The issue of grain loss and waste during the storage of grains has consistently represented a significant challenge to the security of stored grains. The prediction of grain temperature indicates the guidance of granary ventilation, with the objective of reducing the incidence of grain mold and deterioration due to excessive temperature. Current methods of grain temperature prediction mainly include thermal simulation and data-driven prediction. However, the former method has been found to have low accuracy due to an insufficient simulation of the environmental conditions within the grain storage, while the latter has been identified as lacking interpretability, especially for a large amount of historical grain temperature data. In light of the aforementioned limitations, this paper proposes a novel grain temperature prediction model that integrates thermal simulation and machine learning residual correction. The model employs a multi-layer perceptron (MLP) to refine the outputs of the JMAG thermal simulation of the grain storage. This approach not only preserves the interpretability of the thermal simulation but also improves the accuracy of grain temperature prediction.

Zishuai Dai, Yu Wang. Thermal Multi-layer Perceptron Integrated Model for Grain Storage Temperature Forecasting. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 11: 86-95. https://doi.org/10.25236/AJCIS.2024.071112.

[1] Deepak K, Prasanta K. Reducing Postharvest Losses during Storage of Grain Crops to Strengthen Food Security in Developing Countries[J]. Foods, 2017, 6(1):8. 

[2] Ghose, Bishwajit. Food security and food self‐sufficiency in China: from past to 2050[J]. Food & Energy Security, 2015, 3(2).

[3] Zhan L, Du Z X. Thoughts and Suggestions on Promoting Grain Loss Reduction and Waste Reduction through Integrated Food Chain Management[J]. Economic Review, 2021, 01: 90-97.

[4] Zhao X. Research on Strategies for Post-Harvest Grain Conservation and Loss Reduction in China [J]. Rural Economy, 2023, 01: 1-9.

[5] Zhao L, Wang J, Li Z, et al. Quasi-distributed fiber optic temperature and humidity sensor system for monitoring of grain storage in granaries[J]. IEEE Sensors Journal, 2020, 20(16): 9226-9233.

[6] Subrot Panigrahi S, Singh C B, Fielke J, et al. Modeling of heat and mass transfer within the grain storage ecosystem using numerical methods: A review[J]. Drying Technology, 2020, 38(13): 1677-1697.

[7] Duan S, Yang W, Wang X, et al. Forecasting of grain pile temperature from meteorological factors using machine learning[J]. IEEE access, 2019, 7: 130721-130733.

[8] Zhang X, Zhang H, Meng Q. Research on temperature field of wheat grain piles in underground granary [J]. Starch‐Stärke, 2023, 75(7-8): 2200260.

[9] Jia C, Sun D W, Cao C. Computer simulation of temperature changes in a wheat storage bin[J]. Journal of Stored Products Research, 2001, 37(2): 165-177.

[10] Lawrence J, Maier D E, Stroshine R L. Three-dimensional transient heat, mass, momentum, and species transfer in the stored grain ecosystem: Part I. Model development and evaluation[J]. Transactions of the ASABE, 2013, 56(1): 179-188.

[11] Olatunde, G., Atungulu, G. G., Sadaka, S. CFD modeling of air flow distribution in rice bin storage system with different grain mass configurations[J]. Biosystems engineering, 2016, 151: 286-297.

[12] Ni F. Exploration of Temperature Field Prediction Methods in Horizontal Ventilation Processes Based on SVM Optimized by Intelligent Algorithms[J]. Grain Storage, 2017, 46(1): 28-36.

[13] Guo L J, Wang Y X. Research on storage grain temperature prediction based on XGBoost optimization algorithm[J]. Grain and oil, 2022, 35(11): 78-82.

[14] Wang H, Cao Y, Li Y, et al. Establishment of BP neural network prediction model of Pingfangcang grain temperature based on improved particle swarm optimization algorithm [J]. China Journal of Cereals and Oils, 2023, 38(06): 113-118.

[15] Duan S, Yang W, Wang X, et al. Temperature forecasting for stored grain: A deep spatiotemporal attention approach[J]. IEEE Internet of Things Journal, 2021, 8(23): 17147-17160.

[16] Ge L, Chen E. Research on grain storage temperature prediction model based on improved long short-term memory[J]. Journal of Computational Methods in Sciences and Engineering, 2021, 21(5): 1145-1154.

[17] Qu Z, Zhang Y, Hong C, et al. Temperature forecasting of grain in storage: A multi-output and spatiotemporal approach based on deep learning[J]. Computers and Electronics in Agriculture, 2023, 208: 107785.

[18] Jiang W, Jahns T M. Coupled electromagnetic/thermal machine design optimization based on finite element analysis with application of artificial neural network[C]//2014 IEEE Energy Conversion Congress and Exposition (ECCE). IEEE, 2014: 5160-5167.

[19] Rodríguez F, Genn M, Fontán L, et al. Very short-term temperature forecaster using MLP and N-nearest stations for calculating key control parameters in solar photovoltaic generation[J]. Sustainable Energy Technologies and Assessments, 2021, 45: 101085. 

[20] Li Y F, Zhang L L, Cao Y, et al. Determination of thermal conductivity of wheat [J]. Journal of Henan University of Technology (Natural Science Edition), 2010, 31(01): 67-70.

[21] Wang Xue, Zhang Yuan, Lian Feiyu, et al. CFD Simulation of Heat Transfer Model of quasi-static grain Reactor [J]. Agricultural Machinery, 2012(24): 111-114.