Prediction of unconfined compressive strength of carbonate rocks using artificial neural networks

The unconfined compressive strength (UCS) of intact rocks is an important geotechnical parameter for engineering applications. Determining UCS using standard laboratory tests is a difficult, expensive and time consuming task. This is particularly true for thinly bedded, highly fractured, foliated, highly porous and weak rocks. Consequently, prediction models become an attractive alternative for engineering geologists. The objective of study is to select the explanatory variables (predictors) from a subset of mineralogical and index properties of the samples, based on all possible regression technique, and to prepare a prediction model of UCS using artificial neural networks (ANN). As a result of all possible regression, the total porosity and P-wave velocity in the solid part of the sample were determined as the inputs for the Levenberg–Marquardt algorithm based ANN (LM-ANN). The performance of the LM-ANN model was compared with the multiple linear regression (REG) model. When training and testing results of the outputs of the LM-ANN and REG models were examined in terms of the favorite statistical criteria, which are the determination coefficient, adjusted determination coefficient, root mean square error and variance account factor, the results of LM-ANN model were more accurate. In addition to these statistical criteria, the non-parametric Mann–Whitney U test, as an alternative to the Student’s t test, was used for comparing the homogeneities of predicted values. When all the statistics had been investigated, it was seen that the LM-ANN that has been developed, was a successful tool which was capable of UCS prediction.     

Prediction of surface crown pillar stability using artificial neural networks

A relatively novel technique, artificial neural networks (ANN), is used in predicting the stability of crown pillars left over large excavations. Data for the training and verification of the networks were obtained from the literature. Four artificial networks, based on two different architectures, were used. The networks used different numbers of input parameters to predict the stability or failure of crown pillars. Multi‐layer perceptron networks using mine type, dip of orebody, overburden thickness, pillar thickness, pillar length, stope height, backfill height, Rock Mass Rating (RMR) of the host rock and RMR of the orebody showed excellent performance in training and verification. Adding three more variables, namely pillar width, rock density and pillar thickness to width ratio, showed symptoms of over‐learning without degrading performance significantly. Radial basis function networks were capable of predicting crown pillar behaviour on the basis of few input functions. It was shown that mine type, dip and pillar thickness to width ratio can be used for a preliminary estimation of stability. Copyright © 2006 John Wiley & Sons, Ltd.     

Prediction of surface blast patterns in limestone quarries using artificial neural networks

This paper is an application of artificial neural networks (ANNs) in the prediction of the geometry of surface blast patterns in limestone quarries. The built model uses 11 input parameters which affect the design of the pattern. These parameters are: formation dip, blasthole diameter, blasthole inclination, bench height, initiation system, specific gravity of the rock, compressive and tensile strength, Young's modulus, specific energy of the explosive and the average resulting fragmentation size. Detailed data from a previous investigation were used to train and verify the network and predict burden and spacing of a blast. The built model was used to conduct parametric studies to show the effect of blasthole diameter and bench height on pattern geometry.     

Estimation of soil compaction parameters by using statistical analyses and artificial neural networks

This study presents the application of different methods (simple–multiple analysis and artificial neural networks) for the estimation of the compaction parameters (maximum dry unit weight and optimum moisture content) from classification properties of the soils. Compaction parameters can only be defined experimentally by Proctor tests. The data collected from the dams in some areas of Nigde (Turkey) were used for the estimation of soil compaction parameters. Regression analysis and artificial neural network estimation indicated strong correlations (r ² = 0.70–0.95) between the compaction parameters and soil classification properties. It has been shown that the correlation equations obtained as a result of regression analyses are in satisfactory agreement with the test results. It is recommended that the proposed correlations will be useful for a preliminary design of a project where there is a financial limitation and limited time.     

A hierarchical analysis for rock engineering using artificial neural networks

Summary Rock behavior, such as the stability of underground openings, is controlled by many different factors which have varying levels of influence. It is very difficult to identify the relative effect of each factor with traditional methods, such as structural analysis and statistical approaches. This paper introduces a hierarchical analytical method based on the application of neural networks which reveals the different degrees of importance of these factors so as to recognize the key factors. This makes it possible to focus on the key factors and do rock engineering more efficiently. An example is given applying this approach to an underground opening.     

Multi-time-scale analysis of hydrological drought forecasting using support vector regression (SVR) and artificial neural networks (ANN)

In this study, the preprocessing of the gamma test was used to select the appropriate input combination into two models including the support vector regression (SVR) model and artificial neural networks (ANNs) to predict the stream flow drought index (SDI) of different timescales (i.e., 3, 6, 9, 12, and 24 months) in Latian watershed, Iran, which is one of the most important sources of water for the large metropolitan Tehran. The variables used included SDI _t , SDI _t ? 1, SDI _t ? 2, SDI _t ? 3, and SDI _t ? 4 monthly delays. Two variables including SDI _t and SDI _t ? 1 with lower gamma values were identified as the most optimal combination of variables in all drought timescales. The results showed that the gamma test was able to correctly identify the right combination for the forecasting of 6, 9, and 12 months SDI using the ANN model. Also, the gamma test was considered in selecting the appropriate inputs for identifying the values of 9, 12, and 24 months SDI in SVR. The support vector machine approach showed a better efficiency in the forecast of long-term droughts compared to the artificial neural network. In total, among forecasts made for 30 scenarios, the support vector machine model only in scenario 3 of SDI₃, scenario 1 of SDI₆, and scenarios 2 and 3 of SDI₂₄ represented poorer efficiency compared to the artificial neural network (MLP layer), but in other scenarios, the results of SVR had better efficiency.     

Prediction of YoungModulus of coal using artificial neural networks in Qinshui Basin, China

<正>1 Introduction Mechanical properties of coal are most important parameters in controlling fluid storage and flow before and after coal extraction[1-2].Reservoir simulation design and wellbore stability analysis are influenced by elastic and strength character of coal rocks[3].Young’s modulus,and shear modulus are usedwhen deformations in underground mines need to be computed.Thus accurate assessment of     

Prediction of pile settlement using artificial neural networks based on standard penetration test data

In recent years artificial neural networks (ANNs) have been applied to many geotechnical engineering problems with some degree of success. With respect to the design of pile foundations, accurate prediction of pile settlement is necessary to ensure appropriate structural and serviceability performance. In this paper, an ANN model is developed for predicting pile settlement based on standard penetration test (SPT) data. Approximately 1000 data sets, obtained from the published literature, are used to develop the ANN model. In addition, the paper discusses the choice of input and internal network parameters which were examined to obtain the optimum model. Finally, the paper compares the predictions obtained by the ANN with those given by a number of traditional methods. It is demonstrated that the ANN model outperforms the traditional methods and provides accurate pile settlement predictions.     

Prediction of ground subsidence in Samcheok City,Korea using artificial neural networks and GIS

This study shows the construction of a hazard map for presumptive ground subsidence around abandoned underground coal mines (AUCMs) at Samcheok City in Korea using an artificial neural network, with a geographic information system (GIS). To evaluate the factors governing ground subsidence, an image database was constructed from a topographical map, geological map, mining tunnel map, global positioning system (GPS) data, land use map, digital elevation model (DEM) data, and borehole data. An attribute database was also constructed by employing field investigations and reinforcement working reports for the existing ground subsidence areas at the study site. Seven major factors controlling ground subsidence were determined from the probability analysis of the existing ground subsidence area. Depth of drift from the mining tunnel map, DEM and slope gradient obtained from the topographical map, groundwater level and permeability from borehole data, geology and land use. These factors were employed by with artificial neural networks to analyze ground subsidence hazard. Each factor’s weight was determined by the back-propagation training method. Then the ground subsidence hazard indices were calculated using the trained back-propagation weights, and the ground subsidence hazard map was created by GIS. Ground subsidence locations were used to verify results of the ground subsidence hazard map and the verification results showed 96.06% accuracy. The verification results exhibited sufficient agreement between the presumptive hazard map and the existing data on ground subsidence area. An erratum to this article can be found at     

基于人工神经网络的数字识别技术

章介绍了一种利用BP神经网络进行数字识别的方法。首先,对数字进行特征提取,获得采样数据．再对样本数据进行学习和训练,形成良好的网络,然后对与训练数字有所区别的数字进行检测,达到了一定的准确度,表明了该方法在实际应用中具有可行性。本共分为五部分,第一部分对神经网络的基本原理进行了简单介绍,第二部分讲述了反向传播算法的基本原理,第三部分讲述了数字识别的基本原理,第四部分讲述了基于人工神经网络的数字识别的实例,第五部分对上述内容作了简要小结。     

Considerations on strength of intact sedimentary rocks

This study presents the results of laboratory testing of sedimentary rocks under point loading as well as in uniaxial and triaxial compression. From the statistical analysis of the data, different conversion factors relating uniaxial compressive and point loading strength were determined for soft to strong rocks. Additionally, the material constant m_i, an input parameter for the Hoek and Brown failure criterion, was also estimated for different limestone samples by analysing the results from a series of triaxial compression tests under different confining pressures. The uniaxial compressive strength (UCS) of intact rocks, as estimated from the point load index using conversion factors, together with the Hoek–Brown constant m_i, and the Geological Strength Index (GSI) constitute the parameters for the calculation of the strength and deformability of rock masses.     

Prediction of slope stability using multiple linear regression (MLR) and artificial neural network (ANN)

The stability problem of natural slopes, filled slopes, and cut slopes are commonly encountered in Civil Engineering Projects. Predicting the slope stability is an everyday task for geotechnical engineers. In this paper, a study has been done to predict the factor of safety (FOS) of the slopes using multiple linear regression (MLR) and artificial neural network (ANN). A total of 200 cases with different geometric and shear strength parameters were analyzed by using the well-known slope stability methods like Fellenius method, Bishop’s method, Janbu method, and Morgenstern and Price method. The FOS values obtained by these slope stability methods were used to develop the prediction models using MLR and ANN. Further, a few case studies have been done along the Jorabat-Shillong Expressway (NH-40) in India, using the finite element method (FEM). The output values of FEM were compared with the developed prediction models to find the best prediction model and the results were discussed.     

Prediction of rainfall using artificial neural networks for synoptic station of Mashhad: a case study

In this paper, we have utilized ANN (artificial neural network) modeling for the prediction of monthly rainfall in Mashhad synoptic station which is located in Iran. To achieve this black-box model, we have used monthly rainfall data from 1953 to 2003 for this synoptic station. First, the Hurst rescaled range statistical (R/S) analysis is used to evaluate the predictability of the collected data. Then, to extract the rainfall dynamic of this station using ANN modeling, a three-layer feed-forward perceptron network with back propagation algorithm is utilized. Using this ANN structure as a black-box model, we have realized the complex dynamics of rainfall through the past information of the system. The approach employs the gradient decent algorithm to train the network. Trying different parameters, two structures, M₅₃₁ and M₇₄₁, have been selected which give the best estimation performance. The performance statistical analysis of the obtained models shows with the best tuning of the developed monthly prediction model the correlation coefficient (R), root mean square error (RMSE), and mean absolute error (MAE) are 0.93, 0.99, and 6.02 mm, respectively, which confirms the effectiveness of the developed models.     

Prediction of uniaxial compression PFC3D model micro‐properties using artificial neural networks

Artificial neural networks are used to predict the micro‐properties of particle flow code in three dimensions (PFC3D) models needed to reproduce macro‐properties of cylindrical rock samples in uniaxial compression tests. Data for the training and verification of the networks were obtained by running a large number of PFC3D models and observing the resulting macro‐properties. Four artificial networks based on two different architectures were used. The networks used different numbers of input parameters to predict the micro‐properties. Multi‐layer perceptron networks using Young's modulus, Poisson's ratio, uniaxial compressive strength, model particle resolution and the maximum‐to‐minimum particle ratio showed excellent performance in both training and verification. Adding one more variable—namely, minimum particle radius—showed degrading performance. Copyright © 2009 John Wiley & Sons, Ltd.     

逻辑回归和人工神经网络模型在滑坡灾害空间预测中的应用

以三峡坝区到巴东段为研究区,选择坡度、坡向、软弱夹层、水系影响范围和土地利用状况等9项评价指标,分别采用逻辑回归和人工神经网络(ANN)模型,在ArcGIS平台上进行滑坡灾害危险性预测。此外,应用受试者工作特征曲线(receiver operating characteristic curve,ROC曲线)分析方法对两种模型的预测结果进行对比,分析结果表明滑坡危险性预测区划图和实际的滑坡发育情况基本吻合,逻辑回归模型和ANN模型的ROC曲线下面积AUC值分别为0.806和0.799,两种模型的预测结果可以相互验证,且逻辑回归模型的预测精度相对较高。     

Evaluation of dynamic regression and artificial neural networks models for real-time hydrological drought forecasting

In this study, application of a class of stochastic dynamic models and a class of artificial intelligence model is reported for the forecasting of real-time hydrological droughts in the Black River basin in the USA. For this purpose, the Standardized Hydrological Drought Index (SHDI) was adopted in different time scales to represent the hydrological drought index. Six probability distribution functions (PDF) were fitted to the discharge time series to obtain the best fit for SHDI calculation. Then, a dynamic linear spatio-temporal model (DLSTM) and artificial neural network (ANN) were used to forecast SHDI. Although results indicated that both models were able to forecast SHDI in different time scales, the DLSTM was far superior in longer lead times. The DLSTM could forecast SHDI up to 6 months ahead while ANN was only capable of forecasting SHDI up to 2 months ahead appropriately. For short lead times (1–6 months), the DLSTM has performed nearly perfect in test phase and CE oscillates between 0.97 and 0.86 while for ANN modeling, CE is between 0.72 and 0.07. However, the performance of DLSTM and ANN reduced considerably in medium lead times (7–12 months). Overall, the DLSTM is a powerful tool for appropriately forecasting SHDI at short time scales; a major advantage required for drought early warning systems.     

Reservoir parameters determination using artificial neural networks: Ras Fanar field, Gulf of Suez, Egypt

Ras Fanar field is one of the largest oil-bearing carbonate reservoirs in the Gulf of Suez. The field produces from the Middle Miocene Nullipore carbonate reservoir, which consists mainly of algal-rich dolomite and dolomitic limestone rocks, and range in thickness between 400 and 980 ft. All porosity types within the Nullipore rocks have been modified by diagenetic processes such as dolomitization, leaching, and cementation; hence, the difficulty arise in the accurate determination of certain petrophysical parameters, such as porosity and permeability, using logging data only. In this study, artificial neural networks (ANN) are used to estimate and predict the most important petrophysical parameters of Nullipore reservoir based on well logging data and available core plug analyses. The different petrophysical parameters are first calculated from conventional logging and measured core analyses. It is found that pore spaces are uniform all over the reservoirs (17–23%), while hydrocarbon content constitutes more than 55% and represented mainly by oil with little saturations of secondary gasses. A regular regression analysis is carried out over the calculated and measured parameters, especially porosity and permeability. Fair to good correlation (R <65%) is recognized between both types of datasets. A predictive ANN module is applied using a simple forward backpropagation technique using the information gathered from the conventional and measured analyses. The predicted petrophysical parameters are found to be much more accurate if compared with the parameters calculated from conventional logging analyses. The statistics of the predicted parameters relative to the measured data, show lower sum error (<0.17%) and higher correlation coefficient (R >80%) indicating that good matching and correlation is achieved between the measured and predicted parameters. This well-learned artificial neural network can be further applied as a predictive module in other wells in Ras Fanar field where core data are unavailable.     

Estimating compaction parameters of fine- and coarse-grained soils by means of artificial neural networks

This paper presents artificial neural network (ANN) prediction models for estimating the compaction parameters of both coarse- and fine-grained soils. A total number of 200 soil mixtures were prepared and compacted at standard Proctor energy. The compaction parameters were predicted by means of ANN models using different input data sets. The ANN prediction models were developed to find out which of the index properties correlate well with compaction parameters. In this respect, the transition fine content ratio (TFR) was defined as a new input parameter in addition to traditional soil index parameters (i.e. liquid limit, plastic limit, passing No. 4 sieve and passing No. 200 sieve). Highly nonlinear nature of the compaction data dictated development of two separate ANN models for maximum dry unit weight (γ_dmax) and optimum water content (ω_opt). It was found that generalization capability and prediction accuracy of ANN models could be further enhanced by sub-clustered data division techniques.     

Flow estimations for the Sohu Stream using artificial neural networks

In this study, daily rainfall–runoff relationships for Sohu Stream were modelled using an artificial neural network (ANN) method by including the feed-forward back-propagation method. The ANN part was divided into two stages. During the first stage, current flows were estimated by using previously measured flow data. The best network architecture was found to utilise two neurons in the input layer (the delayed flows from the first and second days), two hidden layers, and one output layer (the current flow). The coefficient of determination (R ²) in this architecture was 81.4%. During the second stage, the current flows were estimated by using a combination of previously measured values for precipitation, temperature, and flows. The best architecture consisted of an input layer of 2 days of delayed precipitation, 3 days of delayed flows, and temperature of the current. The R ² in this architecture was calculated to be 85.5%. The results of the second stage best reflected the real-world situation because they accounted for more input variables. In all models, the variables with the highest R ² ranked as the previous flow (81.4%), previous precipitation (21.7%), and temperature.