基于PSR-WSVM模型的边坡位移预测

为建立高精度的边坡位移预测模型，采用相空间重构（PSR）将边坡位移时间序列数据转换为多维数据，同时构造小波核函数改进的支持向量机模型，建立PSR-WSVM模型并应用于边坡位移预测。将PSR-WSVM模型预测结果与传统支持向量机（SVM）模型、小波支持向量机（WSVM）模型和基于相空间重构的支持向量机（PSR-SVM）模型预测结果进行对比，通过平均绝对误差（MAE）、平均绝对误差百分比（MAPE）和均方根误差（RMSE）3个精度评价指标验证PSR-WSVM模型的可行性。工程实例结果表明，PSR-WSVM模型预测结果的3个精度评价指标都优于另外3种模型，边坡位移预测的精度明显提升。     

Subsurface flow measurements using passive flux meters in variably-saturated cold-regions landscapes

To date, passive flux meters have predominantly been applied in temperate environments for tracking the movement of contaminants in groundwater. This study applies these instruments to reduce uncertainty in (typically instantaneous) flux measurements made in a low-gradient, wetland dominated, discontinuous permafrost environment. This method supports improved estimation of unsaturated and over-winter subsurface flows which are very difficult to quantify using hydraulic gradient-based approaches. Improved subsurface flow estimates can play a key role in understanding the water budget of this landscape.     

Solution for a plane strain rough‐walled hydraulic fracture driven by turbulent fluid through impermeable rock

The impact of turbulent flow on plane strain fluid‐driven crack propagation is an important but still poorly understood consideration in hydraulic fracture modeling. The changes that hydraulic fracturing has experienced over the past decade, especially in the area of fracturing fluids, have played a major role in the transition of the typical fluid regime from laminar to turbulent flow. Motivated by the increasing preponderance of high‐rate, water‐driven hydraulic fractures with high Reynolds number, we present a semianalytical solution for the propagation of a plane strain hydraulic fracture driven by a turbulent fluid in an impermeable formation. The formulation uses a power law relationship between the Darcy‐Weisbach friction factor and the scale of the fracture roughness, where one specific manifestation of this generalized friction factor is the classical Gauckler‐Manning‐Strickler approximation for turbulent flow in a rough‐walled channel. Conservation of mass, elasticity, and crack propagation are also solved simultaneously. We obtain a semianalytical solution using an orthogonal polynomial series. An approximate closed‐form solution is enabled by a choice of orthogonal polynomials embedding the near‐tip asymptotic behavior and thus giving very rapid convergence; a precise solution is obtained with 2 terms of the series. By comparison with numerical simulations, we show that the transition region between the laminar and turbulent regimes can be relatively small so that full solutions can often be well approximated by either a fully laminar or fully turbulent solution.     

Geostatistical features of streambed vertical hydraulic conductivities in Frenchman Creek Watershed in Western Nebraska

This study evaluated the spatial variability of streambed vertical hydraulic conductivity (K_v) in different stream morphologies in the Frenchman Creek Watershed, Western Nebraska, using different variogram models. Streambed K_v values were determined in situ using permeameter tests at 10 sites in Frenchman, Stinking Water and Spring Creeks during the dry season at baseflow conditions. Measurements were taken both in straight and meandering stream channels during a 5 day period at similar flow conditions. Each test site comprised of at least three transects and each transect comprised of at least three K_v measurements. Linear, Gaussian, exponential and spherical variogram models were used with Kriging gridding method for the 10 sites. As a goodness-of-fit statistic for the variogram models, cross-validation results showed differences in the median absolute deviation and the standard deviation of the cross-validation residuals. Results show that using the geometric means of the 10 sites for gridding performs better than using either all the K_v values from the 93 permeameter tests or 10 K_v values from the middle transects and centre permeameters. Incorporating both the spatial variability and the uncertainty involved in the measurement at a reach segment can yield more accurate grid results that can be useful in calibrating K_v at watershed or sub-watershed scales in distributed hydrological models.     

A machine learning approach for predicting computational intensity and domain decomposition in parallel geoprocessing

ABSTRACT

High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning.     

地下洞库勘察中的深斜孔钻探和压水试验技术

青岛市黄岛区拟建设海底地下油气储藏库,前期地质勘察工作要求施工倾角45°、深220m的取心钻孔4个,并在孔内分段进行压水试验.本文介绍了如何利用XY-4型金刚石小口径钻机施工基岩深斜孔及压水试验技术.     

Automatic Determination of Stellar Atmospheric Parameters Using Neural Networks and Instance-Based Machine Learning

In this article we show how machine learning methods can beeffectively applied to the problem of automatically predictingstellar atmospheric parameters from spectral information, a veryimportant problem in stellar astronomy. We apply feedforwardneural networks, Kohonen's self-organizing maps andlocally-weighted regression to predict the stellar atmosphericparameters effective temperature, surface gravity and metallicityfrom spectral indices. Our experimental results show that thethree methods are capable of predicting the parameters with verygood accuracy. Locally weighted regression gives slightly betterresults than the other methods using the original dataset asinput, while self-organizing maps outperform the other methods when significant amounts of noise are added. We also implemented a heterogeneous ensemble of predictors, combining the results given by the three algorithms. This ensemble yields better results than any of the three algorithms alone, using both the original and the noisy data.     

Prediction of Stellar Atmospheric Parameters using Instance-Based Machine Learning and Genetic Algorithms

In this article we present a method for the automated prediction of stellar atmospheric parameters from spectral indices. This method uses a genetic algorithm (GA) for the selection of relevant spectral indices and prototypical stars and predicts their properties, using the k-nearest neighbors method (KNN). We have applied the method to predict the effective temperature, surface gravity, metallicity, luminosity class and spectral class of stars from spectral indices. Our experimental results show that the feature selection performed by the genetic algorithm reduces the running time of KNN up to 92%, and the predictive accuracy error up to 35%. This revised version was published online in July 2006 with corrections to the Cover Date.     

Optimization of grid-drilling using computer simulation

A computer simulation method has been developed to find efficient drilling grids for mineral deposits. A well-known ore deposit is used as a model to develop an efficient pattern for undiscovered ore bodies in the same area or in other prospects where similar geometry is suspected. The model for this study is the Austinville, Virginia deposit, a Mississippi Valley-type deposit composed of 17 ore bodies totaling 34 million short tons (30 million metric tons). The method employs a computer program that simulates drilling the model deposit with different patterns, including various levels of follow-up drilling. Follow-up holes are drilled in fences at one half the original spacing around holes in the grid that show ore-grade mineralization. Each pattern is drilled 100 times from random starting locations to provide a range of outcomes of drilling, including the best, worst, and most likely. For this study, patterns of 100 drill holes were composed of 10 fences spaced 1000–5000 feet (305–1524 m) apart, each with 10 holes spaced 200–1000 feet (61–305 m) apart. In all, 25 grids were used with zero to three levels of follow-up drilling. The 600/2000 grid, with drill holes spaced 600 feet (183 m) apart in fences spaced 2000 feet (610 m) apart, was compared with the 200/5000 grid because they represented contrasting outcomes. The 600/2000 grid penetrated many ore bodies consistently but with few multiple hits to individual ore bodies; whereas the 200/5000 grid inconsistently penetrated few ore bodies with many multiple hits. The 600/2000 grid was more efficient than the 200/5000 grid at hitting large ore bodies of 1,000,000 short tons or greater (900,000 metric tons or greater) and was made more effective by adding one cycle of follow-up drilling. The 600/2000 grid had a 97% chance of hitting one or more large ore bodies with at least one drill hole per ore body, and the 200/5000 grid had a 64% chance. Once hit, there was an 82% chance that the largest ore body would be penetrated by three or more holes when using the 600/2000 grid and an 88% chance using the 200/5000 grid.