Research On Intelligent Monitoring System In Grain Situation Based On Bp Neural Network

Grain storage environment is a complex system consisting of a variety of factors, Grain security situation have a closely relationship with microbial activity、temperature、 humidity and CO2concentration,conventional Grain monitor device and prediction method could hardly meet the height and precision of Grain monitoring.A new grain monitoring model was builded on the flexibility of BP neural network.According to the BP neural network characteristics of low precision, slow constringency and local minimum In the learning process,The model used the momentum method to adjust the weight of the grain situation factors,used Rapid-momentum to adjust the learning rate of the grain situation, used L-M algorithm to improve the monitoring network of the grain situation.By collected CO2concentration temperature and humidity information, trained and predicted grain samples and compared the effect with conventional prediction method. Experimental results show that the comprehensive improvement BP neural network can do better in the monitoring of grain situation and meet the monitoring requirements of the grain situation.This paper designs a set of grain situation intelligent monitoring system.Hardware includes the embedded ARM9processor as the core、the controller by single chip microcomputer control unit、information collection module、Peripheral control module and16road gas acquisition channel.Software includes the Grain situation Intelligent monitoring system program、singlechip control-program、Grain situation information acquisition program and the serial communication program.Among them,intelligent monitoring system program key includes the Measurement module、the control module、many regional information fusion module and BP algorithm forecast module.The paper key introduced the the software and hardware design Of the intelligent monitoring system of grain situation、grain situation factor weighted fusion Based on the collection area and BP prediction principle.The experimental results show that monitoring system of grain situation with scene with good stability,convenient collection, real-time on-line detection,etc,For grain situation monitoring has important significance.