The Design And Research Of Master-slave Discrete Laboratory Environmental Monitoring System Based On CAN Bus

The laboratory is an important place for researchers to study and research.In recent years,accidents caused by laboratory safety problems occur frequently and have caused serious consequences.At present,there is a lack of laboratory environment monitoring system in colleges and universities.Most of the research on indoor environment detection system is only aimed at living environment.To solve this problem,this paper proposes a kind of environment monitoring system based on CAN bus,which collects indoor harmful gas NO₂ concentration,temperature,humidity and other information through sensors,and alarms when the environment is abnormal,so as to reduce the occurrence of dangerous accidents.In this paper,a master-slave separate laboratory environment monitoring system based on CAN bus is designed and manufactured.The design of the proposed system is divided into two parts,the hardware system design and the software system design.Using STM32F407 single-chip microcomputer as its core,the hardware system includes master-slave system power management circuits,gas sensor resistance conversion circuit,single chip microcomputer minimum system circuit,SHT30temperature sensor circuit and other peripheral circuit.PCB diagrams are designed and drawn to complete the physical welding.The software system has the transplanted Free RTOS operating system as the general framework of program design,and the communication protocol of CAN bus is used to complete the communication between master system and slave system.The resistance change caused by gas is an important factor affecting the accuracy of the monitoring system.This paper proposes a resistance measurement method based on GA-BP neural network algorithm,which can further reduce the error caused by the linear fitting method.The resistance measurement method based on GA-BP neural network algorithm is implemented,and the design of environment monitoring system based on CAN bus is completed.Compared with the traditional linear fitting algorithm,GA-BP neural network algorithm can get more accurate resistance measurement results.The experimental results show that the resistance measurement error of the linear fitting algorithm is 0.5%,and the resistance measurement error of GA-BP neural network algorithm is 0.1%without increasing the complexity of the system circuit.The system used the porous silicon gas sensor film prepared by the research group to monitor the NO₂ gas.The experimental results showed that the system could measure the concentration of NO₂ gas at the concentration of 1.0 PPM-1.5 PPM,and could obtain the gas measurement results with an error of less than 10%.The process realized by the GA-BP neural network algorithm proposed in this paper can be directly run on the MCU,and the high-precision resistance measurement results can be directly obtained on the system,which can provide certain reference for the design of bridge circuit.