Design And Implementation Of Wireless Monitoring System For Methane Based On Mobile Phone APP

Methane belongs to flammable and explosive gas,which has an extensive range of use.The explosion accidents caused by methane gas happened annually and it suffered great losses.Therefore,how to detect the concentration of methane effectively has become a concerned issue.However,the traditional methane sensors,which are based on the theory of ectrochemistry,catalytic combustion and others are unable to satisfy the demands of people for measurement precision and service life.The traditional detection devices require workers to measure on field,which is inconvenient and unsafety.Focusing on the above problems,this thesis adopted methane sensor which based on the principle of infrared absorption spectroscopy,combined embedded technology which used STM32 as the core of control,wireless communication technology and the mobile phone APP to design a system for monitoring methane.That can use mobile phone APP to realize a remote and real-time measurement in a wireless WiFi environment.Adopting the algorithm of SVM to improve the measurement accuracy of methane sensor in the process of system calibration.The system effectively overcame the problems of poor real-time detection and low precision for traditional methane detection.Firstly,the thesis introduced the basic theory of infrared methane gas detection and the main component of infrared methane sensor.Secondly,the thesis designed the hardware and software for the wireless monitoring system of methane.On the hardware side,the STM32 was used as the core of control to design the module of system power,the module of data collection and the module of wireless communication.On the software side,the programming of STM32 used the language of C,which realized the reading and processing of data for infrared methane sensor and it transfered the processed data to the platform of server by the module of WiFi.Then,the thesis made the mobile phone APP based on the language of Java,which receives and shows the data for methane concentration.Finally,in the process of system calibration,SVM algorithm was introduced to regress.A hybrid optimization algorithm consisting of artificial bee colony(ABC)algorithm and particle swarm optimization(PSO)was proposed to optimize the parameters of SVM.Adopttingoptimized SVM to construct the regression model of ABC-PSO-SVM and comparing with the ABC-SVM and PSO-SVM regression models constructed by single ABC algorithm and PSO algorithm.The result shows that the ABC-PSO-SVM regression model has superior prediction accuracy and the measurement accuracy of methane is higher than the demands of market.