Prediction and Monitoring of stored food grains health using IoT Enable Nodes

Abstract

Accompanying the rapid urbanization, A developing country like India, still monitoring the food grain's health using conventional methods, which detect insects at a very later stage due to which food grain get spoil. The major parameter that should be observed for maintaining the quality of grains are temperature, relative humidity, and carbon dioxide concentration. The observation of these parameters regularly can lead to early detection of the insects in the grains. This paper suggests the state of the art Internet of Things (IoT) based real-time system which observes the warehouse parameter i.e Temperature, Relative Humidity, and Carbon dioxide and predicts the type of insect activity. The hardware Device used in the experiment is designed by keeping in mind the Indian warehouse's conditions and for large scale implementation of this model. The hardware Device contains two sensors (DHT22 and CDM7160), ESP8266, and battery for power supply. DHT22 and CDM7160 sensors are adopted to detect the above parameters i.e temperature, relative humidity, and CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> . The online portal has been created for data analysis and real-time data collection from the sensor nodes. This portal can be easily integrated with the digital portal already present in most of the warehouses. The machine learning algorithm has been deployed on the online server to predict relative humidity on the basis of collected data of relative humidity at a particular temperature. The prediction of relative humidity has been done for the next five days. The predicted relative humidity is used for predicting the type of insect that could attack the stack of the warehouse in the coming future. Based on the predicted results further action will be taken for controlling the food grain quality and prevent the wastage of food grain.