| In recent years,China ’s coal production has maintained sustained and steady growth,and coal will remain the main energy source in China for a long time to come.However,in the process of coal mining,goaf fire seriously threatens the safety of mine production and the safety of coal miners.At present,the monitoring methods of coal spontaneous combustion in goaf mainly include beam tube monitoring method,fiber grating monitoring method,gas index method,etc.,but these monitoring methods have problems such as limited monitoring range,poor real-time performance,high cost and easy damage.Based on these problems,based on Lo Ra wireless communication technology,this thesis uses temperature as the monitoring index to monitor the temperature field in the goaf in real time.Based on this,a temperature prediction model of the goaf is established based on BP neural network,so as to monitor and predict the fire in the goaf.Early detection,early identification,early warning,and early research and judgment.The main work and conclusions of this thesis are as follows :Combined with the actual situation of goaf,the advanced Lo Ra communication technology and automatic control technology are used to realize the remote monitoring and real-time early warning of goaf temperature,and improve the efficiency and accuracy of goaf safety monitoring.Firstly,the overall architecture of the wireless monitoring system includes the host computer,communication base station,switch,gateway and terminal node,and briefly outlines the propagation of wireless instructions between the nodes of the system and the behavioral response between related nodes when the system is working.From the aspect of hardware,the design requirements and selection of the transmission module,temperature sensor and gateway of the terminal node are put forward.The protection shell of the terminal node is designed independently.The protection shell of the terminal node adopts the design of the inner and outer layers.The outer layer provides support and protection,and the inner layer provides dustproof,waterproof,weathering and insulation.Using object-oriented programming,graphical interface design,software engineering management,multi-threaded programming and other technologies to design and develop the host computer software.The main functions of the host computer software are : RF parameter setting function,automatic early warning function,terminal node temperature access function,field strength test function,and routing setting function.Finally,the real-time monitoring of the temperature field in the monitoring area is realized and the temperature field in the monitoring area is visually presented in the form of cloud map.In front of the decision maker,with the real-time collection of temperature data,the temperature field cloud map will also be updated in real time.Then,the functions of the host computer software were tested one by one,and the real-time cloud image of the temperature field in the monitoring area was successfully obtained.The upper computer software applies visualization technology to goaf temperature monitoring and early warning,which improves the perception ability of goaf safety and provides a more intuitive reference for decision makers.In practical application scenarios,Lo Ra signals face great attenuation and interference in the downhole environment.In order to better verify its communication capability,the field test of the communication performance of the monitoring system was carried out.The test covers the communication characteristics of the system in different types of roadways,and deeply explores the transmission distance limit of the terminal node inside the goaf.The communication performance of the monitoring system in different roadways was tested in the track auxiliary transportation crossheading of the 020101 working face of Gujiao Panglong Coal Industry.It was concluded that the reasonable spacing between the two communication base stations in the straight roadway was 180 m,and the reasonable spacing between the communication base stations in the roadway with corners was 125 m.The ultimate transmission distance of the terminal node of the monitoring system in the goaf was tested in the goaf of the 23 lower 10 working face of Jining No.2 Coal Mine of Yankuang Energy Group,and the reasonable spacing between the two terminal nodes in the goaf was 5m.On the basis of the monitoring system,this thesis also designs a goaf temperature prediction model based on BP neural network.The model takes the temperature influencing factors of the past time as the input of the training set,the temperature of the past time as the output of the training set,adjusts to the best training parameters,and takes the temperature influencing factors in the future time as the input of the network simulation.The output result is the temperature prediction value in the future time.The temperature prediction model has been applied in the 3301(lower)working face of Xinglongzhuang Coal Mine,showing high accuracy and reliability.The maximum absolute error between the temperature prediction results and the actual results is 0.32 °C,and the maximum relative error is 0.9 %.The application of temperature prediction model can provide an important reference for temperature monitoring in goaf and provide a strong guarantee for mine safety production.The wireless monitoring system of goaf temperature designed in this thesis has the advantages of low cost,strong adaptability,no monitoring blind area and high sampling frequency.The system can obtain the temperature field in the monitoring area in real time and present it to the decision makers in the form of real-time updated cloud images.In addition,this thesis also develops a goaf temperature prediction model based on BP neural network,which realizes the effect of monitoring and forecasting,and wins more time for decision makers to take measures.These innovative designs provide effective means and scientific basis for the early detection of goaf fires.This thesis has 54 figures,9 tables,and 84 references. |
PDF Full Text Request