A comparative study of ANN and Neuro-fuzzy for the prediction of dynamic constant of rockmass

Physico-mechanical properties of rocks have great significance in all operational parts in mining activities, from exploration to final dispatch of material. Compressional wave velocity (p-wave velocity) and anisotropic behaviour of rocks are two such properties which help to understand the rock response under varying stress conditions. They also influence the breakage mechanism of rock. There are different methods to determine thep-wave velocity and anisotropyin situ and in the laboratory. These methods are cumbersome and time consuming. Fuzzy set theory, Fuzzy logic and Neural Networks techniques seem very well suited for typical geotechnical problems. In conjunction with statistics and conventional mathematical methods, hybrid methods can be developed that may prove to be a step forward in modeling geotechnical problems. Here, we have developed and compared two different models, Neuro-fuzzy systems (combination of fuzzy and artificial neural network systems) and Artificial neural network systems, for the prediction of compressional wave velocity.     

Artificial neural networks and nonlinear regression techniques to assess the influence of slake durability cycles on the prediction of uniaxial compressive strength and modulus of elasticity for carbonate rocks

Understanding rock material characterizations and solving relevant problems are quite difficult tasks because of their complex behavior, which sometimes cannot be identified without intelligent, numerical, and analytical approaches. Because of that, some prediction techniques, like artificial neural networks (ANN) and nonlinear regression techniques, can be utilized to solve those problems. The purpose of this study is to examine the effects of the cycling integer of slake durability index test on intact rock behavior and estimate some rock properties, such as uniaxial compressive strength (UCS) and modulus of elasticity (E) from known rock index parameters using ANN and various regression techniques. Further, new performance index (PI) and degree of consistency (Cd) are introduced to examine the accuracy of generated models. For these purposes, intact rock dataset is established by performing rock tests including uniaxial compressive strength, modulus of elasticity, Schmidt hammer, effective porosity, dry unit weight, p‐wave velocity, and slake durability index tests on selected carbonate rocks. Afterward, the models are developed using ANN and nonlinear regression techniques. The concluding remark given is that four‐cycle slake durability index (I_d4) provides more accurate results to evaluate material characterization of carbonate rocks, and it is one of the reliable input variables to estimate UCS and E of carbonate rocks; introduced performance indices, both PI and Cd, may be accepted as good indicators to assess the accuracy of the complex models, and further, the ANN models have more prediction capability than the regression techniques to estimate relevant rock properties. Copyright © 2011 John Wiley & Sons, Ltd.     

Factors affecting the durability of selected weak and clay-bearing rocks from Turkey, with particular emphasis on the influence of the number of drying and wetting cycles

Weathering can induce a rapid change of rock material from initial rock-like properties to soil-like properties. The resistance of a rock to short-term weathering is described through a durability parameter called the slake durability index. As durability is an important engineering parameter, particularly for weak and clay-bearing rocks, it was assessed by a number of tests. The main purpose of this study is to assess the influence of the number of drying and wetting cycles and controls of mineralogical composition and strength on durability. For this purpose, 141 samples of different types of weak and clay-bearing rocks were selected from different parts of Turkey, and relationships between the above-mentioned rock characteristics were statistically investigated. The samples were subjected to multiple-cycle slake durability testing, X-ray diffraction (XRD) analysis and uniaxial compression testing. In addition, to assess the influence of mineralogical composition on durability, the mineral contents of the original material and the material passing from the drum of the slake durability apparatus after each cycle were also determined by XRD. The results indicate that the type and amount of clay minerals are the main factors influencing the variations of the slake durability index in all samples. The durability of the clay-bearing rocks studied correlates best with the amount of expandable clay minerals. A strong relationship between the uniaxial compressive strength and the fourth-cycle slake durability index is found only for the marls among the rock types studied. Assessment of gradation results of the spoil pile materials consisting of clay-bearing rocks also reveals that the increase in percentage of fines in old piles is indicative of material degradation, as is evident by multiple-cycle slaking. It is emphasized that two-cycle conventional slake durability testing did not appear to offer an acceptable indication of the durability of weak and clay-bearing rocks when compared with multiple-cyclic wetting and drying. Comments on the performance of the test are made that aim to make the testing process and interpretation of the results more reliable.     

Comparison of three artificial neural network approaches for estimating of slake durability index

Slake durability index (I _d2) is an important engineering parameter to assess the resistance of clay-bearing and weak rocks to erosion and degradation. Standard test sample preparation for slake durability test is difficult for some rock types and the test is time-consuming. The paper reports an attempt to define I _d2 using other parameters that are simpler to obtain. In this study, three different artificial neural network approaches, namely feed-forward back propagation (FFBP), radial basis function based neural network (RBNN), and generalized regression neural networks (GRNN) were used for estimating I _d2. The determination coefficient (R ²), root mean square error and mean absolute relative error statistics were used as evaluation criteria of the FFBP, RBNN, and GRNN models. The experimental results were compared with these models. The comparison results indicate that the GRNN models are superior to the FFBP and RBNN models in modeling of the slake durability index (I _d2).     

Assessment of Durability and Weathering State of Some Igneous and Metamorphic Rocks Using Micropetrographic Index and Rock Durability Indicators: A Case Study

Weathering and durability are the key factors of the rock in the suitability and usefulness of different construction materials, building materials and engineering structures. A single test never predicts the entire factor for suitability of rock stone and aggregate in different uses. Thus, variety of physical, mechanical and chemical tests and indices of rocks are widely used to estimate and evaluate the rocks for the suitability of the required purpose. In all the cases, knowledge of durability and weathering properties are the most important along with the strength of the rock. Micropetrographic index and rock durability indicators (dynamic and static) are the one of the best methods to evaluate the rock for weathering and durability. To estimate these indices, variety of tests are performed such as petrographic examination test, point load index, sulfate soundness test, water absorption test, modified aggregate impact value test and test for specific gravity. Slake durability index and impact strength index tests were also performed for correlation with static and dynamic rock durability indicators due to its application and usefulness in the durability and strength of the rock materials. Micropetrographic index was obtained by petrographic examination test and correlated with all the physical and mechanical properties used for find out the durability indicators. The present study is to express the usefulness of these three indices in the classification of weathering and durability classes and estimation of durability indices by slake durability index, impact strength index and micropetrographic index.     

The effect of pH and salty solutions on durability of sandstones of the Aghajari Formation in Khouzestan province,southwest of Iran

Slake durability of rocks is a vital engineering geological property of rock materials that has an important role in the promotion of slope stability, evaluation of rock materials, as well as the estimation of stone degradability. This parameter is related to such factors as mineralogy, physical characteristics of rocks, and the environmental conditions. A major part of south and southwestern Iran embraces the Aghajari Formation whose sandstones are used to form the foundations for some structures in Khouzestan province where they are being extensively used as aggregate(s). In this paper, tests of mineralogical examinations, physical properties, and slake durability of sandstones from Ahwaz and Haftkel anticlines were administered in order to assess the durability and degradability of these rocks. The rock durability of each type was evaluated to be up to 15 cycles in acidic and alkaline watery environments and salt aqueous solutions. The results showed that durability of sandstones under study is related to their petrographical and physical characteristics. The durability index of the sandstone samples was decreased by pH reduction and by increasing the concentration of aqueous solutions. The durability index of Ahwaz samples, too, was decreased in basic solutions while this index increased for Haftkel sandstones with the increase in the pH of solutions. Also, the obtained results illustrated that durability index decreased with the increase in the number of cycles. Based on the results, all sandstones show that the durability in sodium sulfate solution is much lower than that for the sodium chloride. Moreover, the slake durability index of the sandstones is decreased with an increase in the concentration of aqueous solutions. As a final statement, the Ahwaz sandstones show lower resistance to weathering processes than Haftkel sandstones; therefore, the use of rocks as building stones is not recommended here.     

Modeling of tensile strength of rocks materials based on support vector machines approaches

In the predicting of geological variables, artificial neural networks (ANNs) have some drawbacks including possibility of getting trapped in local minima, over training, subjectivity in the determining of model parameters and the components of its complex structure. Recently, support vector machines (SVM) has been found to be popular in prediction studies due to its some advantages over ANNs. Because the least squares SVM (LS‐SVM) provides a computational advantage over SVM by converting quadratic optimization problem into a system of linear equations, LS‐SVM method is also tried in study. The main purpose of this study is to examine the capability of these two SVM algorithms for the prediction of tensile strength of rock materials and to compare its performance with ANN and linear regression (MLR) models. Total porosity, sonic velocity, slake durability index and aggregate impact value were used as input in modeling applications. Favorite performance evaluation measures were employed to assess developed models. The results determined in study indicate that the SVM, LS‐SVM and ANN methods are successful tools for prediction of tensile strength variable and can give good prediction performances than MLR model. Although these three methods are powerful artificial intelligence techniques, LS‐SVM makes the running time considerably faster with the higher accuracy. In terms of accuracy, the LS‐SVM model resulted in error reductions relative to that of the other models. Copyright © 2012 John Wiley & Sons, Ltd.     

An Empirical Correlation of Index Geomechanical Parameters with the Compressional Wave Velocity

The geomechanical strength of rockmass plays a key role in planning and design of mining and civil construction projects. Determination of geomechanical properties in the field as well as laboratory is time consuming, tedious and a costly affair. In this study, density, slake durability index, uniaxial compressive strength (UCS) and P-wave velocity tests were conducted on four igneous, six sedimentary and three metamorphic rock varieties. These properties are crucial and used extensively in geotechnical engineering to understand the stability of the structures. The main aim of this study is to determine the various mechanical properties of 13 different rock types in the laboratory and establish a possible and acceptable correlation with P-wave velocity which can be determined in the field as well as laboratory with ease and accuracy. Empirical equations were developed to calculate the density, slake durability index and UCS from P-wave velocities. Strong correlations among P-wave velocity with the physical properties of different rock were established. The relations mainly follow a linear trend. Student’s ‘t’ test and ‘F’ test were performed to ensure proper analysis and validation of the proposed correlations. These correlations can save time and reduce cost during design and planning process as they represent a reliable engineering tool.     

The influence of carbonate textures and rock composition on durability cycles and geomechanical aspects of carbonate rocks

Acta Geotechnica - One of the most important geotechnical parameters in studying the engineering behavior of a rock mass is slake durability. The major goal of this research is to test how a series...     

Slake durability and mineralogical properties of some pyroclastic and sedimentary rocks

Slake durability of rocks is an important property of rock-mass and rock-materials in geotechnical practice. The slake durability of rocks is closely related to their mineralogical composition. In this paper, mineralogical examinations and slake durability tests for argillaceous clastic rocks, especially pyroclastic rocks, sandstones and mudstones of Neogene Tertiary age from Japan, were performed in order to assess the slake durability and rock alteration process of these rocks as well as to understand the relationship between mineralogy and durability.The mineral composition and textural features of the rocks were studied by means of optical microscopy (OM), X-ray diffractometry (XRD), electron microprobe analysis (EPMA), and scanning electron microscopy (SEM). In addition, the slake durability test was carried out by using the standard testing method of ISRM [Int. J. Rock Mech. Min. Sci. 16 (1979) 148] in distilled water and in the aqueous solutions with dissolved electrolytes of NaCl and CaCl₂.The pyroclastic rocks and tuffaceous sandstone, rich in di-octahedral and tri-octahedral Fe smectite, respectively, show distinctively different slaking behaviors. The pyroclastic rocks show relatively high slaking (Id₂=55.5% and Id₁₀=10.5%) than the tuffaceous sandstone (Id₂=94.1% and Id₁₀=87.8%, refer to text for Id₂ and Id₁₀). This difference in the slake durability observed in these rocks is due to the microscopic occurrences of smectite present in the interspaces between the particles (pyroclastic rocks) and zeolite cementing the interspaces (tuffaceous sandstone) as alteration minerals. In addition, the durability results of tuffaceous sandstone show that the slake durability decreases as the degree of weathering increases (weathered material Id₂=88.7% and Id₁₀=65.3%). Furthermore, two mudstones of Miocene and Pliocene ages, having different clay mineral compositions (smectite vs. illite+chlorite), show the lowest and the highest slake durability among the tested clastic rocks. Hard mudstone shows the highest (Id₂=98.1% and Id₁₀=95.5%) while the soft mudstone shows the lowest (Id₂=33.9% and Id₆=0.4%.) slake durability. Thus, the slake durability of pyroclastic and sedimentary rocks is greatly affected by their mineral composition and texture, and is closely related to their alteration history. Slake durability is also affected by the kind of dissolved electrolyte and its concentration in the aqueous solution, providing some useful information for geotechnical practice.     

Durability Characterization of Marls from the Region of Dalmatia,Croatia

Durability is one of the most important engineering properties of weak and clay-bearing rocks. Weathering can induce a rapid change in rock material from initial properties to soil-like properties. The sensitivity of a rock type against weatherability is usually described by a durability parameter, such as the slake durability index. However, marl resistance is not detected satisfactorily by the durability indices by using slake durability test as suggested by ISRM for two wetting–drying cycles. The results of this study are obtained from samples of compact or laminated eocene marls from region of Dalmatia, Croatia. The samples were subjected to 4 cycles of slake durability, point load tests, determination of dry density, determination of carbonate content and absorption of water. The scatter of data suggests that strength probably has no influence on the durability of marls. On the other hand a separate group of marl samples have a second-cycle slake durability index higher than approximately 85%, and the durability of these samples is classified as “medium-high” to “high”, although the visual inspection of samples after testing, suggests that they should have “medium” to “low” durability classification. According to obtained results these samples of marl fulfil the criterions for the durability classification: a carbonate content lower than approximately 65%, a dry density lower than 2.4 Mg/m³, and values of water absorption higher than 5%.     

Application of Generalized Regression Neural Networks in Predicting the Unconfined Compressive Strength of Carbonate Rocks

Measuring unconfined compressive strength (UCS) using standard laboratory tests is a difficult, expensive, and time-consuming task, especially with highly fractured, highly porous, weak rock. This study aims to establish predictive models for the UCS of carbonate rocks formed in various facies and exposed in Tasonu Quarry, northeast Turkey. The objective is to effectively select the explanatory variables from among a subset of the dataset containing total porosity, effective porosity, slake durability index, and P-wave velocity in dry samples and in the solid part of samples. This was based on the adjusted determination coefficient and root-mean-square error values of different linear regression analysis combinations using all possible regression methods. A prediction model for UCS was prepared using generalized regression neural networks (GRNNs). GRNNs were preferred over feed-forward back-propagation algorithm-based neural networks because there is no problem of local minimums in GRNNs. In this study, as a result of all possible regression analyses, alternative combinations involving one, two, and three inputs were used. Through comparison of GRNN performance with that of feed-forward back-propagation algorithm-based neural networks, it is demonstrated that GRNN is a good potential candidate for prediction of the unconfined compressive strength of carbonate rocks. From an examination of other applications of UCS prediction models, it is apparent that the GRNN technique has not been used thus far in this field. This study provides a clear and practical summary of the possible impact of alternative neural network types in UCS prediction.     

Evaluating selected factors affecting the depth of undercutting in rocks subject to differential weathering

One of the most important considerations in designing cut slopes in sub-horizontal, inter-layered, sedimentary rocks subject to differential weathering is predicting the total depth of undercutting. Undercutting-induced rockfalls are a major problem for many roadways in Ohio. The total depth of undercutting was measured for 59 profiles from 18 cut slope sites in Ohio with references to cut faces of hard rock units containing pre-split blast-hole traces. The presence of blast hole traces ensured that the rock face represented the slope surface that was cut during construction. Additionally, initial design plans were used for reference. Step-wise regression was used to determine the geological, geotechnical, and geometrical factors that have the highest influence on the total depth of undercutting. Selected independent factors used in the regression analysis included the vertical distance of the undercut unit from the slope crest, the relative position of the undercut unit from the slope crest, the total thickness of the undercut unit, the spacing of orthogonal joints within the undercut unit, the slake durability index value of the undercutting unit, the initial slope angle, and the age of the road cut. Factors that showed the most significant correlation with the total depth of undercutting were found to be the vertical distance of the undercut unit from the slope crest, the relative position of the undercut unit from the slope crest, the total thickness of the undercut unit, joint spacing within the undercut unit, and the slake durability index of the undercutting unit. The regression analysis resulted in an R² value of 0.61, with the depth of undercutting correlating most strongly with a closer relative position of the undercut unit to the slope crest above it and to a closer spacing of orthogonal joints within the undercut unit. These results can be attributed to the fact that more porous and fractured rock units closer to the slope crest intercept and transport greater amounts of infiltrating groundwater, producing greater depths of undercutting. Also, closely jointed rock units are more permeable, allowing more groundwater seepage, which also leads to greater undercutting.     

A correlation between Schmidt hammer rebound numbers with impact strength index,slake durability index and P-wave velocity

The main objective of this study was to establish statistical relationship between Schmidt hammer rebound numbers with impact strength index (ISI), slake durability index (SDI) and P-wave velocity. These are important properties to characterize a rock mass and are being widely used in geological and geotechnical engineering. Due to its importance, Schmidt hammer rebound number is considered as one of the most important property for the determination of other properties, like ISI, SDI and P-wave velocity. Determination of these properties in the laboratory is time consuming and tedious as well as requiring expertise, whereas Schmidt hammer rebound number can be easily obtained on site which in addition is non-destructive. So, in this study, an attempt has been made to determine these index properties in the laboratory and each index property was correlated with Schmidt hammer rebound values. Empirical equations have been developed to predict ISI, SDI and P-wave velocity using rebound values. It was found that Schmidt hammer rebound number shows linear relation with ISI and SDI, whereas exponential relation with P-wave velocity. To check the sensitivity of empirical relations, Student’s t test was done to verify the correlation between rebound values and other rock index properties.     

Investigation of the effect of aggregate shape and surface roughness on the slake durability index using the fractal dimension approach

Argillaceous rocks cover about one thirds of the earth's surface. The major engineering problems encountered with weak- to medium-strength argillaceous rocks could be slaking, erosion, slope stability, settlement, and reduction in strength. One of the key properties for classifying and determining the behavior of such rocks is the slake durability. The concept of slake durability index (SDI) has been the subject of numerous researches in which a number of factors affecting the numerical value of SDI were investigated. In this regard, this paper approaches the matter by evaluating the effects of overall shape and surface roughness of the testing material on the outcome of slake durability indices.

For the purpose, different types of rocks (marl, clayey limestone, tuff, sandstone, weathered granite) were broken into chunks and were intentionally shaped as angular, subangular, and rounded and tested for slake durability. Before testing the aggregate pieces of each rock type, their surface roughness was determined by using the fractal dimension. Despite the variation of final values of SDI test results (values of I_d), the rounded aggregate groups plot relatively in a narrow range, but a greater scatter was obtained for the angular and subangular aggregate groups. The best results can be obtained when using the well rounded samples having the lowest fractal values. An attempt was made to analytically link the surface roughness with the I_d parameter and an empirical relationship was proposed. A chart for various fractal values of surface roughness to use as a guide for slake durability tests is also proposed. The method proposed herein becomes efficient when well rounded aggregates are not available. In such condition, the approximate fractal value for the surface roughness profile of the testing aggregates could be obtained from the proposed chart and be plugged into the empirical relation to obtain the corrected I_d value. The results presented herein represent the particular rock types used in this study and care should be taken when applying these methods to different type of rocks.     

泥质膨胀岩崩解物粒径分布与膨胀性关系试验研究

崩解性是膨胀岩最基本的特性之一,但在膨胀岩膨胀性的快速判别指标中少见关于崩解性及崩解物的定量指标。因此,有必要开展膨胀岩崩解物粒径分布特征和崩解性与膨胀性之间关系的研究,以对膨胀岩的快速判别进行补充。以浙江台州黑洞、蛇蟠岛和新疆大阪隧洞泥质岩样品为例,进行了干燥饱和吸水率和干燥饱和崩解试验,并分析了干燥崩解物粒径分布特征、耐崩解性指数及其与基于干燥饱和吸水率的膨胀性判别结果之间的关系。结果表明,膨胀岩膨胀性的强弱与其崩解物的最大含量粒组颗粒粒径、有效粒径和耐崩解性指数呈反相关关系。膨胀岩崩解物粒径分布的差异性对其膨胀性具有一定的指示意义。     

红板岩材料隐式本构模型的建立及其应用研究

针对红板岩材料在岩土工程中所表现的大量模糊的和不确定的因素等特点,基于人工神经网络的学习能力,借助于室内岩石力学试验,进行了对该材料的力学本构特性进行了神经网络模拟研究,提出了隐式本构模型的思想和方法,并通过该方法对该岩石的流变试验结果进行学习,获得了以网络权值结构保存的力学特性知识,由此得到了表征红板岩应力应变本构关系的隐式本构模型。应用结果表明,该方法对岩土类材料本构关系的模拟研究具有很好的应用前景。     

Use of evolutionary computing for modelling some complex problems in geotechnical engineering

In this paper, the feasibility of using evolutionary computing for solving some complex problems in geotechnical engineering is investigated. The paper presents a relatively new technique, i.e. evolutionary polynomial regression (EPR), for modelling three practical applications in geotechnical engineering including the settlement of shallow foundations on cohesionless soils, pullout capacity of small ground anchors and ultimate bearing capacity of pile foundations. The prediction results from the proposed EPR models are compared with those obtained from artificial neural network (ANN) models previously developed by the author, as well as some of the most commonly available methods. The results indicate that the proposed EPR models agree well with (or better than) the ANN models and significantly outperform the other existing methods. The advantage of EPR technique over ANNs is that EPR generates transparent and well-structured models in the form of simple and easy-to-use hand calculation formulae that can be readily used by practising engineers.