Application of the adaptive neuro-fuzzy inference system for prediction of soil liquefaction

The determination of liquefaction potential of soils induced by earthquake is a major concern and an essential criterion in the design process of the civil engineering structures. A purely empirical interpretation of the filed case histories relating to liquefaction potential is often not well constrained due to the complication associated with this problem. In this study, an integrated fuzzy neural network model, called Adaptive Neuro-Fuzzy Inference System (ANFIS), is developed for the assessment of liquefaction potential. The model is trained with large databases of liquefaction case histories. Nine parameters such as earthquake magnitude, the water table, the total vertical stress, the effective vertical stress, the depth, the peak acceleration at the ground surface, the cyclic stress ratio, the mean grain size, and the measured cone penetration test tip resistance were used as input parameters. The results revealed that the ANFIS model is a fairly promising approach for the prediction of the soil liquefaction potential and capable of representing the complex relationship between seismic properties of soils and their liquefaction potential.     

Application of the adaptive neuro-fuzzy inference system for prediction of a rock engineering classification system

The rock engineering classification system is based on six parameters defined by Bieniawski [5], who employed parallel sets of linguistic and numerical criteria that were acknowledged to influence the behaviour of rock masses and the stability of rock structures. Consequently, experts frequently relate rock joints and discontinuities as well as ground water conditions in linguistic terms, with rough calculations. Recently, intelligence system approaches such as artificial neural network (ANN) and neuro-fuzzy methods have been used successfully for time series modelling. Using neuro-fuzzy approaches, which enable the information that is stored in trained networks to be expressed in the form of a fuzzy rule base, would help to overcome this issue. This paper presents the results of a study of the application of neuro-fuzzy methods to predict rock mass rating. We note that the proposed weights technique was applied in this process. We show that neuro-fuzzy methods give better predictions than conventional modelling approaches.     

Caspian Sea level prediction using satellite altimetry by artificial neural networks

The demand for accurate predictions of sea level fluctuations in coastal management and ship navigation activities is increasing. To meet such demand, accessible high-quality data and proper modeling process are critically required. This study focuses on developing and validating a neural methodology applicable to the short-term forecast of the Caspian Sea level. The input and output data sets used contain two time series obtained from Topex/Poseidon and Jason-1 satellite altimetry missions from 1993 to 2008. The forecast is performed by multilayer perceptron network, radial basis function, and generalized regression neural networks. Several tests of different artificial neural network (ANN) architectures and learning algorithms are carried out as alternative methods to the conventional models to assess their applicability for estimating Caspian Sea level anomalies. The results derived from the ANN are compared with observed sea level values and with the forecasts calculated by a routine autoregressive moving average (ARMA) model. Different ANNs satisfactorily provide reliable results for the short-term prediction of Caspian Sea level anomalies. The root mean square errors of the differences between observations and predictions from artificial intelligence approaches can be significantly reduced by about 50 % compared with ARMA techniques.     

Earthquake magnitude prediction by adaptive neuro-fuzzy inference system (ANFIS) based on fuzzy C-means algorithm

This paper investigates the prediction of future earthquakes that would occur with magnitude 5.5 or greater using adaptive neuro-fuzzy inference system (ANFIS). For this purpose, the earthquake data between 1950 and 2013 that had been recorded in the region with 2°E longitude and 4°N latitude in Iran has been used. Thereupon, three algorithms including grid partition (GP), subtractive clustering (SC) and fuzzy C-means (FCM) were used to develop models with the structure of ANFIS. Since the earthquake data for the specified region had been reported on different magnitude scales, suitable relationships were determined to convert the magnitude scales into moment magnitude and all records uniformed based on the relationships. The uniform data were used to calculate seismicity indicators, and ANFIS was developed based on considered algorithms. The results showed that ANFIS-FCM with a high accuracy was able to predict earthquake magnitude.     

A comparative study for the estimation of geodetic point velocity by artificial neural networks

Space geodesy era provides velocity information which results in the positioning of geodetic points by considering the time evolution. The geodetic point positions on the Earth’s surface change over time due to plate tectonics, and these changes have to be accounted for geodetic purposes. The velocity field of geodetic network is determined from GPS sessions. Velocities of the new structured geodetic points within the geodetic network are estimated from this velocity field by the interpolation methods. In this study, the utility of Artificial Neural Networks (ANN) widely applied in diverse fields of science is investigated in order to estimate the geodetic point velocities. Back Propagation Artificial Neural Network (BPANN) and Radial Basis Function Neural Network (RBFNN) are used to estimate the geodetic point velocities. In order to evaluate the performance of ANNs, the velocities are also interpolated by Kriging (KRIG) method. The results are compared in terms of the root mean square error (RMSE) over five different geodetic networks. It was concluded that the estimation of geodetic point velocity by BPANN is more effective and accurate than by KRIG when the points to be estimated are more than the points known.     

Comparison of cokriging and adaptive neuro-fuzzy inference system models for suspended sediment load forecasting

The present research was carried out by using artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), cokriging (CK) and ordinary kriging (OK) using the rainfall and streamflow data for suspended sediment load forecasting. For this reason, the time series of daily rainfall (mm), streamflow (m³/s), and suspended sediment load (tons/day) data were used from the Kojor forest watershed near the Caspian Sea between 28 October 2007 and 21 September 2010 (776 days). Root mean square error, efficiency coefficient, mean absolute error, and mean relative error statistics are used for evaluating the accuracy of the ANN, ANFIS, CK, and OK models. In the first part of the study, various combinations of current daily rainfall, streamflow and past daily rainfall, streamflow data are used as inputs to the neural network and neuro-fuzzy computing technique so as to estimate current suspended sediment. Also, the accuracy of the ANN and ANFIS models are compared together in suspended sediment load forecasting. Comparison results reveal that the ANFIS model provided better estimation than the ANN model. In the second part of the study, the ANN and ANFIS models are compared with OK and CK. The comparison results reveal that CK was a better estimation than the OK. The ANFIS and ANN models also provided better estimation than the OK and CK models.     

A study of slope stability prediction using neural networks

The determination of the non-linear behaviour of multivariate dynamic systems often presents a challenging and demanding problem. Slope stability estimation is an engineering problem that involves several parameters. The impact of these parameters on the stability of slopes is investigated through the use of computational tools called neural networks. A number of networks of threshold logic unit were tested, with adjustable weights. The computational method for the training process was a back-propagation learning algorithm. In this paper, the input data for slope stability estimation consist of values of geotechnical and geometrical input parameters. As an output, the network estimates the factor of safety (FS) that can be modelled as a function approximation problem, or the stability status (S) that can be modelled either as a function approximation problem or as a classification model. The performance of the network is measured and the results are compared to those obtained by means of standard analytical methods. Furthermore, the relative importance of the parameters is studied using the method of the partitioning of weights and compared to the results obtained through the use of Index Information Theory.     

Processing of earthquake catalog data of Western Turkey with artificial neural networks and adaptive neuro<Emphasis>-</Emphasis>fuzzy inference system

Turkey is one of several countries frequently facing significant earthquakes because of its geological and tectonic position on earth. Especially, graben systems of Western Turkey occur as a result of seismically quite active tensional tectonics. The prediction of earthquakes has been one of the most important subjects concerning scientists for a long time. Although different methods have already been developed for this task, there is currently no reliable technique for finding the exact time and location of an earthquake epicenter. Recently artificial intelligence (AI) methods have been used for earthquake studies in addition to their successful application in a broad spectrum of data intensive applications from stock market prediction to process control. In this study, earthquake data from one part of Western Turkey (37–39.30° N latitude and 26°–29.30° E longitude) were obtained from 1975 to 2009 with a magnitude greater than M ≥ 3. To test the performance of AI in time series, the monthly earthquake frequencies of Western Turkey were calculated using catalog data from the region and then the obtained data set was evaluated with two neural networks namely as the multilayer perceptron neural networks (MLPNNs) and radial basis function neural networks (RBFNNs) and adaptive neuro-fuzzy inference system (ANFIS). The results show that for monthly earthquake frequency data prediction, the proposed RBFNN provides higher correlation coefficients with real data and smaller error values.     

灰色系统与神经网络组合模型在地下水水位预测中的应用

灰色GM(1,1)模型与人工BP神经网络对于预测非线性数列变化趋势都具有很好的适用性,但同其他预测方法一样也存在各自的局限性。本文采用灰色GM(1,1)模型与人工神经网络相结合的方法,对GM(1,1)模型预测结果进行了修正。以收集到的某地区1996～2006年的地下水水位埋深数据为算例,计算结果表明,经人工神经网络修正后的灰色系统的预测值比原预测值的预测精度有了很大提高。     

基于ANFIS的单桩竖向极限承载力预测模型

人工神经网络已应用在岩土工程的各个方面。针对常用的BP网络的不足之处,建立了基于自适应神经模糊推理系统（ANFIS）的单桩竖向极限承载力预测模型。利用文献中桩的载荷试验数据来训练ANFIS网络,确定了网络参数。研究结果表明,同常用的BP网络相比,ANFIS预测模型具有学习速度快,拟合能力较好,训练结果唯一等优点,该方法是一种有效地预测单桩极限承载力的方法。     

人工神经网络在爆破块度预测中的应用研究

利用人工神经网络模型对爆破块度进行预测,实验结果表明,该方法是完全可行的。通过对实验样本数据进行归一化处理后再对人工神经网络模型进行训练和预测,其预测精度会得到大大提高。     

Forecasting of groundwater level in hard rock region using artificial neural network

In hardrock terrain where seasonal streams are not perennial source of freshwater, increase in ground water exploitation has already resulted here in declining ground water levels and deteriorating its’ quality. The aquifer system has shown signs of depletion and quality contamination. Thus, to secure water for the future, water resource estimation and management has urgently become the need of the hour. In order to manage groundwater resources, it is vital to have a tool to predict the aquifer response for a given stress (abstraction and recharge). Artificial neural network (ANN) has surfaced as a proven and potential methodology to forecast the groundwater levels. In this paper, Feed-Forward Network based ANN model is used as a method to predict the groundwater levels. The models are trained with the inputs collected from field and then used as prediction tool for various scenarios of stress on aquifer. Such predictions help in developing better strategies for sustainable development of groundwater resources.     

Feasibility of using artificial neural networks to forecast groundwater levels in real time

The purpose of this study is to develop the feed-forward back-propagation neural network (FFBPNN) to estimate the groundwater level (GL) of next hour according the current GL and past precipitation depth in the hillslope. The 72-h precipitation depth and the real-time groundwater levels are used as the model output layer determination variables. The output variables, are type 1, the GL, which has been used in many researches, and type 2, the groundwater level fluctuation (GLF), which is the difference between the current-time and the next-time groundwater level. The order of the water level fluctuation is less than that of the groundwater level by about one order of magnitude (ten times). The landslide area at the downstream of Wu-She Reservoir, Nantou County, Taiwan, is adopted as a field test area. Total 328 cases of Sinlaku typhoon were used to establish the prediction model of real-time GL. Another 327 cases of Jangmi typhoon were adopted to illustrate the model application. The result of model application shows that root-mean-square error of type 2 (=0.104 m) is smaller than that of type 1 (=0.408 m). In conclusion, the forecasting method used GLF gives a much better agreement with the measured values than that of GL.     

基于贝叶斯网络的地下水环境系统决策管理研究

应用贝叶斯网络解决地下水环境管理中具有不确定性的多目标决策问题,通过对决策变量氮肥施用量以及灌溉模式的调控,减少水中的硝酸盐含量,达到既能有效改善水环境又不至使农民经济利益受到损害的目标。通过分析具体的地下水环境管理系统中变量间的相互关系,构建描述变量间不确定性关系的贝叶斯网络模型,其中包括表示其依赖关系的有向无环图和表示其具体概率依赖程度的条件概率表。并在多个水环境管理目标均达到最优的前提下进行概率推理,得到决策变量氮肥施用量以及灌溉模式取不同值时目标变量的概率分布情况。最终确定出能使所有目标均达到最优的合理的水环境管理决策:(1)使用喷灌,将氮肥施用量控制在0.01~0.03 kg/m2;(2)使用漫灌,将氮肥施用量控制在0.01~0.02 kg/m2。     

Prediction of ground vibrations resulting from the blasting operations in an open-pit mine by adaptive neuro-fuzzy inference system

The aim of this study is to predict the peak particle velocity (PPV) values from both presently constructed simple regression model and fuzzy-based model. For this purpose, vibrations induced by bench blasting operations were measured in an open-pit mine operated by the most important magnesite producing company (MAS) in Turkey. After gathering the ordered pairs of distance and PPV values, the site-specific parameters were determined using traditional regression method. Also, an attempt has been made to investigate the applicability of a relatively new soft computing method called as the adaptive neuro-fuzzy inference system (ANFIS) to predict PPV. To achieve this objective, data obtained from the blasting measurements were evaluated by constructing an ANFIS-based prediction model. The distance from the blasting site to the monitoring stations and the charge weight per delay were selected as the input parameters of the constructed model, the output parameter being the PPV. Valid for the site, the PPV prediction capability of the constructed ANFIS-based model has proved to be successful in terms of statistical performance indices such as variance account for (VAF), root mean square error (RMSE), standard error of estimation, and correlation between predicted and measured PPV values. Also, using these statistical performance indices, a prediction performance comparison has been made between the presently constructed ANFIS-based model and the classical regression-based prediction method, which has been widely used in the literature. Although the prediction performance of the regression-based model was high, the comparison has indicated that the proposed ANFIS-based model exhibited better prediction performance than the classical regression-based model.     

Evaluation of blast-induced ground vibrations in open-pit mines by using adaptive neuro-fuzzy inference systems

This study addresses the effects of rock characteristics and blasting design parameters on blast-induced vibrations in the Kangal open-pit coal mine, the Tülü open-pit boron mine, the K?rka open-pit boron mine, and the TKI Çan coal mine fields. Distance (m, R) and maximum charge per delay (kg, W), stemming (m, SB), burden (m, B), and S-wave velocities (m/s, Vs) obtained from in situ field measurements have been chosen as input parameters for the adaptive neuro-fuzzy inference system (ANFIS)-based model in order to predict the peak particle velocity values. In the ANFIS model, 521 blasting data sets obtained from four fields have been used (r ² = 0.57–0.81). The coefficient of ANFIS model is higher than those of the empirical equation (r ² = 1). These results show that the ANFIS model to predict PPV values has a considerable advantage when compared with the other prediction models.     

Using of neural networks for the prediction of nitrate groundwater contamination in rural and agricultural areas

As a neural network provides a non-linear function mapping of a set of input variables into the corresponding network output, without the requirement of having to specify the actual mathematical form of the relation between the input and output variables, it has the versatility for modeling a wide range of complex non-linear phenomena. In this study, groundwater contamination by nitrate, the ANNs are applied as a new type of model to estimate the nitrate contamination of the Gaza Strip aquifer. A set of six explanatory variables for 139 sampled wells was used and that have a significant influence were identified by using ANN model. The Multilayer Perceptrons (MLP), Radial Basis Function (RBF), Generalized Regression Neural Network (GRNN), and Linear Networks were used. The best network found to simulate Nitrate was MLP with six input nodes and four hidden nodes. The input variables are: nitrogen load, housing density in 500-m radius area surrounding wells, well depth, screen length, well discharge, and infiltration rate. The best network found had good performance (regression ratio 0.2158, correlation 0.9773, and error 8.4322). Bivariate statistical test also were used and resulting in considerable unexplained variation in nitrate concentration. Based on ANN model, groundwater contamination by nitrate depends not on any single factor but on the combination of them.     

基于自适应神经-模糊推理系统的软土路基沉降预测模型

针对目前软基沉降预测中最常用的生长曲线模型以及人工神经网络模型的不足,提出将自适应神经模糊推理系统(ANFIS)应用于软基沉降预测。ANFIS将专家的模糊推理过程蕴含于神经网络结构中,使神经网络的结点和权值具有明确的物理意义,避免了传统神经网络工作过程的"黑盒"性。同时该系统可以采用最小二乘法和梯度下降法相结合的混合算法,既具有神经网络的自适应性和学习能力,又克服了它的局部极小值等缺点,预测精度也远高于生长曲线模型。最后用工程实例与生长模型和神经网络模型进行了对比,结果表明:ANFIS模型优于这两种模型,特别是在模拟多输入变量、高维数下软基沉降预测问题时有着独特的优势,具有一定的推广应用价值。