Relocation Of Drift Node After Mine Disaster

As the main energy source in China,coal is the basic raw material for the national economy.However,coal mining has been threatened by various accidents for a long time,so the research on mine disaster relief is of great significance.In the mine disaster rescue mechanism,determining the location of disaster victims and trapped personnel is a top priority,which cannot be separated from the underground information construction and accurate positioning of the mine.The foundation of information construction under the mine is the mine Internet of Things,which connects sensor devices and networks to monitor the mine environment,equipment,and personnel.After a disaster occurs in a mine,relocation faces two problems.One is that the nodes arranged before the disaster may drift and the location is uncertain.If the location information before the disaster is used to locate,positioning errors will occur;The other is the increase of obstacles in the roadway after the mine disaster,the deterioration of the environment,and the occurrence of non line-of-sight propagation of signals,resulting in errors in positioning results.To solve the above two problems,this thesis proposes corresponding algorithms and solutions.First of all,a CWIT algorithm is proposed for relocation of drift nodes after mine disasters.Firstly,experiments are conducted using UWB nodes to obtain the influence ratios of temperature,humidity,and wind power on signal strength values.Then,factors that affect the reliability of the nodes are analyzed.A reliability model is established from three aspects: the surrounding environmental factors of the nodes,the information of the surrounding nodes,and the energy of the nodes.The FAHP-entropy weight method is used to calculate the weight of each factor.Finally,in order to calculate the reliability of drift nodes,the traditional TOPSIS algorithm has been improved to solve the problems such as reverse order.A CWIT algorithm is proposed,which selects the node with the highest reliability as a beacon node to achieve relocation.In the second place,for localization in the NLOS environment,in the mine geometry model,this thesis first proposes an NLOS recognition algorithm RGM,which can quickly determine whether a node is in the NLOS environment.Through simulation analysis,it is determined that when the parameter value is in the range of [2,2.5],the NLOS recognition accuracy of the RGM algorithm is the highest,reaching over 90%.Subsequently,RGM-IWST algorithm is proposed to suppress NLOS.Simulation results show that this algorithm is effective in suppressing NLOS and can improve positioning accuracy.There are 38 figures,6 tables and 94 references in this thesis.