资源描述
英文原文Remnant Roof Coal Thickness Measurement withPassive Gamma Ray Instruments in Coal MinesStephen L. Bessinger and Michael G. NelsonAbstruct:Current underground mining practice often requires that a predetermined amount of coal be left on the roof of the mined-out area. The need to leave such coal occurs on both continuous miner and longwall sections is derived from considerations of ground control, quality control, machine guidance,or simply good operating practice. Efforts at measuring boundary coal thickness have been employed mechanical, nucleonic, and energy adsorption and reflection methods. The nucleonic methods have found application in operations in the United Kingdom,the United States, the former Soviet Union, and Poland. Natural gamma devices are currently the instrument of choice, and several successful installations exist. The calibration of natural gamma background (NGB) instruments must be carefully maintained,and they cannot be used in areas where a NGB radiation is not present. This radiation is ordinarily present in the fine-grained sedimentary rocks that bound many coal seams.I. INTRODUCTIONModern underground coal mining practice often includes leaving coal on the roof of the mine after mining is completed. Roof coal is often left on continuous miner sections for ground control purposes to prevent the failure of an immediate roof that consists of weak, friable rock. Roof coal may also be left in mines where concentrations of sulfur or ash are higher near the top of the seam to reduce the concentrations of these impurities in the salable product. Control of coal quality in this manner is especially advantageous in mines with longwall sections, where a large fraction of the production originates from one general area of the seam, making blending for quality control more difficult.Small amounts of roof coal may also be left for purposes of machine guidance. This practice is common in applications where the coal-cutting machine is to be in an automatic control mode. Longwall face operation in this manner has been demonstrated in the United Kingdom 1, 2, and similar systems have been tested in the United States 3, 4. Leaving a measured amount of roof coal in such applications makes it possible to guide the shearing machine, keeping it in the seam.Leaving both roof and floor coal can enhance both the performance and reliability of the cutting machine by reducing its exposure to the high mechanical stress that is experienced when cutting the rock bordering the seam. This can increase pick life and reduce the wear on all parts of the cutting system 2, 5.The need to leave roof coal leads directly to the need for measurement of the thickness of the coal layer left on the roof. Many methods for making this measurement have been investigated. Manual methods, including drilling and borehole inspection, are time consuming and often unreliable.Many instrumental methods have been investigated, including vibration analysis, pick force sensing, ultrasonic and radar detection, and nucleonic methods, but only the nucleonic methods have been used in actual production. The research conducted by CONSOL Inc. on nucleonic methods will be described in this paper. GAMMA-RAY BACKSCATTER SENSINGThe use of gamma-ray backscatter sensing for machine guidance was suggested as early as 1958 6. An active nucleonic device for coal thickness measurement was proposed in Great Britain in 1961 7 and designed in 1973 8, 9. In this device, a source of gamma radiation (usually cesium 137 or americium 247) is enclosed in a housing that is positioned near the surface to be measured. The gamma rays interact with the coal and rock, and are subject to both Compton scattering and attenuation. The backscattered rays are measured by a gamma detector, and coal thickness is calculated from a calibration curve.Several designs of this type of sensor were tested in England, and a commercial model manufactured by Dowty was tested by CONSOL in West Virginia. A prototype was also tested by NASA in the USBM test mine in Bruceton, PA. In every instance, several problems were encountered. Most significant was the variable effect of the air gap between the sensor and the coal surface. Because of this effect, sensors were designed to operate in contact with the surface, which presented severe difficulties in actual mining operations. In addition, with the low-energy gamma radiation employed, coal thicknesses greater than 200-250 mm (8-10 in) could not be measured. It was also found that any variation of materials in the boundary coal or the immediate roof could significantly but unpredictably alter the calibration. Finally, the presence of an active radiation source in a typical underground mining environment raised concerns of safety and source control. Because of these problems, gamma backscatter sensors have been generally abandoned in favor of other devices11. NATURAL GAMMA BACKGROUND SENSINGDuring the testing of various gamma backscatter sensors, it was observed that in many coal seams, the neighboring rock emits a “natural” gamma radiation 12. It has been shown that this gamma background results from the presence of traces of various radioactive isotopes in the rock. The background is generally high in shale, lower in sandstone, almost absent in limestone, and virtually undetectable in coal. Radiation from the roof rock is attenuated by any coal left in place, according to the well-known exponential attenuation equation 13:Where attenuated intensity in counts per second source intensity in counts per second attenuation coefficient in reciprocal centimeters thickness of attenuating material in centimeters.Intensity I is measured by counting gamma ray emissions in a given time, and coal thickness may be determined from the attenuation equation using an empirically derived attenuation coefficient and known background radiation Io.Although the gamma background varies with the composition of the bordering strata, it is often very consistent over wide areas in a given mine or even a given seam. The attenuation coefficient of coal is also reasonably constant because carbon is by far its major constituent. The gamma background is essentially a planar source, and since the attenuation due to air is much less than that due to coal, the distance from the sensor to the roof is not critical. In most instances, coal thicknesses up to 500 mm (20 in) can be measured.Where the strata bordering the seam has a NGB, the passive gamma sensor provides all the advantages of the active gamma device with none of the associated problems. For these reasons, NGB sensors have become the device of choice, particularly in Great Britain 2. They have also been tested successfully in the United States in many locations such as in the Pittsburgh seam in both Pennsylvania and West Virginia and various seams in Kentucky, Illinois, Wyoming, and New Mexico 10, 14. A typical attenuation curve, which was measured in the Pittsburgh seam, is shown in Fig. 1.Fig. 1. Exponential attenuation curve shown for NGB-1000 instrument.Three significant difficulties may arise with the use of a NGB sensor. First, the strata bordering the coal seam may not have a gamma background, or the background may be too low to facilitate meaningful measurements 15. This condition may be minewide (for example, in a mine with a massive sandstone roof) or may occur sporadically (from sand channels, “false roofs,” or similar conditions). In the first instance, the sensor is simply unusable; in the second, it must be used judiciously with frequent checks of both calibration and accuracy.The second difficulty that may be encountered is an intrinsic variation in the gamma background that does not result from secondary disturbances. Occurrence of this condition is entirely site specific and may be determined only by field measurement. It is also accomodated by frequent checks of the instruments calibration and accuracy. An understanding of the depositional geology of the roof rocks can be useful in assessing the probability of this type of variation.The third difficulty that must always be dealt with arises in the process of deriving information from radiological count data. Because radioactive emission is a random process, the accuracy of information derived from count data is directly related to the number of counts recorded 16. This means that a very accurate coal thickness measurement requires either a very large detector or a very long counting time so that in either case, a large number of counts may be recorded 11. INSTRUMENT TESTINGA variety of gamma detectors were evaluated in both laboratory and underground tests. Two hundred test holes were drilled in the roof of a Pittsburgh seam mine (here known as Mine One). The roof-coal thickness at each hole was determined as accurately as possible, first by observing drill cuttings and then by borescope and fiberscope inspection. After a given detector configuration was found to work satisfactorily in the laboratory, it was tested at the underground site by recording multiple instrument readings at each test hole and comparing these with the known coal thickness at that point.Fig. 2. Correlation of thickness readings.An instrument that had a consistent accuracy of 25 mm was tested underground in 1984. This instrument comprised a cluster of seven gamma detectors, where each was a 25-mm- diameter 50-mm-thick sodium iodide scintillator coupled to a photmultiplier tube. Readings were taken by averaging the counts measured by each detector in a 1-min time period. The detector cluster was shielded by 3.8 mm of lead to omit gamma counts originating from the floor and rib (walls). Final tests of the clustered-detector instrument were conducted in Mine One in 1984 to determine its accuracy when operating on a moving machine. At speeds of 2.5 to 3.0 m/min, the accuracy was still 25 mm. A developmental NGB instrument (the NGB-1000) was also tested in 1984. The NGB-1000 coal thickness sensor (a NASA-designed device) is comprised of a sensing head with a single scintillation crystal (51 102 204 mm) a photomultiplier tube, and a control panel. The control panel provides counts-to-thickness conversion, selectable sampling time (5 to 20 s), and digital thickness display. The sensor is large (228 228 610 mm) and, because of the required shielding, weighs almost 90 kg. The instrument is now permanently approved by the Mine Safety Health Administration (MSHA) for use in underground coal mines.Tests at the Mine One test site showed that the accuracy of the NGB-1000 using a 20-s sampling time was comparable with that of the clustered-detector instrument using a 60-s sample time. Fig. 2 shows correlation plots for the readings of the NGB-1000 with the roof coal thickness at each site as estimated by observation with a borescope. Because of this superior performance, it was decided that the NGB-1000 was the preferable instrument for machine installation at another mine (here known as Mine Two).V. OPERATING INSTALLATIONThe NGB-1000 was installed on a continuous miner in Mine Two. This West Virginia Mine is also in the Pittsburgh seam, and its gamma background levels were found to be almost identical to those of the first mine. Conditions at Mine Two require that 100 to 150 mm (4-6 in) of coal be left at the roof boundary of continuous miner development sections. This roof coal is required because the shale in the immediate roof is friable and weak. In the past, operators have used a rock band that is usually visible near the top of the seam as a guide in maintaining the proper cutting horizon. However, this is not always reliable. Earlier observation showed that the actual thickness of the coal left on the roof varied widely; further, it was noted that occasional, accidental excursions into the immediate roof required supplementary roof control measures such as installation of planks or center bolts. Thus, it was concluded that operators needed a better source of guidance for control of the cutting horizon, and a roof coal thickness sensor was scheduled for installation. The NGB-1000 sensor was installed on a Joy 12CM10 continuous miner in June of 1988. The sensing head was mounted on the cutter boom of the miner, and the control panel was mounted in the operators cab. Power for the sensor was initially derived from an intrinsically safe battery power supply. This worked well for a few weeks, but eventually some battery power supplies were discharged too deeply to allow recharging. Consequently, a request was filed with MSHA to allow the sensor to be powered through intrinsic safety barriers by an electronic power supply connected to machine power. The permit was granted, and the sensor was connected to machine power.After the sensor was connected to machine power, the only operating problem experienced was the occasional failure of cables. A supply of the required cables was made and delivered to the mine so that damaged cables could be quickly replaced. Much of the cable damage that was experienced could be eliminated by slight modifications to the miner during a rebuild, allowing cables to be installed in more protected locations.After the sensor had been in operation for approximately two months, a survey was made to determine its effect on mining operations. A hand-held gamma detector was used to measure roof-coal thickness in 35 locations along the track in the mining development section. The measured coal thicknesses from the survey are plotted in Fig. 3. The point at which the NBG-1000 was installed (block 51) shows clearly, as does the period in which the battery power supplies were not working, blocks 52 and 53. A further indication of the improvement brought about by installation of the NGB-1000 appears in Fig. 4, which shows the population variance among groups of three roof-coal thicknesses, as measured in the survey. Clearly, use of the sensor improves the consistency with which the roof horizon is cut.Fig. 3. Measured coal thicknessIn addition to the improvement in as-mined, roof-coal thickness control, another improvement was also observed. In the first 57 blocks of the track entry, it was noted that the miner had cut into the immediate roof 27 times, requiring corrective action. In 13 instances, center bolting was required; in the remaining 14, planks were installed without center bolting. In the next 14 blocks, which comprised the survey area, only one incident of cutting into the roof was observed.In severe roof-cut incidents, where large areas of immediate roof rock are exposed, additional costs may be generated when more extensive remedial roof control measures are required and when large falls occur that require cleanup, which results in lost production.VI. DISCUSSIONThe NGB-1000 was readily adopted by mine operators as a useful aid to good mining practice. The use of a coal thickness sensor can also result in significant cost savings in a situation such as that described above. Economic benefits derived from the use of a coal thickness sensor result from four factors:1) Higher resource recovery resulting from closer control of the amount of roof coal left after mining2) lower auxiliary roof control costs resulting from reduced incidence of cutting into the roof rock3) higher productivity resulting from reduced time spent in auxiliary roof control4) higher productivity resulting from a reduced level of operator uncertainty during cutting of the roof.Fig. 4. Roof coal thickness varianceIn consultation with mine management personnel, estimates of cost savings derived from these factors were made. Using those estimates, the net present value for a coal thickness sensor and its installation on a continuous miner was calculated by standard methods. Those calculations showed clearly that the installation of the sensor was economically advantageous; the pay-out period was less than one year.REFERENCES1 D. Law, “Auto-steerage-An aid to production: Part one,” Mining Eng.vol. 148, no. 328, pp. 326-335, 1988.2 Anon, Coal Face Automation. Burton-on-Trent: National Coal Board,Mining Res. Development Est., 1984, p. 6.3 T. J. Fisher and E. R. Palowitch, “Overview of the Department of Energys program on the development of automated machinery for underground mining,” Proc. Fourth Con$ Coal Mine Electrotechnol. (Morgantown, WV), Aug. 2-4 , 1978, pp. 33-11-33-15.4 R. E. Pease, “AMEs Longwall automation program,” unpublished paper presented at Longwall USA, June 19-22, 1989, Pittsburgh, PA5 A. E. Bennett, “Automatic steering of shearers,” Mining Technol., vol. 55, no. 631, pp. 181-188, 1973.6 V. G. Segallin and A. A. Rudanovsky, “Stabilization of motion in sinking and extracting machinery with the help of radioactive methods,” Afomnaya Energiya p. 88, Jan. 1958.7 B. J. Greenland, “Radioactive isotope monitoring-Principle and use in steering coal-getting machines,” Colliery Guardian, vol. 209, no. 12, pp. 684-688, 1961.8 P. A. Wood, “Remote and automatic control of Longwall mining,” IEA Rep. ICTIS/TR19, IEA Coal Res., London, June 1982, p. 58.9 V. M. Thomas, “Case study: The development of an instrument to measure coal seam thickness,” in Measurement for Instrumentation and Control (M. G. Mylroi and G. Calvert, Eds.). London, Peter Peregrinus, 1984, pp. 251-279.10 P. Broussard and W. B. Schmidt, “The Longwall automation research project of the U.S. Department of Energy,” Mining Technol., vol. 64, no. 726, pp. 138-143, 1981.11 J. S. Wykes, I. Adsley, L. R. Cooper, and G. M. Croke, “Natural gamma radiation: A steering guide in coal seams,” In?. J. Applied Radiation Isotopes, vol. 34, no. 1, pp. 23-26, 1983.12 Anon, “Coal thickness indicator keeps face machine on current horizon,” Mining J., vol. 294, no. 7656, p. 505, 1980.13 W. H. Tait, Radiation Defecrion. London: Buttenvorths, 1980.14 M. J. Pazuchanics and E. R. Palowitch, “Coal interface sensors for automated mining machines,” in Proc. Fourth Con$ Coal Mine Electrotechnol. (Morgantown, WV), Aug. 2-4, 1978, pp. 33-1-33-1 t.15 D. Hunter, “Computerized shearing aids output,” Coal Age, vol. 62, no. 8, pp.64-68 , 1983.16 G. y. Knoll, Radiation Defection and Measurements. New York, 1979.Author introductionStephen L. Bessinger received the B.S. and M.S.degrees in mining engineering from the Colorado School of Mines. He is also a doctoral degree candidate at West Virginia University in the College of Mineral and Energy Resources.He holds Professional Engineering Registration and various mining supervisory certifications. He is a Senior Research Engineer at the Consol Inc.Research and Development Department. He is responsible for advanced technology longwall mining activities within the Department.Michael G. Nelson received the B.S. degree in metallurgical engineering and an M.S. degree in applied physics, both from the University of Utah.He received the Ph.D. degree in mineral engineering from West Virginia University.He has worked extensively in the application of modem techniques of instrumentation and control in the minerals industries, and has been granted seven patents covering his work in control of coal processing plants, instrumentation, and mining machine automation. His current research interests include the physical and economic modeling of precious metals recovery systems and the reclamation of tailings from placer mines and cyanide leach operations in the Far North. He is currently associate professor of mining engineering in the School of Mineral Engineering at the University of Alaska in Fairbanks.He has 18 years of experience in the minerals industries, including work in copper smelting, steelmaking, zirconium production, Coal mining, and gold recovery. He is president of Alaska Mining Ser
展开阅读全文