Prediction of blasting mean fragment size using support vector regression combined with five optimization algorithms

The main purpose of blasting operation is to produce desired and optimum mean size rock fragments.Smaller or fine fragments cause the loss of ore during loading and transportation,whereas large or coarser fragments need to be further processed,which enhances production cost.Therefore,accurate prediction of rock fragmentation is crucial in blasting operations.Mean fragment size(MFS) is a crucial index that measures the goodness of blasting designs.Over the past decades,various models have been proposed to evaluate and predict blasting fragmentation.Among these models,artificial intelligence(AI)-based models are becoming more popular due to their outstanding prediction results for multiinfluential factors.In this study,support vector regression(SVR) techniques are adopted as the basic prediction tools,and five types of optimization algorithms,i.e.grid search(GS),grey wolf optimization(GWO),particle swarm optimization(PSO),genetic algorithm(GA) and salp swarm algorithm(SSA),are implemented to improve the prediction performance and optimize the hyper-parameters.The prediction model involves 19 influential factors that constitute a comprehensive blasting MFS evaluation system based on AI techniques.Among all the models,the GWO-v-SVR-based model shows the best comprehensive performance in predicting MFS in blasting operation.Three types of mathematical indices,i.e.mean square error(MSE),coefficient of determination(R^(2)) and variance accounted for(VAF),are utilized for evaluating the performance of different prediction models.The R^(2),MSE and VAF values for the training set are 0.8355,0.00138 and 80.98,respectively,whereas 0.8353,0.00348 and 82.41,respectively for the testing set.Finally,sensitivity analysis is performed to understand the influence of input parameters on MFS.It shows that the most sensitive factor in blasting MFS is the uniaxial compressive strength.

The Impacts and Causes of Land Fragmentation on Farm Productivity: Case Review of East African Countries

This report provides an overall assessment of land fragmentation problems in East Africa. Many parts of East Africa have become highly fragmented, putting development systems and activities in these areas at risk of complete collapse. Land fragmentation occurs when land gets converted for agriculture, industrialization, or urbanization, invaded by non-local plants, or enclosed for individual use and by subdividing farmlands into subsequent smaller units called parcels with varying average farm sizes. Fragmentation results from inappropriate agricultural development processes and ineffective land use planning that fails to recognize how farmland is used, and the importance of its interconnected areas. Insecurity of tenure and resource rights are key factors in making this possible. Land fragmentation is one of the key reasons why the ability of most resources in East Africa becomes scarcer, and those remaining become “privatized” by more powerful community members—keen to maintain their access to them. Such individualistic attitudes are new and disadvantage the poorest even further by affecting the traditional customary safety nets and agricultural outputs. Neither the government nor customary governance systems effectively protect resource access for the poorest. This review summary report identifies the key causes, measures, and implications, government interventions, and the common remedies to land fragmentation problems in the East African Countries of Kenya, Uganda, Rwanda, and Tanzania including neighboring Ethiopia, and the Sudan. The findings indicated from 2005 to 2015, the population kept increasing for all the named countries in East Africa with Rwanda and Uganda having a substantial increase in population density. The study review further explores the trend in the performance of agriculture by average farm sizes within the intervals of five years by highlighting their strong linkages and found that the average farm size has declined drastically, especially for Kenya. This can only mean that small farms kept becoming smaller and smaller and that there were more small-scale farmers. The results further depicted that the major and commonly cultivated food crops among the East African countries include maize, sorghum, rice, cassava, sweet potatoes, bananas, Irish potatoes, beans, peas, etc., with maize yields (Mt/ha) in 2003 for Uganda being the highest (1.79 Mt/ha) and the lowest in Rwanda (0.77 Mt/ha) respectively. Therefore, from the review results, recommendations are being made as to how the negative impacts of land fragmentation on agricultural productivity can be reduced or mitigated. One way is by community sensitization and awareness about the importance of land consolidation and its proposition on farm productivity.

烟丝加工过程中的破碎规律研究

为探索烟丝在加工过程中的破碎规律,设计不同切丝宽度和尺寸的烟丝,研究滚筒、流态化、机械输送等关键加工环节前后烟丝的结构变化规律。结果表明,随着烟丝加工进行,各规格烟丝不断破碎,烟丝的特征尺寸降低,破碎度升高,主要表现在>7.10 mm和4.50~7.10 mm等尺寸较长的烟丝质量分数不断降低,<2.00 mm的烟丝质量分数不断升高,不同处理间3.35~4.50 mm的烟丝质量分数变化差异最大。滚筒加工过程中烟丝质量分数的变化率整体较大,其中滚筒干燥过程最大,其次是振动输送、干燥后的提升输送和风送过程,风选和干燥前的提升等过程较小。加工后1.0 mm宽度烟丝、1.0 mm宽度断丝的烟丝、0.8 mm宽度烟丝和0.8 mm宽度断丝的烟丝特征尺寸变幅分别达到1.86 mm、1.09 mm、1.65 mm和1.08 mm,其破碎度分别达到33.51%、26.35%、33.39%和27.43%。综上,>7.10 mm和4.50~7.10 mm等特征尺寸较大及同一断丝或未断丝处理下宽度较窄的烟丝在加工过程中更易破碎,应根据加工的需求,关注特定尺寸的烟丝质量变化规律,在原料消耗允许的范围内,开展烟丝结构调控和工艺优化。

柱状药包侧向爆破下抵抗线对岩石破碎特性的影响研究

为研究抵抗线对柱状药包侧向爆破下岩石破碎特性的影响,采用波动理论,分析了自由面处的应力状态;采用直铜丝电爆炸模拟柱状药包爆破,开展了侧向爆破破岩的物理模型试验;再借助超高速摄像技术,观测了试件表面裂纹网的演化过程;结合爆破漏斗区形态、尺寸和碎块分布特征,研究了不同抵抗线下柱状药包爆破破岩特性。结果表明,抵抗线是控制岩石爆破破裂模式、破坏区形态和碎块块度大小的重要因素。随抵抗线增加,裂纹发育速度与密度减小,裂纹网覆盖范围增加,优先贯穿裂纹由横向转变为环向裂纹。此外,模型试验结果表明,破裂区短轴随抵抗线增加而增加。结合自由面处应力状态,利用最大拉应变准则对此进行了解释。建立了碎块块度与能量单耗之间的关系,研究成果为岩石工程爆破设计优化提供参考。

起爆顺序对台阶岩石破碎块度及爆破振动影响研究

为探究起爆顺序对露天台阶岩石破碎效果以及周围环境的影响,利用LS-DYNA仿真软件分别计算不同起爆顺序下的台阶爆破模型。分析不同起爆顺序下模型的物理力学特征;开展现场试验验证模拟结果,并分析起爆顺序对爆破振动强度及岩石破碎块度的影响特征。研究结果表明:起爆顺序对爆破过程中自由面空间分布产生影响,从而改变应力波的叠加效应,不同起爆顺序下台阶破碎效果及爆破振动强度表现出明显差异。逐孔起爆技术能够有效降低爆破振动,从而减轻对周围环境的不良影响,但其岩石破碎效果并不理想,岩石通过率为81.52%。与逐孔起爆相比,波浪式顺序起爆碎石通过率提高11.68%,增大孔网参数至3 m×7 m时单孔崩落矿量提高16.8%,这可以有效减少炸药消耗量,降低综合成本。然而,其产生的爆破振动强度较大,需要注意对周围环境的影响。

爆炸成型弹丸侵彻不同材料靶板后效破片特性试验研究

为研究靶板材料性能对爆炸成型弹丸(Explosively Formed Projectile,EFP)侵彻靶后破片特性的影响,开展EFP侵彻不同材料靶板(Q235钢、45号钢、装甲钢、2A12铝)后效破片特性试验,采用X光摄影方法观测靶后破片云形态及膨胀尺寸,通过布置多层纤维板获得破片散布特性,并对靶后破片进行回收。研究结果表明:在靶板密度一定的情况下,靶板强度主要影响破片云轴向膨胀能力,对径向膨胀能力影响很小;靶后破片环形毁伤区的飞散角位于20°~25°范围内差别不大,但是靶板背面出口崩落会造成靶后破片飞散角出现极大值,随着钢靶强度的增大,靶后破片径向散布增强,破片总数减小,但是大质量段钢破片数量增多;不同强度钢靶产生的钢破片平均尺寸满足Kipp等提出的基于材料流动应力的碎片尺寸模型。

Rock fragmentation control in opencast blasting

The blasting operation plays a pivotal role in the overall economics of opencast mines.The blasting subsystem affects all the other associated sub-systems,i.e.loading,transport,crushing and milling operations.Fragmentation control through effective blast design and its effect on productivity are the challenging tasks for practicing blasting engineer due to inadequate knowledge of actual explosive energy released in the borehole,varying initiation practice in blast design and its effect on explosive energy release characteristic.This paper describes the result of a systematic study on the impact of blast design parameters on rock fragmentation at three mines in India.The mines use draglines and shoveledumper combination for removal of overburden.Despite its pivotal role in controlling the overall economics of a mining operation,the expected blasting performance is often judged almost exclusively on the basis of poorly defined parameters such as powder factor and is often qualitative which results in very subjective assessment of blasting performance.Such an approach is very poor substitutes for accurate assessment of explosive and blasting performance.Ninety one blasts were conducted with varying blast designs and charging patterns,and their impacts on the rock fragmentation were documented.A high-speed camera was deployed to record the detonation sequences of the blasts.The efficiency of the loading machines was also correlated with the mean fragment size obtained from the fragmentation analyses.

基于裂纹动力学的岩石动力破碎平均块度研究

作为表征动力破碎的重要参数之一,平均破碎块度的研究对于揭示岩石破碎机理具有重要意义。尽管进行了大量理论与实验研究,但是还缺乏从裂纹动力学角度来澄清岩石破碎和块度形成机理。基于动态荷载作用下翼型裂纹扩展模型和J. R. Gladden柱体动力屈曲失稳模型,提出了一种预测岩石平均破碎块度的方法,并探究了应变率对动态强度和平均破碎块度的影响。研究结果表明,随着应变率的增加,动态强度增加,平均破碎块度减小,且应变率依赖性逐渐降低。模型平均破碎块度预测与实验数据吻合良好。

Dynamic fragmentation of microwave irradiated rock

The microwave-assisted rock fragmentation has been proven to be a promising approach in reducing cutting tools wear and improving efficiency in rock crushing and excavation.Thus,understanding the influence of damage induced by microwave irradiation on rock fragmentation is necessary.In this context,cylindrical Fangshan granite(FG)specimens were exposed to microwave irradiation at a power of 6 kW for different durations up to 4.5 min.The damages of the specimens induced by irradiation were quantified by using both X-ray micro-CT scanning and ultrasonic wave measurement.The CT value and Pwave velocity decreased with increase of irradiation duration.The irradiated specimens were then tested using a split Hopkinson pressure bar(SHPB)system to simulate rock fragmentation.A momentum-trap technique was utilized to ensure single-pulse loading on the specimen in SHPB tests,enabling valid fragment size distribution(FSD)analysis.The dependence of dynamic uniaxial compressive strength(UCS)on the irradiation duration and loading rate was revealed.The dynamic UCS increased with increase of loading rate while decreased with increase of irradiation duration.Using the sieve analysis,three fragmentation types were proposed based on FSD,which were dictated by both loading rate and irradiation duration.In addition,an average fragment size was proposed to quantify FSD.The results showed that the average fragment size decreased with increase of loading rate.A loading rate range was identified,where a dramatic reduction of the average fragment size occurred.The dependence of fragmentation on the irradiation duration and loading rate was also discussed.

面向自然崩落法矿山的初始块度预测软件研发及应用

矿石块度是采用自然崩落法开采需要考虑的重要因素。以自然崩落法初始块度预测基本原理为基础,在DIMINE软件平台上采用C++语言开发了初始块度预测软件系统。该系统综合考虑了节理系统的空间几何参数和地应力的作用,支持从块体体积、等效尺寸和块体形状等不同角度对初始块度结果进行分析。以国外某斑岩型铜矿为例,采用该软件对矿体按勘探线分区进行了初始块度预测,并将结果与国外常用的BCF软件的预测结果进行了对比分析。研究结果表明,矿体勘探线间初始块度的大块率在40.8%~74.8%之间,等效尺寸在1.65~2.63 m之间变化。软件预测的结果与BCF软件预测的结果比较接近,变化趋势基本一致,预测结果为自然崩落法矿山后续的开采设计提供了参考依据。

基于深度学习技术的爆堆块度识别方法研究

在露天矿爆破开采过程中,大块率是评价爆破质量的一个重要指标。较高的大块率不仅会大大降低采装效率,同时也增加二次破岩的费用,因此大块率统计是露天矿开采中一项重要工作。针对目前矿山存在的矿岩大块率统计复杂且准确性不高的问题,以乌努格吐山铜钼矿为研究对象,收集了矿区内台阶爆破爆堆图像数据,构建了基于深度学习的爆堆大块率统计模型。首先基于U-net矿岩图像分割模型,初步分割标注处理的数据集,建立了矿岩轮廓初次分割效果图。在残差学习模块基础上,改进Resu-net模型,优化训练标注数据,获得了最终矿岩轮廓分割效果图。最后,采用OpenCV图像处理技术,通过最小外接矩形法确定了爆堆块度尺度信息。结果表明,本研究提出的U-net+Resu-net爆堆块度优化分割模型准确率达到97.84%,爆堆矿岩图像分割数据较准确。通过OpenCV技术与相机单目成像原理相结合的方法,实现了倾斜爆堆矿岩图像的爆堆块度统计。此外,所开发的交互式界面操作简单,可快速统计大块尺寸。满洲里乌努格吐山铜钼矿的应用表明,该方法可高效、准确统计爆堆块度,具有一定的推广价值。

爆堆自适应分层的块度空间分布测量方法研究

爆堆块度分布特征是评价爆破效果的重要指标,针对目前爆堆块度直接与间接法测量方法的不足,提出了一种爆堆自适应分层的块度空间分布测量方法,该方法使用GA-LSSVM模型来预测Weilbull函数的形状参数α、β,并设置多个预测点,生成三维爆堆形态预测曲线。同时对Kuz-Ram块度预测模型参数进行转换和融合,建立了爆堆分层距离预测模型,并将其导入Weilbull-GA-LSSVM爆堆形态预测模型,实现爆堆自动分层。通过现场应用,不断优化分层设计,探究爆堆最佳分层位置,达到爆堆自适应分层的效果。以广东省大排矿山为工程依托开展现场试验,结果表明:(1)Weibull-GA-LSSVM模型能够准确预测爆堆形态,其中爆堆最大前移距离预测结果的平均相对误差仅为5.6%,松散系数预测结果的相对误差多数在9%左右,表现出良好的稳定性。(2)爆堆自动分层模型能够在爆前合理地输出分层距离以及分层层数,保证了爆后现场爆堆的铲装效率。(3)推导出爆堆最佳分层位置距离公式,实现了爆堆自适应分层,爆堆块度分布测量精度明显提高,更接近于爆堆整体块度分布。

爆轰驱动陶瓷面背板碎裂行为研究

为研究陶瓷平板在爆轰载荷下的碎裂行为,采用AUTODYN软件对其爆轰驱动过程进行了数值计算,通过爆轰波与陶瓷平板的作用过程分析及陶瓷破片的特征尺寸统计,得到了陶瓷平板的裂纹拓展规律及陶瓷破片尺度分布规律。结果表明:不同陶瓷平板药室中爆轰波的传播及其与陶瓷平板的作用过程类似,但陶瓷裂纹拓展及形态不同,无拼接陶瓷平板形成网格化裂纹,而拼接陶瓷平板则呈现环向裂纹及径向裂纹;爆轰载荷下陶瓷平板药室形成的陶瓷破片尺度满足Rosin-Rammler分布模型;不同尺度陶瓷破片质量分布的比重由幂指数系数k决定,陶瓷平板的整体损伤程度可用平均特征尺寸系数λ评判,考虑到陶瓷平板药室的低附带损伤特性,选择小尺寸的陶瓷片进行陶瓷平板拼接较合适,研究结果可为陶瓷平板药室在轻型装甲车辆上的应用提供参考。

Damage evolution and fragmentation behavior of pyramid cut blasting under uniaxial compression

Pyramid cut blasting is an essential form of inclined hole cut blasting,but the in-situ stress effect of pyramid cut blasting is rarely studied.Based on the research background of pyramid cut blasting in a deep rock mass,the size,volume,and fragment size distribution of the blasting cavity before and after uniaxial compression were analysed by a model test.Otherwise,the damage and effective stress of the pyramid cut blasting were analysed with LS-DYNA numerical simulation.The results show that the damage and fragmentation of pyramid cut blasting are not only affected by blasting stress wave and blasting gas,but also affected by uniaxial compression.Under the influence of uniaxial compression,the blasting stress wave and blasting gas are more likely to damage the rock mass parallel to the uniaxial compression direction near the connecting line of blasting hole,and make the volume of cavity larger and the fragment rate lower.Additionally,uniaxial compression has a prominent influence during the middle and late stages of blasting.

Parametric study of single confined fragment launch explosive device

In this article, parametric study of single confined fragment launch device was carried out. The configuration proposed was further studied to derive the empirical relationship for effect of fragment size,charge size, confinement thickness on fragment velocity. The simulations were carried out using ANSYSAUTODYNE explicit solver. Fragment velocities were estimated as a function of different parametric combinations of explosive quantities, charge length to diameter ratio, fragment height to diameter ratio,confinement thickness, fragment material and fragment mass. The data was further converted to charge to metal ratio under fragment and confinement. It was observed that, increase in confinement thickness,charge quantity and decrease in fragment height increases the fragment velocity. It is also noted that,charge to metal mass ratio under fragment significantly affects the fragment velocity. At the end, an empirical relationship for fragment velocity interms of all these parameters was established. Using these relations, two velocities 1831.92 m/s and 2523.9 m/s required for NATO STANAG 4496 IM test were estimated. The design parameters for these velocities are presented. Also, the results estimated using the empirical relationship has been compared with published experimental data. Error in the predicted velocities is within the acceptable range. The empirical relationship proposed will be useful for finalization of design of the fragment launch device.

A review of development of better prediction equations for blast fragmentation

The paper reviews the development of prediction formulas for the fragmentation from bench blasting.Much attention has been paid to the Kuz-Ram model,its development and errors,and the mean vs.median misunderstanding.The work by the US Bureau of Mines(USBM)and Chung and Katsabanis are also reviewed,as well as the two Julius Kruttschnitt Mineral Research Centre(JKMRC)models,i.e.the crush zone model(CZM)and the two-component model(TCM),which were developed to cope with the underestimation of blasting fines.The change brought by the Swebrec distribution and the associated Kuznetsov-Cunningham-Ouchterlony(KCO)model is described.Studying distribution-free fragment sizes xP for an arbitrary mass passing P led to the discovery of the fragmentation-energy fan,and with the help of dimensional analysis,to the new fragmentation prediction model xP-frag,which has much lower errors than those of the Kuz-Ram and CZM models.

An estimation model for the fragmentation properties of brittle rock block due to the impacts against an obstruction

Mountain hazards with large masses of rock blocks in motion – such as rock falls, avalanches and landslides – threaten human lives and structures. Dynamic fragmentation is a common phenomenon during the movement process of rock blocks in rock avalanche, due to the high velocity and impacts against obstructions. In view of the energy consumption theory for brittle rock fragmentation proposed by Bond, which relates energy to size reduction, a theoretical model is proposed to estimate the average fragment size for a moving rock block when it impacts against an obstruction. Then, different forms of motion are studied, with various drop heights and slope angles for the moving rock block. The calculated results reveal that the average fragment size decreases as the drop height increases, whether for free-fall or for a sliding or rolling rock block, and the decline in size is rapid for low heights and slow for increasing heights in the corresponding curves. Moreover, the average fragment size also decreases as the slope angle increases for a slidingrock block. In addition, a rolling rock block has a higher degree of fragmentation than a sliding rock block, even for the same slope angle and block volume. Finally, to compare with others' results, the approximate number of fragments is estimated for each calculated example, and the results show that the proposed model is applicable to a relatively isotropic moving rock block.

Fragmentation Analyses of Granular Explosive Under Drop Weight Impact

Fragmental size and distribution of explosive particles play a more important role in the formation of hot-spot than original particles size under drop weight impact.Because the particles breakage and the hot-spots ignition will form in a sequence between fragments and between the fragments and the drop weight surface under the impact.In this paper,the size and distribution of the cyclotetramethylenete tranitramine(HMX)fragments were analyzed by the Laser Particle Size Analyzer Malvern MS2000.The post-analysis results of fragments showed that size distribution of fragments was strongly dependent on drop height.An empirical formula is established to describe the relationship between the average size and drop height.The volume-based probability distribution of explosive fragments was also studied by experiments and theoretical calculations.

Numerical Prediction of Statistical Masonry Wall Fragment Distribution Induced by Blast Loading

The hazard caused by the fragments of damaged structures is usually significant in acci-dental explosions or hostile blast events.A reliable and efficient method to estimate probable fragment size,velocity and launch distance will be useful to assess and design countermeasures to mitigate the possible fragment hazards.This paper presents a numerical method for predicting the size and launch distance of the fragments caused by explosive damage of masonry wall.Numerical simulations with different scaled distances are carried out,and the statistical distribution functions of the fragment size and launch distance in terms of the scaled distance are derived.

流化床热解试验的褐煤碎裂特性研究

褐煤在流化床热解过程中易受热力和机械力影响而发生碎裂形成粉尘,严重影响褐煤梯级加工利用发展,因而需研究褐煤在热力与机械力交互作用条件下的碎裂特性,以期为其热解、燃烧、气化等煤转化利用技术发展提供数据参考。选取内蒙古的2种褐煤,利用微型流化床热解试验装置,研究粒径、加热温度、升温速率、停留时间以及流化数各因素对褐煤碎裂特性的影响。实验研究表明:褐煤热解破碎后产生半焦的粒级主要分布在原粒级及低一档粒级范围内,颗粒的粒度变化率和粉化率随入料粒径增大而均提高,加热终温升高及升温速率增大则均加剧煤样的碎裂程度,停留时间的延长会使颗粒碎裂程度逐渐增大至粒度稳定,流化数的升高会促进颗粒粉化且对细颗粒生成的影响更明显;在流化床热解中热力、机械力分别是产生粉化的主要作用力以及辅助作用力,褐煤受热力作用下的碎裂形式为外围发生细颗粒的剥落、内部破碎为大块颗粒;通过扫描电镜分析不同温度下的半焦孔隙结构可知,随着挥发分的析出则半焦孔隙结构逐渐发育,最终由于集聚胀力与热应力作用而导致孔隙结构被破坏直至坍塌、碎裂。