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 31 毫秒
1.
Durability is a notion that is integrated with the performance of stabilized pavement materials. Also, because it can be quantified and measured, it carries significant influence on the design of pavements. This study focuses on using support vector machine, a machine learning algorithm, in assessing the performance of stabilized aggregate bases subject to wet–dry cycles. Support Vector Regression (SVR) is a statistical learning algorithm that is applied to regression problems and is gaining popularity in pavement and geotechnical engineering. In our study, SVR was shown to be superior to the least‐squares (LS) method. Results of this study show that SVR significantly reduces the mean‐squared error (MSE) and improves the coefficient of determination (R2) compared to the widely used LS method. Copyright © 2011 John Wiley & Sons, Ltd.  相似文献   

2.
Bathymetric information for shallow coastal/lake areas is essential for hydrological engineering applications such as sedimentary processes and coastal studies. Remotely sensed imagery is considered a time-effective, low-cost, and wide-coverage solution for bathymetric measurements. This study assesses the performance of three proposed empirical models for bathymetry calculations in three different areas: Alexandria port, Egypt, as an example of a low-turbidity deep water area with silt-sand bottom cover and a depth range of 10.5 m; the Lake Nubia entrance zone, Sudan, which is a highly turbid, unstable, clay bottom area with water depths to 6 m; and Shiraho, Ishigaki Island, Japan, a coral reef area with varied depths ranging up to 14 m. The proposed models are the ensemble regression tree-fitting algorithm using bagging (BAG), ensemble regression tree-fitting algorithm of least squares boosting (LSB), and support vector regression algorithm (SVR). Data from Landsat 8 and Spot 6 satellite images were used to assess the performance of the proposed models. The three models were used to obtain bathymetric maps using the reflectance of green, red, blue/red, and green/red band ratios. The results were compared with corresponding results yielded by two conventional empirical methods, the neural network (NN) and the Lyzenga generalised linear model (GLM). Compared with echosounder data, BAG, LSB, and SVR results demonstrate higher accuracy ranges from 0.04 to 0.35 m more than Lyzenga GLM. The BAG algorithm, producing the most accurate results, proved to be the preferable algorithm for bathymetry calculation.  相似文献   

3.
The determination of the compaction parameters such as optimum water content (wopt) and maximum dry unit weight (γdmax) requires great efforts by applying the compaction testing procedure which is also time consuming and needs significant amount of work. Therefore, it seems more reasonable to use the indirect methods for estimating the compaction parameters. In recent years, the artificial neural network (ANN) modelling has gained an increasing interest and is also acquiring more popularity in geotechnical engineering applications. This study deals with the estimation of the compaction parameters for fine‐grained soils based on compaction energy using ANN with the feed‐forward back‐propagation algorithm. In this study, the data including the results of the consistency tests, standard and modified Proctor tests, are collected from the literature and used in the analyses. The optimum structure of a network is determined for each ANN models. The analyses showed that the ANN models give quite reliable estimations in comparison with regression methods, thus they can be used as a reliable tool for the prediction of wopt and γdmax. Copyright © 2010 John Wiley & Sons, Ltd.  相似文献   

4.
In this study, the preprocessing of the gamma test was used to select the appropriate input combination into two models including the support vector regression (SVR) model and artificial neural networks (ANNs) to predict the stream flow drought index (SDI) of different timescales (i.e., 3, 6, 9, 12, and 24 months) in Latian watershed, Iran, which is one of the most important sources of water for the large metropolitan Tehran. The variables used included SDI t , SDI t ? 1, SDI t ? 2, SDI t ? 3, and SDI t ? 4 monthly delays. Two variables including SDI t and SDI t ? 1 with lower gamma values were identified as the most optimal combination of variables in all drought timescales. The results showed that the gamma test was able to correctly identify the right combination for the forecasting of 6, 9, and 12 months SDI using the ANN model. Also, the gamma test was considered in selecting the appropriate inputs for identifying the values of 9, 12, and 24 months SDI in SVR. The support vector machine approach showed a better efficiency in the forecast of long-term droughts compared to the artificial neural network. In total, among forecasts made for 30 scenarios, the support vector machine model only in scenario 3 of SDI3, scenario 1 of SDI6, and scenarios 2 and 3 of SDI24 represented poorer efficiency compared to the artificial neural network (MLP layer), but in other scenarios, the results of SVR had better efficiency.  相似文献   

5.
This paper presents the use of the Low Memory Locality Sensitive Hashing (LMLSH) technique operating in Euclidean space to build a data structure for the Defense Meteorological Satellite Program (DMSP) satellite imagery database. The LMLSH technique finds satellite image matches in sublinear search time. The texture feature vectors of the images are extracted using pyramid-structured wavelet transform coupled with Gaussian central moment technique. These feature vectors and families of hash functions, drawn randomly and independently from a Gaussian distribution, are used to build hash tables. Given a query, the hash tables are used to pull out the best matches to that query and this is done in a sublinear search time complexity. When tested, our algorithm has proven to be approximately twenty six times faster than the Linear Search (LS) algorithm. In addition, the LMLSH algorithm searches about two percent of the entire database randomly to find the possible matches to any given query without loss of accuracy compared to the absolute best matches returned by its LS counterpart.  相似文献   

6.
The numerical algorithm of calculation of Lyapounov coefficients (L k) of any order is developed. The apparatus of analytical calculations is not used in this algorithm. The proposed algorithm is of use for usual computer languages and allows us to find the numerical value of L k for any k and to make complete qualitative analyses of dynamic models on the plane.  相似文献   

7.
一种基于核学习的储集层渗透率预测新方法   总被引:2,自引:1,他引:2  
基于核学习的支持向量机,是一种采用结构风险最小化原则代替传统经验风险最小化原则的新型统计学习方法,具有完备的理论基础。这里提出了核学习技术在储集层非均质特性描述中渗透率参数预测的新用途。在复杂地层中,基于支持向量机的智能和自适应模式识别能力而建立了常规测井多参数信息输入的渗透率预测模型,然后对实际油田储集层渗透率进行了预测。与常规线性回归模型预测结果相对比,所提出的方法更易于使用,很少受不确定因素的影响,并具有较强的信息整合能力以及更高的预测准确性和可信度。  相似文献   

8.
Statistical learning algorithms provide a viable framework for geotechnical engineering modeling. This paper describes two statistical learning algorithms applied for site characterization modeling based on standard penetration test (SPT) data. More than 2700 field SPT values (N) have been collected from 766 boreholes spread over an area of 220 sqkm area in Bangalore. To get N corrected value (Nc), N values have been corrected (Nc) for different parameters such as overburden stress, size of borehole, type of sampler, length of connecting rod, etc. In three‐dimensional site characterization model, the function Nc=Nc (X, Y, Z), where X, Y and Z are the coordinates of a point corresponding to Nc value, is to be approximated in which Nc value at any half‐space point in Bangalore can be determined. The first algorithm uses least‐square support vector machine (LSSVM), which is related to a ridge regression type of support vector machine. The second algorithm uses relevance vector machine (RVM), which combines the strengths of kernel‐based methods and Bayesian theory to establish the relationships between a set of input vectors and a desired output. The paper also presents the comparative study between the developed LSSVM and RVM model for site characterization. Copyright © 2009 John Wiley & Sons, Ltd.  相似文献   

9.
In this study, we developed multiple hybrid machine-learning models to address parameter optimization limitations and enhance the spatial prediction of landslide susceptibility models. We created a geographic information system database, and our analysis results were used to prepare a landslide inventory map containing 359 landslide events identified from Google Earth, aerial photographs, and other validated sources. A support vector regression (SVR) machine-learning model was used to divide the landslide inventory into training (70%) and testing (30%) datasets. The landslide susceptibility map was produced using 14 causative factors. We applied the established gray wolf optimization (GWO) algorithm, bat algorithm (BA), and cuckoo optimization algorithm (COA) to fine-tune the parameters of the SVR model to improve its predictive accuracy. The resultant hybrid models, SVR-GWO, SVR-BA, and SVR-COA, were validated in terms of the area under curve (AUC) and root mean square error (RMSE). The AUC values for the SVR-GWO (0.733), SVR-BA (0.724), and SVR-COA (0.738) models indicate their good prediction rates for landslide susceptibility modeling. SVR-COA had the greatest accuracy, with an RMSE of 0.21687, and SVR-BA had the least accuracy, with an RMSE of 0.23046. The three optimized hybrid models outperformed the SVR model (AUC = 0.704, RMSE = 0.26689), confirming the ability of metaheuristic algorithms to improve model performance.  相似文献   

10.
Numerical simulations are essential tools for studying tsunami generation and evolution and finite-element (FE) methods are widely used, especially because of their capability in modeling water waves in basins with complex bathymetry and irregular coastlines. This paper presents the numerical simulation of an historical Italian tsunami that affected the Tyrrhenian coasts of Calabria and Sicily on 5 February 1783 following a strong destructive earthquake that was the first of a terrible sequence of seismic shocks terrifying the Calabrian population for more than two months. The numerical model is an FE model based on the nonlinear nondispersive shallow-water approximation of the Navier-Stokes equations. Since FE discretization schemes may lead to solutions undesirably affected by noise over coarse grids, in this study numerical noise is controlled by suitably smoothing the FE solution at regular time steps t s. The performance of our smoothing algorithm is tested for significant linear cases for which an analytical solution is available.  相似文献   

11.
Nonlinear complex behavior of pore-water pressure responses to rainfall was modelled using support vector regression (SVR). Pore-water pressure can rise to disturbing levels that may result in slope failure during or after rainfall. Traditionally, monitoring slope pore-water pressure responses to rainfall is tedious and expensive, in that the slope must be instrumented with necessary monitors. Data on rainfall and corresponding responses of pore-water pressure were collected from such a monitoring program at a slope site in Malaysia and used to develop SVR models to predict pore-water pressure fluctuations. Three models, based on their different input configurations, were developed. SVR optimum meta-parameters were obtained using k-fold cross validation and a grid search. Model type 3 was adjudged the best among the models and was used to predict three other points on the slope. For each point, lag intervals of 30 min, 1 h and 2 h were used to make the predictions. The SVR model predictions were compared with predictions made by an artificial neural network model; overall, the SVR model showed slightly better results. Uncertainty quantification analysis was also performed for further model assessment. The uncertainty components were found to be low and tolerable, with d-factor of 0.14 and 74 % of observed data falling within the 95 % confidence bound. The study demonstrated that the SVR model is effective in providing an accurate and quick means of obtaining pore-water pressure response, which may be vital in systems where response information is urgently needed.  相似文献   

12.
Creepex, defined as the systematic deviation of the magnitude of a single earthquake from the linear orthogonal regression between local magnitude ML and coda duration magnitude Md, calculated for the whole region, is used as a measure of the frequency content of the seismic sources in the Italian region. Predominantly high-frequency events are found in the two areas of Quaternary tectonic shortening in North-Central Italy and in the Calabrian Arc. This result, confirmed by two independent statistical tests, is in agreement with the global pattern obtained from the study of the regression between body-wave magnitude, mb, and surface-wave magnitude Ms: systematic shift to high frequencies in the energy release of seismic sources located in subduction zones and to low frequencies in zones of spreading. The analysis of the correlation between the patterns of heat flow and of seismic source spectral properties indicates that these source properties, in general, do not reflect thermal conditions in the lithosphere, but rather represent the result of tectonic processes.  相似文献   

13.
The creep strain is proportional to the logarithm of the time under load, and is proportional to the stress and the temperature. At higher temperatures the creep rate falls off less rapidly with time, and the creep strain is proportional to a fractional power of time, with exponent increasing as the temperature increases and reaching a value ∼1/3 at temperatures, of about 0.5 T m. At these temperatures the creep increases with stress according to a power greater than unity and possibly exponentially increases with temperature as (−U/kT), where U is an activation energy and k is Boltzman’s constant. There are different methods to determine the creep strain and the energy of Jog (B) such as by experimental methods and multivariate regression analysis etc. These methods are cumbersome and time consuming. In conjunction with statistics and conventional mathematical methods, a hybrid method can be developed that may prove a step forward in modeling geotechnical problems. In the present investigation, Artificial Neural Network (ANN) technique and Co-active neuro-fuzzy inference system (CANFIS) backed Genetic algorithm technique have been used for the prediction of creep strain and energy of Jog (B), and a comparative study has made between the two models.  相似文献   

14.
An accurate estimation of flow using different models is an issue for water resource researchers. In this study, support vector regression (SVR) and gene expression programming (GEP) models in daily and monthly scale were used in order to simulate Gamasiyab River flow in Nahavand, Iran. The results showed that although the performance of models in daily scale was acceptable and the result of SVR model was a little better, their performance in the daily scale was really better than the monthly scale. Therefore, wavelet transform was used and the main signal of every input was decomposed. Then, by using principal component analysis method, important sub-signals were recognized and used as inputs for the SVR and GEP models to produce wavelet-support vector regression (WSVR) and wavelet-gene expression programming. The results showed that the performance of WSVR was better than the SVR in such a way that the combination of SVR with wavelet could improve the determination coefficient of the model up to 3% and 18% for daily and monthly scales, respectively. Totally, it can be said that the combination of wavelet with SVR is a suitable tool for the prediction of Gamasiyab River flow in both daily and monthly scales.  相似文献   

15.
Northeast India region is one of the most seismically active areas in the world. Events data for the period 1897–2010, used in this study has been largely compiled from global ISC, NEIC and GCMT databases. Historical seismicity catalogue of Gupta et al (1986) and some events data from the bulletins of India Meteorological Department are also used. Orthogonal regression relations for conversion of body and surface wave magnitudes to M w,HRVD based on events data for the period 1978–2006 have been derived. An Orthogonal Standard Regression (OSR) relationship has also been obtained for scaling of intensity estimates to M w,NEIC using 126 global intensity events with intensity VI or greater during the period 1975–2010.  相似文献   

16.
Yadav  Ganesh  Singh  R. B.  Anand  Subhash  Pandey  B. W.  Mohanty  Ashutosh  Dash  Sushree Sangita 《GeoJournal》2021,87(4):469-483

Ambient air pollution, particularly in the urban environment of developing countries, has turned out to be a major health risk factor. We explore the compounded impact of age sensitivity, exposure, poverty, co-morbidity, etc., along with composite air pollution in determining morbidity and health burden of people in Lucknow, India. This cross-sectional study is confined to analyse respiratory health status across different socio-economic and geographic locations using n = 140 in-depth questionnaire method. We used mean daily ambient air pollution data of PM10, PM2.5, SO2, and NO2 for the 2008–2018 period. We used the ecological model framework to assess the risk at different hierarchical levels and compounded severity on a spatial scale. We also used Logistic regression model with log odds and odds ratio to analyze the association of risks outcomes with composite air pollution scores calculated using the principal component analysis method. There is a strong association of location-specific respiratory disease prevalence with an overall 32 percent prevalence. The prevalence of ecological model 1 (individual domain) is 4.3 percent, while ecological model 2 (community domain) has the highest prevalence at 32.4 percent. The logistic regression model shows that respiratory disease load is positively associated with age sensitivity (P < .001) and composite pollution level (P < .001). For another model with suffocation as the outcome variable, composite pollution level (P < .001) and exposure (P < .001) are positively associated. Optimum interventions are required at Ecological models 1, 2, and 3 levels for better respiratory health outcomes.

  相似文献   

17.
In the estimation of momentum fluxes over land surfaces by the bulk aerodynamic method, no unique value of the drag coefficient (C D) is found in the literature. The drag coefficient is generally estimated from special observations at different parts of the world. In this study an attempt is made to estimate drag coefficient over the western desert sector of India using data sets of Monsoon Trough Boundary Layer Experiment (MONTBLEX) during the summer monsoon season of 1990. For this purpose, the fast and slow response data sets obtained simultaneously from a 30 m high micro-meteorological tower at Jodhpur are used. All the observations used in this study are confined to a wind speed regime of 2.5–9.0 ms−1. A comparison of momentum fluxes computed by eddy correlation (direct estimation) with profile and bulk aerodynamic (C D = 3.9 × 10−3, Garratt, 1977) methods revealed that though the nature of variation of the fluxes by all these methods is almost similar, both the indirect methods give an under-estimated value of the fluxes. The drag coefficient is estimated as a function of wind speed and surface stability by a multiple regression approach. An average value of the estimated drag coefficient is found to be of the order of 5.43 × 10−3. The estimated value ofC D is validated with a set of independent observations and found to be quite satisfactory. The recomputed momentum fluxes by bulk aerodynamic method using the estimated drag coefficient are in close agreement with the directly estimated fluxes.  相似文献   

18.
Earthquake hazard zonation of Sikkim Himalaya using a GIS platform   总被引:2,自引:1,他引:1  
An earthquake hazard zonation map of Sikkim Himalaya is prepared using eight thematic layers namely Geology (GE), Soil Site Class (SO), Slope (SL), Landslide (LS), Rock Outcrop (RO), Frequency–Wavenumber (F–K) simulated Peak Ground Acceleration (PGA), Predominant Frequency (PF), and Site Response (SR) at predominant frequencies using Geographic Information System (GIS). This necessitates a large scale seismicity analysis for seismic source zone classification and estimation of maximum earthquake magnitude or maximum credible earthquake to be used as a scenario earthquake for a deterministic or quasi-probabilistic seismic scenario generation. The International Seismological Center (ISC) and Global Centroid Moment Tensor (GCMT) catalogues have been used in the present analysis. Combining b-value, fractal correlation dimension (Dc) of the epicenters and the underlying tectonic framework, four seismic source zones are classified in the northeast Indian region. Maximum Earthquake of M W 8.3 is estimated for the Eastern Himalayan Zone (EHZ) and is used to generate the seismic scenario of the region. The Geohazard map is obtained through the integration of the geological and geomorphological themes namely GE, SO, SL, LS, and RO following a pair-wise comparison in an Analytical Hierarchy Process (AHP). Detail analysis of SR at all the recording stations by receiver function technique is performed using 80 significant events recorded by the Sikkim Strong Motion Array (SSMA). The ground motion synthesis is performed using F–K integration and the corresponding PGA has been estimated using random vibration theory (RVT). Testing for earthquakes of magnitude greater than M W 5, a few cases presented here, establishes the efficacy and robustness of the F–K simulation algorithm. The geohazard coverage is overlaid and sequentially integrated with PGA, PF, and SR vector layers, in order to evolve the ultimate earthquake hazard microzonation coverage of the territory. Earthquake Hazard Index (EHI) quantitatively classifies the terrain into six hazard levels, while five classes could be identified following the Bureau of Indian Standards (BIS) PGA nomenclature for the seismic zonation of India. EHI is found to vary between 0.15 to 0.83 quantitatively classifying the terrain into six hazard levels as “Low” corresponding to BIS Zone II, “Moderate” corresponding to BIS Zone III, “Moderately High” belonging to BIS Zone IV, “High” corresponding to BIS Zone V(A), “Very High” and “Severe” with new BIS zones to Zone V(B) and V(C) respectively.  相似文献   

19.
The bivalve Pisidium amnicum (Müller 1774) is a common species in several European freshwater ecosystems. However, few Iberian watersheds are colonized by this species, and the River Minho estuary is possibly the Iberian aquatic ecosystem with the larger population. In October 2004–2007, investigations on spatial and temporal variations in P. amnicum abundance and biomass were carried out at 16 sites along the River Minho tidal freshwater wetlands. Mean abundance and biomass per site ranged from 0 to 750 ind m−2 and 0 to 7.42 g AFDW m−2, respectively. A clear decrease in the spatial distribution, abundance, and biomass was observed during the 4-year assessment. Furthermore, a stepwise multiple regression model showed that organic matter and conductivity explained 50.2% of the variation in P. amnicum abundance (R 2 = 0.502, F [2, 15] = 7.569, p = 0.005). Ecological knowledge is essential to the implementation of future conservation plans for P. amnicum, and the results of this study are of paramount importance to identify habitats that should be protected in order to preserve this species and provide scientific reference that may be useful in the development of management and/or restoration plans.  相似文献   

20.
The main objectives of this study were to investigate conditions for stable and metastable liquid immiscibility in dry borosilicate synthetic systems and to evaluate effects of temperature and bulk melt composition on two-liquid element partitioning and boron speciation. To distinguish between the stable immiscibility and quench heterogeneity, we used high-temperature centrifuge phase separation. For the case of stable liquid immiscibility, silica-rich (LS) and borate-rich (LB) conjugate liquids formed two distinct layers separated by a sharp meniscus. The liquids were quenched into glasses, which were analysed by electron microprobe. Some of the glasses were also studied by Raman spectroscopy. We used several synthetic mixtures along the danburite-anorthite (CaB2Si2O8-CaAl2Si2O8) and danburite-reedmergnerite (CaB2Si2O8-NaBSi3O8) joins. In addition, we studied four complex, six-component, Mg-bearing compositions with variable Na2O and Al2O3 contents. The experiments show that the width of the LS-LB miscibility gap decreases more rapidly with the B-Al substitution (in the danburite-anorthite join) than with the Ca-Na substitution, implying that interactions between network-forming elements have a greater effect on borate-silicate unmixing than the nature of network-modifying cations. Ca and Mg partition strongly to the depolymerised borate-rich liquid with LB-LS partition coefficients of ∼40 and higher. On the other hand, two-liquid partition coefficients of Na and Al in most cases are close to 1 and show complex variations with temperature and bulk melt composition. Raman spectra of LB glasses quenched at different temperatures suggest that the proportion of trigonal boron in bulk boron content decreases with decreasing temperature. The change in boron speciation appears to affect Al and Na two-liquid partitioning in such a way that at low temperatures, the latter element becomes more compatible with LS.  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