Flood forecasting based on an artificial neural network scheme

Natural Hazards - Nowadays, floods have become the widest global environmental and economic hazard in many countries, causing huge loss of lives and materials damages. It is, therefore, necessary...     

基于遗忘因子的BP神经网络水文实时预报方法

在应用神经网络进行洪水预报时,因洪水系统随着河道上游来流、区间降雨、河床演变等因素的动态变化,其特性并不总是按照基本相同的规律变化,对这类系统的参数辨识,要求算法具有较强的实时跟踪能力,以适应模拟或预测洪水运动变化过程的要求。在BP神经网络模型的基础上,运用最小二乘递推算法,引入时变遗忘因子实时跟踪模型中时变参数的变化,建立了神经网络在非线性系统中动态系统输入、输出数据间的映射关系。计算实例表明:该法对参数的快速时变具有较快的跟踪能力和较高的辨识精度,是一种非常实用的水文实时预报方法。     

基于人工神经网络模型的岩石特性预测

近年来,软计算技术被用作替代的统计工具。如人工神经网络（ANN）被用于开发预测模型来估计所需的参数。在本研究中,通过利用冲击钻进过程中的一些钻进参数（气压、推力、钻头直径、穿透率）和所产生的声级,建立了预测岩石性质的神经网络模型。在实验室中所产生的数据,用于开发预测岩石特性（如单轴抗压强度、耐磨性、抗拉强度和施密特回弹数）的神经网络模型,并使用各种预测性能指标对所建模型进行检验,结果表明人工神经网络模型适用于岩石性质的预测。     

Dynamic routing modeling for flash flood forecast in river system

A real-time flood-forecasting method coupled with the one-dimensional unsteady flow model was developed for the Danshuei River system in northern Taiwan. Based on the flow at current time, the flow at new time is calculated to provide the water stage forecasting during typhoons. Data, from two typhoons in 2000: Bilis and Nari, were used to validate and evaluate the model capability. First, the developed model was applied to validate and evaluate with and without discharge corrections at the Hsin-Hai Bridge in Tahan Stream, Chung-Cheng Bridge in Hsintien Stream, and Sir-Ho Bridge in the Keelung River. The results indicate that the calculated water stage profiles approach the observed data. Moreover, the water stage forecasting hydrograph with discharge correction is close to the observed water stage hydrograph and yields a better prediction than that without discharge correction. The model was then used to quantify the difference in prediction between different methods of real-time water stage correction. The model results reveal that water stages using the 1–6 h forecast with real-time stage correction exhibits the best lead times. The accuracy for 1–3 h lead time is higher than that for 4–6 h lead time, suggesting that the flash flood forecast in the river system is reasonably accurate for 1–3 h lead time only. The method developed is effective for flash flood forecasting and can be adopted for flood forecasting in complicated river systems.     

An artificial neural network model for flood simulation using GIS: Johor River Basin, Malaysia

Flooding is one of the most destructive natural hazards that cause damage to both life and property every year, and therefore the development of flood model to determine inundation area in watersheds is important for decision makers. In recent years, data mining approaches such as artificial neural network (ANN) techniques are being increasingly used for flood modeling. Previously, this ANN method was frequently used for hydrological and flood modeling by taking rainfall as input and runoff data as output, usually without taking into consideration of other flood causative factors. The specific objective of this study is to develop a flood model using various flood causative factors using ANN techniques and geographic information system (GIS) to modeling and simulate flood-prone areas in the southern part of Peninsular Malaysia. The ANN model for this study was developed in MATLAB using seven flood causative factors. Relevant thematic layers (including rainfall, slope, elevation, flow accumulation, soil, land use, and geology) are generated using GIS, remote sensing data, and field surveys. In the context of objective weight assignments, the ANN is used to directly produce water levels and then the flood map is constructed in GIS. To measure the performance of the model, four criteria performances, including a coefficient of determination (R ²), the sum squared error, the mean square error, and the root mean square error are used. The verification results showed satisfactory agreement between the predicted and the real hydrological records. The results of this study could be used to help local and national government plan for the future and develop appropriate (to the local environmental conditions) new infrastructure to protect the lives and property of the people of Johor.     

MERLIN: a flood hazard forecasting system for coastal river reaches

Natural Hazards - This study presents MERLIN, an innovative flood hazard forecasting system for predicting discharges and water levels at flood prone areas of coastal catchments. Discharge...     

Assessment of a conceptual hydrological model and artificial neural networks for daily outflows forecasting

Artificial neural networks (ANNs) are used by hydrologists and engineers to forecast flows at the outlet of a watershed. They are employed in particular where hydrological data are limited. Despite these developments, practitioners still prefer conventional hydrological models. This study applied the standard conceptual HEC-HMS’s soil moisture accounting (SMA) algorithm and the multi layer perceptron (MLP) for forecasting daily outflows at the outlet of Khosrow Shirin watershed in Iran. The MLP [optimized with the scaled conjugate gradient] used the logistic and tangent sigmoid activation functions resulting into 12 ANNs. The R ² and RMSE values for the best trained MPLs using the tangent and logistic sigmoid transfer function were 0.87, 1.875 m³ s^?1 and 0.81, 2.297 m³ s^?1, respectively. The results showed that MLPs optimized with the tangent sigmoid predicted peak flows and annual flood volumes more accurately than the HEC-HMS model with the SMA algorithm, with R ² and RMSE values equal to 0.87, 0.84 and 1.875 and 2.1 m³ s^?1, respectively. Also, an MLP is easier to develop due to using a simple trial and error procedure. Practitioners of hydrologic modeling and flood flow forecasting may consider this study as an example of the capability of the ANN for real world flow forecasting.     

基于神经网络的混沌时间序列预测

应用混沌方法对时间序列观测数据进行处理,计算出最大lyapunov指数,得到最大可预报时间尺度。在此基础上,建立人工神经网络预测预报混沌时间序列的模型。结合实例,对该预测方法进行了计算验证。     

南江滑坡群体积的BP神经网络模型与预测

基于南江县境内244个典型土质滑坡统计样本,利用BP神经网络模型,采用3种不同的方案(基于不同的评价参数)对滑坡体积进行预测。方案一选取坡高、坡度、坡向、高程、植被覆盖率、岩层倾向、岩层倾角等7项评价参数;方案二选取坡高、坡度、坡向、岩层倾向、岩层倾角等5参数;方案三选取坡高、坡度、坡向等3参数。研究结果表明:3种方案建立的BP神经网络模型都具有较高的可靠性,其预测结果都可以较好地逼近真实滑坡体积值,BP神经网络能有效应用到滑坡体积预测中;3种方案预测值与实际值基本吻合,且两者间的相关系数分别为0.87083,0.90826,0.86119,评价参数的合理选择对滑坡体积预测的准确性有着重要的影响;方案二的相关系数最高,其预测准确性最好,这表明坡高、坡度、坡向、岩层倾向、岩层倾角是影响滑坡体积的重要因素,植被覆盖率和高程为其次要影响因素。     

Real-time flood forecasting of the Tiber river in Rome

An adaptive, conceptual model for real-time flood forecasting of the Tiber river in Rome is proposed. This model simulates both rainfall-runoff transformations, to reproduce the contributions of 37 ungauged sub-basins that covered about 30% of the catchment area, and flood routing processes in the hydrographic network. The adaptive component of the model concerns the rainfall-runoff analysis: at any time step the whole set of the model parameters is recalibrated by minimizing the objective function constituted by the sum of the squares of the differences between observed and computed water surface elevations (or discharges). The proposed model was tested through application under real-time forecasting conditions for three historical flood events. To assess the forecasting accuracy, to support the decision maker and to reduce the possibility of false or missed warnings, confidence intervals of the forecasted water surface elevations (or discharges), computed according to a Monte Carlo procedure, are provided. The evaluation of errors in the prediction of peak values, of coefficients of persistence and of the amplitude of confidence intervals of prediction shows the possibility to develop a flood forecast model with a lead time of 12 h, which is useful for civil protection actions.     

基于模糊人工神经网络识别的水质评价模型

人工神经网络和模糊识别理论作为模拟生物体的信息处理系统,在实践中得到广泛的应用,且各有所长,将二者相结合,构造出模糊人工神经识别网络,从而使识别系统的柔性处理能力得到很大提高.阐述了模型和方法,将其用于长江支流沱江枯水期的水质综合评价,结果表明,模糊人工神经网络综合评价具有客观性和实用性.     

Uncertainty analysis of HEC-HMS model using the GLUE method for flash flood forecasting of Mekerra watershed,Algeria

In this paper, the lumped quasi-distributed hydrological model HEC HMS is used to simulate the rainfall–runoff process of the Mekerra watershed, located in the northwest of Algeria. The model parameters’ uncertainty and the predictive intervals were evaluated with the generalized likelihood uncertainty estimation (GLUE) approach. According to the results, good simulations were obtained with different values of variables for many sets of parameters generated randomly by the Monte Carlo procedure, which is known as Equifinality. After the analysis, only the hydraulic conductivity at saturation parameter appears well defined, taking values within a limited range. In addition, results indicated that combinations of likelihood measures associated with multiple and different periods of observations reduce a posterior uncertainty of estimated parameters and predictive intervals in some degree. Overall, the GLUE analysis showed that there is a significant uncertainty associated with hydrological modelling of watershed Mekerra, to a great extent due to multiple sources of errors.     

Prediction of copper and chromium concentrations in bean leaves based on an artificial neural network model

The assessment of copper and chromium concentrations in plants requires the quantification of a large number of soil factors that affect their potential availability and subsequent toxicity and a mathematical model that predicts their relative concentrations in plants. While many soil characteristics have been implicated as altering copper and chromium availability to plants in soil, accurate, rapid and simple predictive models of metal concentrations are still lacking for soil and plant analysis. In the current study, an artificial neural network model was developed and applied to predict the exposure of bean leaves (BL) to high concentrations of copper and chromium versus some selected soil properties (pH, soil electrical conductivity and dissolved organic carbon). A series of measurements was performed on soil samples to assess the variation of copper and chromium concentrations in BL versus the soil inputs. The performance of the artificial neural network model was then evaluated using a test data set and applied to predict the exposure of the BL to the metal concentration versus the soil inputs. Correlation coefficients of 0.99981 and 0.9979 for Cu and 0.99979 and 0.9975 for Cr between the measured and artificial neural networks predicted values were found, respectively, during the testing and validation procedures. Results showed that the artificial neural network model can be successfully applied to the rapid and accurate prediction of copper and chromium concentrations in BL.     

Validation of an artificial neural network model for landslide susceptibility mapping

The aim of this study was to validate an artificial neural network model at Youngin, Janghung, and Boeun, Korea, using the geographic information system (GIS). The factors that influence landslide occurrence, such as the slope, aspect, curvature, and geomorphology of topography, the type, material, drainage, and effective thickness of soil, the type, diameter, age, and density of forest, distance from lineament, and land cover were either calculated or extracted from the spatial database and Landsat TM satellite images. Landslide susceptibility was analyzed using the landslide occurrence factors provided by the artificial neural network model. The landslide susceptibility analysis results were validated and cross-validated using the landslide locations as study areas. For this purpose, weights for each study area were calculated by the artificial neural network model. Among the nine cases, the best accuracy (81.36%) was obtained in the case of the Boeun-based Janghung weight, whereas the Janghung-based Youngin weight showed the worst accuracy (71.72%).     

基于水流阻力与图论的河网连通性评价

河网连通是区域防洪、供水和生态安全的重要基础。基于图论将河网概化为图模型,考虑不同类型河道输水能力差异,以河道水流阻力倒数表征水流通畅度,并以河道水流通畅度为权值,借助ArcGIS平台建立河网图模型加权邻接矩阵,再利用MATLAB工具进行顶点水流通畅度计算和河网连通度分析,从而实现对河网连通性的定量化分析。以太湖流域嘉兴平原河网为例,进行河道疏浚前后的河网连通程度定量评价,结果表明河道疏浚后河网连通度明显增加。     

公路软基沉降函数干涉神经网络预测模型

建立了基于函数干涉神经网络的公路软基沉降预测模型。工程实例表明,所建议的模型外延性好,而且,可以由较短预压期内沉降观测资料预测远期沉降发展,与传统沉降预测模型相比具有显著的优越性,工程应用前景广阔。     

煤矿立井井筒非采动破裂的人工神经网络预测

应用人工神经网络的基本原理,建立了一个基于神经网络的煤矿立井井筒非采动破裂的预测系统,实现了立井井筒破裂预测的智能化。最后将神经网络预测结果与数值计算结果对比,认为应用人工神经网络对立井井筒破裂时间的预测比较准确、实用。