| Dangerous gas leakage is extremely easy to occur in the production workshops of modern petrochemical enterprises,and the leakage problem cannot be completely stopped.Therefore,people have to start from the initial stage of dangerous gas leakage and conduct online monitoring on the places prone to leakage.If a dangerous gas leakage occurs in a workshop,and can be accurately detected and positioned at the first moment,it can guide different petrochemical workers to take emergency measures quickly.Therefore,this thesis designed a set of online monitoring system for the location of dangerous gas leakage,and studied the relevant gas source location algorithm for the system.Firstly,aiming at the urgent demand of petrochemical maintenance personnel for mobile terminal of monitoring system,this thesis designs an embedded mobile terminal monitoring platform and successively designs monitoring nodes,routing nodes and coordinator nodes according to the modular design idea.Based on the S5PV210 development board as the hardware platform,the embedded mobile terminal uses various gas sensors and wind speed and direction sensors to collect real-time environmental information of the petrochemical workshop,realizes the transmission of gas data through ZIGBEE wireless sensor network,and completes the design of QT monitoring interface and device driver based on the embedded Linux system.Secondly,in order to accurately locate the gas source,based on the two common gas diffusion models,the general expressions of the gas turbulence diffusion models suitable for both the calm and the wind were derived,and several classical gas source localization algorithms were introduced based on this model.Then,in view of the nonlinear characteristics of the gas diffusion model under wind conditions and the reduction of positioning accuracy of the NLS algorithm caused by rough initial value estimation,this thesis proposes a gas source positioning algorithm based on particle filter for NLS(PF-NLS)initial value optimization.This algorithm uses the estimated results of NLS algorithm as the initial value,and accurately estimates the position and strength of the gas source through PF algorithm.Finally,the accurate location of the gas source can be realized under the condition that the strength of the gas source is unknown.Simulation results show that the PF-NLS algorithm is superior to the NLS algorithm when the gas source strength is unknown,and the algorithm has higher positioning accuracy than the NLS algorithm in windy conditions,and further improves the convergence speed of the algorithm by optimizing the initial value.At the same time,the positioning performance of PF-NLS algorithm was analyzed under different wind speed and measurement noise conditions,and the practicability of the algorithm was verified.Finally,the system location function is realized by PC Kingview monitoring platform,the data communication between Kingview and MATLAB is realized based on OPCtechnology,and the accurate location of gas source is realized by PF-NLS algorithm,and the location results are presented on the Kingview interface.The monitoring platform can display the gas concentration status of monitoring nodes in each region in real time,conveniently view the concentration data curve of each node,and provide the corresponding workshop parameter report form,which is convenient for petrochemical managers to timely query the dangerous gas leakage information and make a quick and accurate decision.The results of this study can be widely applied to the search and location of toxic and harmful gas leakage sources,the monitoring of hazardous environment,the detection and warning of fire sources. |
PDF Full Text Request