Research On Monitoring System For Dangerous Goods Warehouse Based On Internet To Things Technology

Chemical industry is the basic industry of the country,it is very important to keep the security of chemical industry.Especially the dangerous goods of chemical industry,in all aspects of the production,transportation and storage,more safety measures to prevent accidents.As the dangerous goods store,that is the longest aspect to get in touch with dangerous goods,the hidden danger of the warehouse directly threatens the safety of employees and the safety of the property.It is an urgent problem to be solved in the development of dangerous goods warehouse that how to improve the monitoring ability of warehouse.Based on this fact,the intelligent security monitoring system for warehouse has been designed to improve the efficient and intelligent of warehouse monitoring,which takes advantage of embedded technology,multi-sensor information fusion technology and neural network algorithm.Firstly,the main functions and the key parameters of the monitoring system are analyzed in this thesis,and the monitoring lower computer is completed which is based on control chip stm32.The lower computer is designed to include the humidity measurement module,the eight point temperature measuring module,the speed adjusting of the temperature regulating module and the alarm module of the doppler sensor.Using C# to carry on the software design of the upper computer,the remote monitoring and the monitoring data record of the upper computer are completed.Secondly,this thesis also studies the communication protocol between the upper computer and lower computer,the protocol of Light Wight IP(LWIP)is analyzed and ported to the hardware platform.Transport data on internet is encapsulated in a unified format,added a custom type information.And some problems is solved,such as retransmission,sticky package,disconnection detection and so on.In the monitoring of the overall environment of the warehouse need to use the information fusion algorithm,so information fusion algorithm is studied in this thesis the,and the Back Propagation(BP)neural network algorithm is selected for specific analysis.The environment data is preprocessed,then umber of input of the BP neural network is 3,the number output is 1,and the number of nodes in the hidden layer is 6 according to system mathematical model.Write MATLAB program to achieve BP neural network algorithm.Under the test of 1000 groups of samples,the prediction accuracy is up to 98.9%.The test results show that the BP neural network can accurately predict the overall environmental safety of the warehouse.Finally,the MATLAB program generates the dynamic library file,which is called by the host computer to complete the prediction function.The monitoring system realizes the monitoring of the temperature and humidity of the warehouse environment,and uses the information fusion algorithm to accurately judge the security situation,and can remotely adjust the environment.