Functional networks in real-time flood forecasting—a novel application

Functional networks were recently introduced as an extension of artificial neural networks (ANNs). Unlike ANNs, they estimate unknown neuron functions from given functional families during the training process. Here, we applied two types of functional network models, separable and associativity functional networks, to forecast river flows for different lead-times. We compared them with a conventional artificial neural network model, an ARMA model and a simple baseline model in three catchments. Results show that functional networks are flexible and comparable in performance to artificial neural networks. In addition, they are easier and quicker to train and so are useful tools as an alternative to artificial neural networks. These results were obtained with only the simplest structures of functional networks and it is possible that a more detailed study with more complex forms of the model will improve even further on these results. Thus we recommend that the use of functional networks in discharge time series modelling and forecasting should be further investigated.     

Application of artificial neural networks in optimal tuning of tuned mass dampers implemented in high-rise buildings subjected to wind load

High-rise buildings are usually considered as flexible structures with low inherent damping. Therefore, these kinds of buildings are susceptible to wind-induced vibration. Tuned Mass Damper (TMD) can be used as an effective device to mitigate excessive vibrations. In this study, Artificial Neural Networks is used to find optimal mechanical properties of TMD for high-rise buildings subjected to wind load. The patterns obtained from structural analysis of different multi degree of freedom (MDF) systems are used for training neural networks. In order to obtain these patterns, structural models of some systems with 10 to 80 degrees-of-freedoms are built in MATLAB/SIMULINK program. Finally, the optimal properties of TMD are determined based on the objective of maximum displacement response reduction. The Auto-Regressive model is used to simulate the wind load. In this way, the uncertainties related to wind loading can be taken into account in neural network’s outputs. After training the neural network, it becomes possible to set the frequency and TMD mass ratio as inputs and get the optimal TMD frequency and damping ratio as outputs. As a case study, a benchmark 76-story office building is considered and the presented procedure is used to obtain optimal characteristics of the TMD for the building.     

Estimation of river flow by artificial neural networks and identification of input vectors susceptible to producing unreliable flow estimates

Reliable river flow estimates are crucial for appropriate water resources planning and management. River flow forecasting can be conducted by conceptual or physical models, or data-driven black box models. Development of physically-based models requires an understanding of all the physical processes which impact a natural process and the interactions among them. Since identification of the relationships among these physical processes is very difficult, data-driven approaches have recently been utilized in hydrological modeling. Artificial neural networks are one of the widely used data-driven approaches for modeling hydrological processes. In this study, estimation of future monthly river flows for Guvenc River, Ankara is conducted using various artificial neural network models. Success of artificial neural network models relies on the availability of adequate data sets. A direct mapping from inputs to outputs without consideration of the complex relationships among the dependent and independent variables of the hydrological process is identified. In this study, past precipitation, river flow data, and the associated month are used to predict future river flows for Guvenc River. Impacts of various input patterns, number of training cycles, and initial values assigned to the weights of the connections are investigated. One of the major weaknesses of artificial neural networks is that they may fail to generate good estimates for extreme events, i.e. events that do not occur at all or often enough in the training data set. It is very important to be able to identify such unlikely events. A fuzzy c-means algorithm is used in this study to cluster the training and validation input vectors into regular and extreme events so that the user will have an idea about the risk of the artificial neural network model to generate unreliable results.     

Nonlinear inversion of electrical resistivity imaging using pruning Bayesian neural networks

Conventional artificial neural networks used to solve electrical resistivity imaging (ERI) inversion problem suffer from overfitting and local minima. To solve these problems, we propose to use a pruning Bayesian neural network (PBNN) nonlinear inversion method and a sample design method based on the K-medoids clustering algorithm. In the sample design method, the training samples of the neural network are designed according to the prior information provided by the K-medoids clustering results; thus, the training process of the neural network is well guided. The proposed PBNN, based on Bayesian regularization, is used to select the hidden layer structure by assessing the effect of each hidden neuron to the inversion results. Then, the hyperparameter α _k , which is based on the generalized mean, is chosen to guide the pruning process according to the prior distribution of the training samples under the small-sample condition. The proposed algorithm is more efficient than other common adaptive regularization methods in geophysics. The inversion of synthetic data and field data suggests that the proposed method suppresses the noise in the neural network training stage and enhances the generalization. The inversion results with the proposed method are better than those of the BPNN, RBFNN, and RRBFNN inversion methods as well as the conventional least squares inversion.     

Long‐lead seasonal rainfall forecasting using time‐delay recurrent neural networks: a case study

A temporal artificial neural network‐based model is developed and applied for long‐lead rainfall forecasting. Tapped delay lines and recurrent connections are two different components that are used along with a static multilayer perceptron network to design a time‐delay recurrent neural network. The proposed model is, in fact, a combination of time‐delay and recurrent neural networks. The model is applied in three case studies of the Northwest, West, and Southwest basins of Iran. In addition, an autoregressive moving average with exogenous inputs (ARMAX) model is used as a baseline in order to be compared with the time‐delay recurrent neural networks developed in this study. Large‐scale climate signals, such as sea‐level pressure, that affect the rainfall of the study area are used as the predictors in the models, as well as the persistence between rainfall data. The results of winter‐spring rainfall forecasts are discussed thoroughly. It is demonstrated that in all cases the proposed neural network results in better forecasts in comparison with the statistical ARMAX model. Moreover, it is found that in two of three case studies the time‐delay recurrent neural networks perform better than either recurrent or time‐delay neural networks. The results demonstrate that the proposed method can significantly improve the long‐lead forecast by utilizing a non‐linear relationship between climatic predictors and rainfall in a region. Copyright © 2007 John Wiley & Sons, Ltd.     

Application of neural networks in natural periods identification and simulation of prefabricated buildings response

The paper deals with an application of neural networks for detection of natural periods of vibrations of prefabricated, medium height buildings. The neural network technique is also used to simulate the dynamic response at selected floor of one of the analysed buildings subject to seismic loading induced by explosives in a nearby quarry. Both the training and testing patterns were formulated on the basis of measurements performed on actual structures. The results of neural network identification of natural periods of the considered buildings obtained with different soil, geometrical and stiffness parameters are compared with the results of experiments. The application of back-propagation neural networks enables us to identify the natural periods of the buildings with accuracy quite satisfactory for engineering practice. The experimental and generated data of vibration displacements are compared and much clearer comparison is given on the phase plane: displacements versus velocities. It was stated that a good generalization takes place both with respect to displacements and velocities.     

基于BP神经网络的有限特征参数砌体结构震害等级推演

详细的建筑结构特征参数是得到合理地震易损性分析结果的基础.本文给出了一种结合已有地震易损性分析成果,在具备有限特征参数的情况下,利用BP神经网络进行单体或群体结构震害等级推演的方法.以陕西省渭南市607栋设防砌体易损性评估结果为样本构建了一个3层BP神经网络模型,并对北京市海淀区近2万栋设防砌体不同地震烈度下的可能破坏...     

Monthly total sediment forecasting using adaptive neuro fuzzy inference system

Accurate forecasting of sediment is an important issue for reservoir design and water pollution control in rivers and reservoirs. In this study, an adaptive neuro-fuzzy inference system (ANFIS) approach is used to construct monthly sediment forecasting system. To illustrate the applicability of ANFIS method the Great Menderes basin is chosen as the study area. The models with various input structures are constructed for the purpose of identification of the best structure. The performance of the ANFIS models in training and testing sets are compared with the observed data. To get more accurate evaluation of the results ANFIS models, the best fit model structures are also tested by artificial neural networks (ANN) and multiple linear regression (MLR) methods. The results of three methods are compared, and it is observed that the ANFIS is preferable and can be applied successfully because it provides high accuracy and reliability for forecasting of monthly total sediment.     

Real‐time recurrent learning neural network for stream‐flow forecasting

Various types of neural networks have been proposed in previous papers for applications in hydrological events. However, most of these applied neural networks are classified as static neural networks, which are based on batch processes that update action only after the whole training data set has been presented. The time variate characteristics in hydrological processes have not been modelled well. In this paper, we present an alternative approach using an artificial neural network, termed real‐time recurrent learning (RTRL) for stream‐flow forecasting. To define the properties of the RTRL algorithm, we first compare the predictive ability of RTRL with least‐square estimated autoregressive integrated moving average models on several synthetic time‐series. Our results demonstrate that the RTRL network has a learning capacity with high efficiency and is an adequate model for time‐series prediction. We also investigated the RTRL network by using the rainfall–runoff data of the Da‐Chia River in Taiwan. The results show that RTRL can be applied with high accuracy to the study of real‐time stream‐flow forecasting networks. Copyright © 2002 John Wiley & Sons, Ltd.     

Seismic buffer recognition using wavelet-based features and neural classification

One of the first operations in a seismic signal processing system applied to earthquake data is to distinguish between valid and invalid records. Since valid signals are characterized by a combination of their time and frequency properties, wavelets are natural candidates for describing seismic features in a compact way. This paper develops a seismic buffer pattern recognition technique, comprising wavelet-based feature extraction, feature selection based on the mutual information criterion, and neural classification based on feedforward networks. The ability of the wavelet transform to capture discriminating information from seismic data in a small number of features is compared with alternative feature reduction techniques, including statistical moments. Three different variations of the wavelet transform are used to extract features: the discrete wavelet transform, the single wavelet transform and the continuous wavelet transform. The mutual information criterion is employed to select a relatively small set of wavelets from the time–frequency grid. Firstly, it is determined whether wavelets can capture more informative data in an equal number of features compared with other features derived from raw data. Secondly, wavelet-based features are compared with features selected based on prior knowledge of class differences. Thirdly, a technique is developed to optimize wavelet features as part of the neural network training process, by using the wavelet neural network architecture. The automated classification techniques developed in this paper are shown to perform similarly to human operators trained for this function. Wavelet-based techniques are found to be useful, both for preprocessing of the raw data and for extracting features from the data. It is demonstrated that the definition of wavelet features can be optimized using the classification wavelet network architecture.     

Benchmark on the accuracy and efficiency of several neural network based phase pickers using datasets from China Seismic Network

Seismic phase pickers based on deep neural networks have been extensively used recently, demonstrating their advantages on both performance and efficiency. However, these pickers are trained with and applied to different data. A comprehensive benchmark based on a single dataset is therefore lacking. Here, using the recently released DiTing dataset, we analyzed performances of seven phase pickers with different network structures, the efficiencies are also evaluated using both CPU and GPU devices. Evaluations based on F₁-scores reveal that the recurrent neural network (RNN) and EQTransformer exhibit the best performance, likely owing to their large receptive fields. Similar performances are observed among PhaseNet (UNet), UNet++, and the lightweight phase picking network (LPPN). However, the LPPN models are the most efficient. The RNN and EQTransformer have similar speeds, which are slower than those of the LPPN and PhaseNet. UNet++ requires the most computational effort among the pickers. As all of the pickers perform well after being trained with a large-scale dataset, users may choose the one suitable for their applications. For beginners, we provide a tutorial on training and validating the pickers using the DiTing dataset. We also provide two sets of models trained using datasets with both 50 Hz and 100 Hz sampling rates for direct application by end-users. All of our models are open-source and publicly accessible.     

A comparative study of three neural network forecast combination methods for simulated river flows of different rainfall—runoff models

Abstract

The performances of three artificial neural network (NN) methods for combining simulated river flows, based on three different neural network structures, are compared. These network structures are: the simple neural network (SNN), the radial basis function neural network (RBFNN) and the multi-layer perceptron neural network (MLPNN). Daily data of eight catchments, located in different parts of the world, and having different hydrological and climatic conditions, are used to enable comparisons of the performances of these three methods to be made. In the case of each catchment, each neural network combination method synchronously uses the simulated river flows of four rainfall—runoff models operating in design non-updating mode to produce the combined river flows. Two of these four models are black-box, the other two being conceptual models. The results of the study show that the performances of all three combination methods are, on average, better than that of the best individual rainfall—runoff model utilized in the combination, i.e. that the combination concept works. In terms of the Nash-Sutcliffe model efficiency index, the MLPNN combination method generally performs better than the other two combination methods tested. For most of the catchments, the differences in the efficiency index values of the SNN and the RBFNN combination methods are not significant but, on average, the SNN form performs marginally better than the more complex RBFNN alternative. Based on the results obtained for the three NN combination methods, the use of the multi-layer perceptron neural network (MLPNN) is recommended as the appropriate NN form for use in the context of combining simulated river flows.     

Prediction of the spatial distribution of the modified mercalli intensity using neural networks

This paper presents the development of an adaptive, non-parametric forecast model for the direct prediction of the spatial distribution of the Modified Mercalli Intensity (MMI) corresponding to an earthquake scenario. The model is based on recent advances in neural networks computation, and is constructed through supervised learning using historical earthquake and regional geological data as training sets. A MMI forecast model for moderate earthquakes with magnitudes between 6 and 7 was developed based on data from the Loma Prieta, Coalinga and Morgan Hill earthquakes. For these data sets, the neural networks forecast model is shown to have excellent data synthesis capability; multiple sets of data can be encapsulated by a relatively simple network architecture. Limited comparison of forecasts made by the neural networks model and conventional models demonstrates that improved accuracy can be achieved. Implementation and operational advantages of the neural networks approach such as general input features, minimum preconceived knowledge of the data sets, the ability to learn and to adapt incrementally and the autonomous and automatic synthesis of the structure underlying the data sets, have been illustrated.     

Neural networks for quick earthquake damage estimation

This paper proposes the use of neural networks to predict damage due to earthquakes from the indices of recorded ground motion. Since the relationship between ground motion indices and resulting damage is difficult to express in mathematical form, neural networks are conveniently applied for this problem. Simulated earthquake ground motions are used to have a well-distributed data set and the ductility factor from non-linear analysis of two single-degree-of-freedom structural models is used to represent the damage. A sensitivity analysis procedure is described to identify qualitatively the input parameters that have a greater influence on the damage. The result of the trained neural network is then verified by using several recorded earthquake ground motions. It is found that some instability in the prediction can occur. Instability occurs when input values exceed the range of the training data. The neural network model using PGA and SI as input give the best performance in the recall tests using actual earthquake ground motion, demonstrating the usefulness of neural network models for the quick estimation of damage through earthquake intensity monitoring.     

Neural networks in auroral data assimilation

Data assimilation is an essential step for improving space weather forecasting by means of a weighted combination between observational data and data from a mathematical model. In the present work data assimilation methods based on Kalman filter (KF) and artificial neural networks are applied to a three-wave model of auroral radio emissions. A novel data assimilation method is presented, whereby a multilayer perceptron neural network is trained to emulate a KF for data assimilation by using cross-validation. The results obtained render support for the use of neural networks as an assimilation technique for space weather prediction.     

Committee neural networks for porosity and permeability prediction from well logs

Neural computing has moved beyond simple demonstration to more significant applications. Encouraged by recent developments in artificial neural network (ANN) modelling techniques, we have developed committee machine (CM) networks for converting well logs to porosity and permeability, and have applied the networks to real well data from the North Sea. Simple three‐layer back‐propagation ANNs constitute the blocks of a modular system where the porosity ANN uses sonic, density and resistivity logs for input. The permeability ANN is slightly more complex, with four inputs (density, gamma ray, neutron porosity and sonic). The optimum size of the hidden layer, the number of training data required, and alternative training techniques have been investigated using synthetic logs. For both networks an optimal number of neurons in the hidden layer is in the range 8–10. With a lower number of hidden units the network fails to represent the problem, and for higher complexity overfitting becomes a problem when data are noisy. A sufficient number of training samples for the porosity ANN is around 150, while the permeability ANN requires twice as many in order to keep network errors well below the errors in core data. For the porosity ANN the overtraining strategy is the suitable technique for bias reduction and an unconstrained optimal linear combination (OLC) is the best method of combining the CM output. For permeability, on the other hand, the combination of overtraining and OLC does not work. Error reduction by validation, simple averaging combined with range‐splitting provides the required accuracy. The accuracy of the resulting CM is restricted only by the accuracy of the real data. The ANN approach is shown to be superior to multiple linear regression techniques even with minor non‐linearity in the background model.     

长短时记忆神经网络在地电场数据处理中的应用

作为深度学习方法的一种,长短时记忆神经网络(LSTM)是一种信号处理的重要方法.本文基于实际观测地电场数据来合成训练集,对特定结构的长短时记忆神经网络进行训练,将训练所得网络对测试集数据进行测试后,将网络应用至实际观测数据.结果显示,经过训练的网络很好地学到了训练集样本的特征,对测试集数据的信噪比压制了约20 dB,并过滤了人为添加的特定频率的干扰成分,对实际观测数据处理后得到明显的日变、半日变以及半月变、月变、半年变、年变等潮汐响应,表明长短时记忆神经网络可以有效应用于地电场数据处理研究.     

Deep neural network application for 4D seismic inversion to changes in pressure and saturation: Optimizing the use of synthetic training datasets

In this work, we tackle the challenge of quantitative estimation of reservoir dynamic property variations during a period of production, directly from four-dimensional seismic data in the amplitude domain. We employ a deep neural network to invert four-dimensional seismic amplitude maps to the simultaneous changes in pressure, water and gas saturations. The method is applied to a real field data case, where, as is common in such applications, the data measured at the wells are insufficient for properly training deep neural networks, thus, the network is trained on synthetic data. Training on synthetic data offers much freedom in designing a training dataset, therefore, it is important to understand the impact of the data distribution on the inversion results. To define the best way to construct a synthetic training dataset, we perform a study on four different approaches to populating the training set making remarks on data sizes, network generality and the impact of physics-based constraints. Using the results of a reservoir simulation model to populate our training datasets, we demonstrate the benefits of restricting training samples to fluid flow consistent combinations in the dynamic reservoir property domain. With this the network learns the physical correlations present in the training set, incorporating this information into the inference process, which allows it to make inferences on properties to which the seismic data are most uncertain. Additionally, we demonstrate the importance of applying regularization techniques such as adding noise to the synthetic data for training and show a possibility of estimating uncertainties in the inversion results by training multiple networks.     

A Modified Neural Network for Improving River Flow Prediction/Un Réseau de Neurones Modifié pour Améliorer la Prévision de L'Écoulement Fluvial

Abstract

Artificial neural network (ANN) models provide huge potential for simulating nonlinear behaviour of hydrological systems. However, the potential of ANN is yet to be fully exploited due to the problems associated with improving the model generalization performance. Generalization refers to the ability of a neural network to correctly process input data that have not been used for calibrating the neural network model. In the hydrological context, better generalization performance implies higher precision of forecasting. The primary objectives of this study are to explore new measures for improving the generalization performance of an ANN-based rainfall–runoff model, and to evaluate the applicability of the new measures. A modified neural network model (entitled goal programming (GP) neural network) for modelling the rainfall–runoff process has been developed, in which three enhancements are made as compared to the widely-used backpropagation (BP) network. The three enhancements are (a) explicit integration of hydrological prior knowledge into the neural network learning; (b) incorporation of a modified training objective function; and (c) reduction of network sensitivity to input errors. Seven watersheds across a range of climatic conditions and watershed areas in China were selected for examining the alternative networks. The results demonstrate that the GP consistently outperformed the BP both in the calibration and verification periods and three proposed measures yielded improvement of performance.