Neural networks and landslide susceptibility: a case study of the urban area of Potenza

For those working in the field of landslide prevention, the estimation of hazard levels and the consequent production of thematic maps are principal objectives. They are achieved through careful analytical studies of the characteristics of landslide prone areas, thus, providing useful information regarding possible future phenomena. Such maps represent a fundamental step in the drawing up of adequate measures of landslide hazard mitigation. However, for a complete estimation of landslide hazard, meant as the degree of probability that a landslide occurs in a given area, within a given space of time, detailed and uniformly distributed data regarding their incidence and causes are required. This information, while obtainable through laborious historical research, is usually partial, incomplete and uneven, and hence, unsatisfactory for zoning on a regional scale. In order to carry this out effectively, the utilization of spatial estimation of the relative levels of landslide hazard in the various areas was considered opportune. These areas were classified according to their levels of proneness to landslide activity without taking recurrence periods into account. Various techniques were developed in order to obtain upheaval numerical estimates. The method used in this study, which was applied in the area of Potenza, is based on techniques derived from artificial intelligence (Artificial Neural Network—ANN). This method requires the definition of appropriate thematic layers, which parameterize the area under study. These are recognized by means of specific analyses in a functional relationship to the event itself. The parameters adopted are: slope gradient, slope aspect, topographical index, topographical shape, elevation, land use and lithology.     

Geoacoustic inversion using adaptive neuro-fuzzy inference system

The geoacoustic parameters form significant input for underwater acoustic propagation studies and geoacoustic modeling. Conventional inversion techniques commonly used as indirect approach for extraction of geoacoustic parameters from acoustic or seismic data are computationally intensive and time-consuming. In the present study, we have tried to exploit the advantage of soft computing techniques like, reasoning ability of fuzzy logic and learning abilities of neural networks, in inversion studies. The network model based on the combined approach called adaptive neuro-fuzzy inference system (ANFIS), is found to be very promising in inversion of the acoustic data. The network model once built is capable of invert a few thousand data sets instantaneously, to a reasonably good accuracy. In the case of conventional approaches, repetition of the entire inversion process with each new data set is required. A limited number of sensor’s data are sufficient for simulation of the network model and provides an advantage to use short hydrophone array data. Inversion results of a few hundred test data sets, representing different geoacoustic environments, show the prediction error is much less than 0.01 g/cc, 10 m/s, 10 m and 0.1 against first layer’s density, compressional sound speed, thickness and attenuation respectively for a three-layer geoacoustic model. However, the error is relatively large for the second- and third-layer parameters, which need to be improved. The model is efficient, robust and inexpensive.     

基于卷积神经网络的地震预警震级持续估算方法研究

为提高地震预警震级快速持续估算结果的准确性,本文构建了基于多种地震动特征参数的卷积神经网络震级估算CNN-M模型.该模型基于日本KiK-net强震动观测记录,利用其P波触发后3～ 10s时间窗内的幅值参数、周期参数、烈度参数、信噪比参数共11种地震波特征参数以及震中距参数作为输入.本文所建立的CNN-M模型随着地震发生...     

基于网格搜索法优化支持向量机的围岩稳定性分类模型

为科学评价围岩稳定性,本次研究借助支持向量机(SVM)处理小样本、非线性问题能力强的特性,对围岩的稳定性进行了分类。选取16组围岩数据作为学习样本,以岩石质量指标、岩石单轴饱和抗压强度、完整性系数、结构面强度系数和地下水渗水量5个指标作为模型输入,围岩稳定程度为模型输出,建立了基于支持向量机的围岩稳定性分类模型。为增强模型的推广性能,提高其预测准确率,运用改进的网格搜索方法(GSM)寻找最优的支持向量机参数,并对8组围岩数据进行预测,并同BP神经网络模型的预测结果进行对比。结果表明,建立的GSM-SVM模型对预测样本的评判结果与实际结果一致,其预测精度较BP神经网络有很大的提升。     

Evaluation of total load sediment transport formulas using ANN

The calculated results from various sediment transport formulas often differ from each other and from measured data. Some parameters in the sediment transport formulas are more effective than others to estimate total sediment load. In this study, an Artificial Neural Network （ANN） model is trained using four dominant parameters of sediment transport formulas. ANN models are able to reveal hidden laws of natural phenomena such as sediment transport process. The results of ANN and some total bed material load sediment transport formulas have been compared to indicate the importance of variables which can be used in developing sediment transport formulas. To train ANN, average flow velocity, water surface slopes, average flow depth, and median particle diameter are used as dominant parameters to estimate total bed material load. Two hundreds and fifty samples are used to train the ANN model. Twenty-four sets of field data not used in the training nor calibration of ANN are used to compare or verify the accuracy of ANN and some well-known total bed material load formulas. The test results show that the ANN model developed in this study using minimum number of dominant factors is a reliable and uncomplicated method to predict total sediment transport rate or total bed material load transport rate. Results show that the accuracy of formulas in descending order are those by Yang （1973）, Laursen （1958）, Engelund and Hansen （1972）, Ackers and White （1973）, and Toffaleti （1969）. These results are similar to those made by ASCE （1982） based on laboratory and field data not used in this paper. Study results also show that the formulas based on physical laws of sediment transport, like those formulas that were developed based on power concept, are more accurate than other formulas for estimating total bed material sediment load in rivers.     

Modeling River Stage‐Discharge Relationships Using Different Neural Network Computing Techniques

One of the most important problems in hydrology is the establishment of rating curves. The statistical tools that are commonly used for river stage‐discharge relationships are regression and curve fitting. However, these techniques are not adequate in view of the complexity of the problems involved. Three different neural network techniques, i. e., multi‐layer perceptron neural network with Levenberg‐Marquardt and quasi‐Newton algorithms and radial basis neural networks, are used for the development of river stage‐discharge relationships by constructing nonlinear relationships between stage and discharge. Daily stage and flow data from three stations, Yamula, Tuzkoy and Sogutluhan, on the Kizilirmak River in Turkey were used. Regression techniques are also applied to the same data. Different input combinations including the previous stages and discharges are used. The models' results are compared using three criteria, i. e., root mean square errors, mean absolute error and the determination coefficient. The results of the comparison reveal that the neural network techniques are much more suitable for setting up stage‐discharge relationships than the regression techniques. Among the neural network methods, the radial basis neural network is found to be slightly better than the others.     

Prediction of Density Flow Plunging Depth in Dam Reservoirs: An Artificial Neural Network Approach

Experimental findings and observations indicate that plunging flow is related to the formation of bed load deposition in dam reservoirs. The sediment delta begins to form in the plunging region where the inflow river water meets the ambient reservoir water. Correct estimation of dam reservoir flow, plunging point, and plunging depth is crucial for dam reservoir sedimentation and water quality issues. In this study, artificial neural network (ANN), multi‐linear regression (MLR), and two‐dimensional hydrodynamic model approaches are used for modeling the plunging point and depth. A multi layer perceptron (MLP) is used as the ANN structure. A two‐dimensional model is adapted to simulate density plunging flow through a reservoir with a sloping bottom. In the model, nonlinear and unsteady continuity, momentum, energy, and k–ε turbulence equations are formulated in the Cartesian coordinates. Density flow parameters such as velocity, plunging points, and plunging depths are determined from the simulation and model results, and these are compared with previous experimental and model works. The results show that the ANN model forecasts are much closer to the experimental data than the MLR and mathematical model forecasts.     

Geomorphometric feature analysis using morphometric parameterization and artificial neural networks

This paper presents a semi-automatic method using an unsupervised neural network to analyze geomorphometric features as landform elements. The Shuttle Radar Topography Mission (SRTM) provided detailed digital elevation models (DEMs) for all land masses between 60°N and 57°S. Exploiting these data for recognition and extraction of geomorphometric features is a challenging task. Results obtained with two methods, Wood's morphometric parameterization and the Self Organizing Map (SOM), are presented in this paper.Four morphometric parameters (slope, minimum curvature, maximum curvature and cross-sectional curvature) were derived by fitting a bivariate quadratic surface with a window size of 5 by 5 to the SRTM DEM. These parameters were then used as input to the two methods. Wood's morphometric parameterization provides point-based features (peak, pit and pass), line-based features (channel and ridge) and area-based features (planar). Since point-based features are defined as having a very small slope when their neighbors are considered, two tolerance values (slope tolerance and curvature tolerance) are introduced. Selection of suitable values for the tolerance parameters is crucial for obtaining useful results.The SOM as an unsupervised neural network algorithm is employed for the classification of the same morphometric parameters into ten classes characterized by morphometric position (crest, channel, ridge and plan area) subdivided by slope ranges. These terrain features are generic landform element and can be used to improve mapping and modeling of soils, vegetation, and land use, as well as ecological, hydrological and geomorphological features. These landform elements are the smallest homogeneous divisions of the land surface at the given resolution. The result showed that the SOM is an efficient scalable tool for analyzing geomorphometric features as meaningful landform elements, and uses the full potential of morphometric characteristics.     

An elliptical basis function network for classification of remote sensing images

An elliptical basis function (EBF) network is employed in this study for the classification of remotely sensed images. Though similar in structure, the EBF network differs from the well-known radial basis function (RBF) network by incorporating full covariance matrices and employing the expectation-maximization (EM) algorithm to estimate the basis functions. Since remotely sensed data often take on mixture-density distributions in the feature space, the network not only possesses the advantage of the RBF mechanism, but also utilizes the EM algorithm to compute the maximum likelihood estimates of the mean vectors and covariance matrices of a Gaussian mixture distribution in the training phase. Experimental results show that the EM-based EBF network is more effective in training and simpler in structure than an RBF network constructed for the same task.The research was supported by grant 40101021 from the Natural Science Foundation of China, and grant 2002AA135230 from Hi-Tech research and development program of China. The authors would like to thank the reviewers for their valuable comments.