Acquisition And Processing Of Underground Critical Data In Digital Mine

With the rapid development of economy,the demand of mineral resource is increasing sharply and different casualty accidents are in rising trend.Safety accidents occur frequently in underground mines,and the mined out space due to long-term mining is one of the main hazard source under the shaft.If large number of mined out spaces exist in a mine,the mining conditions will deteriorate;the mine props will become deformed and the maintenance of adjacent quarries and alleys will be difficult.In serious situation,large area in-break,rock movement and gushing water,etc may occur under the shaft.Therefore,in the construction of digital mine,the collection of three-dimensional data of underground mined out space is a key and difficult point.With the development of three-dimensional laser scan technology and gradual decline of the instrument,its application in different fields must be widespread,relevant research will go deeper,and the application of CMS in data collection of underground mined out space must have a good prospect.At present,most of the scholars divide the three-dimensional scan system into airborne laser radar system and ground laser radar system according to different carriers.However,the current research of underground three-dimensional laser scan technology is relatively few,and it is generally included in the ground three-dimensional laser scan technology to research.The sweeping survey content of underground three-dimensional laser scan technology is broad:besides sweeping the mined out space for mining underground mine,it may also be used for sweeping shaft and drifting data required by the digital mine,internal conditions of solution cavity,underground factory building of hydropower station,tunnel(drawing water,highway,railway)and other underground projects,especially the unsafe regions where the personnel cannot get in.It is easy to find out the characteristics of ground object on a point,line or surface,and the preprocessing algorithm of many data is based on these characteristics.However,due to the complex and irregular shape of underground space of underground space and that many places cannot be seen with naked eyes,thus the scan method and data processing become more complex.In three-dimensional reconstruction,the preprocessing of data plays the role of a connecting link between the preceding and the following,and the performance of this step directly influences the quality of three-dimensional model.Aiming at this specialty of the underground project,this paper selects preprocessing of point cloud sata of three-dimensional laser scan as the main research content.The main research content of this paper is as follows:1.Analyze and research the basic framework of digital mine,method of getting three-dimensional data of the mine,current status of application of three-dimensional laser scan in mine both at home and abroad,technical route of processing the three-dimensional laser scan data of the mine,establishment and visualization of the three-dimensional model,etc.2.Research the structure,principle and application of three-dimensional laser scan system;give a detailed introduction of the principle,performance,its operation method under the mine and data processing flow in early stage of CMS;analyze the characteristics and advantages of CMS in the data collection of underground mined out space;display the specific application of CMS and compare the performance of various kinds of three-dimensional laser scan systems.3.Analyze and research the source of error and influencing factors of CMS;design experiment to demonstrate the assessment method of its internal and external accuracy and draw a conclusion;aiming at the fact of low external accuracy,make quantitative analysis of the errors under different scan methods,through the experiment,this paper discovers the tilting direction,side rolling direction and azimuth angle errors in the scan and measurement of CMS;summarize the rules of each error in different angle and at different distance and put forward the correction methods.4.Research the de-noise method of point cloud data of CMS collection;make primary study of the exclusion algorithm of isolated point according to the accuracy requirement of the mined out space;and analyze the smoothing algorithm after de-noising.5.Design multi-angle scan method of the mined out space,formulation and algorithm of the identical point;research the combination algorithm of point set to point set:four elements method,iterative closest point algorithm ICP and its improved algorithm;make primary study of the data fusion theory which is based on the point cloud combination of cluster theory,including inspection of overlapping areas,calculation of initial core point and cluster fusion.6.Based on the research of theory of completing the point cloud data cavity,provide a simplified algorithm for completing the point cloud data cavity:including the recognition,clearing and projection of boundary points of point cloud data cavity;triangularization of simple cavity,simplification of complex cavity;triangulation of cavity,updating and creation of top point and triangle.7.Based on overall review of the feasibility,advantages and disadvantages of different types of simplified algorithm of point cloud data,provide the assessment system of the simplified algorithm of point cloud data;analyze the advantages and disadvantages of current data simplified algorithm in the simplification of point cloud data in mined out space according to the data collection method of CMS and shape of mined out space;put forward simplified algorithm of point cloud data of underground space based on the slice,including slice processing,determination of threshold value,fitting of slice border line,drawing characteristic points of slice border line,etc;and bring forth improved method of deflection angle to get the characteristic points of slice border line.Through the empirical checking of the data in mined out space,this algorithm is considered to be feasible and its simplification rate is very high.To sum up,this paper takes the research and realization of the three-dimensional geographical information system of copper mine in Mountain Dahong as the background,the data collection and processing of underground mined out space of CMS as the mainline,the experiment as basis and practical application as premise.It researches the principle,data collection method and data output of CMS and designs accuracy assessment method.The experiment discovers the system error and this paper establishes error correction model.Centering on the preprocessing content of point cloud data of three-dimensional laser scan and combining with the characteristics of three-dimensional space in the mined out space,it researches the de-noising and smoothing algorithm of point cloud data,point cloud combination alignment and point cloud data fusion algorithm;puts forward a point cloud data fusion algorithm based on cluster;makes primary study of simplified algorithm of point cloud data based on slice;and brings forth improved method of deflection angle to conduct the algorithm of getting characteristic points of the border.This paper explores and establishes a set of relatively complete methods in data collection and correction,preprocessing of point cloud data,which aims at providing high-quality,high-efficiency measurement data for establishing three-dimensional model and application in mined out space and offer valuable reference to the research and application of three-dimensional laser scan technology in the scan and measurement of underground space.