A case study from Koyulhisar (Sivas-Turkey) for landslide susceptibility mapping by artificial neural networks

A case study for the use of an artificial neural network (ANN) model for landslide susceptibility mapping in Koyulhisar (Sivas-Turkey) is presented. Digital elevation model (DEM) was first constructed using ArcGIS software. Relevant parameter maps were created, including geology, faults, drainage system, topographical elevation, slope angle, slope aspect, topographic wetness index, stream power index, normalized difference vegetation index and distance from roads. Finally, a landslide susceptibility map was constructed using the neural networks. The drawbacks of the method are discussed but as the validation procedures used confirmed the quality of the map produced, it is recommended the use of ANN may be helpful for planners and engineers in the initial assessment of landslide susceptibility.      

Modeling the slake durability index using regression analysis, artificial neural networks and adaptive neuro-fuzzy methods

Clay bearing, weathered and other weak rocks cause major problems in engineering practice due to their interactions with water. The slake durability index (I _d2) is an important tool used to assess the resistance of these rocks to erosion and degradation, but sample preparation for this test is tedious. The paper reports an attempt to define I _d2 through statistical models using other parameters that are simpler to obtain. The main objective of this study was to define the best empirical relationship between the I _d2 and the point load strength index (I _s(50)), dry unit weight (γ _d) and fractal dimension (D) parameters of eight rock types by applying general multiple linear regression (GLM), artificial neural networks (ANN) and adaptive neuro-fuzzy inference systems (ANFIS). The models obtained were evaluated using the R ², MSE, MARE and d parameters. The results indicate that the relationships between I _d2 and γ _d, I _s(50) and D were best obtained using ANN, followed by GLM and ANFIS. It is concluded that ANN modelling is a fast and practical method of establishing I _d2.     

An artificial neural network approach to inhomogeneous soil slope stability predictions based on limit analysis methods

The assessment of soil slope stability is an important task in geotechnical designs. This study uses finite element upper bound (UB) and lower bound (LB) limit analysis (LA) methods to investigate inhomogeneous soil slope stability on the basis of the conventional Mohr–Coulomb parameters. The obtained stability numbers are presented in inhomogeneous soil slope stability charts. In order to minimize manual reading errors when using the chart solutions, an artificial neural network (ANN) is employed to develop a stability assessment tool for the slopes investigated in this paper. The slope stability analysis using the ANN-based tool is convenient, and the predictions it provides are highly accurate.     

Evaluating landslide deposits along the Tsurukawa fault zone, Japan, using magnetic susceptibility

 This paper describes the results of in-situ measurements of magnetic susceptibility in landslide deposits along the Tsurukawa fault zone in central Japan. The measured magnetic susceptibility values range from 0.4 to 9.6×10^–3 (in SI) and correspond to the proportions of the two component materials, weathered volcanic ash and faulted rock fragments. The study shows that landslide deposits along the Tsurukawa fault zone are composed of varying proportions of weathered volcanic ash. The results contrast with some general assumptions concerning landslides along fault zones in Japan. Received: 16 April 1999 · Accepted: 10 August 1999     

混凝土早期弹性模量的预测RBF模型

混凝土的早期弹性模量对施工的进度和工程的可靠度有重要的影响。借鉴RBF神经网络具有很强的非线性映射功能,在测定混凝土早期强度的基础上利用RBF神经网络对其弹性模量进行预测。讨论了RBF神经网络的拓扑结构和修正算法。通过对检验结果进行分析比较,证明了利用RBF网络能对混凝土早期的弹性模量进行精确的预测。     

Comparative analysis of artificial neural networks and evolutionary programming in roof deterioration modeling

Modeling of roof performance and deterioration has been done in the past years by means of regression and condition rating as part of the management of civil infrastructure. However, in recent years artificial neural networks (ANNs) have also been used to model the performance of civil infrastructure systems such as pavements. The evolutionary algorithm (EA) method is one the most current methods that can also be used to predict the performance of infrastructure systems. This paper presents the comparative analyses of ANNs and EA methods in predicting and modeling the performance of a roofing system.     

Fire analysis of steel frames with the use of artificial neural networks

The paper presents an alternative approach to the modelling of the mechanical behaviour of steel frame material when exposed to the high temperatures expected in fires. Based on a series of stress-strain curves obtained experimentally for various temperature levels, an artificial neural network (ANN) is employed in the material modelling of steel. Geometrically and materially, a non-linear analysis of plane frame structures subjected to fire is performed by FEM. The numerical results of a simply supported beam are compared with our measurements, and show a good agreement, although the temperature-displacement curves exhibit rather irregular shapes. It can be concluded that ANN is an efficient tool for modelling the material properties of steel frames in fire engineering design studies.     

Comparison of construction cost estimating models based on regression analysis,neural networks,and case-based reasoning

Adequate estimation of construction costs is a key factor in construction projects. This paper examines the performance of three cost estimation models. The examinations are based on multiple regression analysis (MRA), neural networks (NNs), and case-based reasoning (CBR) of the data of 530 historical costs. Although the best NN estimating model gave more accurate estimating results than either the MRA or the CBR estimating models, the CBR estimating model performed better than the NN estimating model with respect to long-term use, available information from result, and time versus accuracy tradeoffs.     

Mapping rework causes and effects using artificial neural networks

Rework can have adverse effects on the performance and productivity of construction projects. Techniques such as artificial neural networks (ANN) are widely used for prediction and classification problems and thus can be used to map the causes and effects of rework. The traditional back propagation neural network and general regression neural network data from 112 Hong Kong construction projects are used to examine the influence of rework causes on the various project performance indicators such as cost overrun, time overrun, and contractual claims. The results from this research could be used to develop forecasting systems and appropriate intelligent decision support frameworks for enhancing performance in construction projects. Furthermore, analysis of the neural network results indicates that the general regression neural network architecture is better suited for modelling rework causes and their impacts on project performance.

Les travaux de reprise peuvent avoir des effets néfastes sur les performances et la productivité dans les projets de construction. Des techniques comme les réseaux neuronaux artificiels (AAN) sont largement utilisés pour résoudre les problèmes de prévision et de classification et peuvent donc servir à cartographier les causes et les effets des travaux de reprise. Les données de réseaux neuronaux classiques à rétropropagation et les données de réseaux neuronaux à régression générale provenant de 112 projets de construction à Hong Kong sont utilisées pour examiner l'influence des travaux de reprise sur les divers indicateurs de performances de projets, comme les dépassements de coûts, les dépassements de délais et les réclamations contractuelles. Les résultats de cette recherche pourraient servir à développer des systèmes de prévision et des cadres appropriés et intelligents de soutien à la décision pour améliorer les performances de projets de construction. En outre, l'analyse des résultats de réseaux neuronaux indique que l'architecture du réseau neuronal à régression générale convient mieux à la modélisation des causes des travaux de reprise et à leurs conséquences sur les performances des projets.

Mots cle´s: projet de construction, dépassement des coûts, productivité, performances de projets, travaux de reprise, dépassement des délais     

基于灰色关联分析的神经网络方法在边坡稳定分析中的应用

采用灰色关联度对影响边坡稳定性的各因素进行分析,将得到的主要因素作为BP网络的输人参数对边坡稳定性进行预测,并用L-M方法对BP网络进行改进,通过与传统的分析方法比较,证明该方法能够满足工程要求,方便可行.     

基于遗传神经网络的公路工程造价估算分析

从影响公路造价的影响因素中提取特征因子为参数,用历史数据为依据,用遗传神经网络建立了公路造价快速估测模型,最后用实例验证了遗传神经网络模型在公路工程造价估测中的优良效果。     

基于BP人工神经网络的抗剪强度参数值反演

利用F-SPW5软件和BP人工神经网络工具建立了新疆碎石类土深基坑开挖位移反分析的专家系统计算模型,并将其反演得到的参数用于与随机变量统计法得到的c,φ的取值范围进行验证,确认了研究成果的可靠性。     

基于Hopfield神经网络的结构优化分析

探讨了Hopfield网络在土木工程优化中应用的途径，并简要介绍了与Hopfield网络密切相关的模拟退火技术，介绍了用于工程结构优化的连续型Hopfield网络的能量函数，该能量函数能够准确地描述结构优化问题。同时，探讨了Hopfield网络用于工程结构优化的机理。然后，研究了基于CHNN的工程结构优化在Matlab上的实践，探讨了网络参数的选取。最后，通过算例表明方法的有效性。     

基于人工神经网络的混凝土实时强度影响因素敏感性分析

准确地预测出混凝土材料在使用过程的实时强度对于正确评估结构安全性有着重要的意义。影响混凝土材料实时强度的主要因素包括龄期、环境类别、水灰比、胶凝材料用量等等。采用人工神经网络进行混凝土实时强度影响因素敏感性分析。首先,对影响混凝土实时强度的各类因素进行分析,确定敏感性因素。其次,针对龄期敏感性因素,建立两个神经网络,一个神经网络的输入变量包含龄期,另一个不包含龄期,将训练好的两个神经网络用同组数据进行测试,比较两组测试结果,以此来确定龄期因素对混凝土强度的敏感性。采用上述方法分别对环境类别、水灰比、胶凝材料用量等因素进行敏感性分析。最后,通过比较确定混凝土龄期、环境类别、水灰比为影响混凝土实时强度的敏感性因素。     

Optimizing overbreak prediction based on geological parameters comparing multiple regression analysis and artificial neural network

Underground mining becomes more efficient due to the technological advancements of drilling and blasting methods and the developing of highly productive mining methods that facilitate easier access to ore. In the perspective of maximizing productivity in underground mining by drilling and blasting methods, overbreak control is an essential component. The causing factors of overbreak can simply divided as blasting and geological parameters and all of the factors are nonlinearly correlated. In this paper, the blasting design of the tunnel was fixed as the standard blasting pattern and the research focus on effects of geological parameters to the overbreak phenomenon. 49 sets of rock mass rating (RMR) and overbreak data were applied to linear and nonlinear multiple regression analysis (LMRA and NMRA) and artificial neural network (ANN) to predict overbreak as input and output parameters, respectively. The performance of LMRA, NMRA, and optimized ANN models was evaluated by comparing coefficient correlations (R²) and their values are 0.694, 0.704 and 0.945, respectively, which means that the relatively high level of accuracy of the optimized ANN in comparison with LMRA and NMRA. The developed optimum overbreak predicting ANN model is suitable for establishing an overbreak warning and preventing system and it will utilize as a foundation reference for a practical drift blasting reconciliation at mines for operation improvements.     

基于概率神经网络的离心式制冷机故障诊断

使用概率神经网络(PNN)对制冷系统7种常见故障进行诊断,包括系统故障和局部故障。详细介绍了应用PNN建立故障诊断模型以及平滑因子寻优过程,并探索了样本规模对最佳平滑因子和诊断正确率的影响。将PNN与人工神经网络中最常用的误差反向传播(BP)神经网络进行比较,结果表明,PNN网络的诊断正确率比BP网络诊断正确率高3.48%,且诊断耗时更短,并且PNN网络的单次训练结果更可靠。尽管2种网络的训练结果均显示系统故障比局部故障更难以被识别,但使用PNN网络进行诊断时,系统故障的诊断正确率明显高于BP网络的诊断正确率。     

Benefit of splines and neural networks in simulation based structural reliability analysis

Simulation based algorithms are often used to calculate an accurate value for the system reliability of complex systems. These concepts are very appealing because of their inherent simplicity. This is even more emphasized when used in combination with an implicit limit state function, for which the outcome is calculated by means of a finite element analysis. One of the major disadvantages however is the large number of simulations required to obtain an accurate estimate of the failure probability. This might result in an unrealistic processing time, making the method unusable for practical purposes. To meet this disadvantage, reliability analysis based on simulation methods in combination with an adaptive low order polynomial response surface are often used. The applicability has been demonstrated extensively. Ideally, no functional form is preset. The objective of this paper is to further increase the efficiency of simulation based reliability algorithms. Therefore the low order polynomial response surfaces are extended using neural networks and splines. The reliability framework is presented, compared with traditional response surface methods and commented extensively. The overall behaviour of the technique is addressed referring to several benchmark examples.     

GIS based statistical and physical approaches to landslide susceptibility mapping (Sebinkarahisar,Turkey)

The case study presents GIS-aided statistically and physically based landslide susceptibility mapping in the landslide-prone Avutmus district of Sebinkarahisar (Giresun, Turkey). Field investigations, analysis of geological data and laboratory tests suggested that two important factors have acted together to cause sliding: ground water pressures and toe erosion. Frequency ratio (FR) and stability index mapping (SINMAP) were used to create the landslide susceptibility maps based on a landslide inventory; distance from drainage systems, faults and roads; slope angle and aspect; topographic elevation and topographical wetness index; and vegetation cover. Validation of the models indicated high quality susceptibility maps with the more realistic results were obtained from the statistically based FR model.     

人工神经网络在结构工程中的应用和发展

依据神经网络原理及其自身的特点，对其应用在结构优化设计、结构分析及可靠度分析等方面进行了综述和研究，并在此基础上分析了神经网络在结构工程中的研究方向。