Design And Implementation Of Stored Tobacco Mildew Warning System Based On NB-IoT

Tobacco leaves are the material basis of the tobacco industry.Tobacco leaves must pass through a long period of storage alcohol treatment before they can meet the quality requirements.Tobacco leaves are prone to mildew in the process of storage alcoholization.If they cannot be found in time for disposal,it will cause great losses.To reduce losses,the tobacco industry has been studying the law of tobacco leaf mildew and the environmental monitoring system applicable to tobacco leaf warehouses.The traditional tobacco leaf storage environment monitoring system has the problems of low efficiency,high cost,difficult deployment,etc.,and basically only monitors the temperature and humidity information,it is difficult to detect tobacco leaf mildew in time,and cannot meet the needs of the tobacco industry.Therefore,based on NB-IoT technology,this subject designed and implemented a set of highly efficient,easy-todeploy,and targeted tobacco leaf storage environment monitoring system to reduce the burden on warehouse managers,find anomalies in time,and reduce the losses caused by mildew.In recent years,in order to meet the long-distance and low power consumption requirements of the Internet of Things market,the narrowband Internet of Things(NBIoT)has emerged.The combination of the Internet of Things and artificial intelligence provides new means for many industry problems.The use of NB-IoT technology to optimize the storage tobacco environment monitoring method,and the use of artificial neural networks to identify the storage tobacco mold mildew is the research idea of this topic.According to the needs of tobacco storage management and the characteristics of tobacco mildew,the overall structure design of the storage tobacco mildew early warning system is completed.The system is mainly composed of a data collection terminal and a monitoring server.In the realization of the data collection terminal,a variety of gas sensors are selected to form the sensor array,STM32F103RCT6 is selected as the main control chip,and Ghostyu NB200 is selected as the communication module,and the corresponding circuit schematic diagram is designed.After the process is completed,the hardware part is implemented;then the software of the data collection terminal is designed and developed,and finally the software and hardware design and implementation of the data collection terminal are completed.The BP neural network is used to establish the tobacco leaf mildew state recognition model,and the tobacco leaf mildew state recognition algorithm is deeply studied.On the monitoring server,the B / S architecture was used to build a tobacco leaf storage environment monitoring platform,which completed the data reception,database design,and page implementation in turn.Warehouse workers can log in to the platform to check the temperature,humidity,carbon dioxide,ethanol and other gas concentrations of the storage environment and the mildew status of the tobacco leaves,and to understand the storage conditions of the tobacco leaves in detail,so as to discover problems in time.After actual measurement,the system meets the requirements of tobacco leaf storage environment monitoring,can better reflect the mildew state of tobacco leaves,stable operation,and has high practical value.