A hybrid method for grade estimation using genetic algorithm and neural networks

In the present paper, a new hybrid method is proposed for grade estimation. In this method, the multilayer perceptron (MLP) network is trained using the combination of the Levenberg–Marquardt (LM) method and genetic algorithm (GA). Having a few samples for grade estimation, it is difficult to get a proper result using some function approximation methods like neural networks or geostatistical methods. The neural network training methods are very sensitive to initial weight values when there are a few samples as a training dataset. The main objective of the proposed method is to resolve this problem. Here, our method finds the optimal initial weights by combining GA and LM method. Having the optimal initial values for weights, the local minima are avoided in the training phase and subsequently the neural network sustainability is trained optimally. Furthermore, the hybrid method is applied for grade estimation of Gol-e-Gohar iron ore in south Iran. The proposed method shows significant improvements compared to both conventional MLP and Kriging method. The efficiency of the proposed method gets more highlighted when the training data set is small.     

Parameter estimations of a storm surge model using a genetic algorithm

A genetic algorithm was used to optimize the parameters of the two-dimensional Storm Surge/Tide Operational Model (STORM) to improve sea level predictions of storm surges. The model was then tested using data from Typhoon Maemi, which landed on the Korean Peninsula in 2003. The following model parameters were used: the coefficients for bottom drag, background horizontal diffusivity, Smagorinsky’s horizontal viscosity, and sea level pressure scaling. The simulation results using the optimized parameters improved sea level predictions. This study demonstrates that parameter optimizations and their adequate applications are essential for improving model performance.     

Recharge rate estimation in the Mountain karst aquifer system of Figeh spring, Syria

Figeh watershed spring is one of the important groundwater aquifer, which is considered a major source for drinking waters of Damascus city and countryside. The origin identification and recharge estimates of groundwater are significant components of sustainable groundwater development in this Mountain karst aquifer of Figeh spring. During the period 2001–2009, monthly groundwater and precipitation samples were taken and the isotopic compositions of δ¹⁸O, δ²H, and chloride contents were analyzed to identify groundwater origins and to estimate recharge rates. The δ¹⁸O, δ²H of the groundwater show that the groundwater recharge is of meteoric origin. The chloride mass balance (CMB) method was used to quantify recharge rates of groundwater in the Mountain karst aquifer of Figeh spring. The recharge rate varies from 192 to 826 mm/year, which corresponds to 43 and 67% of the total annual rainfall. Recharge rates estimated by CMB were compared with values obtained from other methods and were found to be in good agreement. This study can be used to develop effective programs for groundwater management and development.     

基于十进制遗传算法的含水层参数的反演

含水层参的反演是一个复杂的非线性优化问题,针对传统二进制遗传算法收敛性能差的缺陷,提出了反演含水层参数的十进制遗传算法。以直线隔水边界附近的井流模型为例,讨论了十进制遗传算法在含水层参数反演中的应用,并与二进制遗传算法的进行比较。结果表明,该方法在含水层参数的反演中不仅是可行的,而且具有较好的确定性和较高的精度;与二进制遗传算法相比,十进制遗传算法的收敛性较好,省时高效,且表示较为自然,容易引入相关领域知识。同时,结合实例的分析结果得出种群的规模对算法的收敛性没有明显的影响。     

Parameter identification of minifrac numerical tests using a gradient boosting-based proxy model and genetic algorithm

In recent years, the petroleum industry has devoted considerable attention to studying fluid flow inside fracture channels due to the discovery of naturally fractured reservoirs. The behavior prediction of these reservoirs is a well-known challenging task, in which the initial stage consists of identifying reservoir hydromechanical parameters. This work proposes an artificial intelligence-based approach to identify hydromechanical parameters from borehole injection pressure curves acquired through minifrac tests. This approach combines proxy modeling with a stochastic optimization algorithm to match observed and predicted borehole pressure curves. Therefore, a gradient boosting-based proxy model is built to predict borehole pressure curves, considering a proper strategy to develop time series modeling. Moreover, a Bayesian optimization algorithm is applied to compute the gradient boosting hyperparameters. In this optimization scenario, this paper proposes an appropriate objective function established from the assumed time series prediction strategy and the k-fold cross-validation. Finally, a genetic algorithm is adopted to identify unknown hydromechanical parameters, solving an inverse problem. Based on the proposed workflow, a study of the importance of the hydromechanical parameters is developed. To assess the methodology applicability, the approach is employed to identify parameters in synthetic and field minifrac tests. The results present how this approach can adequately identify hydromechanical parameters of hydraulic fracturing problems.     

用水位恢复数据反演越流承压含水层参数

水文地质勘测中,由于水位恢复阶段没有人力和机械因素干扰,由测量数据可以画出平滑的曲线,更适宜于分析水文地质参数,但经典非稳定流Theis公式和Theis水位恢复法所依据的若干假设条件在实际中难以满足。针对越流承压含水层具有补给条件而很快稳定以及非稳定抽水阶段流量常“抖动”的特点,论证了利用稳定流量和Hantush-Jacob越流模型计算的可行性,并用扩展的卡尔曼滤波器反演水文地质参数,经文献中历史数据和实地抽水试验检验结果表明,由该方法反演的预测值和实测值吻合良好。     

Groundwater flow and residence time in a karst aquifer using ion and isotope characterization

In order to identify the origin of the main processes that affect the composition of groundwater in a karstic aquifer, a hydrogeochemical and isotopic study was carried out of water from numerous observation wells located in Sierra de Gador, a semiarid region in SE Spain. Several natural and anthropogenic tracers were used to calculate groundwater residence time within this complex aquifer system. Analysis of major ions enabled the principal geochemical processes occurring in the aquifer to be established, and the samples were classified into four distinctive solute groups according to this criterion. Dissolution of carbonate rocks determines the chemical composition of less mineralized water. In another group, the concurrent dissolution of dolomite and precipitation of calcite in gypsum-bearing carbonate aquifer, where the dissolution of relatively soluble gypsum controls the reaction, are the dominant processes. Marine intrusion results in highly mineralized waters and leads to base exchange reactions. The groundwater enrichment of minor and trace elements allowed classification of the samples into two classes that are linked to different flow patterns. One of these classes is influenced by a slow and/or deep regional flow, where the temperature is generally elevated. The influence of sulphate reduces by up to 40 % the barium concentration due to the barite precipitation. Isotope data (T, ¹⁴C) confirm the existence of recent local flows, and regional flow system, and ages of ground water may reach 8000 years. The importance of gypsum dissolution in this aquifer is proved by the δ³⁴S content.     

Specific yield of phreatic variation zone in karst aquifer with the method of water level analysis

Regime of groundwater level is a comprehensive reflection of the hydrogeological environment from the perspective of groundwater. Based on the analysis of the water-level change of 65 groundwater monitoring points from 1987 to 1990, it is found that intermittent cones of depression came into being due to groundwater exploitation in Guilin during the observation period. The buried depth of groundwater in the drawdown cones, the annual variation of water level and specific yield have higher values. Improvement has been made to the formula of infiltration coefficient of precipitation. By using the precipitation response data recorded at every 15 minutes for water level of No. 9 borehole which is near Zengpiyan Cave, the specific yield of phreatic variation zone is indirectly calculated by using the modified formula. The results are range from 0.012 to 0.462 and the spatial distribution of specific yield is ascertained. These make up the deficiency that empirical values cannot be categorized based on the actual conditions. What’s more, the widely used Aviriyanover’s empirical formula is poorly applicable to karst area. This is due to its strict requirement for outside conditions, such as shallow buried depth, homogeneous aquifer medium and small hydraulic gradient.     

基于松鼠搜索算法的跨孔电阻率溶洞探测

针对在基建工程前期探测中,传统物探反演方法对溶洞等不良地质条件的探测精度低的问题,提出了一种基于松鼠搜索算法的跨孔电阻率溶洞探测反演方法,用于改善传统的基于Tikhonov正则化的灵敏度迭代法对初值、噪声敏感、容易陷入局部最优等缺陷。采用小型、大型和串珠状溶洞3种数值算例,对不同智能搜索算法和灵敏度迭代法的探测结果进行了对比分析,并开展了室内模型试验和现场实验验证。研究结果表明:基于松鼠搜索算法的反演方法收敛速度快、精确度高,可显著提高跨孔电阻率溶洞的探测精度。     

基于遗传算法的BP神经网络计算岩溶水安全开采量

随着淮北市相山区岩溶水开采量不断增大,区内岩溶水水位降落漏斗范围不断增大,为保障岩溶水的安全开采与地质环境安全,进行本区岩溶水安全开采量计算十分必要。目前神经网络模型已被广泛应用于岩溶水水位动态计算,但由于网络全局寻优能力不理想,网络训练容易陷入局部极小值,导致网络泛化能力不理想。针对人工神经网络的不足,利用遗传算法(GA)对较为常用的BP神经网络权值、阈值进行优化,将此方法应用于相山区岩溶水水位动态的预测,并以该区岩溶水临界开采水位为控制条件,经模型计算得到相山区岩溶水多年平均安全开采量为3 001.7×10~4m~3。计算结果表明:与BP神经网络相比,GA-BP神经网络具有更高的预测精度,遗传算法可以有效提高BP网络的泛化能力。     

Soil parameter identification using a genetic algorithm

This paper is dedicated to the identification of constitutive parameters of the Mohr–Coulomb constitutive model from in situ geotechnical measurements. A pressuremeter curve and the horizontal displacements of a sheet pile wall retaining an excavation are successively used as measurements. Two kinds of optimization algorithms are used to minimize the error function, the first one based on a gradient method and the second one based on a genetic algorithm. The efficiency of each algorithm related to the error function topology is discussed. Finally, it is shown that the use of a genetic algorithm to identify the soil parameters seems particularly suitable when the topology of the error function is complex. Copyright © 2007 John Wiley & Sons, Ltd.     

Enhancement of aquifer vulnerability indexing using the analytic-element method

Indexing methods are used for the evaluation of aquifer vulnerability and establishing guidelines for the protection of ground-water resources. The principle of the indexing method is to rank influences on groundwater to determine overall vulnerability of an aquifer to contamination. The analytic element method (AEM) of ground-water flow modeling is used to enhance indexing methods by rapidly calculating a potentiometric surface based primarily on surface-water features. This potentiometric map is combined with a digital-elevation model to produce a map of water-table depth. This is an improvement over simple water-table interpolation methods. It is physically based, properly representing surface-water features, hydraulic boundaries, and changes in hydraulic conductivity. The AEM software, SPLIT, is used to improve an aquifer vulnerability assessment for a valley-fill aquifer in western New York State. A GIS-based graphical user interface allows automated conversion of hydrography vector data into analytic elements.     

考虑溶洞顶板自重时桩端持力岩层安全厚度计算方法

通过对溶洞顶板持力层模型的合理简化,进一步完善了已有研究成果的溶洞顶板持力层抗冲切、抗剪切厚度公式,考虑自重影响因素。提出了可以考虑有效宽度、自重等影响因素的单向板、双向板抗拉弯破坏溶洞顶板持力层厚度公式。该公式力学概念清晰,计算简便。进一步结合工程实际中各种参数的取值范围,按照常规抗冲切、抗剪切和抗拉弯破坏模式,得出一般情况下溶洞持力层顶板抗冲切、抗剪切净厚度大于2.5～3.5倍桩径和抗拉弯破坏顶板总厚度大于5.0～5.5倍桩径即可满足设计要求。该研究可供岩溶区桥梁桩基工程设计和施工参考。     

A parsimonious distributed model for simulating transient water flow in a high-relief karst aquifer

A mathematical model of a highly heterogeneous functioning karst aquifer is described. The aquifer is in a high-relief karst massif and, as is common for such locations, data are scarce and there are no borehole, piezometer or pumping-test data. The scarcity of data in this case required a parsimonious approach to ensure that the level of complexity of the model was commensurate with the amount, type and quality of the available data. Parsimony also requires the model to include the minimum essential components that account adequately for the data, which in this and similar cases are the functional dualities of the karst system: duality in recharge, flow and discharge. The model is three-dimensional (3D) in the sense that the aquifer is discretized into 3D voxels, although the flow is one-dimensional (1D) and vertical in the vadose zone, and horizontal and two-dimensional (2D) in the saturated zone. The parsimonious model was designed by coupling a 1D unsaturated gravity-driven flow along the vertical (along each column of voxels that discretize the aquifer) and a 2D unconfined Darcy flow in the saturated zone. In the context of this type of aquifer, preferential recharge through the network of karst conduits implies a rapid rise in the water table, the location and extension of which are model parameters. The karst springs are simulated by drains. The methodology, which is completely general, is illustrated by application to the karst aquifer in the Sierra de las Nieves mountains in southern Spain.     

Assessing flow paths in a karst aquifer based on multiple dye tracing tests using stochastic simulation and the MODFLOW-CFP code

Karst systems show high spatial variability of hydraulic parameters over small distances and this makes their modeling a difficult task with several uncertainties. Interconnections of fractures have a major role on the transport of groundwater, but many of the stochastic methods in use do not have the capability to reproduce these complex structures. A methodology is presented for the quantification of tortuosity using the single normal equation simulation (SNESIM) algorithm and a groundwater flow model. A training image was produced based on the statistical parameters of fractures and then used in the simulation process. The SNESIM algorithm was used to generate 75 realizations of the four classes of fractures in a karst aquifer in Iran. The results from six dye tracing tests were used to assign hydraulic conductivity values to each class of fractures. In the next step, the MODFLOW-CFP and MODPATH codes were consecutively implemented to compute the groundwater flow paths. The 9,000 flow paths obtained from the MODPATH code were further analyzed to calculate the tortuosity factor. Finally, the hydraulic conductivity values calculated from the dye tracing experiments were refined using the actual flow paths of groundwater. The key outcomes of this research are: (1) a methodology for the quantification of tortuosity; (2) hydraulic conductivities, that are incorrectly estimated (biased low) with empirical equations that assume Darcian (laminar) flow with parallel rather than tortuous streamlines; and (3) an understanding of the scale-dependence and non-normal distributions of tortuosity.     

Multiple-pumped-well aquifer test to determine the anisotropic properties of a karst limestone aquifer in Pasco County, Florida, USA

Groundwater-level data from an aquifer test utilizing four pumped wells conducted in the South Pasco wellfield in Pasco County, Florida, USA, were analyzed to determine the anisotropic transmissivity tensor, storativity, and leakance in the vicinity of the wellfield. A weighted least-squares procedure was used to analyze drawdowns measured at eight observation wells, and it was determined that the major axis of transmissivity extends approximately from north to south and the minor axis extends approximately from west to east with an angle of anisotropy equal to N4.54°W. The transmissivity along the major axis ${\left( {T_{{\xi \xi }} } \right)}$ is 14,019 m² day^–1, and the transmissivity along the minor axis ${\left( {T_{{\eta \eta }} } \right)}$ is 4,303 m² day^–1. The equivalent transmissivity $T_{e} = {\left( {T_{{\xi \xi }} T_{{\eta \eta }} } \right)}^{{1 \mathord{\left/ {\vphantom {1 2}} \right. \kern-0em} 2}} = 7,767{{\text{m}}^{2} } \mathord{\left/ {\vphantom {{{\text{m}}^{2} } {{\text{day}}^{{ - {\text{1}}}} }}} \right. \kern-0em} {{\text{day}}^{{ - {\text{1}}}} }$ , and the ratio of anisotropy is 3.26. The storativity of the aquifer is 7.52?×?10^?4, and the leakance of the overlying confining unit is 1.37?×?10^?4 day^?1. The anisotropic properties determined for the South Pasco wellfield in this investigation confirm the results of previous aquifer tests conducted in the wellfield and help to quantify the NW–SE to NE–SW trends for regional fracture patterns and inferred solution-enhanced flow zones in west-central Florida.     

Multi-scale characterization of a complex karst and alluvial aquifer system in southern Germany using a combination of different tracer methods

Water suppliers face major challenges such as climate change and population growth. To prepare for the future, detailed knowledge of water resources is needed. In southern Germany, the state water supplier Zweckverband Landeswasserversorgung provides 3 million people with drinking water obtained from a complex karst and alluvial aquifer system and the river Danube. In this study, a combination of different tracing techniques was used with the goal of a multi-scale characterization of the aquifer system and to gain additional knowledge about groundwater flow toward the extraction wells in the Danube Valley. For the small-scale characterization, selected groundwater monitoring wells were examined using single-borehole dilution tests. With these tests, a wide range of flow behavior could be documented, including fast outflow within just a few hours in wells with good connection to the aquifer, but also durations of many weeks in low-permeability formations. Vertical flow, caused by multiple flow horizons or uprising groundwater, was detected in 40% of the tested wells. A regional multi-tracer test with three injections was used to investigate the aquifer on a large scale. For the highly karstified connection between a swallow hole and a spring group, high flow velocities of around 80 m/h could be documented. Exceptionally delayed arrivals, 250 and 307 days after the injection, respectively showing maximum velocities of 0.44 and 0.39 m/h, were observed in an area where low-permeability sediments overlay the karst conduits. With the chosen methods, a distinct heterogeneity caused by the geological setting could be documented on both scales.

     

利用滨海地下水潮汐效应估计含水层参数:以北海半岛为例

通过分析滨海含水层中观测孔的水位动态资料,确定含水层的水力性质和某些参数,或推测含水层在海底的延伸长度和露头处是否存在覆盖层,是水文地质学家一直很感兴趣的问题之一。文中充分分析了广西北海半岛滨海地区水文地质条件,将研究区第一个承压含水层系统概化为具有上覆弱透水层越流和含水层海底露头覆盖层的垂向三层结构的不稳定流模型。应用该模型的解析解对广西北海半岛滨海含水层系统的观测数据进行了分析,拟合效果很好,并利用最小平方罚函数法估算了研究区含水层的参数。     

Characterisation of rapid infiltration flows and vulnerability in a karst aquifer using a decomposed fluorescence signal of dissolved organic matter

The fluorescent properties of dissolved organic matter (DOM) have been used as natural tracers in various water systems. However, only few studies have focused on groundwater systems, specifically, on karst systems. The aim of this study was to develop the use of the DOM fluorescence signal as a natural tracer, considering the multiple compartments of a karst system. This method was applied to the Lez hydrosystem, which supplies the city of Montpellier with drinking water. The hydrodynamics and hydrochemistry of the spring were monitored beginning March 2006. The DOM fluorescence was measured by the excitation–emission matrix spectroscopy technique. The analysis of the total fluorescence signal confirms the efficiency of this tool to trace rapid infiltration flows. Moreover, the decomposition of the signal into different fluorophores complements the information provided by the total signal. Indeed, the fluorescence emitted by the humic compounds seems to be the ideal tool for identifying rapid infiltration flows. Nevertheless, the fluorescence of protein-like compounds is better correlated with the inflow of faecal bacteria at the outlet. This decomposition of the fluorescence signal is an interesting way to provide information on both the rapid infiltration flow as well as the vulnerability of the karst aquifers.