Fractal characteristics and its application in electromagnetic radiation signals during fracturing o

Intemational Joumal of Mining Science and Technoloyy 22(2012)255-258Contents lists available at SciVerse ScienceDirectInternational Journal of Mining Science and TechnoloELSEVIERjournalhomepagewww.elsevier.com/locate/ijmstfractal characteristics and its application in electromagnetic radiation signalsduring fracturing of coal or rockang Chao…*,Ⅺ u Jiankun,zhaoⅪ Xiaoxia, Wei mingyaoSchool of Safety Science and Engineeringg Technical University, Fuxin 123000, Chinahool of Safety Engineerng China University of Mining 8 Technolog. Xuzhou 221008, ChinaSchool of Land Resource Engineering Kunming University of Science and Technology. Kuning 650093, chinaARTICLE INFOABSTRACTThe present study analyzed the electromagnetic radiation(Emr)time series of the destructionofReceived 12 August 2011coal or rock sample under uniaxial loading and the monitoring process in working face by means of frac-Received in revised form 29 August 2011al geometry, and results of the correlation dimension change curve of EMR time series were obtained. InAvailable online 28 March 2012he meantime. the current study also sought the fractal charactenstic to the emr signals by contrast tothe change curve of EMR signals and explored the precursory phenomenon of rock burst. This paper con-cluded the main findings as followed: the EMR time series of the destruction process of coal or rock sarmetic radiationunder uniaxial loading and the monitorng process in working face corresponded to fractal; theactal characteristiccorrelation dimension of EMR time series reflected the process of coal or rock damage deforrelat ion dimensionis, the inner damage of coal or rock made a change from random to order In the field application, theRock burstorrelation dimrved as a new index of forecasting the coal or rock dynamic disaster.e 2012 Published by Elsevier B V. on behalf of China University of Mining Technologyof EMR signals cannot be described by traditional geometrye they are complex unsmooth and undifferentiable. How-Rock burst is one of the most severe natural disasters in thebeer usehe fractal theory can abstractly describe complex curvesworlds mining industry and threatens the life security of miners. which are self-similar, unsmooth and undifferentiable. In orderIn particular. the rock burst is extremely severe in China which is to forecast rock burst, this study analyzed the EMR signals curveone of the biggest countries in coal production and consumption by means of fractal geometry theory, which can be used to study1]. with the increase in the mining intensity and depth of coal the fractal characteristic of EMR and the precursory phenomenonmines, the geology and mining condition have become more and of rock burst (9 In this paper, we calculated and analyzed EMRmore complicated. what is more, the intensity and frequency of time series by selecting signals from the destruction process of coalrock burst occurrence has also increased. Therefore prevention of or rock sample under uniaxial loading and the monitorirock burst is of great practical value and realistic significance.in coal face. and then studied the correlation between thea EMR signals are generated when the coal or rock deforms and of EMR fractal dimension and rock burst risk thectures under the action of load, EMR is formed by charged proposed in this study shows great theoretical and practical valuesarticles moving at changing rate which occurs in the process of to forecasting rock burst.charge transfer and fracture spread caused by the different partsof coal rock shifting gears and deformation in non-uniform motion2-5]. Researches show that the EMR signals are relatively high in a 2. Fundamental principle of fractal theoryperiod before the rock burst can occur. After that period, the valuesare relatively low. Within this period, the values of EMR are either 2.1. Fundamental principleequal to, or close to or exceed the critical value then rock burstll occur 15). EMR technique, the method that uses short periodhanges of EMR signals before rock burst, is widely applied to lar structure. with the help of self-similar principle, this paperforecast rock burst [6-8apperceives the fine structure in disordered phenomena, discoversThe curves, in research object of traditional geometry, are the regularity and provides the brand- new language as well asmooth, differentiable and have the characteristic of length. The quantitative description. Fractal is generally regarded as a totaltermTe:+8613591993947chara中国煤化工 -nings with the develoCNMHG95-2686)s- see front matter e 2012 Published by Elsevier B V. on behalf of China University of Mining Technology.structure and shape, but also to information, function and time. the value of m increasing to certain degree the slope tendsThe self-similarity of statistical fractal is described in certain range stabilize. ie saturated, then the saturated slope is the correlationof scale which is named as non-scale range [91dimension D,, at this moment, the value of m is the smallestThe fractal characteristic of mechanical behavior is common in embedding dimension. The slope of line in the curve increases withnature. Some researches show that fractal characteristics exist from the addition of value m, not focusing on some certain value whichearths plates to grain failure[10]. The damage evolution process of shows that the time series is randomrock is also a kind of fractal while fractal dimension is a character-istic parameter reflecting the degree of damage to the rock [ll]. The 3. Fractal study of EMR experimentation during fracture of coalinteger dimension extending to fractional dimension in the fractalgeometry portrays the complexity of fractal set in the use of fractalor rockdimension and describes the fractal set quantitatively. the correla-ion dimension, which is measured from experimental data, is used 3. 1. Experimental methodto describe the fractal characteristic of meR time series andresearches on the evolution rule of coal or rock failure process.his experimental system consists of a SANS electro-hydraulicservo loading system, a DiSP-24 acoustic instrument, an electro-2.2. Calculation method of correlation dimensionmagnetic dynamic monitoring system, a data acquisition systemand an EMR shield system.Based on the meaning of fractal dimension, the givCoal or rock samples were obtained from a coal mine which hadis usually extended to m-dimension( m 2 3)in order to show 50 mm x 100 mm dimensions of our cylindrical samples werefatal rock bursts. TIthe information hidden in time series this method is called time.The experimental loading mode was axial displacement controldelay method, ie, the delay coordinate method [ 12,131. This method and the loading velocity was 0. 1 mm/min.algorithm firstly uses correlation integral method(c-C method)to calculate the delay time and then selects the suitable embed-ding dimension m and delay time r[12, 14 As for the raw data, it 3. 2. Results and discussionembeds in high dimension and reconstructs the phase space, andFive samples were tested in thehen calculates the correlation dimension D, by saturated correla- The relationships among the timetion dimension method (G-p method )For any given positive number r. the count of dot pairs(x, x,) in Fig. sity (referred to as energy)如四The G-P method is described as follows 114. 15]:EMR incombined in n points whose distance is less than the value ofEMR signals were generated during the process of differentin the embedding phase space can be calculated The ratio of this deformations and fracturing of coal or rock at uniaxial comprescount accounting for the total count M(M-1)/2 is expressed as: sion In the initial loading stages, patulous pores and fractures ofloaded coal or rock were gradually closed in the compression direCm(r)=M(M-1)naiM∑Hr-Ⅸx1-X),r>0(1) ton, while the MER signals increased considerably. And then EMRsignals reduced in the elastic stage and had a smoothzone. Increasing with the addition of stress, cracks of cohere Cm(r) is a cumulative distribution function, indicating the grew quickly until the main fracture, and EMR signalsdistance between the two points is less than r in the chaotic attrac- greatly. In the residual deformation stage MER signals decreasedtor of phase space. The value M=IN-(m-1)t] is the number of graduallyembedded points in m-dimensional phase space And H(x) is theAs noted in the general tendency, the EMR signals increasedHeaviside step function as follows:with the increasing stress degree of deformation and failure, thatis, before peak value stress. with an increase in loading stressH(x0X<0(2) and deformation damage degree, EMR signals increased. This indicated that EMR signals were directly proportional to the stress, At3.3. Fractal characteristic of EMR time seriesBased on the calculation method mentioned in Section 2. weprogrammed in Matlab by collecting the EMR time series duringWhile calculating the correlation dimension, if the value r is too failure process under loading, and calculated the correlationlarge, then the distances between any two points in phase space dimension of EMR time series on every stageare all less than r, and then c(r)=l and in C(r)=0. if the valueThe results show that the lower correlation dimension of emris too small, then C(r)=0 and In C(r)=-o Only when r changes signals of coal or rock under low stress reflected less micro-nupturein a certain range, then there will be the expressionin the initial loading stage. The correlation dimension of EMRsignalsincreased with the addition of stress, which suggests that thetures mainly with small scale microcracks in coal or rock were wellwhere the value of d is the dimension of state manifold in m- distributed. And then the correlation dimension began to decreasedimensional phase space.with the continuous loading, which demonstrated that large scaleThe correlation dimension D is defined asmicrocracks increased gradually and diverted to the principal rup-ture plor rock turmed from ran-中国煤化工pped to the minimumvalue bC Goal or rock and at thC Wang et al. /International Joumal of Mining Science and Technology 22(2012)28257100003000040002(00Time (si(a) Curve of stress load as a function of timeb) EMR pulses count in fracture(c)EMR energy in fractureFig 1. EMR characteristics of fractures at uniaxial stress loads.M=1.0Fig. 2. Correlation dimension change curve of EMR time seriesFig 3. EMK change curve of No 1409 working face.Based on the observation from the change of correlationsion of emr time series, we found that the correlation dimeflected well the process of coal or rock damage and themcurve of correlation dimension from February 3rd to March 18th[151. We marked three mining-induced seismicities with theirespective occurrence time and magnitude. In the change curveof correlation dimension could be taken as the precursor of main there were abnormal changes of correlation dimensions aroundfracture and used to forecast the coal or rock dynamic disasterthe point of mining-induced seismicity, which manifested thatthe correlation dimension firstly gradually increased and then de-4. EMR time series analysis of coal facecreased. The mining-induced seismicity or rock burst often occursin the process of reduction stage or drastic fluctuation stage. AsRock bursts occurred hundred times in Huafeng Coal Mine of shown in Fig. 2, this change was similar to the change of correlaXinwen Mining Industry Group since 1991. It poses the greatest tion dimension of EMR time series collected during the failure pro-reat to coal production safety. Now, in working face of Huafeng cess of coal or rock under loading. From the correlation dimensionCoal Mine, the EMR testing equipment named KBD7 on-line elec- curve, we found out that the curve changed greatly from March 2rdomagnetic radiation monitor, which was invented and developed to March 5th, which was the precursor of mining-induced seismicsee In this paper, we took the No. 1409 working face as our case of mo 2 burst, and then took measures by injecting water in theby China University of Mining and Technology, is being usedorcontinuously monitor and forecast rock burst.mining roadway from March 6th to March 8th, which effectively中国煤化工 micity or rock burst.udy. The data(taking the collected data in each working time-he average value of thean unite)collected from the KBD7 were used to calculate the cor- monCNMHGadway of Norelation dimension of EMR time series, and then obtained the relation dimension change curve of EMR time series around the.s in accord with the corchange curve of correlation dimension of each shift. Fig. 2 is thepoints of mining-induced seismicities. After water injection, the Open Fund of Laboratory in China University of Mining and TechEMR intensity decreased significantly and the fluctuation changed nology(No. 2010-I1-004). The authors are grateful to Wang Jing.slightly.the master of engineering of China University of Mining and TechFrom experimental analyses and field application, the correla- nology for her help in part of the work, and to all the coal minestion dimension of EMR time series can well reflect the process of mentioned in the paper for providing experimental environments.coal or rock damage deformation and can be a new index combined with other indexes to forecast the coal or rock dynamic Referencesdisaster in the field application.[1] Qian MG. Shi Pw. Mining pressure and strata control, Xuzhou: China[2] Frid V Electromagnetic radiation method for rock and gas outburst(1) This study condudes that electromagnetic radiations aregenerated during the deforming and fracturing of coal or Wang Er, He ko an experim ntal srudy of the ck romagnetic ems iohrock under the action of load. The EMR technique, using43(1}131-7the short period changes of EMR signals before rock burst, 14] Frid V. Vozoff K Electromagnetic radiation induced by mining rock failure. IntJwidely applied to forecast rock burst.oal geol 2005:641)57-652]The EMR time series, collected from the destruction processnder uniaxial loading, can be characterized by the fractal [6] Wangju XF, Zhao El Liu ZT. Study of electromagnetic characteristics ofand reflected well the process of coal or rock damage defor-stress distribution andes in the mining of gob- surrounded coalmation, that is, the inner damage of coal or rock turns from pI Dou LM. L CP. Mu Z G:o Ms Prevention and forecasting of rock burstrandom to order the change curve of correlation dimension hazards in coal mines Mining Sci Tech 2009; 19(5): 585-91of EMR time series collected from the working face is consisal and gas outbursttent with the change curve of EMR intensityTeh2000293):225-99!(3)In the current study, we used only qualitative analysis forrock burst by technology of electromagnetic emission(EME) J Liaoning Techprediction of rock burst in coalmine by analyzing the changeurve of EMR intensity or correlation dimension of EMR timedy of fracture system anisotropic Hydrogeod Engeries collected from working face. The quantitative analysis [11]Sun I. Rock mechanics and engineering in the 21st century. Beijing: Chinashould be studied future to research on the relationshipetween the rock burst risk and the anomaly change of 12 Ly H Lu jA chaote time senes analysisEMR intensity or its correlation dimension, as a new index [131 xie HP. Introduction to fractal rock mechanics. Beijing: Science Press: 1996.to forecast the coal or rock dynamic disaster. The correlationberger P, Procaccia L Measuring the strangeness of strange attractors.dimension of EMr time series should be combined withD1s83:9:1892Nonlinear characteristic analysis of EME signal during dynamicalother indexes in field applications.process of coal or rock. Xuzhou: China University of Mining a116] Li YH. Liu JP. ZhaO XD. Yang Y]. Study on b-value and fractal dimension ofAcknowledgerocess. Rock Soil117 Yin XG, Li SL Tang HY, Pei JL Study on quiet period and its fractalThe work presented in this paper was supported by the Fundamental Research Funds for the Central Universities in china2009:22A02):3383-90University of Mining and Technology(No. 2010QNB23)and the中国煤化工CNMHG