偏最小二乘回归神经网络的矿坑涌水量预测     

陈南祥  曹连海  李梅  黄强 《吉林大学学报(地球科学版)》2005,35(6):766-770

影响矿坑充水的因素多且复杂,矿坑涌水量预测模型主要考虑降水、地表水、引水灌溉等影响因素,因变量和自变量的关系比较复杂。将偏最小二乘回归与神经网络耦合,建立了矿坑涌水预报模型。模型将自变量利用偏最小二乘回归处理,提取对因变量影响强的成分,既可以克服变量之间的相关性问题,又可以降低神经网络的输入维数,并能较好地解决非线性问题,提高了模型的学习能力和表达能力。以河南鹤壁八矿涌水量为例,建立了基于偏最小二乘回归和神经网络耦合的矿坑涌水量预测模型。计算验证表明,该类模型具有较高的预报精度和推广应用价值。  相似文献   

中国北方山丘区城镇地下水资源可持续利用——以吉林省磐石市为例   总被引：4，自引：2，他引：2  

姜纪沂  曹剑峰  谢晓峰  王利刚  平建华 《吉林大学学报(地球科学版)》2005,35(6):756-761

如何使有限的地下水资源发挥其最大的综合效益,保障城镇社会经济的可持续发展是山丘区中小城镇当前面临的重要问题。以实现磐石市城市地下水资源可持续利用为目的,建立地下水流数值模型,以地下水位降深、埋深为约束条件,优化市政供水开采布局,进行地下水资源开发潜力评价,进而提出了城市地下水资源可持续开发利用对策。  相似文献   

Integration of remotely sensed C factor into SWAT for modelling sediment yield     

Xianfeng Song  Zheng Duan  Yasuyuki Kono  Mingyu Wang 《水文研究》2011,25(22):3387-3398

The C factor, representing the impact of plant and ground cover on soil loss, is one of the important factors of the Modified Universal Soil Loss Equation (MUSLE) in the Soil and Water Assessment Tool (SWAT) to model sediment yield. The daily update of C factors in SWAT was originally determined by land use types and plant growth cycles. This does not reflect the spatial variation of C values that exists within a large land use area. We present a new approach to integrate remotely sensed C factors into SWAT for highlighting the effect of detailed vegetative cover data on soil erosion and sediment yield. First, the C factor was estimated using the abundance of ground components extracted from remote sensing images. Then, the gridding data of the C factor were aggregated to hydrological response units (HRUs), instead of to land use units of SWAT. In the end, the C factor values in HRUs were integrated into SWAT to predict sediment yield by modifying the ysed subroutine. This substitution work not only increases the spatial variation of the C factor in SWAT, but also makes it possible to utilize other sources of C databases rather than those from the United States. The demonstration in the Dage basin shows that the modified SWAT produces reasonable results in water flow simulation and sediment yield prediction using remotely sensed C values. The Nash–Sutcliffe efficiency coefficient (E_NS) and R² for surface runoff range from 0·69 to 0·77 and 0·73 to 0·87, respectively. The coefficients E_NS and R² for sediment yield were generally above 0·70 and 0·60, respectively. The soil erosion risk map based on sediment yield prediction at the HRU level illustrates instructive details on spatial distribution of soil loss. Copyright © 2011 John Wiley & Sons, Ltd.  相似文献   

Quantification of peatland water storage capacity using the water table fluctuation method          下载免费PDF全文

Marc‐André Bourgault  Marie Larocque  Michelle Garneau 《水文研究》2017,31(5):1184-1195

Peat specific yield (S_Y) is an important parameter involved in many peatland hydrological functions such as flood attenuation, baseflow contribution to rivers, and maintaining groundwater levels in surficial aquifers. However, general knowledge on peatland water storage capacity is still very limited, due in part to the technical difficulties related to in situ measurements. The objectives of this study were to quantify vertical S_Y variations of water tables in peatlands using the water table fluctuation (WTF) method and to better understand the factors controlling peatland water storage capacity. The method was tested in five ombrotrophic peatlands located in the St. Lawrence Lowlands (southern Québec, Canada). In each peatland, water table wells were installed at three locations (up‐gradient, mid‐gradient, and down‐gradient). Near each well, a 1‐m long peat core (8 cm × 8 cm) was sampled, and subsamples were used to determine S_Y with standard gravitational drainage method. A larger peat sample (25 cm × 60 cm × 40 cm) was also collected in one peatland to estimate S_Y using a laboratory drainage method. In all sites, the mean water table depth ranged from 9 to 49 cm below the peat surface, with annual fluctuations varying between 15 and 29 cm for all locations. The WTF method produced similar results to the gravitational drainage experiments, with values ranging between 0.13 and 0.99 for the WTF method and between 0.01 and 0.95 for the gravitational drainage experiments. S_Y was found to rapidly decrease with depth within 20 cm, independently of the within‐site location and the mean annual water table depth. Dominant factors explaining S_Y variations were identified using analysis of variance. The most important factor was peatland site, followed by peat depth and seasonality. Variations in storage capacity considering site and seasonality followed regional effective growing degree days and evapotranspiration patterns. This work provides new data on spatial variations of peatland water storage capacity using an easily implemented method that requires only water table measurements and precipitation data.  相似文献   

Hydrologic processes that govern stormwater infrastructure behaviour          下载免费PDF全文

Caitlin G. Eger  David G. Chandler  Charles T. Driscoll 《水文研究》2017,31(25):4492-4506

Using water budget data from published literature, we demonstrate how hydrologic processes govern the function of various stormwater infrastructure technologies. Hydrologic observations are displayed on a Water Budget Triangle, a ternary plot tool developed to visualize simplified water budgets, enabling side‐by‐side comparison of green and grey approaches to stormwater management. The tool indicates ranges of hydrologic function for green roofs, constructed wetlands, cisterns, bioretention, and other stormwater control management structures. Water budgets are plotted for several example systems to provide insight on structural and environmental design factors, and seasonal variation in hydrologic processes of stormwater management systems. Previously published water budgets and models are used to suggest appropriate operational standards for several green and grey stormwater control structures and compare between conventional and low‐impact development approaches. We compare models, characterize and quantify water budgets and expected ranges for green and grey infrastructure systems, and demonstrate how the Water Budget Triangle tool may help users to develop a data‐driven approach for understanding design and retrofit of green stormwater infrastructure.  相似文献   

Did annual run‐off and low flow decrease with reduced forestry practices in Japan?     

Hikaru Komatsu  Tomonori Kume  Yoshinori Shinohara  Yoshiyuki Miyazawa  Kyoichi Otsuki 《水文研究》2010,24(17):2440-2451

Japan developed large areas of coniferous plantations for timber production between the 1950s and 1970s; however, forestry practices such as thinning, pruning, and harvesting in most of the plantations have declined since the 1980s. Researchers speculated that reduced forestry practices could reduce run‐off and therefore available water resources. As a countermeasure to this potential risk, many local governments have introduced local taxes to stimulate forestry practices in the plantations. However, no studies have presented evidence for decreased annual run‐off and/or low flow in watersheds where forestry practices have declined. As a starting point for assessing this risk, this study examined potential changes in the annual run‐off and low flow in the Terauchi watershed. A large area of this watershed was covered with coniferous plantations. We first surveyed the annual investment in forestry operations and the number of forest owners in the city of Amagi. (Note that Amagi includes the Terauchi watershed.) Both decreased during the period 1979–2007, indicating reduced forestry practices. The frequency distribution of plantation tree ages in the watershed also suggested reduced forestry practices. After excluding the effect of precipitation, we examined potential changes in the annual run‐off and low flow during the period 1979–2007. We did not observe significant decreases in the annual run‐off and low flow during the period. Thus, the risk of decreased water resources might be less than expected, and countermeasures to the risk should be reconsidered. Copyright © 2010 John Wiley & Sons, Ltd.  相似文献   

Assessing the benefit of satellite-based Solar-Induced Chlorophyll Fluorescence in crop yield prediction     

《International Journal of Applied Earth Observation and Geoinformation》2020

Large-scale crop yield prediction is critical for early warning of food insecurity, agricultural supply chain management, and economic market. Satellite-based Solar-Induced Chlorophyll Fluorescence (SIF) products have revealed hot spots of photosynthesis over global croplands, such as in the U.S. Midwest. However, to what extent these satellite-based SIF products can enhance the performance of crop yield prediction when benchmarking against other existing satellite data remains unclear. Here we assessed the benefits of using three satellite-based SIF products in yield prediction for maize and soybean in the U.S. Midwest: gap-filled SIF from Orbiting Carbon Observatory 2 (OCO-2), new SIF retrievals from the TROPOspheric Monitoring Instrument (TROPOMI), and the coarse-resolution SIF retrievals from the Global Ozone Monitoring Experiment-2 (GOME-2). The yield prediction performances of using SIF data were benchmarked with those using satellite-based vegetation indices (VIs), including normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and near-infrared reflectance of vegetation (NIRv), and land surface temperature (LST). Five machine-learning algorithms were used to build yield prediction models with both remote-sensing-only and climate-remote-sensing-combined variables. We found that high-resolution SIF products from OCO-2 and TROPOMI outperformed coarse-resolution GOME-2 SIF product in crop yield prediction. Using high-resolution SIF products gave the best forward predictions for both maize and soybean yields in 2018, indicating the great potential of using satellite-based high-resolution SIF products for crop yield prediction. However, using currently available high-resolution SIF products did not guarantee consistently better yield prediction performances than using other satellite-based remote sensing variables in all the evaluated cases. The relative performances of using different remote sensing variables in yield prediction depended on crop types (maize or soybean), out-of-sample testing methods (five-fold-cross-validation or forward), and record length of training data. We also found that using NIRv could generally lead to better yield prediction performance than using NDVI, EVI, or LST, and using NIRv could achieve similar or even better yield prediction performance than using OCO-2 or TROPOMI SIF products. We concluded that satellite-based SIF products could be beneficial in crop yield prediction with more high-resolution and good-quality SIF products accumulated in the future.  相似文献   

Kinematic analysis of pile behavior for improvement of slope stability in fractured and saturated Hoek–Brown rock masses          下载免费PDF全文

C.B. Qin  S.C. Chian  C.Y. Wang 《国际地质力学数值与分析法杂志》2017,41(6):803-827

Because of a multitude of steep slopes being constructed adjacent to roadways, there is greater concern of landslide occurrence, particularly in instances where poor geomaterials are present. Installation of piles along the slope is one commonly adopted method. This paper presents the assessment of the stability of a rock slope with stabilizing piles based on kinematic analysis. The pile effect is introduced with a resultant lateral force and a moment. Upper bound solutions of the pile's lateral force are derived with a log‐spiral rotational failure mechanism. The slope performance based on the bearing capacity of surcharge loading is also discussed with consideration of pore water pressure. In order to substantiate the derived theoretical solutions, numerical analysis with optimization technique is carried out. Results demonstrate that rock materials with high quality are conducive to ensure slope stability. Reduced lateral force on the pile is produced with lower rock weight, slope height, and surcharge loading. Finally, the safety factor and stability coefficient are discussed to complete the evaluation of the slope stability. Copyright © 2016 John Wiley & Sons, Ltd.  相似文献   

Analysis of piles subjected to lateral soil movements using a three‐dimensional displacement approach          下载免费PDF全文

Hiroyoshi Hirai 《国际地质力学数值与分析法杂志》2016,40(2):235-268

An analytical approach using the three‐dimensional displacement of a soil is investigated to provide analytical solutions of the horizontal response of a circular pile subjected to lateral soil movements in nonhomogeneous soil. The lateral stiffness coefficient of the pile shaft in nonhomogeneous soil is derived from the rocking stiffness coefficient that is obtained from the analytical solution, taking into account the three‐dimensional displacement represented in terms of scalar potentials in the elastic three‐dimensional analysis. The relationship between horizontal displacement, rotation, moment, and shear force of a pile subjected to lateral soil movements in nonhomogeneous soil is obtainable in the form of the recurrence equation. For the relationship between the lateral pressure and the horizontal displacement, it is assumed that the behavior is linear elastic up to lateral soil yield, and the lateral pressure is constant under the lateral soil yield. The interaction factors between piles subjected to both lateral load and moment are calculated, taking into account the lateral soil movement. The formulation of the lateral displacement and rotation of the pile base subjected to lateral loads in nonhomogeneous soils is presented by taking into account the Mindlin equation and the equivalent thickness for soil layers in the equivalent elastic method. For lateral movement, lateral pressure, bending moment, and interaction factors, there are small differences between results obtained from the 1‐D and the 3‐D displacement methods except a very flexible pile. Copyright © 2015 John Wiley & Sons, Ltd.  相似文献   

凡纳滨对虾(Litopenaeus vannamei)形态性状与净肉重和出肉率的关系   总被引：1，自引：0，他引：1  

张倩  王全超  于洋  张成松  黄皓  李富花  相建海 《海洋与湖沼》2018,49(3):653-661

为了探究凡纳滨对虾(Litopenaeus vannamei)表型性状与净肉重、出肉率的关系,本研究以5月龄的650尾凡纳滨对虾为材料,测量了全长、体长、头胸甲和各体节的长宽高等24个表型性状,根据体重、净肉重计算出肉率,通过相关分析、逐步回归分析和通径分析,计算了各性状间的相关系数、通径系数和决定系数,分析了24个表型性状与净肉重和出肉率的关系。结果表明,相关系数均达到极显著水平(P0.01);通过逐步回归分析建立了以净肉重为因变量,13个性状为自变量的回归方程;同时建立了以出肉率为因变量,9个性状为自变量的回归方程。通径分析结果显示,体长对净肉重的直接影响最大,其次是第3腹节长和第1腹节宽;对出肉率的正向直接影响最大的是体长,其次是第2腹节高和第3腹节长,头胸甲长、宽、高以及尾节长与出肉率呈明显负相关。决定系数分析结果与通径分析结论基本一致,在所分析的24个形态性状中,体长和第3腹节长与净肉重和出肉率密切相关。在实际生产中,可以通过测量体长和第3腹节长等形态性状间接实现对净肉重、出肉率的选择,选育优良品种。  相似文献