共查询到20条相似文献，搜索用时 312 毫秒
1.
Hydrological models used for flood prediction in ungauged catchments are commonly fitted to regionally transferred data. The key issue of this procedure is to identify hydrologically similar catchments. Therefore, the dominant controls for the process of interest have to be known. In this study, we applied a new machine learning based approach to identify the catchment characteristics that can be used to identify the active processes controlling runoff dynamics. A random forest (RF) regressor has been trained to estimate the drainage velocity parameters of a geomorphologic instantaneous unit hydrograph (GIUH) in ungauged catchments, based on regionally available data. We analyzed the learning procedure of the algorithm and identified preferred donor catchments for each ungauged catchment. Based on the obtained machine learning results from catchment grouping, a classification scheme for drainage network characteristics has been derived. This classification scheme has been applied in a flood forecasting case study. The results demonstrate that the RF could be trained properly with the selected donor catchments to successfully estimate the required GIUH parameters. Moreover, our results showed that drainage network characteristics can be used to identify the influence of geomorphological dispersion on the dynamics of catchment response. 相似文献
2.
Seree Supharatid 《水文研究》2003,17(15):3085-3099
This paper presents the applicability of neural network (NN) modelling for forecasting and filtering problems. The multilayer feedforward (MLFF) network was first constructed to forecast the tidal‐level variations at the mouth of the River Chao Phraya in Thailand. Unlike the well‐known conventional harmonic analysis, the NN model uses a set of previous data for learning and then forecasting directly the time‐series of tidal levels. It was found that lead time of 1 to 24 hourly tidal levels can be predicted successfully using only a short‐time hourly learning data. The MLFF network was further used to establish a stage–discharge relationship for the tidal river. The results show a considerably better performance of the NN model over the conventional models. In addition, the stage–discharge relationship obtained by the NN model can indicate reasonably well the important behaviour of the tidal influences. Copyright © 2003 John Wiley & Sons, Ltd. 相似文献
3.
Self‐organizing maps (SOMs) have been successfully accepted widely in science and engineering problems; not only are their results unbiased, but they can also be visualized. In this study, we propose an enforced SOM (ESOM) coupled with a linear regression output layer for flood forecasting. The ESOM re‐executes a few extra training patterns, e.g. the peak flow, as recycling input data increases the mapping space of peak flow in the topological structure of SOM, and the weighted sum of the extended output layer of the network improves the accuracy of forecasting peak flow. We have investigated an ESOM neural network by using the flood data of the Da‐Chia River, Taiwan, and evaluated its performance based on the results obtained from a commonly used back‐propagation neural network. The results demonstrate that the ESOM neural network has great efficiency for clustering, especially for the peak flow, and super capability of modelling the flood forecast. The topology maps created from the ESOM are interesting and informative. Copyright © 2007 John Wiley & Sons, Ltd. 相似文献
4.
The present study aims to develop a hybrid multi‐model using the soft computing approach. The model is a combination of a fuzzy logic, artificial neural network (ANN) and genetic algorithm (GA). While neural networks are low‐level computational structures that perform well dealing with raw data, fuzzy logic deal with reasoning on a higher level by using linguistic information acquired from domain experts. However, fuzzy systems lack the ability to learn and cannot adjust themselves to a new environment. Moreover, experts occasionally make mistakes and thus some rules used in a system may be false. A network type structure of the present hybrid model is a multi‐layer feed‐forward network, the main part is a fuzzy system based on the first‐order Sugeno fuzzy model with a fuzzification and a defuzzification processes. The consequent parameters are determined by least square method. The back‐propagation is applied to adjust weights of network. Then, the antecedent parameters of the membership function are updated accordingly by the gradient descent method. The GA was applied to select the fuzzy rule. The hybrid multi‐model was used to forecast the flood level at Chiang Mai (under the big flood 2005) and the Koriyama flood (2003) in Japan. The forecasting results are evaluated using standard global goodness of fit statistic, efficient index (EI), the root mean square error (RMSE) and the peak flood error. Moreover, the results are compared to the results of a neuro‐genetic model (NGO) and ANFIS model using the same input and output variables. It was found that the hybrid multi‐model can be used successfully with an efficiency index (EI) more than 0·95 (for Chiang Mai flood up to 12 h ahead forecasting) and more than 0·90 (for Koriyama flood up to 8 h ahead forecasting). In general, all of three models can predict the water level with satisfactory results. However, the hybrid model gave the best flood peak estimation among the three models. Therefore, the use of fuzzy rule base, which is selected by GA in the hybrid multi‐model helps to improve the accuracy of flood peak. Copyright © 2009 John Wiley & Sons, Ltd. 相似文献
5.
Abstract Researchers have used various physical, chemical, or topographic features to define estuaries, based on the needs of their particular subject. The principal features of estuaries are the tides that influence their water stages; thus, the boundaries of an estuary can be determined based on whether the water stage is subject to tidal influence. However, the water stage is also influenced by the upstream river discharge. A hydrograph of water stage will therefore include both non-stationary and nonlinear features. Here, we use the Hilbert-Huang Transform (HHT), which allows us to process such non-stationary and nonlinear signals, to decompose the water-stage hydrographs recorded at different gauging stations in an estuary into their intrinsic mode function (IMF) components and residuals. We then analyse the relationships between the frequencies of IMFs and known tidal components. A frequency correlation indicates that the water stage of the station is subject to tidal influences and is located within the estuary. The spatial distribution of the stations that are subject to tidal influences can then be used to define the estuary boundaries. We used data from gauging stations in the estuary region of Taiwan's Tanshui River to assess the feasibility of using the HHT to define an estuary. The results show that the HHT is a dependable and easy method for determining the boundaries of an estuary. Citation Chen, Y.-C., Kao, S.-P., and Chiang, H.-W., 2013. Defining an estuary using the Hilbert-Huang transform. Hydrological Sciences Journal, 58 (4), 841–853. 相似文献
6.
Much of the nonlinearity and uncertainty regarding the flood process is because hydrologic data required for estimation are often tremendously difficult to obtain. This study employed a back‐propagation network (BPN) as the main structure in flood forecasting to learn and to demonstrate the sophisticated nonlinear mapping relationship. However, a deterministic BPN model implies high uncertainty and poor consistency for verification work even when the learning performance is satisfactory for flood forecasting. Therefore, a novel procedure was proposed in this investigation which integrates linear transfer function (LTF) and self‐organizing map (SOM) to efficiently determine the intervals of weights and biases of a flood forecasting neural network to avoid the above problems. A SOM network with classification ability was applied to the solutions and parameters of the BPN model in the learning stage, to classify the network parameter rules and to obtain the winning parameters. The outcomes from the previous stage were then used as the ranges of the parameters in the recall stage. Finally, a case study was carried out in Wu‐Shi basin to demonstrate the effectiveness of the proposal. Copyright © 2009 John Wiley & Sons, Ltd. 相似文献
7.
Tidal straining effect on sediment transport dynamics in the Huanghe (Yellow River) estuary was studied by field observations
and numerical simulations. The measurement of salinity, suspended sediment concentration, and current velocity was conducted
during a flood season in 1995 at the Huanghe river mouth with six fishing boats moored at six stations for 25-h hourly time
series observations. Based on the measurements, the intra-tidal variations of sediment transport in the highly turbid river
mouth was observed and the tidal straining effect occurred. Our study showed that tidal straining of longitudinal sediment
concentration gradients can contribute to intra-tidal variability in sediment stratification and to asymmetries in sediment
distribution within a tidal cycle. In particular, the tidal straining effect in the Huanghe River estuary strengthened the
sediment-induced stratification at the flood tide, thus producing a higher bottom sediment concentration than that during
the ebb. A sediment transport model that is capable of simulating sediment-induced stratification effect on the hydrodynamics
in the bottom boundary layers and associated density currents was applied to an idealized estuary to demonstrate the processes
and to discuss the mechanism. The model-predicted sediment processes resembled the observed characteristics in the Huanghe
River estuary. We concluded that tidal straining effect is an important but poorly understood mechanism in the transport dynamics
of cohesive sediments in turbid estuaries and coastal seas. 相似文献
8.
A numerical modeling study of the influence of the lateral flow on the estuarine exchange flow was conducted in the north passage of the Changjiang estuary. The lateral flows show substantial variabilities within a flood-ebb tidal cycle. The strong lateral flow occurring during flood tide is caused primarily by the unique cross-shoal flow that induces a strong northward (looking upstream) barotropic force near the surface and advects saltier water toward the northern part of the channel, resulting in a southward baroclinic force caused by the lateral density gradient. Thus, a two-layer structure of lateral flows is produced during the flood tide. The lateral flows are vigorous near the flood slack and the magnitude can exceed that of the along-channel tidal flow during that period. The strong vertical shear of the lateral flows and the salinity gradient in lateral direction generate lateral tidal straining, which are out of phase with the along-channel tidal straining. Consequently, stratification is enhanced at the early stage of the ebb tide. In contrast, strong along-channel straining is apparent during the late ebb tide. The vertical mixing disrupts the vertical density gradient, thus suppressing stratification. The impact of lateral straining on stratification during spring tide is more pronounced than that of along-channel straining during late flood and early ebb tides. The momentum balance along the estuary suggests that lateral flow can augment the residual exchange flow. The advection of lateral flows brings low-energy water from the shoal to the deep channel during the flood tide, whereas the energetic water is moved to the shoal via lateral advection during the ebb tide. The impact of lateral flow on estuarine circulation of this multiple-channel estuary is different from single-channel estuary. A model simulation by blocking the cross-shoal flow shows that the magnitudes of lateral flows and tidal straining are reduced. Moreover, the reduced lateral tidal straining results in a decrease in vertical stratification from the late flood to early ebb tides during the spring tide. By contrast, the along-channel tidal straining becomes dominant. The model results illustrate the important dynamic linkage between lateral flows and estuarine dynamics in the Changjiang estuary. 相似文献
9.
Classified real-time flood forecasting by coupling fuzzy clustering and neural network 总被引：2，自引：0，他引：2
Minglei REN Bende WANG Qiuhua LIANG Guangtao FU 《国际泥沙研究》2010,25(2):134-148
This paper presented a new classified real-time flood forecasting framework by integrating a fuzzy clustering model and neural network with a conceptual hydrological model. A fuzzy clustering model was used to classify historical floods in terms of flood peak and runoff depth, and the conceptual hydrological model was calibrated for each class of floods. A back-propagation (BP) neural network was trained by using real-time rainfall data and outputs from the fuzzy clustering model. BP neural network provided a rapid on-line classification for real-time flood events. Based on the on-line classification, an appropriate parameter set of hydrological model was automatically chosen to produce real-time flood forecasting. Different parameter sets was continuously used in the flood forecasting process because of the changes of real-time rainfall data and on-line classification results. The proposed methodology was applied to a large catchment in Liaoning province, China. Results show that the classified framework provided a more accurate prediction than the traditional non-classified method. Furthermore, the effects of different index weights in fuzzy clustering were also discussed. 相似文献
10.
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. 相似文献
11.
径向基神经网络（RBFNN）具有结构简单、学习速度快、不易陷入局部极小等优点,能够有效地提高电阻率层析成像反演的收敛速度和求解质量.本文针对电阻率层析成像反演的非线性特征,提出了一种基于汉南-奎因信息准则（HQC）的正交最小二乘法（OLS）学习算法（HQOLS）.该算法通过计算HQC的最优值来自动选择RBFNN的网络结构,避免了传统OLS学习算法中阈值参数的设定,保证了网络的泛化性能.通过比较聚类法、梯度法、OLS和HQOLS等学习算法的反演性能,构建了基于RBFNN的电阻率层析成像反演模型.数值仿真和模型反演的结果表明,该方法实现简单,在准确性上优于BP反演,成像质量优于传统最小二乘法反演. 相似文献
12.
Comparison of artificial neural network models for hydrologic predictions at multiple gauging stations in an agricultural watershed 总被引：1，自引：0，他引：1
This paper reports on an evaluation of the use of artificial neural network (ANN) models to forecast daily flows at multiple gauging stations in Eucha Watershed, an agricultural watershed located in north‐west Arkansas and north‐east Oklahoma. Two different neural network models, the multilayer perceptron (MLP) and the radial basis neural network (RBFNN), were developed and their abilities to predict stream flow at four gauging stations were compared. Different scenarios using various combinations of data sets such as rainfall and stream flow at various lags were developed and compared for their ability to make flow predictions at four gauging stations. The input vector selection for both models involved quantification of the statistical properties such as cross‐, auto‐ and partial autocorrelation of the data series that best represented the hydrologic response of the watershed. Measured data with 739 patterns of input–output vector were divided into two sets: 492 patterns for training, and the remaining 247 patterns for testing. The best performance based on the RMSE, R2 and CE was achieved by the MLP model with current and antecedent precipitation and antecedent flow as model inputs. The MLP model testing resulted in R2 values of 0·86, 0·86, 0·81, and 0·79 at the four gauging stations. Similarly, the testing R2 values for the RBFNN model were 0·60, 0·57, 0·58, and 0·56 for the four gauging stations. Both models performed satisfactorily for flow predictions at multiple gauging stations, however, the MLP model outperformed the RBFNN model. The training time was in the range 1–2 min for MLP, and 5–10 s for RBFNN on a Pentium IV processor running at 2·8 GHz with 1 MB of RAM. The difference in model training time occurred because of the clustering methods used in the RBFNN model. The RBFNN uses a fuzzy min‐max network to perform the clustering to construct the neural network which takes considerably less time than the MLP model. Results show that ANN models are useful tools for forecasting the hydrologic response at multiple points of interest in agricultural watersheds. Copyright © 2008 John Wiley & Sons, Ltd. 相似文献
13.
This paper documents a numerical modeling study to calculate the residence time and age of dissolved substances in a partially mixed estuary. A three-dimensional, time-dependent hydrodynamic model was established and applied to the Danshuei River estuarine system and adjacent coastal sea in Taiwan. The model showed good agreement with observations of surface elevation, tidal currents and salinity made in 2002. The model was then applied to calculate the residence time and age distribution response to different freshwater discharges with and without density-induced circulations in the Danshuei River estuarine system. Regression analysis of model results reveals that an exponential equation can be used to correlate the residence time to change of freshwater input. The simulated results show it takes approximately 10, 4.5, and 3 days, respectively, for a water parcel that has entered the headwaters of the estuary to be transported out of the estuary under low, mean, and high flow conditions with density-induced circulation. The calculated age with density-induced circulation is less than that without density-induced circulation. The age of the surface layer is less than that at the bottom layer. Overall the study shows that freshwater discharges are the important factors in controlling the transport of dissolved substances in the Danshuei River estuarine system. 相似文献
14.
The morphodynamics of shallow, vertically well-mixed estuaries, characterised by tidal flats and deeper channels, have been
investigated. This paper examines what contributes to flood/ebb-dominant sediment transport in localised regions through a
2D model study (using the TELEMAC modelling system). The Dyfi Estuary in Wales, UK has been used as a case study and, together
with idealised estuary shapes, shows that shallow water depths lead to flood dominance in the inner estuary whilst tidal flats
and deep channels cause ebb dominance in the outer estuary. For medium sands and with an artificially ‘flattened’ bathymetry
(i.e. no tidal flats), the net sediment transport switches from ebb-dominant to flood-dominant where the parameter a/h (local tidal amplitude ÷ local tidally averaged water depth) exceeds 1.2. Sea level rise will reduce this critical value
of a/h and also reduce the ebb-directed sediment transport significantly, leading to a flood-dominated estuarine system. A similar
pattern, albeit with greater transport, was simulated with tidal flats included and also with a reduced grain size. This suggests
that analogous classifications for flood/ebb asymmetry of the tide in estuaries as a whole may not represent the local sediment
transport in sufficient detail. Through the Dyfi simulations, the above criterion involving a/h is shown to be complicated further by augmented flow past a spit at the estuary mouth which gives rise to a self-maintaining
scour hole. Simulations of one year of bed evolution in an idealised flat-bottomed estuary, including tidal flow past a spit,
recreate the flood/ebb dominance on either side of the spit and the formation of a scour hole in between. The erosion rate
at the centre of the hole is reduced as the hole deepens, suggesting the establishment of a self-maintaining equilibrium state. 相似文献
15.
16.
Interaction between suspended sediment and tidal amplification in the Guadalquivir Estuary 总被引：1，自引：1，他引：0
Water level records at two stations in the Guadalquivir Estuary (Spain), one near the estuary mouth (Bonanza) and one about 77 km upstream (Sevilla), have been analysed to study the amplification of the tide in the estuary. The tidal amplification factor shows interesting temporal variation, including a spring-neap variation, some extreme low values, and especially the anomalous behaviour that the amplification factor is larger during a number of periods. These variations are explained by data analysis combined with numerical and analytical modelling. The spring-neap variation is due to the quadratic relation between the bottom friction and the tidal flow velocity. The river flood events are the direct causes of the extreme low values of the amplification factor, and they trigger the non-linear interaction between the tidal flow and suspended sediment transport. The fluvial sediment input during a river flood causes high sediment concentration in the estuary, up to more than 10 g/l. This causes a reduction of the effective hydraulic drag, resulting in stronger tidal amplification in the estuary for a period after a river flood. After such an event the tidal amplification in the estuary does not always fall back to the same level as before the event, indicating that river flood events have significant influence on the long-term development of this estuary. 相似文献
17.
Coupling the Xinanjiang model to a kinematic flow model based on digital drainage networks for flood forecasting 总被引：1，自引：0，他引：1
The Xinanjiang model, which is a conceptual rainfall‐runoff model and has been successfully and widely applied in humid and semi‐humid regions in China, is coupled by the physically based kinematic wave method based on a digital drainage network. The kinematic wave Xinanjiang model (KWXAJ) uses topography and land use data to simulate runoff and overland flow routing. For the modelling, the catchment is subdivided into numerous hillslopes and consists of a raster grid of flow vectors that define the water flow directions. The Xinanjiang model simulates the runoff yield in each grid cell, and the kinematic wave approach is then applied to a ranked raster network. The grid‐based rainfall‐runoff model was applied to simulate basin‐scale water discharge from an 805‐km2 catchment of the Huaihe River, China. Rainfall and discharge records were available for the years 1984, 1985, 1987, 1998 and 1999. Eight flood events were used to calibrate the model's parameters and three other flood events were used to validate the grid‐based rainfall‐runoff model. A Manning's roughness via a linear flood depth relationship was suggested in this paper for improving flood forecasting. The calibration and validation results show that this model works well. A sensitivity analysis was further performed to evaluate the variation of topography (hillslopes) and land use parameters on catchment discharge. Copyright © 2009 John Wiley & Sons, Ltd. 相似文献
18.
Contemporary hydrodynamics and morphological change are examined in a shallow microtidal estuary, located on a wave-dominated coast (Port Stephens, NSW, Australia). Process-based numerical modelling is undertaken by combining modules for hydrodynamics, waves, sediment transport and bathymetry updates. Model results suggest that the complex estuarine bathymetry and geometry give rise to spatial variations in the tidal currents and a marked asymmetry between ebb and flood flows. Sediment transport paths correspond with tidal asymmetry patterns. The SE storms significantly enhance the quantities of sediment transport, while locally generated waves by the westerly strong winds also are capable of causing sediment entrainment and contribute to the delta morphological change. The wave/wind-induced currents are not uniform with flow over shoals driven in the same direction as waves/winds while a reverse flow occurring in the adjacent channel. The conceptual sediment transport model developed in this study shows flood-directed transport occurs on the flood ramp while ebb-directed net transport occurs in the tidal channels and at the estuary entrance. Accretion of the intertidal sand shoals and deepening of tidal channels, as revealed by the model, suggest that sediment-infilling becomes advanced, which may lead to an ebb-dominated estuary. It is likely that a switch from flood- to ebb-dominance occurs during the estuary evolution, and the present-day estuary acts as a sediment source rather than sediment sink to the coastal system. This is conflictive to the expectation drawn from the estuarine morphology; however, it is consistent with previous research suggesting that, in an infilling estuary, an increase in build-up of intertidal flats/shoals can eventually shift an estuary towards ebb dominance. Thus, field data are needed to validate the result presented here, and further study is required to investigate a variety of estuaries in the Australian area. 相似文献
19.
In many engineering problems, such as flood warning systems, accurate multistep‐ahead prediction is critically important. The main purpose of this study was to derive an algorithm for two‐step‐ahead forecasting based on a real‐time recurrent learning (RTRL) neural network that has been demonstrated as best suited for real‐time application in various problems. To evaluate the properties of the developed two‐step‐ahead RTRL algorithm, we first compared its predictive ability with least‐square estimated autoregressive moving average with exogenous inputs (ARMAX) models on several synthetic time‐series. Our results demonstrate that the developed two‐step‐ahead RTRL network has efficient ability to learn and has comparable accuracy for time‐series prediction as the refitted ARMAX models. We then investigated the two‐step‐ahead RTRL network by using the rainfall–runoff data of the Da‐Chia River in Taiwan. The results show that the developed algorithm can be successfully applied with high accuracy for two‐step‐ahead real‐time stream‐flow forecasting. Copyright © 2003 John Wiley & Sons, Ltd. 相似文献
20.
A three-dimensional hydrodynamic model is used to investigate intra-tidal and spring–neap variations of turbulent mixing, stratification and residual circulation in the Chesapeake Bay estuary. Vertical profiles of salinity, velocity and eddy diffusivity show a marked asymmetry between the flood and ebb tides. Tidal mixing in the bottom boundary layer is stronger and penetrates higher on flood than on ebb. This flood–ebb asymmetry results in a north–south asymmetry in turbulent mixing because tidal currents vary out of phase between the lower and upper regions of Chesapeake Bay. The asymmetric tidal mixing causes significant variation of salinity distribution over the flood–ebb tidal cycle but insignificant changes in the residual circulation. Due to the modulation of tidal currents over the spring–neap cycle, turbulent mixing and vertical stratification show large fortnightly and monthly fluctuations. The stratification is not a linear function of the tidal-current amplitude. Strong stratification is only established during those neap tides when low turbulence intensity persists for several days. Residual circulation also shows large variations over the spring–neap cycle. The tidally averaged residual currents are about 50% stronger during the neap tides than during the spring tides. 相似文献