J. Cent. South Univ. (2017) 24: 1252-1262

DOI: 10.1007/s11771-017-3529-2

Particle size distribution of coal and gangue after impact-crush separation

YANG Dao-long(杨道龙)¹, LI Jian-ping(李建平)¹, DU Chang-long(杜长龙)¹,

ZHENG Ke-hong(郑克洪)², LIU Song-yong(刘送永)¹

1. School of Mechatronic Engineering, China University of Mining & Technology, Xuzhou 221116, China;

2. College of Mines and Earth Sciences, University of Utah, Salt Lake City, UT84112-0114, USA

Central South University Press and Springer-Verlag Berlin Heidelberg 2017

Abstract: Based on the separation and backfilling system of coal and gangue, the mineral material impact experiments were conducted utilizing the hardness difference between coal and gangue according to the uniaxial compression experiments. The broken coal and gangue particles were collected and screened by different size meshes. The particle size distributions of coal and gangue under different impact velocities were researched according to the Rosin-Rammler distribution. The relationships between separation indicators and impact velocities were discussed. It is found from experiments that there is a fully broken boundary of coal material. The experimental results indicate that the Rosin-Rammler distribution could accurately describe the particle size distribution of broken coal and gangue under different impact velocities, and there is a minimum overlap region when the impact velocity is 12.10 m/s which leads to the minimum mixed degree of coal and gangue, and consequently the benefit of coal and gangue separation.

Key words: Rosin-Rammler distribution; impact crush; separation indicator; coal and gangue; separation and backfilling system

1 Introduction

Separating coal and gangue underground directly not only improves the production efficiency [1], but also protects the environment around the coal mine. Especially, the gangue separated from raw coal could be used to backfill goaf [2,3]. It solves the problem of gangue accumulation, and provides raw materials underground backfilling, meanwhile it reduces the cost of filling and conforms to the green mining concept [4-6].

Coal and gangue can be separated based on the different hardnesses, densities, surface grayscales and so on. The separation methods mainly include mechanical separation method, surface image method, ray transmission method and fluidized bed flotation method. Many scholars put forward measures to separate coal and gangue. XU and ZHANG [7] designed a novel apparatus to realize the on-line non-contact measurement of ash in the raw coal during the process of conveying based on the attenuation of dual energy γ-rays transmission. In 2003, XU et al [8] described a fuzzy neural network (FNN) used for the identification of coal and gangue, and put forward the niche genetic algorithm (NGA) based on a proposed practical niche operator for the offline training of FNN. DING and LASKOWSKI [9] investigated the reverse flotation of a subbituminous coal/gangue mixture and the effect of various factors on the reverse flotation. LUO et al [10] constructed a vibrated fluidized bed for separating dry fine coal particles from unwanted gangue particles, and performed an experimental investigation of vibrational energy transmission and interaction between vibration and gas flow. MA and LI [11] established underground separation system of coal and gangue, mainly including screening process, sorting process and coal returning process, based on the analysis of the underground gangue source. HE et al [12] investigated the separation performance of a raw coal sample from south africa based on the dense gas-solid fluidized bed beneficiation technique, and his study provided an effective flow sheet for achieving the high-efficiency separation performance on raw coal from arid areas. HE et al [13] proposed a new method based on support vector machine (SVM) and texture analysis to increase the efficiency in the separation of gangue from coal, and the experimental results proved that the accuracy of the differentiation between coal and gangue through SVM was high. XU and WANG [14] prompted a separation method using image segment based on the analysis of grey value of digital images. MIAO and ZHANG [15] put forward the realistic requirement and significance of coal-gangue separation underground on the basis of study progress of solid backfilling coal mining technology, and put forward the basic principles of system design of coal-gangue separation and backfilling coal mining. ZHENG et al [16, 17] tested and analyzed the size distribution of coal dust formed by different impacting forms using the Rosin-Rammler distribution function, and found out that each of the regression curves of the particle size distribution was a straight line and the linear regression was very good in the double-logarithmic coordinate system.

The above researches provide significant references for separation and filling system of coal and gangue underground. We have devoted much effort to investigate mechanical separation method [18-24] and surface image method [25] of separating coal and gangue underground, and put forward a suggestion that the mechanical separation method is feasible for underground separation by considering the narrow and lightless working conditions of coal mining, while surface image method requires high surface clearness, ray transmission method is not security for coal mining, and fluidized bed flotation method adopts larger or heavier equipments which can be used underground hardly. Mechanical separation method is based on the different hardnesses between coal and gangue, and separation device can be changed based on the working space needed. However, previous studies of our team only considered that whether the coal and gangue were broken or broken under a certain particle size [26], and less researched the particle size distribution of coal and gangue after mechanical separation.

Therefore, based on a novel separation and backfilling system of coal and gangue, the mineral material experiments were conducted to research the particle size distribution of coal and gangue. The broken coal and gangue particles were collected and screened by different size meshes. The relationships and influences between the separation indicator and velocity were put forward according to the Rosin-Rammler distribution [27], which is a reference to mechanical separation method.

2 Principle

The separation and backfilling system of coal and gangue in the fully mechanized coal mining face is shown in Fig. 1. It includes three parts, separation system, conveyor system and backfilling system.

The separation system breaks and screens raw coal. The conveyor system conveys the screened-out coal to underground bunker and the screened-out gangue to backfilling system. The gangues from separation system, tunnel excavation and ground accumulation are threw into goaf utilizing belt conveyor of throwing gangue by backfilling system. The working process of separation and backfilling system is shown in Fig. 2.

The raw coals are mined by shearer and conveyed to the separation system by scraper conveyor. In separation system, the raw coals are accelerated by high-speed belt conveyor and impact on the crushing plate. According to the different hardnesses of coal and gangue, there is a certain velocity that leads to the coal broken while gangue unbroken or little broken after impacting with crushing plate. The broken coals drop down from sifter and are conveyed into underground bunker by belt conveyor. The unbroken gangues remain on the screen and are conveyed into belt conveyor of throwing gangue in the backfilling system by the chute and belt conveyor in the conveyor system, and then filled into the goaf with others coming from the tunnel excavation and the ground accumulation.

Fig. 1 Separation and backfilling system of coal and gangue:

Fig. 2 Working process of separation and backfilling system

Separation method of separation and backfilling system is impact-crush separation. Separation principle of separation system is shown in Fig. 3. Coals and gangues mixed in the raw coal are accelerated by high-speed belt conveyor, and impacted on the crushing plate with high velocity and then fall into the sifter. Because of different hardnesses between coal and gangue, fragmentation degree of coal and gangue is also different under the same impact velocity. Therefore, there may be a certain impact velocity at which the low hardness coals are broken into pieces and the high hardness gangues stay the same.

Fig. 3 Separation principles of coal and gangue:

Separation and backfilling system could separate coal and gangue underground directly which decreases transmission cost, improves economic and environmental benefit of coal mining, and conforms to the concept of green mining. It was discovered in practical application that coal was not easy to be broken when velocity of high-speed belt conveyor was low and gangue was broken when velocity was high. Therefore, it is important to find an optimal velocity benefit for coal and gangue separation.

3 Theoretical model

Evaluation indicators for coal and gangue separation were put forward and the Rosin-Rammler distribution was used to evaluate the impact-crush separation effect.

Rosin-Rammler distribution was utilized to evaluate particle size distribution of coal and gangue after impact-crush separation, cumulative distribution function of Rosin-Rammler distribution can be calculated as [28, 29]

                       (1)

where F(d) is cumulative distribution function expressing particle diameter less than d, d is characteristic particle diameter expressing the minimum mesh size which allows particle to pass through, d' is the shape parameter (also called as median diameter) expressing the median particle diameter when the particle size at a mass fraction of 0.5 oversize [30], n is the scale parameter describing the width of distribution, with a wider distribution corresponding to a smaller n, and vice versa.

Discrete degree of particle size distribution after impact-crush increases with the scale parameter n, which means that the major discrete degree of coal and gangue after impact-crush leads to the major mixed gangue ratio and fine coal ratio. It reduces the effect of impact-crush separation.

After logarithm transformation, the logarithmic cumulative distribution function is obtained from Eq. (1) and shown as Eq. (2).

              (2)

According to Eq. (2), there is a linear relationship between cumulative distribution function, F(d) and characteristic particle diameter, d in the logarithmic coordinates. Defining a as the material crushing parameter and a = -n lnd' [31], expressing the small particles amount of broken coal and gangue. Number of small particles increases with material crushing parameter, a. It is indicated that lump coal ratio deceases and the fine coal ratio increases with material crushing parameter, a.

After derivation transformation, the density distribution function, f(d) is obtained from Eq. (1) and shown in Eq. (3).

                 (3)

There is a peak value in the density distribution function f(d) which is called as the mode diameter, d₀ [32] and shown in Eq. (4).

                            (4)

Mode diameter, d₀, expresses the largest proportion of particle size among the hole particle size range, which shows the particle size distribution effectively cooperating with scale parameter, n and material crushing parameter, a. Therefore, scale parameter, n, material crushing parameter, a and mode diameter, d₀ are the separation indicators that evaluate clump coal ratio, mixed gangue ratio and fine coal ratio for coal and gangue separation effect, respectively.

4 Experimental methods

4.1 Experimental material

Coal and gangue from Shi-tun mine in China were selected as experiment material. Three coal specimens were drilled around underground mining working face and three gangue specimens were drilled around transportation roadway as uniaxial compression experimental material. Sizes of all specimens were 50 mm in diameter and 105-110 mm in height. In order to prepare for uniaxial compression experiment, end faces of six specimens were grinded smoothly in laboratory, and became standard uniaxial compressive specimens of which the height was 100 mm and diameter was 50 mm.

Raw coal was screened on the ground, and coals with particle size from 100 mm to 120 mm were picked out as the impact experimental material. Gangues from transportation roadway tunneling process with the same particle size as coal were picked out as other impact experimental material. According to the mineral material impact experiment required, materials were divided into ten groups (five of coal and five of gangue) and weight of each group was 50 kg. Above experimental materials demanded no weathering phenomenon and out of the mine in recent three days.

4.2 Uniaxial compression experiment

Uniaxial compression experiment is mainly to test compressive strength, Poisson ratio and elasticity modulus of coal and gangue specimens. Uniaxial compression experiments were conducted on SANS test system as shown in Fig. 4. The minimum sampling time of this system was 50 μs, the maximum axial force of vertical hydraulic cylinder was 1700 kN and confining pressure was under 45 MPa.

Before experiments, the specimen was placed smooth and steady. Necessary parameters for uniaxial compression experiments were set by bundled software of SANS Power Test_DOOC. Advance speed of vertical hydraulic cylinder was 1 mm/min, and contact force, displacement and peak force were reset. Then, vertical hydraulic cylinder was adjusted downward finely and change of contact force was checked constantly. Until contact force had a tiny change, fine adjustment of vertical hydraulic cylinder was stopped, and contact force, displacement and peak force were reset again.

Fig. 4 SANS test system (a) and specimen placement (b)

Uniaxial compression experiments were conducted according to the order of specimen number. When experiments began, vertical hydraulic cylinder was started and data were collected until the specimen was fractured. Experimental reports were obtained after stopping sampling.

After experiments, the relationship of load force and displacement was calculated, and compressive strength was the average maximum stress. Compressive strength, Poisson ratio and elasticity modulus of coal and gangue specimens are shown in Table 1.

Table 1 Characteristics of coal and gangue

The Protodikonov’s hardness coefficient could be obtained based on compressive strength shown in Table 1. The relationship between the Protodikonov’s hardness coefficient and compressive strength is shown in Eq. (5). The Protodikonov’s hardness coefficients of coal and gangue are 1.58 and 6.57, respectively, and the hardness of gangue is 4.16 times as high as that of coal, which indicates that there is a large hardness difference between coal and gangue. In consequence, impact-crush separation method could be used to realize separating coal and gangue.

f = R/10                                     (5)

where R is uniaxial compressive strength.

4.3 Mineral material impact experiment

Experimental method and process are shown in Fig. 5. In order to obtain the relationships between broken particle size distributions and impact velocities, five different impact velocities were applied to ten groups of coal and gangue impact experiments.

Before experiments, angle α called crushing plate angle between crushing plate and horizontal plane was adjusted to realize the normal impact between particle and crushing plate, which is shown in Fig. 6, where v is initial velocity; v_t is instantaneous velocity of particle; v_x is velocity component in x-direction and v_x=v; v_y is velocity component in y-direction. Particles were accelerated with high-speed belt, and then particles left high-speed roller and did horizontal projectile motion with initial velocity. At last, particles impacted on the crushing plate. Horizontal distance between crushing plate centre and high-speed roller centre is 650 mm. According to horizontal projectile motion law and geometric relationships shown in Fig. 6, crushing plate angle α is obtained from Eq. (6) and values under different velocities are shown in Table 2.

                           (6)

where r is particle radius, r=55 mm; g is gravitational acceleration, g=9.81 m/s²; l is horizontal distance between crushing plate centre and high-speed roller centre, l=650 mm; v is initial velocity(impact velocity) of particles, v=6, 8, 10, 12 and 14 m/s; α is acute angle between crushing plate and horizontal plane.

When mineral material impact experiment began, put a group of mineral material into funnel and then started impact-crushing separation device and opened funnel. Impact-crush separation device is shown in Fig. 7. Speed of high-speed belt accelerator was adjusted by the electric frequency converter to obtain different impact velocities. Particles were conveyed by conveying belt and accelerated by high-speed belt, and then particles impacted on the crushing plate and fell into the aggregating box. On the same impact velocity, low hardness coal was broken into small pieces and high hardness gangue kept their original lumpiness, and then screening coal and gangue to achieve separation.

Fig. 5 Experimental method and process:

Fig. 6 Positional relation between the particle and crushing plate

Table 2 Crushing plate angles under different impact velocities

After the experiments, closed power supply of impact-crush separation device, collected the broken materials, screened the broken particles using different mesh size, then weighed and recorded experimental results.

5 Result and discussion

5.1 Rosin-Rammler distribution model

Experimental results are shown in Table 3, where d is characteristic particle diameter, F(d) is cumulative distribution function and v is impact velocity. The relationships between cumulative distribution function, F(d) and impact velocity, v of coal and gangue are shown in Figs. 8 and 9, respectively.

In Figs. 8 and 9, cumulative distribution functions of coal and gangue increase with impact velocity. But it is difficult to find a direct relationship between particle size distribution and impact velocity. Therefore, the logarithmic cumulative distribution functions of coal and gangue were obtained based on the experimental results and Eq. (2), and shown in Figs. 10 and 11, respectively. The correlation coefficients of fitting formulas in Figs. 10 and 11 are greater than 0.98 which indicates a well fitting degree.

The function lines of logarithmic coal cumulative distribution present divergent shape and low regularity, and distance between those lines increase with the logarithmic characteristic particle diameter in Fig. 10. However, the function lines of logarithmic gangue cumulative distribution present convergence shape and strong regularity, and distance between those lines decrease with the logarithmic characteristic particle diameter in Fig. 11. It indicates that the properties of the five groups of coal samples have some otherness, coal could come from the same coal seam but on the different working faces. The gangue used in experiments is all from transportation roadway tunneling process and there is little otherness of gangue material. Although there is some otherness of coal samples, it has a little influence on the experiment results and it does not make a fatal mistake due to the large hardness difference between coal and gangue.

Fig. 7 Experimental device for impact-crush separation of coal and gangue:

Table 3 Experiment results of coal (f=1.14) and gangue (f=3.27)

Fig. 8 Relationships between cumulative distribution function, F(d) and impact velocity, v, of coal

Fig. 9 Relationships between cumulative distribution function, F(d) and impact velocity, v, of gangue

Fig. 10 Logarithmic cumulative distribution functions of coal

Fig. 11 Logarithmic cumulative distribution functions of gangue

5.2 Analysis of separation indicators

The fitting formulas of logarithmic cumulative distribution function are obtained using least squares method and are shown in Table 4. The change law of scale parameter, n, material crushing parameter, a and shape parameter, d' with the impact velocities are shown in Figs. 12, 13 and 14, respectively.

In Fig. 12, the coal scale parameter increases floatingly and the gangue scale parameter decreases straightly with the impact velocity. But the coal scale parameter is always less than the gangue scale parameter, which indicates that the discrete degree of broken coal size distribution is larger than that of the gangue under the same impact velocity. The discrete degree difference of particle size distribution between the broken coal and gangue is narrowed with the impact velocity increasing. There is a raise of coal scale parameter when impact velocity increases to 8 m/s. It indicates that there is a big range of coal particle size distribution, and the not fully broken particles and the broken particles are all mixed in the broken coal. When impact velocity is 6 m/s, the coal are not fully broken with a few small particles generating. When impact velocity is between 10 to 14 m/s, all the coal are fully broken with no big particles remaining. Therefore, impact velocity of 8 m/s is the fully-broken velocity boundary of coal. The gangue scale parameter decreases straightly which indicates that there is no obvious fully broken velocity boundary.

Table 4 Fitting formulas of logarithmic cumulative distribution function

Fig. 12 Relationships between scale parameter, n, and impact velocity, v

Fig. 13 Relationships between material crushing parameter, a, and impact velocity, v

Fig. 14 Relationships between shape parameter, d', and impact velocity, v

In Fig. 13, the coal material crushing parameter likes a wave but changes slightly with the impact velocity, while the absolute value of gangue material crushing parameter decreases with the impact velocity and there is a linear relationship between the gangue material crushing parameter and the impact velocity. The absolute value of coal material crushing parameter is always less than that of gangue. It indicates that the number of small coal particles is stable in a certain range and does not increase with impact velocity. But the number of small gangue particles does not act like coal and increases with impact velocity. When impact velocity is 8 m/s, there is a fall valley of coal material crushing parameter. This is associated with the coal scale parameter and the reason is no longer discussed here. The difference of material crushing parameter between the coal and gangue is explained by the fact that there is a large physical properties difference between coal and gangue, which is benefit for coal and gangue separation.

In Fig. 14, the shape parameters of broken coal and gangue both decrease with impact velocity, and the shape parameters of broken coal are always less than that of gangue. It indicates that particle sizes of broken coal and gangue decrease with impact velocity. The decrease of gangue shape parameter is fluctuating and nearly linear, while coal shape parameters decrease rapidly firstly and then slow down, which indicates that the coal was not fully broken and the large particles improved the shape parameter when impact velocity was low (6 m/s, 8 m/s). The increase of impact velocity leads to the increase of coal fragmentation degree which reduces large particles rapidly. The particle size of broken coal remains in a certain range although large particles were all broken into pieces when the impact velocity is major. Therefore, the change of coal shape parameter is non-significant with the impact velocity increasing.

Combining with Figs. 12, 13 and 14, yet the change of each separation indicator is different with impact velocity, it can be obtained that the separation indicators of coal are always less than that of the gangue (using absolute value) under the same impact velocity, and there is a fully-broken velocity boundary of coal material. The particle size distribution of broken coal is too wide to separate coal and gangue effectively. Therefore, impact velocity must be larger than the fully- broken boundary velocity, but large impact velocity will increase the number of small particles which increases the mixed gangue ratio and fine coal ratio. For this reason, the analysis of particle size density distribution is conducted to find out the suitable impact velocity.

5.3 Analysis of particle size density distribution

The density distribution functions of broken coal and gangue are obtained according to Eq. (3) as shown Table 5. The relationships between density distribution function and impact velocity are shown in Fig. 15. The values in Fig. 15 are the characteristic particle diameters of curve peaks.

In Fig. 15, the peak values of density distribution function increase with the impact velocity, and the curve crest moves toward left with increasing impact velocity. The peak values of broken coal increase observably but the offset and the curve span decrease. However, the peak values of broken gangue increase non-significantly compared with that of coal. It indicates that the coal particles concentrate toward the range of 0-50 mm with impact velocity increasing. There are overlap regions between the density distribution curves of coal and gangue. The increased overlap region leads to the higher mixed degree of coal and gangue and worse separation effect.

According to Fig. 15, the peak values of density distribution curves are the mode diameters of broken coal and gangue. The mode diameters under different impact velocities are acquired based on Eq. (4) and the relationships between the mode diameter, d₀ and the median diameter, d′(shape parameter) are shown in Fig. 16.

In Fig. 16, the mode diameter of coal and gangue is less than the median diameter under the same impact velocity. The variation tendency gangue mode diameter and median diameter is the same and decreases with impact velocity. The variation tendency of coal mode diameter and median diameter is not the same when impact velocity is small (v=6 and 8 m/s), and the variation tendency becomes almost the same when impact velocity gets large (v>10 m/s). It indicates that large particles break into pieces and number of large particles decreases rapidly, and number of small particles increases but amount of the broken small particles is less than large particles. The coal mode diameter decreases with impact velocity as a stair-step shape due to the fully broken boundary of coal material.

The intersecting coordinates of density distribution curves of broken coal and gangue under the same impact velocity are obtained based on the Table 5 and Fig. 15, and then the relationships between the overlap region and impact velocity are also acquired and shown in Table 6 and Fig. 17. The d_c expresses the characteristic particle diameter of intersecting point and f(d_c) stands for the density distribution function of intersecting point.

In Fig. 17, the overlap regions of density distribution curve of coal and gangue decrease rapidly and then increase slightly with increasing impact velocity. It indicates that there is a certain impact velocity when the overlap regions is the minimum. The overlap region equation is acquired using the least squares method and the correlation coefficient is 0.9700, meaning a well fitting degree. The minimum extreme value is 49.13 while impact velocity is 12.10 m/s. The minimum overlap region of density distribution curve leads to the minimum mixed degree of coal and gangue, and it is the most benefit velocity for coal and gangue separation when impact velocity is 12.10 m/s.

As the limitations and shortcomings of this work, it focuses on the relationships between the particle size distribution and impact velocities of coal and gangue, but impact velocity is only one of the several main factors affecting impact-crush separation of coal and gangue. Hardness difference, mineral mixture ratio and particle size distribution of raw coal are also the main factors affecting separation results. Because of the experiment materials and funds limits, the separation experiments based on the hardness difference and the mineral mixture ratio will be carried out in the future.

Table 5 Density distribution functions of broken coal and gangue

Fig. 15 Density distribution functions of broken coal and gangue

Fig. 16 Relationships between mode diameter d₀ and median diameters d′

Table 6 Intersecting coordinates and overlap regions of density distribution function curves

Fig. 17 Relationship between overlap region and impact velocity

6 Conclusions

1) When the impact velocity was higher than 10 m/s and the mesh size was bigger than 80 mm, the zero loss coal ratio was realized in the impact-crush separation although the mixed gangue ratio was larger than 40%, which proves that the impact-crush separation is feasible. The cumulative distribution fitting curve of broken coal and gangue was obtained using least squares method. The all correlation coefficients of fitting equations were greater than 0.98, which indicated that the Rosin-Rammler distribution could accurately describe the particle size distribution of broken coal and gangue under different impact velocities.

2) The scale parameter of broken coal increases floatingly with impact velocity, while the material crushing parameter likes a wave but changes slightly with impact velocity. The scale parameter of broken gangue decreases while the material crushing parameter increases with impact velocity. All the shape parameters of broken coal and gangue increase with impact velocity. Under the same impact velocity, the scale parameter and shape parameter of broken coal are always less than that of gangue, and the material crushing parameter of broken coal is larger than that of gangue. There is a fully-broken velocity boundary of coal material, and the impact velocity must be higher than the fully broken boundary velocity for a better coal and gangue separation.

3) The peak values of density distribution function increase with the impact velocity. The mode diameter of coal and gangue are always less than the median diameter under the same impact velocity. The variation tendency of gangue mode diameter decreases with the impact velocity which is the same as that of median diameter. The variation tendency of coal mode diameter and median diameter is not the same when the impact velocity is low (v=6 and 8 m/s), and the variation tendency becomes almost the same when the impact velocity is high (v>10 m/s). The coal mode diameter decreases as stair-step shape with the impact velocity due to the fully-broken velocity boundary of coal material. The overlap regions of density distribution curve decrease rapidly and then increases slightly with the impact velocity. The minimum extreme value of overlap region equation is 49.13 when the impact velocity is 12.10 m/s, which leads to the minimum mixed degree of coal and gangue, and it is the most benefit velocity for coal and gangue separation.

References

[1] TRIPATHY S K, BHOJA S K, KUMAR C R, SURESH N. A short review on hydraulic classification and its development in mineral industry [J]. Powder Technology, 2015, 270: 205-220.

[2] WANG Xin-min, ZHAO Bin, ZHANG Chuan-shu, ZHANG Qin-li. Paste-like self-flowing transportation backfilling technology based on coal gangue [J]. Mining Science and Technology, 2009, 19(2): 137-143.

[3] GUO Guang-li, ZHA Jian-feng, MIAO Xie-xing, WANG Qiang, ZHANG Xian-ni. Similar material and numerical simulation of strata movement laws with long wall fully mechanized gangue backfilling [J]. Procedia Earth and Planetary Science, 2009, 1(1): 1089-1098.

[4] QIAN Ming-gao, MIAO Xie-xing, XU Jia-lin. Green mining of coal resources harmonizing with environment [J]. Journal of China Coal Society, 2007, 32(1): 1-7. (in Chinese)

[5] MIAO Xie-xing, QIAN Ming-gao. Research on green mining of coal resources in China: current status and future prospects [J]. Journal of Mining and Safety Engineering, 2009, 26(1): 1-14. (in Chinese)

[6] XIE He-ping, WANG Jin-hua, SHEN Bao-hong, LIU Jian-zhong, JIANG Peng-fei, ZHOU Hong-wei, LIU Hong, WU Gang. New idea of coal mining: Scientific mining and sustainable mining capacity [J]. Journal of China Coal Society, 2012, 37(7): 1069-1079. (in Chinese)

[7] XU Qi, ZHANG Yong-chuan. A novel automated separator based on dual energy gamma-rays transmission [J]. Measurement Science and Technology, 2000, 11(9): 1383-1388.

[8] XU Qi, LI Kong, CHENG Jing-jing. Adaptive fuzzy pattern recognition of coal and gangues based on niche genetic algorithm [J]. Journal of Huazhong University of Science and Technology, 2003, 31(12): 22-24. (in Chinese)

[9] DING Ke-jian, LASKOWSKI J S. Coal reverse flotation, Part I: Separation of a mixture of subbituminous coal and gangue minerals [J]. Minerals Engineering, 2006, 19(1): 79-86.

[10] LUO Zhen-fu, FAN Mao-ming, ZHAO Yue-min, TAO Xiuxiang, CHEN Qing-ru, CHEN Zeng-qiang. Density-dependent separation of dry fine coal in a vibrated fluidized bed [J]. Powder Technology, 2008, 187(2): 119-123.

[11] MA Li-qiang, LI Yong-sheng. The clean production technology of replacement of coal gangue in coal mine [J]. Journal of China Coal Society, 2010, 35(5): 816-819. (in Chinese)

[12] HE J, ZHAO Y, HE Ya-qun, LUO Z, DUAN Chen-long. Separation performance of raw coal from South Africa using the dense gas-solid fluidized bed beneficiation technique [J]. Journal-South African Institute of Mining and Metallurgy, 2012, 113(7): 1-8.

[13] HE Min, WANG Pei-pei, JIANG Hui-hui. Recognition of coal and stone based on SVM and texture [J]. Computer Engineering and Design, 2012, 33(3): 1117-1121. (in Chinese)

[14] XU Jie, WANG Feng-wen. Study of automatic separation system of coal and gangue by ir image recognition technology [M]. Advances in Automation and Robotics, Berlin Heidelberg: Springer, 2012, 2: 87-92.

[15] MIAO Xie-xing, ZHANG Ji-xiong. Key technologies of integration of coal mining-gangue washing-backfilling and coal mining [J]. Journal of China Coal Society, 2014, 39(8): 1424-1433. (in Chinese)

[16] ZHENG Gang-biao, KANG Tian-he, YIN Zhi-hong, ZHANG Chi-ming, DUAN Jun. Research on dust size distribution of coal impacted with different forms [J]. Journal of Mining and Safety Engineering, 2007, 24(1): 96-100. (in Chinese)

[17] ZHENG Gang-biao, KANG Tian-he, CHAI Zhao-yun, YIN Zhi-hong. Applied the Rosin-Rammler Distribution function to study on the law of coal dust particle-size distribution [J]. Journal of Taiyuan University of Technology, 2006, 37(3): 317-319. (in Chinese)

[18] LI Jian-ping, DU Chang-long, XU Long-jiang. Impactive crushing and separation experiment of coal and gangue [J]. Journal of the China Coal Society, 2011, 36(4): 687-690. (in Chinese)

[19] LI Jian-ping, DU Chang-long, BAO Jian-wei. Direct-impact of sieving coal and gangue [J]. Mining Science and Technology, 2010, 20(4): 611-614.

[20] LUO Chen-xu, DU Chang-long, XU Long-jiang, ZHENG Ke-hong. Fractal distribution studies of a rotary crushing mechanism [J]. Recent Patents on Mechanical Engineering, 2014, 7(1): 44-51.

[21] ZHENG Ke-hong, DU Chang-long, YANG Dao-long. Orthogonal test and support vector machine applied to influence factors of coal and gangue separation [J]. International Journal of Coal Preparation and Utilization, 2014, 34(2): 75-84.

[22] LI Jian-ping, YANG Dao-long, DU Chang-long. Evaluation of an underground separation device of coal and gangue [J]. International Journal of Coal Preparation and Utilization, 2013, 33(4): 188-193.

[23] LI Jian-ping, ZHENG Ke-hong, DU Chang-long. The distribution discipline of impact crushed on coal and gangue [J]. Journal of China Coal Society, 2013, 38(Supplement 1): 54-58. (in Chinese)

[24] GAO Kui-dong, DU Chang-long, WANG Hao-xiang , ZHANG Shi-bo. An efficient of coal and gangue recognition algorithm [J]. International Journal of Signal Processing, Image Processing and Pattern Recognition, 2013, 4(6): 345-354.

[25] ZHENG Ke-hong, DU Chang-long, LI Jian-ping, QIU Bing-jing, YANG Dao-long. Underground pneumatic separation of coal and gangue with large size (≥50 mm) in green mining based on the machine vision system [J]. Powder Technology, 2015, 278: 223-233.

[26] DU Chang-long, JIANG Hong-xiang, LIU Song-yong. Research on variable linear vibration screen with flexible screen face of separator for coal and gangue underground [J]. Journal of China Coal Society. 2013, 38(3): 493-497.

[27] SELBIG W R, FIENEN M N. Regression modeling of particle size distributions in urban storm water: advancements through improved sample collection methods [J]. Journal of Environmental Engineering, 2014, 138(12): 1186-1193.

[28] KASHANI A, NICOLAS R S, QIAO G G, JANNIE S J, JOHN L P. Modelling the yield stress of ternary cement–slag–fly ash pastes based on particle size distribution [J]. Powder Technology, 2014, 266(6): 203-209.

[29] STOYAN D. Weibull, RRSB or extreme-value theorists? [J]. Metrika, 2013, 76(2): 153-159.

[30] ALDERLIESTEN M. Mean particle diameters. Part VII. The rosin-rammler size distribution: physical and mathematical properties and relationships to moment-ratio defined mean particle diameters [J]. Particle and Particle Systems Characterization, 2013, 30(3): 244-257.

[31] KONDOLF G M, ADHIKARI A. Weibull vs. lognormal distributions for fluvial gravels [J]. Journal of Sedimentary Research, 2000, 70(3): 456-460.

[32] FENG Jian-rong, LIU Zhi-he, WANG Cheng-shou, LI Zhi-hong, Experimental study on the critical parameter on the crush of mixture of coal and gangue [J]. Journal of Taiyuan University of Technology, 2006, 37(1): 42-43. (in Chinese)

(Edited by DENG Lü-xiang)

Cite this article as: YANG Dao-long, LI Jian-ping, DU Chang-long, ZHENG Ke-hong, LIU Song-yong. Particle size distribution of coal and gangue after impact-crush separation [J]. Journal of Central South University, 2017, 24(6): 1252-1262. DOI: 10.1007/s11771-017-3529-2.

Foundation item: Project(2012AA062102) supported by High-Tech Research and Development Program of China; Project(KYLX_1379) supported by the Innovation Training Project of Graduate Student in Jiangsu Province, China; Project supported by the Priority Academic Program Development of Jiangsu Higher Education Institutions, China

Received date: 2016-02-22; Accepted date: 2016-07-10

Corresponding author: LI Jian-ping, Professor; Tel: +86-13505205658; E-mail: jdljping@cumt.edu.cn