首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 437 毫秒
1.
In the Himalayan regions, precipitation-runoff relationships are amongst the most complex hydrological phenomena, due to varying topography and basin characteristics. In this study, different artificial neural networks (ANNs) algorithms were used to simulate daily runoff at three discharge measuring sites in the Himalayan Kosi River Basin, India, using various combinations of precipitation-runoff data as input variables. The data used for this study was collected for the monsoon period (June to October) during the years of 2005 to 2009. ANNs were trained using different training algorithms, learning rates, length of data and number of hidden neurons. A comprehensive multi-criteria validation test for precipitation-runoff modeling has been undertaken to evaluate model performance and test its validity for generating scenarios. Global statistics have demonstrated that the multilayer perceptron with three hidden layers (MLP-3) is the best ANN for basin comparisons with other MLP networks and Radial Basis Functions (RBF). Furthermore, non-parametric tests also illustrate that the MLP-3 network is the best network to reproduce the mean and variance of observed runoff. The performance of ANNs was demonstrated for flows during the monsoon season, having different soil moisture conditions during period from June to October.  相似文献   

2.
In this paper, sparse data problem in neural network and geostatistical modeling for ore-grade estimation was addressed in the Nome offshore placer gold deposit. The problem of sparse data arises because of the random data division into training, validation, and test subsets during ore-grade modeling. In this regard, the possibility of generating statistically dissimilar data subsets by random data division was also explored through a simulation exercise. A combined approach of data segmentation and application of a Kohonen network then was used to solve the data division problem. Two neural networks and five kriging models were applied for grade modeling. The neural network was trained using an early stopping method. Performance evaluation of the models was carried out on the test data set. The study results indicated that all the models that were investigated in this study performed almost equally. It was also revealed that by using the secondary variable watertable depth the neural network and the kriging models slightly improved their prediction precision. Further, the overall R 2 of the models was poor as a result of high nugget (noisy) component in ore-grade variation.  相似文献   

3.
This study evaluates the performances of two distinct linear and non-linear models for simulating non-linear rainfall–runoff processes and their applications to flood forecasting in the Navrood River basin, Iran. Due to the excellent capacity of the artificial neural networks [multilayer perceptron (MLP)] and Volterra model, these models were used to approximate arbitrary non-linear rainfall–runoff processes. The MLP model was trained using two different training algorithms. The Volterra model was applied as a linear model [the first-order Volterra (FOV) model] and solved using the traditional ordinary least-square (OLS) method. Storm events within the Navrood River basin were used to verify the suitability of the two models. The models’ performances were evaluated and compared using five performance criteria namely coefficient of efficiency, root mean square error, error of total volume, relative error of peak discharge, and error of time for peak to arrive. Results indicated that the non-linear MLP models outperform the linear FOV model. The latter was ineffective because of the non-linearity of the rainfall–runoff process. Moreover, the OLS method is inefficient when the FOV model has many parameters that must be estimated.  相似文献   

4.
Paleolimnological information is often extracted from diatom records using weighted averaging calibration and regression techniques. Larger calibration sample sets yield better inferences because they better characterize the environmental characteristics and species assemblages of the sample region. To optimize inferred information from fossil assemblages, however, it is worth knowing if fewer calibration samples can be used. Furthermore, confidence in environmental reconstructions is greater if we consider the relative importance of (A) similarity between fossil and calibration assemblages and (B) how well fossil taxa respond to the environmental variable of interest. We examine these issues using ~200-year sediment profiles from four Minnesota lakes and a 145-lake surface sediment training set calibrated for total phosphorus (TP). Training set sample sizes ranging from 10 to 145 were created through random sample selection, and models based on these training sets were used to calculate diatom-inferred (DI) TP data from fossil samples. Relationships between DI-TP variability and sample size were used to determine the minimum sample size needed to optimize the model for paleo-reconstruction. Similarly, similarities between fossil and modern assemblages were calculated for each size training set. Finally, fossil and modern assemblages were compared to determine whether older fossil samples had poorer similarity with modern analogs. More than 50–80 samples, depending on lake, were needed to stabilize variability in DI-TP results, and >110 training set samples were needed to minimize modern-fossil assemblage dissimilarities. Dissimilarities appeared to increase with sample age, but only one of the four studied cores displayed a significant trend. We have two recommendations for future studies: (1) be cautious when dealing with smaller training sets, especially if they are used to interpret older fossil assemblages and (2) understand how well fossil taxa are attuned to the variable of interest, as it is critical to evaluating the quality of the diatom-inferred data.  相似文献   

5.
Artificial neural networks were applied to simulate runoff from the glacierized part of the Waldemar River catchment (Svalbard) based on hydrometeorological data collected in the summer seasons of 2010, 2011 and 2012. Continuous discharge monitoring was performed at about 1 km from the glacier snout, in the place where the river leaves the marginal zone. Averaged daily values of discharge and selected meteorological variables in a number of combinations were used to create several models based on the feed‐forward multilayer perceptron architecture. Due to specific conditions of melt water storing and releasing, two groups of models were established: the first is based on meteorological inputs only, while second includes the preceding day's mean discharge. Analysis of the multilayer perceptron simulation performance was done in comparison to the other black‐box model type, a multivariate regression method based on the following efficiency criteria: coefficient of determination (R2) and its adjusted form (adj. R2), weighted coefficient of determination (wR2), Nash–Sutcliffe coefficient of efficiency, mean absolute error, and error analysis. Moreover, the predictors' importance analysis for both multilayer perceptron and multivariate regression models was done. The performed study showed that the nonlinear estimation realized by the multilayer perceptron gives more accurate results than the multivariate regression approach in both groups of models.  相似文献   

6.
Assuming a study region in which each cell has associated with it an N-dimensional vector of values corresponding to N predictor variables, one means of predicting the potential of some cell to host mineralization is to estimate, on the basis of historical data, a probability density function that describes the distribution of vectors for cells known to contain deposits. This density estimate can then be employed to predict the mineralization likelihood of other cells in the study region. However, owing to the curse of dimensionality, estimating densities in high-dimensional input spaces is exceedingly difficult, and conventional statistical approaches often break down. This article describes an alternative approach to estimating densities. Inspired by recent work in the area of similarity-based learning, in which input takes the form of a matrix of pairwise similarities between training points, we show how the density of a set of mineralized training examples can be estimated from a graphical representation of those examples using the notion of eigenvector graph centrality. We also show how the likelihood for a test example can be estimated from these data without having to construct a new graph. Application of the technique to the prediction of gold deposits based on 16 predictor variables shows that its predictive performance far exceeds that of conventional density estimation methods, and is slightly better than the performance of a discriminative approach based on multilayer perceptron neural networks.  相似文献   

7.
Bathymetry is an important variable in scientific and operational applications. The research objectives in this study were to estimate bathymetry based on derivative reflectance spectra used as input to a multilayer perceptron artificial neural network (ANN) and to evaluate the efficacy of field and simulated training/testing data. ANNs were used to invert reflectance field data acquired in optically shallow coastal waters. Results indicate that for the simulation‐based models, nonderivative spectra yielded more accurate bathymetry retrievals than the derivative spectra used as ANN input. However, for the empirical field‐based models, derivative spectra were superior to nonderivative spectra as ANN input. This study identifies circumstances under which derivative spectra are useful in bathymetry estimation, and thus increases the likelihood of obtaining accurate inversions.  相似文献   

8.
A TEST OF SIGNIFICANCE FOR PARTIAL LEAST SQUARES REGRESSION   总被引：1，自引：0，他引：1  
Partial least squares (PLS) regression is a commonly used statistical technique for performingmultivariate calibration, especially in situations where there are more variables than samples. Choosingthe number of factors to include in a model is a decision that all users of PLS must make, but iscomplicated by the large number of empirical tests available. In most instances predictive ability is themost desired property of a PLS model and so interest has centred on making this choice based on aninternal validation process. A popular approach is the calculation of a cross-validated r~2 to gauge howmuch variance in the dependent variable can be explained from leave-one-out predictions. Using MonteCarlo simulations for different sizes of data set, the influence of chance effects on the cross-validationprocess is investigated. The results are presented as tables of critical values which are compared againstthe values of cross-validated r~2 obtained from the user's own data set. This gives a formal test forpredictive ability of a PLS model with a given number of dimensions.  相似文献   

9.
Spatiotemporal kriging (STK) is recognized as a fundamental space-time prediction method in geo-statistics. Spatiotemporal regression kriging (STRK), which combines space-time regression with STK of the regression residuals, is widely used in various fields, due to its ability to take into account both the external covariate information and spatiotemporal autocorrelation in the sample data. To handle the spatiotemporal non-stationary relationship in the trend component of STRK, this paper extends conventional STRK to incorporate it with an improved geographically and temporally weighted regression (I-GTWR) model. A new geo-statistical model, named geographically and temporally weighted regression spatiotemporal kriging (GTWR-STK), is proposed based on the decomposition of deterministic trend and stochastic residual components. To assess the efficacy of our method, a case study of chlorophyll-a (Chl-a) prediction in the coastal areas of Zhejiang, China, for the years 2002 to 2015 was carried out. The results show that the presented method generated reliable results that outperform the GTWR, geographically and temporally weighted regression kriging (GTWR-K) and spatiotemporal ordinary kriging (STOK) models. In addition, employing the optimal spatiotemporal distance obtained by I-GTWR calibration to fit the spatiotemporal variograms of residual mapping is confirmed to be feasible, and it considerably simplifies the residual estimation of STK interpolation.  相似文献   

10.
Most previous research on areas with abundant rainfall shows that simulations using rainfall-runoff modes have a very high prediction accuracy and applicability when using a back-propagation(BP), feed-forward, multilayer perceptron artificial neural network(ANN). However, in runoff areas with relatively low rainfall or a dry climate, more studies are needed. In these areas—of which oasis-plain areas are a particularly good example—the existence and development of runoff depends largely on that which is generated from alpine regions. Quantitative analysis of the uncertainty of runoff simulation under climate change is the key to improving the utilization and management of water resources in arid areas. Therefore, in this context, three kinds of BP feed-forward, three-layer ANNs with similar structure were chosen as models in this paper.Taking the oasis–plain region traverse by the Qira River Basin in Xinjiang, China, as the research area, the monthly accumulated runoff of the Qira River in the next month was simulated and predicted. The results showed that the training precision of a compact wavelet neural network is low; but from the forecasting results, it could be concluded that the training algorithm can better reflect the whole law of samples. The traditional artificial neural network(TANN) model and radial basis-function neural network(RBFNN) model showed higher accuracy in the training and prediction stage. However, the TANN model, more sensitive to the selection of input variables, requires a large number of numerical simulations to determine the appropriate input variables and the number of hidden-layer neurons. Hence, The RBFNN model is more suitable for the study of such problems. And it can be extended to other similar research arid-oasis areas on the southern edge of the Kunlun Mountains and provides a reference for sustainable water-resource management of arid-oasis areas.  相似文献   

11.
Most calibration data sets used to infer past environmental conditions from biological proxies are derived from many sites. An alternative strategy is to derive the calibration data set from within a single site. Transfer functions derived from such intra-site calibration data sets are usually applied to fossil assemblages from the focal lake, but a recent development has been to apply these transfer functions to other sites. Transfer functions derived from intra-site calibration data sets can have impressive cross-validation performance, but that gives little indication of their performance when applied to other sites. Here, we develop transfer functions for lake depth from intra-lake chironomid calibration data sets in Norway and Alaska and test the resulting models by cross-validation and against known depth in external lakes. Lake depth is a statistically significant predictor of chironomid assemblages at all these lakes, and most intra-lake transfer functions perform reasonably well under cross-validation, but their performance against external data is erratic. Downcore reconstructions from transfer functions developed on different lakes are dissimilar. Ignoring the poorly performing transfer functions, only 3 of 14 downcore reconstructions are statistically significant. Few assemblages downcore had good modern analogues in the calibration data set, even when the core was from the same lake as the calibration data set. We conclude that intra-site calibration data sets can find site-specific rather than general relationships between species and the environment and thus should be applied with care and to external sites only after careful and critical validation.  相似文献   

12.
Resource estimation of a placer deposit is always a difficult and challenging job because of high variability in the deposit. The complexity of resource estimation increases when drill-hole data are sparse. Since sparsely sampled placer deposits produce high-nugget variograms, a traditional geostatistical technique like ordinary kriging sometimes fails to produce satisfactory results. In this article, a machine learning algorithm—the support vector machine (SVM)—is applied to the estimation of a platinum placer deposit. A combination of different neighborhood samples is selected for the input space of the SVM model. The trade-off parameter of the SVM and the bandwidth of the kernel function are selected by genetic algorithm learning, and the algorithm is tested on a testing data set. Results show that if eight neighborhood samples and their distances and angles from the estimated point are considered as the input space for the SVM model, the developed model performs better than other configurations. The proposed input space-configured SVM model is compared with ordinary kriging and the traditional SVM model (location as input) for resource estimation. Comparative results reveal that the proposed input space-configured SVM model outperforms the other two models.  相似文献   

13.
Topodata: Brazilian full coverage refinement of SRTM data   总被引：2，自引：0，他引：2  
This work presents the selection of a set of geostatistical coefficients suitable for a unified SRTM data refinement from 3″ to 1″ through kriging over the entire Brazilian territory. This selection aimed at data potential for geomorphometric derivations, given by the preservation of detailed geometric characteristics of the resulting digital elevation models (DEM), which are sensitive to refining procedures. The development contained a long-term experimentation stage, when data refinement through kriging was locally developed to support distinct regional projects, followed by a unified selection stage, where the acquired experience was applied to select a single and unified interpolation scheme. In this stage, the selected geostatistical models with promising performances were tested for unified refinement on 40 Brazilian areas with distinct geological settings. Tested areas encompass reliefs varying from mountainous to plain. The effects of data preparation were observed on the perception of patterns (texture and roughness), as well as of singularities (edges, peaks, thalwegs etc.). Results were evaluated mainly through the examination of shaded reliefs, transects and perspectives observed in different scales. Terrains with low slopes and small amplitudes had their DEM promptly affected by the refining methods, as opposed to mountainous terrains. The evaluation, unambiguously confirmed by all consulted interpreters, converged into a refining model with outstanding performance in all tested conditions.  相似文献   

14.
This study addresses the question of what diatom taxa to includein a modern calibration set based on their relative contribution in apalaeolimnological calibration model. Using a pruning algorithm for ArtificialNeural Networks (ANNs) which determines the functionality of individual taxa interms of model performance, we pruned the Surface Water Acidification Project(SWAP) pH-diatom data-set until the predictive performance of thepruned set (as assessed by a jackknifing procedure) was statistically differentfrom the initial full-set. Our results, based on the validation at each5% data-set reduction, show that (i) 85% of the taxa canbe removed without any effect on the pH model calibration performance, and (ii)that the complexity and the dimensionality reduction of the model by theremoval of these non-essential or redundant taxa greatly improve therobustness of the calibration. A comparison between the commonly usedmarginal criteria for inclusion (species tolerance andHill's N2) and our functionality criterion shows that the importance ofeach taxon in an ANN palaeolimnological model calibration does not appear todepend on these marginal characteristics.  相似文献   

15.
Multiple Random Walk Simulation consists of a methodology adapted to run fast simulations if close-spaced data are abundant (e.g., short-term mining models). Combining kriging with the simulation of random walks attempts to approximate traditional simulation algorithm results but at a computationally faster way when there is a large amount of conditioning samples. This paper presents this new algorithm illustrating the situations where the method can be used properly. A synthetic study case is presented in order to illustrate the Multiple Random Walk Simulation and to analyze the speed and goodness of its results against the ones from using Turning Bands Simulation and Sequential Gaussian Simulation.  相似文献   

16.
A landslide susceptibility analysis is performed by means of Artificial Neural Network (ANN) and Cluster Analysis (CA). This kind of analysis is aimed at using ANNs to model the complex non linear relationships between mass movements and conditioning factors for susceptibility zonation, in order to identify unstable areas. The proposed method adopts CA to improve the selection of training, validation, and test records from data, managed within a Geographic Information System (GIS). In particular, we introduce a domain-specific distance measure in cluster formation. Clustering is used in data pre-processing to select non landslide records and is performed on the whole dataset, excluding the test set landslides. Susceptibility analysis is carried out by means of ANNs on the so-generated data and compared with the common strategy to select random non-landslide samples from pixels without landslides. The proposed method has been applied in the Brembilla Municipality, a landslide-prone area in the Southern Alps, Italy. The results show significant differences between the two sampling methods: the classification of the test set, previously separated and excluded from the training data, is always better when the non-landslide patterns are obtained using the proposed cluster sampling. The case study validates that, by means of a domain-specific distance measure in cluster formation, it is possible to introduce expert knowledge into the black-box modelling method, implemented by ANNs, to improve the predictive capability and the robustness of the models obtained.  相似文献   

17.
A baseline climatology is required in evaluating climate variability and changes on regional and local scales. Gridded climate normals, i.e. averages over a 30‐year period, are of special interest since they can be readily used for validation of climate models. This study is aimed at creating an updated gridded dataset for Swedish monthly temperature normals over the period 1971–2000, based on standard 2‐m air temperature records at 510 stations in mainland Sweden. Spatial trends of the normal temperatures were modelled as functions of latitude, longitude and elevation by multiple linear regression. The study shows that the temperature normals are strongly correlated with latitude throughout the year and especially in cold months, while elevation was a more important factor in June and July. Longitude played a minor role and was only significant in April and May. Regression equations linking temperature to latitude, longitude and elevation were set up for each month. Monthly temperature normals were detrended by subtracting spatial trends given by the regressions. Ordinary kriging was then applied to both original data (simple method) and de‐trended data (composite method) to model the spatial variability and to perform spatial gridding. The multiple regressions showed that between 82% (summer) and 96% (winter) of the variance in monthly temperature normals could be explained by latitude and elevation. Unexplained variances, i.e. the residuals, were modelled with ordinary kriging with exponential semivariograms. The composite grid estimates were calculated by adding the multiple linear trends back to the interpolated residuals at each grid point. Kriged original temperature normals provided a performance benchmark. The cross–validation shows that the interpolation errors of the normals are significantly reduced if the composite method rather than the simple one was used. A gridded monthly dataset with 30‐arcsecond spacing was created using the established trends, the kriging model and a digital topographic dataset.  相似文献   

18.
Mineral-potential mapping is the process of combining a set of input maps, each representing a distinct geo-scientific variable, to produce a single map which ranks areas according to their potential to host mineral deposits of a particular type. The maps are combined using a mapping function that must be either provided by an expert (knowledge-driven approach), or induced from sample data (data-driven approach). Current data-driven approaches using multilayer perceptrons (MLPs) to represent the mapping function have several inherent problems: they are highly sensitive to the selection of training data; they do not utilize the contextual information provided by nondeposit data; and there is no objective interpretation of the values output by the MLP. This paper presents a new approach by which MLPs can be trained to output values that can be interpreted strictly as representing posterior probabilities. Other advantages of the approach are that it utilizes all data in the construction of the model, and thus eliminates any dependence on a particular selection of training data. The technique is applied to mapping gold mineralization potential in the Castlemaine region of Victoria, Australia, and results are compared with a method based on estimating probability density functions.  相似文献   

19.
洪水过程的特征指标不仅包括洪水量级,还包括时间、形态、动力学等指标。现有模型和方法重点关注洪水量级指标的模拟,对其他指标的模拟仍有待深入。如何实现对洪水过程所有特征指标的模拟已成为目前洪水预报的技术瓶颈。论文采用4种机器学习模型(多元线性回归、多层感知器、随机森林和支持向量机)对淮河上游长台关流域59场降雨—洪水场次7个特征指标(洪水总量、洪峰流量、洪水历时、洪峰时间偏度、高流量历时占比、涨洪和落洪速率)进行模拟,评估不同模型对不同洪水类型和特征指标的模拟效果。结果显示：① 长台关流域洪水过程可分为3类,第1类洪量中等、历时长且洪峰出现时间偏前(16场);第2类洪量低、形态矮胖且洪峰出现时间靠后(34场);第3类洪量大、涨落水迅速、形态尖瘦(9场)。② 时间指标模拟效果最优,动力学指标模拟效果最差。多元线性回归和随机森林模拟效果随所有特征指标数值的增加而增强;支持向量机的模拟效果随着洪水历时指标数值的增加而降低,随着其余特征指标数值的增加而增强;多层感知器模拟效果随着洪水总量、洪峰流量、高流量历时占比和涨洪速率等指标值的增加而增强。③ 从各类型洪水特征模拟精度来看,4种模型对第3类洪水特征模拟均为最佳,第2类最差;随机森林在第1类和第3类洪水特征模拟中效果最优,支持向量机对第2类洪水特征模拟效果最优。④ 从综合模拟精度来看,支持向量机效果最优,然后依次为随机森林、多层感知器和多元线性回归。上述4种模型率定和验证期相对误差分别为23%和98%、21%和109%、37%和75%、41%和102%。研究可为流域洪水过程深度挖掘和防洪措施制定提供参考和借鉴。  相似文献   

20.
Jeuken  Rick  Xu  Chaoshui  Dowd  Peter 《Natural Resources Research》2020,29(4):2529-2546

In most modern coal mines, there are many coal quality parameters that are measured on samples taken from boreholes. These data are used to generate spatial models of the coal quality parameters, typically using inverse distance as an interpolation method. At the same time, downhole geophysical logging of numerous additional boreholes is used to measure various physical properties but no coal quality samples are taken. The work presented in this paper uses two of the most important coal quality variables—ash and volatile matter—and assesses the efficacy of using a number of geostatistical interpolation methods to improve the accuracy of the interpolated models, including the use of auxiliary variables from geophysical logs. A multivariate spatial statistical analysis of ash, volatile matter and several auxiliary variables is used to establish a co-regionalization model that relates all of the variables as manifestations of an underlying geological characteristic. A case study of a coal mine in Queensland, Australia, is used to compare the interpolation methods of inverse distance to ordinary kriging, universal kriging, co-kriging, regression kriging and kriging with an external drift. The relative merits of these six methods are compared using the mean error and the root mean square error as measures of bias and accuracy. The study demonstrates that there is significant opportunity to improve the estimations of coal quality when using kriging with an external drift. The results show that when using the depth of a sample as an external drift variable there is a significant improvement in the accuracy of estimation for volatile matter, and when using wireline density logs as the drift variable there is improvement in the estimation of the in situ ash. The economic benefit of these findings is that cheaper proxies for coal quality parameters can significantly increase data density and the quality of estimations.

  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号