Research And Application Of Optimal Location Method Of Mine Water Inrush Monitoring Network Based On Internet Of Things

In recent years,with the advancement of industrialization,informatization and intelligence of the coal industry,the coal mining technology has been continuously improved,and the total number of mine accidents has shown a downward trend,but serious water inrush accidents causing casualties still occur from time to time.Based on the current technology can not accurately predict the time and place of mine water inrush,this thesis proposes a monitoring network for mine water inrush with networked water level sensors in the mine roadway,and studies the optimal layout method of mine water inrush monitoring network.By optimizing the layout of the monitoring network for mine water inrush,the depth of water accumulation and the spread scope of water inrush in mine roadway can be monitored in real time,and the occurrence of mine water inrush can be monitored within the monitoring response time,which provides a new technical method for the prevention and control of mine water disasters.In this thesis,the optimal layout of the monitoring network for mine water inrush is transformed into the optimal location of water level sensors.Firstly,in view of the serious consequences of mine disastrous water inrush,it is proposed to monitor the occurrence of water disaster from the perspective of the whole mine for the first time,and creatively abstract the optimal location problem of water level sensors as the Location Set Covering Problem(LSCP).Taking the drainage capacity of different areas of the mine as the critical value of water inrush,the water inrush process at different locations is simulated by SWMM software.According to the simulation data,the monitoring scope of water level sensors at different positions is obtained.Taking the monitoring response time as the coverage radius and minimizing the number of water level sensors as the goal,a single objective optimal location mathematical model of water level sensors based on the LSCP is established.The greedy algorithm is used to solve the model,and the experimental analysis and verification are carried out.Secondly,on the basis of the above single objective,an objective is added,that is,the average time when water inrush is monitored,the multi-objective optimal location mathematical model of water level sensors is established,and the NSGA-Ⅱ multi-objective genetic algorithm is optimized and improved by using the approximate solution of the single objective optimal location model.Finally,a multi constraint and multi-objective optimal location model of water level sensor with cost(number of sensors)and average monitoring time as objective functions under multiple constraints such as mining area,risk level of mine water inrush and fixed-point of installation is established,and the improved NSGA-Ⅱ multi-objective genetic algorithm is used to solve it,enhancing the practicability of the monitoring network for mine water inrush.The main research contents and achievements are as follows:(1)Starting from Internet of Things(Io T)in mine,the concept of monitoring network for mine water inrush is proposed by summarizing and analyzing the key technologies of Io T in mine(perception layer technology,network layer technology and application layer technology).Its main feature is to monitor the water flow depth in real time through the networked deployment of water level sensors in the mine roadway,so as to provide data support for early warning before and emergency rescue after the mine water inrush.It is proposed to optimize the monitoring network for mine water inrush from the perception layer,network layer and application layer.This thesis focuses on the optimization of the perception layer of the monitoring network for mine water inrush.On the basis of a large number of investigations on traditional water level sensors,a new submersible mine water level sensor was developed.The device has better waterproof performance than the traditional water level sensor,and is more suitable for network deployment in the mine roadway.The key factors affecting the location of water level sensors are analyzed,and the factors that can be processed numerically are extracted,which provides quantitative parameters for the establishment of the optimal location model of water level sensors.(2)The location covering problem in facility location theory is summarized,and the Location Set Covering Problem(LSCP)is studied in detail.The location characteristics of mine water level sensors are analyzed,and an optimal location method of water level sensors based on LSCP is proposed.This method can monitor the occurrence of disastrous water inrush from the perspective of the whole mine with the least water level sensors in the monitoring response time.Firstly,the basic concepts of potential monitoring point,water inrush point and monitoring response time T are defined;Secondly,according to the layout of mine excavation engineering and roadway traverse points,the mine roadway topology is established.Take the drainage capacity of the area where the water inrush point is located as the initial value of the water inrush,simulate the water inrush through SWMM software,analyze and process the simulation data,and obtain the spread scope of the water inrush points and the monitoring scope of potential monitoring points;Then,taking the monitoring response time T as the coverage radius and minimizing the number of water level sensors as the goal,a single objective optimal location mathematical model of mine water level sensors based on the LSCP is proposed,and the greedy algorithm is used to solve the model.(3)Based on the above single objective optimal location method of water level sensor,taking the minimum number of water level sensors and the shortest average monitoring time as the objective function,a multi-objective optimal location method of water level sensors is proposed,and the mathematical model is established,then the relevant variables and parameters are defined in detail.At the same time,combined with the greedy algorithm for solving the single objective optimal location model in Chapter3,the non-dominated sorting multi-objective genetic algorithm with elite strategy(NSGA-Ⅱ)is optimized and improved,and a multi-objective genetic algorithm with set covering model coupled with NSGA-Ⅱ is proposed,which improves the convergence and operation efficiency of the algorithm.The optimal location method can be a powerful supplement to the single objective optimal location method in the previous chapter,solve the contradiction between time and cost,and obtain the Pareto optimal solution set.(4)In order to better meet the actual needs of prevention and control for mine water,a multi-constraint and multi-objective optimal location model of water level sensors is established,such as mining area,risk zoning of water inrush and fixed-point of installation.Mine water inrush often occurs in the mining area,and the prevention and control for mine water is more focused on the monitoring and prevention in the mining area.Therefore,the constraints in the mining area are more in line with the actual requirements for monitoring mine water inrush.With the restriction of risk level of water inrush,the deployment of water level sensors focuses on the area with high probability of water inrush,and the higher the risk level of water inrush,the shorter the monitoring response time,which improves the practicability of the optimal location model.The constraints of fixed-point of installation mainly consider the location of monitoring points for the daily water inflow in the mine,so that the optimal layout scheme of the monitoring network is not only for the disastrous water inrush,but also for the the daily water inflow in the mine,which expands the practicability of the monitoring network for mine water inrush.(5)Taking Beixinyao Coal Mine in Shanxi Province as the research area,the optimal location of the water level sensors is studied by using the multi-constraint and multiobjective optimization location method,the mathematical model is established,and the multi-objective genetic algorithm coupled with NSGA-Ⅱ is used to solve the problem.Based on the optimal layout of monitoring network for water inrush in Beixinyao mine,the monitoring effect of the monitoring network is analyzed,and the problem of dynamic moving water level sensor with the advance of mine working-face is discussed and the application scenario of monitoring data is introduced.