Detection Model And System Implementation Of Granary Storage Quantity Based On The Pressure Data Statistics

Grain storage quantity monitoring is one of the important measure to guarantee food security,and the traditional methods of monitoring cannot satisfy the needs such as food management and macro-control.Therefore,the food supervision department urgently needs a food supervision technology and system with simple operation and accurate results.In this paper,the relationship between the statistics of pressure data at the bottom of the granary and the quantity of stored grain and the theory of deep belief network were deeply studied,then a detection model of the quantity of stored grain based on the statistics of pressure data was constructed and the corresponding detection system was developed.The main work of this paper is as follows:(1)For the layout of sensor had two rings at the bottom of the granary,through the analysis of the relationship between the statistics of pressure data at the bottom of the granary and the quantity of stored grain,deduced the relation model of pressure data statistics and grain storage quantity.On this basis,combined with the deep belief network,a grain quantity detection model based on the pressure data statistics was proposed.(2)In order to improve the generalization ability of the stored grain quantity detection model,a new model training loss function was proposed by combining L2 norm and custom deviation function,and the modeling optimization of the stored grain quantity detection model based on deep belief network was proposed.The experimental results verified the feasibility and effectiveness of the proposed optimization model.Aiming at the weak generalization ability of the grain quantity detection model based on the pressure data statistics in the actual detection,the new model training loss function was set based on the L2 norm and the custom deviation function,and the original model was optimized.The actual test results verified the feasibility and effectiveness of the proposed optimization model.(3)According to the characteristics of the pressure measurement data of the pressure sensor at the bottom of the granary,a data preprocessing method combined the pressure data statistics with Pauta criterion was proposed,and a specific data preprocessing algorithm was given.The practical results show that the proposed method can improve the accuracy of the detection model.(4)Combined with the actual demand of grain quantity detection,a detection system of grain storage quantity based on pressure data statistics was developed,which had a variety of functions such as detection,modeling,data storage and statistical analysis and so on.It can meet the needs of automatic detection and data analysis of grain quantity stored in granaries.