Digital monitoring of grain conditions in large-scale bulk storage facilities based on spatiotemporal distributions of grain temperature

Abstract

Managing large-scale facilities for storing bulk grain is time-consuming, labour intensive, and often difficult to be thorough. This paper presents a computer algorithm for using temperature data to remotely monitor and inspect stored grain in large bulk storage facilities. The algorithm is based on the analysis of the spatiotemporal distributions of the temperature field in the stored grain. The characteristics and irregularities of the temperature field were analysed to detect changes in grain quantity (inventory) and quality. The algorithm was implemented in computer software and tested on 234,300 sets of temperature data from 592 different grain depots in 10 provinces in China. The average accuracy of correctly identifying grain quality and inventory problems was 94%. • 234,300 sets of temperature data in large grain storage facilities were analysed. • Spatiotemporal distributions of grain temperature were related to grain conditions. • An algorithm was developed to identify grain quality and quantity issues in storage. • The average accuracy of identification of 592 grain storage facilities was 94%.