Research On Energy Consumption Balance And Localization Of Downhole Safety Monitoring Networks

Safety monitoring system is designed to provide real-time monitoring for the environment of coal mine underground operation to ensure personnel safety.Based on wired network,all current safety monitoring systems have their disadvantages including poor scalability,complex wiring,high costs and maintenance fees,thus hindering the development of the system.Wireless sensor networks(WSNs)have been widely applied in a multitude of domains thanks to their lower cost,smaller node size,self-organization and other advantages.The introduction of WSNs to downhole safety monitoring system can effectively address the defects of wired systems to increase monitoring level,laying the groundwork for the construction of Smart Mine in the future.In order to ensure the reliability of monitoring,underground WSNs must have precise positioning.With precise positioning,users can easily carry out effective monitoring of specific sites,take timely and accurate emergency or rescue measures in the event of an emergency or disaster.Stable and reliable is the premise to ensure the precise positioning of WSNs,and the node energy consumption is easy to cause instability of the network.How to deploy and manage WSNs rationally in a harsh underground environment so as to reduce energy consumption of nodes and extend the network life means a huge challenge for the research on WSNs.Built on previous research findings,this paper conducts further in-depth studies in energy consumption balance and node localization of underground WSNs,with"The Development and Application of Integrated Automation System for Coal Mines in Southern China"(2011H6027),a major scientific and technical project in industry-academy-research cooperation carried out in Fujian Province,as its background.The following are the main tasks and innovations:1.This paper proposes an energy consumption balance strategy based on the combination of multi-coverage with clustering optimization and a more effective roadway zoning method according to existing network partition.On the premise that multi-coverage is fulfilled,a roadway is divided into several isosceles triangles,and sensor nodes are deployed at the vertex of the triangle.The cluster head adopts a mechanism of working alternately,and the optimal size of cluster can be worked out through the analysis of energy consumption model to help network energy consumption get balanced,improving node energy utilization and prolonging the network life.2.This paper presents a deployment strategy of multi-sink node subsection optimization based on node distance optimization.An optimal distance for node placement can be obtained under the condition that the number of nodes is given so as to achieve a balance of energy consumption.On the basis of deployment in optimal distance,this paper also proposed the multi-sink node subsection optimization deployment.The placement of sinks at each monitoring area section can shorten node forwarding distance and reduce energy consumption,further extending the service life;also,linear programming is used to solve for the optimal number of sinks and subsections,respectively.3.This paper suggests an optimal forwarding strategy for node data based on the uniform distribution of underground WSNs.For underground WSNs with uniformly deployed nodes,the current method of data transferring in a single and stepwise manner is easier to cause the "energy-hole" problem.The method of data distribution and transmission is used in this paper to avoid uneven network energy consumption,allowing nodes to forward different amounts of data to any successor nodes and working out optimal forwarding destination node and data volume through optimization.4.This paper proposes an iterative weight-centroid localization algorithm based on Radio Signal Strength Indications(RS SI)clustering.For the impact of downhole noise interference and roadway reflection,K-means clustering is used to effectively filter out RSSI.Parameters of signal path loss are acquired dynamically in real time to get comparatively ideal RSSI ranging.Iterative weigh-centroid method is proposed as existing centroid algorithm can't get high localization accuracy.Weighted iteration is not only an embodiment of the influence of beacon node,but also can get higher localization accuracy.5.This paper brings forward a kind of underground personnel localization algorithm built on weighted Monte Carlo method.Considering the shortcomings of current underground personnel localization method,an improved localization method has been proposed using the Monte Carlo thought.In the stage of sampling initialization,sampling areas are gradually built according to the size of RSSI and combined with the Monte Carlo method to further narrow the range,thereby improving successful sampling rate.During the filtering phase,the quality of samples within the t area can be further increased,with reference to the t-1 sampling area,after two filtrations.During the resampling phase,the number of samples can be reduced comparing with the set of invalid samples and the selected samples are weighted to improve localization accuracy of unknown nodes.