Search Documents

By E. Bane Kroeger

For many years, researchers around the world have been investigating coal pillar stability. Many have focused on trying to optimize the size of the pillars by examining stable and failed pillars in underground coal mines. For mines that are already in operation or companies that are opening a new mine adjacent to an existing mine where the pillar dimensions and stability can easily be determined, this is the preferred technique. However, when a new mine is planned for an area where there has been relatively little mining, then samples of the coal need to be drilled and tested for strength. Because of the fractured nature of coal, this lab testing is often problematic and may often yield poor results because of the limited size and number of samples that can be obtained from the core. To remedy this problem, research was conducted in the labs at SIUC to determine if improvements could be made when optimizing the size of pillars for underground coal mining in Illinois. Through a greater understanding of the relationship between dimensions and strengths of coal samples tested in the lab, it was hoped the in situ strength of coal pillars could be determined with greater accuracy. This could allow the extraction ratio to be increased and more of the coal resources in Illinois recovered without impacting the safety of the mine workers. Many different agencies around the world have published recommended minimum numbers of samples for strength determination based on the size of the coal samples. For two inch cubes, uniaxial testing of at least seven samples is recommended and this minimum number decreases to four when testing four inch cubes. However, the laboratory testing of coal from two seams thus far has suggested this number should be almost doubled to accurately determine the coal strength. Testing also showed there is a pronounced decrease in the variation in the strength of the samples tested as the dimensions of the sample were increased. From the testing results thus far, it is recommended that a sample size versus strength curve be constructed for every coal seam for Illinois. Creating such a curve will better determine both the minimum number of samples and minimum sample dimensions for that seam or region of a particular scam. A second set of tests were conducted in the laboratory to determine the increase in strength with reduction in height for samples with the same width and length. In Illinois, typical pillar dimensions are 40 to 80 feet (12.2 to 24.4 m) wide and long, with coal thickness varying from about 4 to 9 feet (1.22 to 2.74 m). This provides width/height ratios ranging from 4.5 to 17.5. Coal samples having width/height ratios ranging from 0.5 to 13 have been tested thus far. As with most pillar design formulae, the strength of the Murphysboro coal versus width/height ratio was linear and closely matched shape factors published by other researchers. The laboratory testing thus far has identified three factors that affect the strength of laboratory samples, the number of discontinuities, the angles of critical discontinuities, and the shape of the samples. The factor that probably had the most pronounced effect on the strength of the coal samples was the angle of critical discontinuities. This testing has confirmed that there is still a gap that needs to be bridged between the strength obtained from laboratory testing and the in situ coal strength predicted.

Jan 1, 2004

By D. N. Cleaver

Like other influence function methods, the SWIFT subsidence prediction program, developed within the Mineral Resources Engineering Department at the University of Nottingham, requires calibration to regional data in order to produce accurate predictions of ground movements. Previously, this software had been solely calibrated to give results consistent with the Subsidence Engineer's Handbook (NCB, 1975). This approach was satisfactory for the majority of cases based in the United Kingdom, upon which the calibration was based. However, in certain circumstances within the UK and, almost always, in overseas case studies, the predictions did not correspond to observed patterns of ground movement. Therefore, in order that SWIFT, and other subsidence prediction packages, can be considered more universal, an improved and adaptable method of regional calibration must be incorporated. This paper describes the analysis of a large database of case histories from the UK industry and international publications. Observed maximum subsidence, mining geometry and Geological Index for several hundred cases have been statistically analysed in terms of developing prediction models. The models developed can more accurately predict maximum subsidence than previously used systems but also, are capable of indicating the likely range of prediction error to a certain degree of probability. Finally, the paper illustrates how this statistical approach can be incorporated as a calibration system for the influence function program, SWIFT.

Jan 1, 1995

By Chengguo Zhang, Bruce Hebblewhite, Faham Tahmasebinia, lsmet Canbulat

"Coal burst is a dynamic release of energy within the rock (or coal) mass leading to high velocity expulsion of the broken/ failed material into mine openings. This phenomenon has been recognised as one of the most catastrophic failures associated with the coal mining industry, which can often lead to injuries and fatalities of miners as well as significant production losses. This paper aims to examine the mechanisms contributing to coal burst occurrence, with an emphasis on the energy release concept. In this study, a numerical modelling study has been conducted to evaluate the roles and contributions of difference energy components. The energy analysis presented in this paper can help to improve the understanding of energy release mechanisms especially under Australian conditions.INTRODUCTIONCoal burst is one of the most hazardous problems encountered in underground coal mines, which occurs at different locations in a variety of mining systems and operations. This phenomenon always involves a violent and dynamic energy release with large rock mass/coal deformation and ejection that can cause severe damage to openings, equipments and may result in fatalities and injuries.The first coal burst occurrence was reported in Britain in 1738 (Dou 2001). Since then, international experiences are available in Canada, South Africa, USA, former Soviet Union, China, India, France, Germany, Poland and Czechoslovakia. Rice (1935) classified coal bursts as two general types, namely excessive pressure bumps and shock bumps and he further mentioned that pressure bumps are caused when pillar stress exceeds bearing strength. Shock bumps are induced by the breaking of thick, massive strata at a considerable distance above the coal-bed, causing the immediate roof to transmit a shock wave to the coal."

Jan 1, 2016

By Michael Alber

Resin grouted rock bolts are the major support measure in coal mining. They are rather easy and foolproof to install, take immediately load and are the least time consuming support measure for controlling gateroad deformation. The use of only resin grouted bolts allows fast gateroad development which is essential for economical resource exploitation. The design of reinforcement / support systems for coal mining typically bases on empirical methods, which work fine when encountering known or usual ground conditions. However, coal mining approaches depths where little experience exists and numerical modeling is often the only method to estimate ground behavior. This paper discusses the geotechnical properties relevant to numerical model and focuses particularly on input values for modeling resin grouted rock bolts for safe gate design.

Jan 1, 2006

By Michael Murphy, Khaled M. Mohamed, Berk Tulu

"Failures of coal ribs are major hazards in underground coal mines. For the period of 2010 to 2014, rib falls were the leading cause of groundfall fatalities and accounted for 35% of all groundfall-related fatalities. Currently, the National Institute for Occupational Safety and Health (NIOSH) is conducting a research project to develop an engineering-based design methodology for rib control in underground coal mines to reduce rib fall accidents and fatalities. Numerical models can be used as part of the design methodology provided the models are appropriately calibrated and verified.This paper presents a coal rib model to simulate the mechanism of coal deformation and fracturing that can lead to stress-driven rib falls. The proposed coal rib model was calibrated against published data from an instrumented rib site at the West Cliff Mine in Australia. Fracturing of the coal rib was defined by critical VALUES (of plastic shear strain of the coal material and rib displacement. A critical plastic shear strain of 2.8% and a sudden increase of rib displacement of 30 to 45 mm were sufficient to develop a fracture plane in the rib model. Accordingly, a discontinuous rib deformation was achieved and it compared favorably against field monitoring data.INTRODUCTIONCoal ribs are the vertical walls of coal formed when entries are developed in a coal bed. Failures of coal ribs are a major hazard in underground coal mines. Furthermore, failing ribs can contribute indirectly to roof and floor instability by increasing opening widths across intersections and entryways. Over the last 18 years, there has been a continual occurrence of injuries and fatalities in underground coal mines due to rib falls. For the period of 2010 to 2014, rib falls (excluding longwall face failure and bursts) were the leading cause of ground fall fatalities and accounted for 35% of all groundfall-related fatalities (MSHA, 2015)."

Jan 1, 2016

By Monika Patrycja Polak - Micewicz

This paper presents an approach for estimating the fatigue life for the base of powered roof supports (shield) manufactured by the FAZOS roof support company. FAZOS?s powered roof supports are designed to operate on an inclined coal seam in full caving with a medium, strong roof strength. This shield support is utilized in conjunction with a shearer and face conveyor for a complete longwall mining system. To design shield support, it is essential for contractor to know the physical (hardness, toughness, etc.) and mechanical properties of material (tensile strength, Young?s modulus, etc.) used for the shield support as well as the geotechnical properties of the mine site. The fatigue strength of the shield support material depends on the mean stress, alternating stress, cycle frequency, transport dimensions, mine environment and other factors. In order to estimate the fatigue life for a powered roof support system, it is assumed in this paper that the support must keep the roof stable in the mine under variable load cycles. The approach selected in this study contains Goodman?s model (Goodman, 1899) for all blocks in the design housing which is subjected to sinusoidal load. On this basis, the degree of fatigue damage was calculated and compared to the results of experimental tests. It will be shown that there is a significant detrimental effect on the fatigue life of the base of the powered roof supports due to geometrical notches. The Ramberg-Osgood model and damage accumulation rule (Palmgren-Miner) (Miner M., 1945) were used and numerical analyses were conducted, utilizing ANSYS finite element software. These analyses play a very important role in longwall casing design. The proposed method allows assessment of the yield state and forecast of the structure cracking, taking plastic strains occurrence into account. The presented method is universal and can be applied for fatigue life estimation of other subassemblies of powered roof supports. The final portion of the paper presents results of variable amplitude loading simulations on the S355N and S690Q steel used to make powered roof supports. The results of the physical tests were compared with the results from finite element methods. The proposed approach to calculate fatigue life gives the best correlation between the experimental results and calculated values of fatigue life for the powered roof support base. Finally, the design process uses a simple approach to predict fatigue life in a short time, reducing design costs and providing longer service life for roof supports.

Jan 1, 2012

By David Bigby

Room and pillar mining with full pillar extraction is one of the most hazardous systems of underground coal mining, a significant hazard being the potential for sudden goaf override. The Autowarning Telltale (AWTT) is a new development for pillar extraction operations, based on the well proven and simple Rockbolting Telltale safety monitoring device. It is now being applied in room and pillar coal mines in India, with significant potential for application worldwide. The Autowarning Telltale is an intrinsically safe low-cost, self-contained unit installed into the roof either during pillar development or prior to pillar extraction. It provides a highly visible warning via flashing Light Emitting Diodes (LEDs) when a preset level of roof movement has been exceeded. The paper provides a full description of the instrument and two case histories of its initial application at fully-mechanized Indian sites using the split and fender technique.

Jan 1, 2011

By Raymond M. Stateham

The Bureau of Mines has developed a portable, lightweight device for determining the integrity of resin-grouted roof bolts. The instrument functions by sending a known pulse of ulstrasonic energy into the bolt and comparing the amount of energy reflected hack with the original measured pulse. Where the bolt is properly bonded or coupled with the rock, the energy passes through the bolt-resin-rock interface and dissipates in the rock. Where the bolt is not properly coupled, the energy is reflected back along the bolt to the instrument. The ratio of reflected energy to the pulse approximates the fraction of the bolt's surface that is improperly bonded. The instrument is easy to use. The operator holds the sensor head of the device firmly over the head of the bolt to be tested, causing the pulse of ultrasonic energy to be imparted to the bolt. The device then listens for reflected energy. This cycle is repeated about 200 times per second until an evaluation of the bolt installation is made and indicated by colored lights in the display window. Bolts having 50% or less of their surface bonded are considered had and cause a red light to he illuminated. Bolts with 50% to 75% of their surface bonded cause a yellow light to be lit and are considered marginal. Bolts having more than 75% of their surface bonded are considered good and cause a green light to be lit. The indicator lamp remains on for about 4 seconds and then goes off, indicating that the instrument is ready for another test.

Jan 1, 1982

By Nikolay Boykov

Subsurface imaging technologies are part of the electromagnetic (EM) geophysical methods. EM technologies such as radio imaging method (RIM), ground penetrating radar (GPR), and drill string radar (DSR) have great application potentials in the mining industry. However, the success for applying the EM methods to subsurface observations is largely dependant on the correct evaluation of the electrical properties (i.e., resistivity and dielectric constant) of the rocks in the domain of interest. The depth of penetration of the EM signal in coal, as well as the electrical contrast between coal and subsurface boundary rocks or anomalies, are critical parameters for any successful radar system. Based on the measured electrical properties of the coal seam, the proper operating frequency of the radar system must be chosen in order to successfully detect features such as the coal-roof rock boundary, dykes, and voids ahead of mining. The electrical properties of the rocks in a coal seam are frequency dependent. Even though there has been a great advancement in the hardware and data processing software in the past twenty years, there have not been many examinations of the high-frequency properties of coal measure rocks. In this paper, the laboratory setups, the testing procedure, and case studies will be presented for a wide spectrum of frequencies for coal measure rock samples collected from various US coal fields. Tests have been conducted to study the different conditions, including moisture content, that affect the EM properties.

Jan 1, 2006

By A. T. Iannacchione

Underground stone mining represents an emerging sector for the U.S. mining industry. As this expansion takes mines under deeper cover and as more efficient mining methods are utilized, adequate stone pillar design methods will become more important. It is the purpose of this paper to examine current design practices and to discuss issues for safe mine layouts so that a rational first approach towards balancing the demands for increased production can be weighed against increased risk to worker safety from rib instabilities and pillar failures. Seventy-two stone mine pillar designs were examined. Pillars with width-to- height ratios below 1.5 and subjected to excessive stress levels appear more likely to fail. When width-to- height ratios fall below 1 .0, defects in the pillars, such as through-going discontinuities, can have a significant influence on stability. Discontinuity persistence, dip, material properties, and orientation are important factors controlling pillar strength. The influence of discontinuity dip, a characteristic easily identified in the field, was examined so that its presence could be accounted for in developing generalized guidelines for pillar design.

Jan 1, 1999

By Scott A. Shapkoff

High strength cable strand supports are predominately used as a supplemental means of ground control in longwall and development operations in the U.S. The advantages of cable strand supports have been documented and include flexibility, strength, and ease of installation. Since the majority of cable supports installed in the U.S. today are partially grouted, and this same segment of supports has no inherent installed tension, the supports are considered currently arbitrarily strictly supplemental and therefore, cannot currently be deemed primary support (in most mining districts). This paper details and discusses two distinctly different configurations and the technical advantages of active partially grouted cable supports for certain primary support applications. The evolution of the designs for both systems is highlighted. Field data illustrates the amount of active load in relationship to installed torque and the affects that component variables have on this load measurement. Roof control patterns using active cable supports are examined as alternatives for combination patterns using primary and supplemental systems. The importance of maintaining consistent installation procedures plays an important role in the level of performance of the system. Economics comparing the installation of active cables with other primary/supplemental patterns is also included.

Jan 1, 2006

By Damon P. Vandermaat

Coupon testing allows for the rapid understanding of a material?s likely behavior in service conditions, allowing engineers and operators to make informed decisions about the likely service-life of parts when subjected to field conditions. Bend testing has been identified as the best approach for exploring rock bolt susceptibility to stress corrosion cracking (SCC) in underground mining. The bend test specimens used were 2?3 mm thick slivers of steel, cut from the outer edge of a rock bolt. For the specimens, 1350 grade steel was chosen due to its wide use in the industry. The jigs themselves are small assemblies made of stainless steel that are capable of imparting a bending load on the specimens. The jigs were installed at a coal mine located in the Western Coals West of Sydney, Australia, in such a way as to expose the test specimen to water dripping from the mine roof. It is expected that this approach will yield results pointing to the likely SCC mechanisms affecting the service life of rock bolts, as well as demonstrate the jigs? usefulness as an early warning system for SCC. Furthermore, the coupon tests will be correlated against the results of large-scale laboratory tests.

Jan 1, 2012

By Edward A. Barrett

A significant number of injuries resulting from fa118 of roof and rib may be avoided by modifying the reactive behavior of miners as they encounter potentially hazardous ground conditions. In 1984-85, the Bureau investigated barriers which prevent miners from recognizing and correcting groundfall hazards. One basic problem from which these barriers originate is the inability of miners to recognize dangerous roof and rib conditions. This study is a report of experimental research on the use of stereo scopic (3-D slides for presenting visual cues associated with hazardous roof and rib conditions. Findings reveal that, in comparison to conventional slides, 3-D provides for significantly greater hazard recognition among both experienced and inexperienced miners. Details such as the presence of t cracks and gaps in the surrounding rock are more accurately and realistically perceived. Acceptance by the mine training community has been affirmative and is additionally encouraged by the fact that 3-D slides stimulate trainees' overall interest.

Jan 1, 1986

By Sudhir Kumar Kashyap

In underground activities ground control is a challenging problem. It affects safety, production, and efficiency. As per statistics of accident data, ?fall of roof/sides? is one of the major causes of mine accidents. In many cases, our experiences and understanding of soil and rock behavior still fall short of being able to predict how the ground will behave. Presently, empirical approaches to design are widely used in estimating mine support parameters. Under these circumstances, expert judgment plays an important role, and such accidents can be obviated with accurate measurement and optimization of data and analysis using the neuro-fuzzy technique of artificial intelligence. Recently, a neuro-fuzzy hybrid approach has become one of the major areas of interest in engineering fields because it gets the benefits of neural networks as well as those of fuzzy logic systems, and it removes the individual disadvantages by combining them on the common features. Neural network and fuzzy logic technologies have common features, such as a distributed representation of knowledge, the ability to handle data with uncertainty and imprecision, etc., like in ground control. Fuzzy logic technique has tolerance for imprecision of data, while neural networks have tolerance for noisy data. In this paper we have focused an intelligent technique, i.e., neuro-fuzzy technique, to approximate the setting load given to the standing support (props) erected for the purpose of supporting freshly exposed roof during underground mining. We have used twelve input variables of rock parameters, and after having trained using a neural network with a sigmoidal function as an activation function, simulation was done to find the output parameter, i.e., the pre-load (setting load)to be applied on props. Outputs of the neural network were again fed to the fuzzy system together with the twelve parameters, incorporating five triangular membership functions for each parameter. Final optimum output was approximated using the MATLAB program, which was found to be satisfactory. Neuro-fuzzy technique has better performance over individual neural networks or fuzzy logic technique.

Jan 1, 2012

By Eric G. Zahl

Researchers at the Spokane Research Laboratory of the National institute for Occupational Safety and health are evaluating stress and displacement measurements in a longwall coal mine. The objective is to determine the best indicators of coal bump hazards during operations so that mine foremen and technical stall can make more informed intervention decisions to protect miners from coal bumps. Instruments were clustered in two panels, a yield pillar, and the immediate roof and floor of a two-entry longwall coal mine. Data were gathered continuously from a variety of instruments as mining proceeded toward and past the instrument cluster. Researchers and the mine's technical staff evaluated these data with respect to coal bumps and bump mechanisms. Selected graphs were then presented to mine foremen on a periodic basis to test their usefulness in identifying hazards. Of particular interest was the use of horizontal stress data from the immediate roof. To measure changes in horizontal stress, a new technique was devel¬oped for installing a grouted biaxial stressmeter in the roof. An evaluation of the development of this stress monitoring system in producing data beneficial to mine foremen is presented.

Jan 1, 2000

By Hani Mitri

Understanding the interaction between rock bolts and underground rock movement is critical for safe and cost effective underground excavation design. Although early research on this subject involved a balance of theoretical analysis and field measurement, recent work has been heavily focused on analytical and numerical studies. This paper describes technologies that have the potential to redress the balance through instrumentation of rock bolts deployed under routine operating conditions in underground mines. Two new products are presented in this paper. Firstly the U-cell coupler is a load cell device that can be routinely coupled to the threaded end of a bolt prior to installation. It measures the ?series? load at the head of the bolt. Secondly, the d-REBAR involves an array of small-diameter, long-base length displacement sensors recessed into grooves along the entire length of the bolt. Both new instruments are interfaced with on-board digital signal conditioning and telemetry. Methodologies for the deployment of the new instrumentation and guidelines are presented for the interpretation of results based on data obtained from field trials. Successful field installations demonstrate the viability and practicality of these new technologies and, moreover, provide important insight in to rock bolt/rock mass interaction.

Jan 1, 2012

By M. Karmis

Underground mining will disturb the natural equilibrium of the rock mass, causing significant stress redistributions in the vicinity of the excavation with corresponding horizontal and vertical displacements. Subsidence of the ground occurs when these displacements propagate from the mine opening, through the overlying strata, to the surface. Such ground movements may take either of two general forms: a sudden and sometimes violent collapse, or a more gradual and progressive ground movement. The former, characterized by localized failure of roof or pillar in old, shallow room and pillar workings, is rather unpredictable and difficult to analyze. As a result, most control measures are directed toward the latter type of movement, which is associated with the fracture and flow of the strata overlying longwall and high-extraction room and pillar systems, resulting in horizontal and vertical strains in the affected area. Such ground movements can often cause considerable surface disturbances ranging from simple land settlement to severe structural damage. Due to the increasing incidence of this phenomenon in more populated areas and the rising awareness of the problem, it has become necessary to establish a balance between the development of mineral resources and the protection of the environment and society. This paper presents a summary of the major findings derived from a substantial research effort on mining subsidence carried out, during the past few years, in the southern Appalachian coalfield. The research program encompassed the following major tasks: identification of the most significant factor influencing ground movements above mined openings, collection of subsidence case studies for the southern Appalachian coalfield, development of subsidence and strain prediction techniques for that region and initiation of a subsidence monitoring program in southwestern Virginia.

Jan 1, 1984

By Zhang Liandong

Roof control in thick-seam mining has long been a difficult problem in mine design and coal mining. In the early 1980's, through ground control instrumentation and investigation the rules of roof strata movement were determined, based on which the sliding roof bar powered support was properly designed and successfully applied to roof control to thick seam mining with caving. Underground trials in Huating Coal Mine produced satisfying results. The technique is being employed in many Chinese coal mines and won the national price for technical advance.

Jan 1, 1990

By Martin Bolesta

The use of strata injection as a means of stabilizing brittle and heavily stressed ground as well as faulted zones is now state-of-the art technology. Users are obliged to select the right combination from a wide range of injection products and equipment in order to ensure that the operation will be a success. Laboratory tests and a fundamental understanding of the technical aspects of injection products not only help in drawing up the correct injection procedure but have in fact become indispensable for planning the most reliable and cost-effective solution.

Jan 1, 2001

By Yuanping Cheng, Jingyu Jiang, Qiang Zhang

"With the increase of mining depths, deep-seated coal seams face threats from high ground stresses, high gas pressures, high gas content, and low seam permeability, which significantly increase the likelihood of coal and gas outbursts. Challenges continue to grow in coalbed methane (CBM) extraction fields, especially for short-distance gassy multiple-seams. Research has proved that the most effective way to increase coal seam permeability is to decrease the ground stress. Protective seam mining technology is a surrounding rock control technology, which can effectively reduce the ground stress and increase the permeability of the protected coal seam. In this paper, we try to use upper protective seam mining technology to simultaneously extract coal and gas. Findings are based on a FLAC3D simulation of stress variation during the mining of the No. 12 coal seam in the Daxing coal mine, Tiefa Coalfield, China. Results indicate that, after mining the upper protective seam, the pressure-relief effects of the rock mass generated three belts and five zones below the protective seam, providing the space and time for gas pre-extraction engineering. Within the protected area, the numerical simulation indicates that the stress of the No. 13/14 coal seam reduces 95.84% and 88.25% respectively, while the expansion deformation of the protective seams reach 6.87 ‰ and 6.70 ‰ respectively, which demonstrate that the protective mining technology can provide a significant pressure-relief effect to the outburst-prone coal seam. Combining stress relief in this manner with pressure-relief from engineered gas drainage (performed by crossing boreholes from the floor rock roadway) applied before the mining of the protective seam, it would not only be beneficial to CBM drainage but also can reduce the gas outburst risk of the protected seams.INTRODUCTION Internationally, since France used protective seam mining technology to control coal and gas outburst in 1939, it has been promoted and applied in many countries (?????, 1966). Since 1958, China has taken the mining test and engineering practice of protective seam mining technology in the Beipiao, Tianfu, Nantong, Zhongliangshan, Songzao, Huainan, Huaibei, and other coalfields (Yu, 1992). Long-term theoretical research and mining practice of outbursts and dangerous coal seams show that protective seam mining technology is the most effective, safest, and most economical outburst prevention measure (Cheng, Fu, and Yu, 2009; Hu et al., 2008; Cheng and Yu, 2007). Under the condition of high gas content coal seams, the safety and efficiency of simultaneous extraction of coal and gas can be realized through a reasonable mining sequence and an effective gas extraction method (Cheng and Yu, 2003). The in situ stress distribution of the protective seam (No. 1 coal seam) and protected seam (No. 2 coal seam) during protective seam mining in the Haishiwan coalfield formation (Figure 1) (Li et al., 2014). A comparison of the stress and displacement (five zones and three belts) of the rock mass below the protective seam, and in a multiple-seam context in the Dalong Mine, is shown in Figure 2 (Jiang et al., 2015). After mining protection coal in a close-distance seam group, the small spacing between the coal seams causes the protective coal and the protected coal between part of the rock to crack through each other until the protective layer of gob is through the area. This provides a desorption of gas to the protective layer of the mining space channel, making the coal seam permeability increase to several hundred times and improving the gas discharge capacity (Zhai, 2008)."

Jan 1, 2017