| The large-scale mining of coal resources has broken the initial stress balance state of surrounding rock,resulting in collapse,fracture and bending deformation of overlying strata.When the overburden movement spreads to the surface,a series of mining damage phenomena will occur,such as surface subsidence,railway deformation,building damage and so on.The development characteristics of overburden fracture field can be inversed through physical simulation experiment.If digital image processing technology can be used to realize the automatic extraction of mining fractures in the fractures in the process of experiment,it can significantly reduce the error manual interpretation and realize the accurate identification of fracture distribution range,shape and development height.Therefore,it is of practical significance to carry out the research on the law of mining surface deformation and automatic identification of overburden fracture field.Taking 15235 working face of Jiulong mine in Fengfeng mining area as the engineering background,using various research means and methods such as field measurement,theoretical analysis,3DEC discrete element numerical simulation and similar material experiment,this paper analyzes the characteristics and laws of mining surface movement,determines the rock movement angle parameters in the test area,inverts the prediction parameters of surface movement and deformation,and establishes the relationship model between surface subsidence rate and mining degree based on Logistic function.In the process of physical simulation experiment,the Orthophoto image of the model is obtained by using the multi baseline close-range photogrammetry method,the inter-class variance function considering the weight coefficient is constructed,and the optimal threshold expression of image segmentation is determined.Based on the research and development platform of MATLAB,the calculation program is compiled to realize the effective segmentation of target image foreground and background information and the automatic recognition of model fracture field.Finally,the pixel method and fractal theory are used to quantitatively analyze and evaluate the identified fracture field.The main work of this paper is as follows:(1)Field measurement and Research on the law of surface movement and deformation:combined with the geological and mining conditions of 15235 working face,the surface movement observation station is arranged and monitored periodically.The surface movement and deformation curve is drawn,the characteristics of surface dynamic movement and deformation are analyzed,and the angular parameters of surface subsidence are determined,in which the boundary angle is 52.7,the average value of the leading influence angle is 65.95,and the lag angle of maximum subsidence speed is 69.1.Combined with the subsidence speed of the maximum subsidence point,the duration of surface movement is determined,in which the initial period,active period and decline period are 120 d,30d and 420 d respectively.(2)Inversion of surface and overburden migration law based on 3DEC: according to the geological and mining conditions in the test area,a 3DEC discrete element numerical analysis model is established to study the migration characteristics and law of overburden and surface.Firstly,according to the measured data,the reliability and rationality of the numerical calculation model are tested(including the model grid size,boundary condition setting,block and joint mechanical parameter setting).According to the mining results of six simulated working faces,the surface deformation information of the main section of the model is extracted,and the process of working face from extremely insufficient mining→insufficient mining→full mining is inversed,The dynamic development and change process of surface movement and deformation.Then,according to the subsidence characteristics of the maximum subsidence point,the relationship model between the mining degree of the working face and the surface subsidence rate based on logistic function is constructed.From this,the critical value of mining degree from extremely insufficient mining to insufficient mining is0.34 and the limit subsidence rate of surface is 0.72.Finally,according to the numerical simulation results of subsidence and horizontal movement,the predicted parameters of probability integral are inversed.(3)Automatic identification and quantitative analysis of mining overburden fracture field: using the close-range photogrammetry method,the automatic identification and quantitative analysis of mining overburden fracture field based on similar material model are carried out.Taking the experimental working face as the prototype,the simulation experiment of similar materials is designed.The Orthophoto Image of the research area is obtained by splicing and correcting the close-range photogrammetric images obtained in batch by the fixed focus camera.Aiming at the deficiency of Canny algorithm in image information threshold segmentation,an adaptive threshold segmentation optimization algorithm is proposed to realize the automatic recognition of apparent fracture field of similar material model.Based on the research and development platform of MATLAB,the quantitative analysis of overburden fracture field with different pushing progress of working face is carried out by using pixel method and fractal theory.The results show that the information of fracture field is extracted completely,and the optimized Canny algorithm can realize the accurate identification of mining overburden fracture field in physical simulation experiment.38 pictures,14 tables,61 references... |
PDF Full Text Request