The introduction of neural network system and its applications in rock engineering

The neural network system has been developing very fast recently. It has been widely used in many industries such as automation, nuclear power plant, chemical industry, etc. Neural network systems have a great advantage in dealing with problems in which many factors influence the process and result, and the understanding of this process is poor, and there are experimental data or field data. In rock engineering, many problems are of this nature. In this paper, a brief introduction to neural network systems is given. Problems such as what is a neural network, how it works and what kind of advantages it has are discussed. After this, several applications in rock engineering, made by us, are presented. Case 1 is ore boundary delineation. In this case, the rock are divided into three classes, i.e.: (1) waste rock; (2) semi-ore; and (3) ore for mining purposes. The neural network system built can judge whether it is ore, semi-ore or waste rock along the borehole according its corresponding geophysical logging data, such as gamma-ray, gamma-gamma, neutron and resistivity. Case 2 is aggregate quality prediction. In this case, the quality parameters: (1) impact value; (2) abrasion value I; and (3) abrasion value II are predicted by using a neural network system based on density, point load, content of quarts and content of brittle minerals. Case 3 is rock indentation depth prediction. In this case, the rock indentation depth under indentation load is predicted by the established neural network system based on the indentation load on rock, indenter type and rock mechanical properties, such as uniaxial compressive strength, Young's modulus. Poisson's ratio, critical energy release rate and density. In all these cases, the neural network systems have been applied successfully. The testing results are satisfactory and better than the existing techniques.     

Texture Segmentation Based on Image Model and Neural Network

INTRODUCTIoNImagetextureanalysisisanimportantpartofresearchin-tocomputervision.Forobjectidentificationandunderstand-ing,textureanalysisisthebasisoftheresearchwork.Ingeneral,texturecanberegardedasakindofstructurethatconsistsofmanytextureelementsorpatternswhicharemoreorlesssimilar,i.e.,primitivesthatformtexturesandspatialdependenceorinteractionbetweentheprimitives.Ex-tensiveresearchhasbeendoneintextureanalysis,andrefer-ence(Haralick,l979)hasmadeathoroughsurvey.Amongthesetextureanalysismetho…     

Channelized and hillslope sediment transport and the geomorphology of mountain belts

This paper uses the results of landscape evolution models and morphometric data from the Andes of northern Peru and the eastern Swiss Alps to illustrate how the ratio between sediment transport on hillslopes and in channels influences landscape and channel network morphologies and dynamics. The headwaters of fluvial- and debris-flow-dominated systems (channelized processes) are characterized by rough, high-relief, highly incised surfaces which contain a dense and hence a closely spaced channel network. Also, these systems tend to respond rapidly to modifications in external forcing (e.g., rock uplift and/or precipitation). This is the case because the high channel density results in a high bulk diffusivity. In contrast, headwaters where landsliding is an important sediment source are characterized by a low channel density and by rather straight and unstable channels. In addition, the topographies are generally smooth. The low channel density then results in a relatively low bulk diffusivity. As a consequence, response times are greater in headwaters of landslide-dominated systems than in highly dissected drainages. The Peruvian and Swiss case studies show how regional differences in climate and the litho-tectonic architecture potentially exert contrasting controls on the relative importance of channelized versus hillslope processes and thus on the overall geomorphometry. Specifically, the Peruvian example illustrates to what extent the storminess of climate has influenced production and transport of sediment on hillslopes and in channels, and how these differences are seen in the morphometry of the landscape. The Swiss example shows how the bedding orientation of the bedrock drives channelized and hillslope processes to contrasting extents, and how these differences are mirrored in the landscape. An erratum to this article can be found at     

裂隙网络模拟与REV尺度研究

研究了利用平面四边形模拟节理岩体三维裂隙网络的方法。在生成裂隙网络时,同时考虑结构面几何参数和力学参数的随机性。利用裂隙网络研究了确定岩体REV尺度的指标。在此基础上,编制了岩体裂隙网络模拟程序、裂隙网络图形输出程序和岩体REV尺度指标分析程序。通过算例分析,验证了研究成果及程序的合理性,并得出了岩体REV尺度约为各组裂隙中最大迹长期望值3~4倍的结论。研究成果为后续计算岩体等效力学参数奠定了基础。     

基于模糊神经网络的义县萤石成矿预测

成矿预测的模糊综合评判方法中评价因素权重确定时具有较强的人为因素,模糊神经网络方法可以建立在无模型基础上,自动获取输入与输出的关系,在专家确定权重初始值的情况下,通过网络训练确定出最优权重值,这在一定程度上减少了人为因素的影响.应用该方法对义县萤石成矿远景区进行了评价,其评价结果与实际一致.表明该方法可推广到其他区域、其他矿种的预测,但应根据预测对象的实际情况建立控矿因素集和相应的隶属度函数.     

三维地震属性参数在煤层厚度预测中的应用

依托“西部煤炭资源高精度三维地震勘探技术”项目工程,对晋城某矿南翼大巷东南区5m×5m×1ms的三维地震数据体,采用三维地震属性参数预测煤层厚度及其变化规律：沿3煤层、15煤层10ms时窗提取地震属性42种,根据钻孔资料,计算出煤厚与地震属性相关系数;从中优选出相关系数大于0．35的地震属性,其中3煤层9个、15煤层10个;然后进行地震属性互相关分析,优选出与3煤、15煤层厚度相关系数较大的4种属性,建立预测煤厚的BP神经网络模型,分别选取3煤层12个、15煤层4个实测数据作为学习训练和测试样本,以钻孔地震属性作为学习样本,对网络进行训练,最终获得全区煤层厚度。经与预留钻孔成果资料对比,预测精度较高,结果可用。     

改进BP神经网络在岩质边坡稳定性分析中的应用

使用MATLAB软件中的NNtool工具箱建立了一个通过添加动量改进的BP神经网络模型,用以分析岩质边坡的稳定性。通过对天津某工程的边坡进行分析来检验这个模型的可靠度。根据现场调查的实际情况分析,该边坡岩体比较破碎,因此文中选择了几个有代表性的岩石完整性参数作为输入层参数,通过收集的一些类似边坡的参数对所建立的神经网络模型进行训练,以保证此模型适合被用来分析该工程。结果证明,所建立的模型是可靠的,此种方法在分析岩质边坡稳定性的的实践中具有广泛的使用性。     

一种新的可靠度计算方法及其在工程中的应用

利用神经网络拟合极限状态方程、几何法计算结构可靠度,并推导了基于神经网络的几何法计算公式。最后比较了该法与其他可靠度计算方法的精度以及计算效率,并将该法应用于实际工程当中。     

基于地震反演方法的奥陶系顶部含隔水层探测

在山西阳泉泊里矿区,太原组K₂灰岩是15号煤层上部主要的含水层,查明其富水分布特征对上下组煤层安全开采至关重要。为了准确得到K₂灰岩的富水分布区域,首先,利用常规的波阻抗反演获取精确的K₂灰岩空间展布特征。然后,结合皮尔逊相关系数法与交叉验证−逐步回归法优选出9种地震属性,构成网络的训练数据。此外,引入适合于时序数据处理且能够捕捉测井曲线前后相关性的长短期记忆神经网络(LSTM),构建智能化、多变量LSTM视电阻率预测模型,以精确地预测研究区视电阻率进而得到地层富水性分布特征。同时,分别利用常规多属性回归算法与多变量LSTM模型在井点位置建立电阻率测井曲线与地震属性井旁道之间的映射关系。最后,将井点处训练好的网络模型推广至无井区得到全区视电阻率体,根据视电阻率值的高低、矿区地质构造与陷落柱发育情况圈定灰岩富水区。实际数据的测试结果表明：与常规多属性回归算法相比,多变量LSTM模型预测误差小,与测井相关系数高,说明多变量LSTM模型可以更加精确地预测出工区视电阻率,在含煤地层的富水性预测中有较好的应用价值。\t\t\t\t