based on the analysis of certain agro-environmental variables and the implementation of predictive models to generate projections related to the behavior of the crop in each of its productive stages.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Application of IoT and Prediction Algorithms in the Analysis of Agronomic Variables in Coffee Crops

  • Cesar Osimani,
  • William Ruiz Martinez,
  • Jaime Andrés Arevalo

摘要

The Colombian agricultural sector faces significant challenges, one of which is the lack of implementation of technologies that allow for the technological advancement of production processes related to crops such as coffee, as these are currently still managed manually and empirically. The research presented proposes the development of a model called the “Digital Coffee Grower,” which aims to technically streamline and systematize all production processes currently performed manually by coffee growers on their crops. The methodology used consisted of an exploratory study where the implementation of a wireless sensor network (WSN) was carried out, which uses the LoRaWAN communication protocol to communicate the sensor nodes with the base station to collect data on certain agro-environmental variables, subsequently these data are sent from the Gateway or base station to the router; Then the data is uploaded to an Internet of Things (IoT) platform where said data will be stored, purified and transformed to later be analyzed and serve as support for the decision-making process. Among the main results obtained we have: 1) The configuration and implementation of a wireless sensor network 2) The connectivity and data transmission between sensor nodes, base station, router and Internet of Things (IoT) platform 3) Collection and storage of data on the IoT platform. 4) Collection, cleaning, debugging and transformation of data of agro-environmental variables. 5) Development of a web application to monitor and analyze agro-environmental variables in real time through a dashboard. 6) Evaluation of the collected data using a predictive model that uses the algorithm known as Random Forest to obtain a more precise and stable prediction. In conclusion, this research provides important evidence on the technification of agriculture and the possibility of improving the productive processes of crops such as coffee based on the analysis of certain agro-environmental variables and the implementation of predictive models to generate projections related to the behavior of the crop in each of its productive stages.