ISSN 1009-6248CN 61-1149/P 双月刊

主管单位:中国地质调查局

主办单位:中国地质调查局西安地质调查中心
中国地质学会

    • 中文核心期刊
    • CSCD收录期刊
    • 中国科技核心期刊
    • Scopus收录期刊
高级检索

基于GM(1,1)与BP神经网络模型的西安市地下水位动态特征及趋势预测研究

李培月, 梁豪, 杨俊岩, 田艳, 寇晓梅

李培月,梁豪,杨俊岩,等. 基于GM(1,1)与BP神经网络模型的西安市地下水位动态特征及趋势预测研究[J]. 西北地质,xxxx,x(x): x−xx. doi: 10.12401/j.nwg.2024118
引用本文: 李培月,梁豪,杨俊岩,等. 基于GM(1,1)与BP神经网络模型的西安市地下水位动态特征及趋势预测研究[J]. 西北地质,xxxx,x(x): x−xx. doi: 10.12401/j.nwg.2024118
LI Peiyue,LIANG Hao,YANG Junyan,et al. Dynamic Characteristics and Trend Prediction of Groundwater Level in Xi’an City, China Based on GM (1, 1) and BP Neural Network Models[J]. Northwestern Geology,xxxx,x(x): x−xx. doi: 10.12401/j.nwg.2024118
Citation: LI Peiyue,LIANG Hao,YANG Junyan,et al. Dynamic Characteristics and Trend Prediction of Groundwater Level in Xi’an City, China Based on GM (1, 1) and BP Neural Network Models[J]. Northwestern Geology,xxxx,x(x): x−xx. doi: 10.12401/j.nwg.2024118

基于GM(1,1)与BP神经网络模型的西安市地下水位动态特征及趋势预测研究

基金项目: 国家重点研发计划项目课题“土壤–地下水污染时空演化规律及主控因子”(2023YFC3706901)、国家自然科学基金面上项目“大型灌区地下水多场协同作用下典型农业污染物迁移转化机制研究”(42472316)联合资助。
详细信息
    作者简介:

    李培月(1984–),男,教授,博士生导师,主要从事地下水文学与水资源研究。E–mail:lipy2@163.com

  • 中图分类号: P641

Dynamic Characteristics and Trend Prediction of Groundwater Level in Xi’an City, China Based on GM (1, 1) and BP Neural Network Models

  • 摘要:

    地下水是干旱与半干旱地区极其珍贵的自然资源,地下水动态的精准预测与评估关乎着地下水资源的有效保护与合理利用。该研究根据西安市2010~2020年地下水位监测数据,系统分析了西安市地下水位年际、年内动态变化特征,探究了影响地下水位动态的主要因素,通过SPSS对影响地下水位动态的降水量和开采量两个主要因素进行相关性分析,并基于GM(1,1)灰度预测模型和BP神经网络模型对地下水位变动趋势进行了预测。结果表明:①2010~2016年,地下水位整体上呈下降趋势,2016~2020年期间,得益于地下水压采和供水设施的不断优化完善,地下水位呈回升趋势。②降水和人为开采均对西安市地下水位变动具有显著影响;地下水位埋深是决定受降水影响程度的关键因素,其中河漫滩地区最为敏感,阶地次之,黄土塬区较弱。地下水开采量与地下水位埋深具有更强的相关性。这凸显了其在调控地下水位动态变化中的主导地位。③地下水位预测结果显示,随着地下水开采量呈现出逐年下降的趋势,研究区地下水整体处于波动上升趋势。本研究对西安市地下水动态的影响因素及预测趋势进行了研究,对地下水资源管理和可持续发展具有重要参考价值。

    Abstract:

    Groundwater is exteremely important in arid and semiarid regions, and the core of its effective protection and rational utilization lies in accurate prediction and evaluation of groundwater dynamics, based on which protection, utilization, and planning strategies are formulated. Based on groundwater level monitoring data from 2010 to 2020 in Xi'an City, this study systematically analyzed the inter-annual and intra-annual dynamic changes in groundwater levels, investigated the main factors influencing groundwater dynamics, and conducted a correlation analysis using SPSS on the two primary factors affecting groundwater dynamics: precipitation and extraction volume. Furthermore, the study utilized the GM (1,1) grey prediction model and the BP neural network model to forecast the trend of groundwater level changes. The results indicate that: ① From 2010 to 2016, the groundwater level showed an overall decreasing trend. However, from 2016 to 2020, due to the yearly reduction in extraction volume and continuous optimization and improvement of water supply facilities, the groundwater level exhibited a rising trend. ② Both precipitation and human extraction significantly impact the groundwater level fluctuations in Xi'an. The depth of the groundwater level is a crucial factor determining the degree of influence from precipitation, with river floodplains being the most sensitive, followed by terraces, and loess plateaus showing the weakest response. The correlation between groundwater extraction volume and groundwater depth is stronger, highlighting its dominant role in regulating groundwater level dynamics. ③ Groundwater level predictions suggest that as groundwater extraction continues to decline annually, the overall groundwater in the study area is on a fluctuating upward trend. This study has conducted research on the influencing factors and prediction trends of groundwater dynamics in Xi'an, which has important reference value for groundwater resource management and sustainable development.

  • 各玛龙银多金属矿位于青海省海西州都兰县境内,矿区中心点坐标为35°42′00″N,98°42′00″E,位于那更康切尔大型银矿以南10 km处。自2016年以来,矿区开展了银多金属矿找矿工作,共圈出了8条含矿蚀变带,27条银多金属矿体,成因为浅成低温热液型,但成矿物质和热液来源一直未查明(张先超等,2017王婧等,2020)。以往通过1/1万地磁测量,在矿区东南角圈定一处正负相伴的椭圆状磁异常M1,含矿蚀变带围绕M1磁异常呈环状分布,但磁异常与成矿关系未查明(裴有生等,2021)。本文主要是结合地物化资料的综合分析,来分析M1磁异常起因及其找矿潜力。

    各玛龙银多金属矿地处青海都兰县热水乡西南约120 km处,处于东昆仑成矿带的东段(图1任家祺等,2019薛长军等,2019)。区域上分布着那更康切尔大型银矿床、坑德弄舍大型多金属矿床、以及尕芝麻金矿点、叶陇沟金矿点等,成矿地质条件优越(杨涛等,2017国显正等,2019徐崇文等,2020秦阳等,2020谢升浪等,2020a, 2020b谷子成等,2021)。

    图  1  各玛龙矿区大地构造位置图(薛长军等,2019
    1.主缝合带;2.次缝合带;3.新元古代-早古生代缝合带俯冲方向,一侧有齿者为单向俯冲,两侧有齿者为双向俯冲;4.晚古生代-早古生代缝合带俯冲方向;5.A型俯冲带;6.构造单元界线;7.一级构造单元编号;8.二-三级构造单元编号;9.研究区;10.河流、湖泊
    Figure  1.  Geotectonic location map of Gemalong mining area

    矿区出露地层较简单,主要为晚三叠统鄂拉山组(T3e)、第四系。鄂拉山组为一套陆相酸性火山碎屑岩建造夹少量正常沉积岩,为中酸性角砾熔岩、安山岩、英安岩、流纹岩、凝灰岩夹少量灰绿色、灰黄色细砂岩(图2图3)。

    图  2  各玛龙矿区地质简图(裴有生等,2021
    1.第四系;2.鄂拉山组砂岩;3.鄂拉山组英安岩;4.鄂拉山组流纹岩;5.鄂拉山组晶屑凝灰岩;6.鄂拉山组安山岩;7.鄂拉山组火山角砾岩;8.早三叠世二长花岗岩;9.早三叠世钾长花岗岩;10.早三叠世花岗闪长岩;11.早三叠世花岗斑岩;12.晚三叠世花岗闪长斑岩;13.矿(化)体类型及编号;14. 钻孔位置及编号;15.断层;16. 1∶1万地磁异常(M1)
    Figure  2.  Geological sketch map of Gemalong mining area
    图  3  各玛龙矿区主要岩性显微镜下照片
    a.安山岩;b.晶屑凝灰岩;c.流纹岩;d.花岗斑岩;e.二长花岗岩;f.花岗闪长斑岩(1.斜长石;2.石英;3.角闪石)
    Figure  3.  Micrograph of the main lithology in Gemalong mining area

    矿区主要发育近东西向、近南北向、北东向三组断裂构造,其中近南北向断裂为主要控矿和容矿构造。

    矿区岩浆活动强烈,发育三叠世火山岩和侵入岩,其中火山岩分布在矿区南部和东部,岩性为安山岩、火山角砾岩、流纹岩、英安岩,从基性向中性演变。侵入岩分布在矿区北部,主要以早三叠世-晚三叠世侵入岩为主,为浅成-超浅成花岗闪长岩、二长花岗岩、钾长花岗岩、花岗斑岩、花岗闪长斑岩,一般呈岩株或岩脉状产出(张志颖,2019)。

    矿区已圈出8条含矿蚀变带,其中Ⅰ-Pb、Ⅱ-AgAuPb、Ⅲ-AgAuCuPbZn、Ⅳ-PbZn、Ⅵ-AgAu位于M1磁异常西侧,Ⅴ-PbAgAu、Ⅶ-Cu、Ⅷ-AuAg位于M1磁异常内,围绕M1磁异常由外至内存在低中温-中高温的成矿分带性,其中Ⅶ-Cu矿带位于M1磁异常中心位置,其含矿岩性为花岗闪长斑岩,圈出1条厚14.4 m的Cu矿化体,Cu平均品位为0.15%,最高品位为0.36%;Ⅲ-AgAuCuPbZn矿带为本区银金主矿带,控制长达1.5 km,控制斜深达600 m,厚60 m,含矿岩性为构造角砾岩,围岩为二长花岗岩和花岗斑岩。

    矿区共圈出27条银多金属矿体,主要矿种为银,伴生矿种主要有金、铜、铅、锌。矿体呈脉状,长为68~1050 m,厚为0.65~5.69 m,控制斜深120~600 m,Au品位为0.86×10−6~3.79 ×10−6,平均品位为0.86×10−6;Ag品位为45.3×10−61120×10−6,平均品位为326×10−6;Pb品位0.3%~3.52%,平均品位为0.42%;Zn品位为0.53%~2.86%,平均品位为0.65%;Cu品位为0.21%~1.72%,平均品位为0.37%。

    矿石结构主要为自形-半自形粒状结构、交代残留结构、乳滴状结构(图4),构造主要为细脉状构造,偶见稀疏浸染状构造。矿石矿物主要有黄铜矿、黄铁矿、方铅矿、闪锌矿、白铁矿、毒砂等,脉石矿物主要有石英、冰长石、方解石、绢云母、绿泥石、绿帘石等。

    图  4  各玛龙矿区主要矿石显微镜下照片
    a.黄铜矿呈团块状分布于毒砂间;b.方铅矿中呈乳滴状的辉银矿;c.黄铜矿与黄铁矿、白铁矿一起分布;d.方铅矿、黄铁矿、黄铜矿分布于石英颗粒间
    Figure  4.  Microscopic photographs of the main ores in Gemalong mining area

    矿区发育一套斑岩型-浅成低温热液成矿系统的叠加热液蚀变,M1磁异常西侧的Ⅰ、Ⅱ、Ⅲ、Ⅳ、Ⅵ矿带发育以硅化、高岭土化、冰长石化、绿泥石化为主的典型低硫型浅成低温热液蚀变(图5a图5b);M1磁异常内的Ⅴ、Ⅶ、Ⅷ矿带地表发育面状的泥化,浅部钻孔内发育黄铁绢英岩化(图5c图5d)。

    图  5  各玛龙矿区围岩蚀变照片
    a.Ⅲ矿带硅化、高岭土化;b.Ⅲ矿带围岩绿泥石、绿帘石化;c.Ⅴ矿带黄铁绢英岩化;d.M1磁异常地表泥化
    Figure  5.  Photographs of wall rock alteration in Gemalong mining area

    经吉林大学孙丰月团队等学者对M1磁异常西侧Ⅲ-AgAuCuPbZn矿带分析研究,认为矿床成因为浅成低温热液型,矿床形成主要受近南北向构造控制,成矿以Ag、Au、Cu、Pb、Zn为主,主成矿温度分布在134.2~204.3 ℃,成矿深度为1.34~2.28 km,成矿热液以岩浆热液为主并伴随有大气降水的参与,其物质来源于晚三叠世陆相火山-岩浆作用(张先超等,2017王婧等,2020)。

    M1磁异常内的Ⅶ-Cu矿带与构造关系不大,含矿岩性为花岗闪长斑岩,黄铜矿、黄铁矿呈稀疏浸染状分布,与斑岩型铜矿特征类似。

    矿区的大部分岩石磁性都较弱,其中鄂拉山组的凝灰岩、流纹岩、安山岩、火山角砾岩和早三叠世的花岗岩、二长花岗岩、花岗岩闪长岩、闪长玢岩等岩体磁性整体不高,磁化率均值在11.54×10−6×4πSI~105.13×10−6×4πSI,最高值为1127.13×10−6×4πSI,其剩余磁化强度在6.28×10−3~301.09×10−3A/m(表1)。但变化梯度大,推测与构造、热液等活动影响着磁性矿物的形成有关(孙忠实等,2005冯军等,2022)。

    表  1  各玛龙地区磁性参数特征一览表
    Table  1.  List of magnetic parameter characteristics in Gemalong mining area
    岩性 磁化率κ范围 (10−6×4π·SI) 磁化率κ平均值 (10−6×4π·SI) 剩磁范围Jr (10−3A/m) 剩磁Jr平均值 (10−3A/m)
    英安岩 32.71~1061.65 261.48 5.85~350.42 85.2
    安山岩 2.17~112.31 43.26 1.33~439.35 84.18
    二长花岗岩 6.77~31.50 31.50 4.49~67.19 16.19
    流纹岩 0.22~48.73 23.09 3.73~16.25 7.45
    闪长岩 13.57~86.26 13.57 4.39~22.23 8.09
    火山角砾岩(岩芯) 2.87~24.71 11.54 1.70~12.52 6.28
    晶屑凝灰岩 13.03~81.39 39.21 3.53~16.95 9.35
    花岗斑岩 5.2~41.19 22.01 4.88~12.09 8.90
    银金矿石 0.12~33.76 18.69 4.43~15.55 9.56
    花岗闪长斑岩 384.36~931.06 512.48 1.63~27.81 11.63
    下载: 导出CSV 
    | 显示表格
    表  2  钻孔岩芯磁性参数特征一览表
    Table  2.  List of magnetic parameter characteristics in Drill
    岩性 位置 标本数(块) 磁化率κ范围 (10−6×4π·SI)
    最小值 最大值 平均值
    黄铜矿化、磁黄铁矿化
    花岗闪长斑岩
    QZ003钻孔34.35~96.3 m 11 302.39 1209.58 621.79
    碎裂岩 QZ003钻孔96.3~103.65 m 7 177.46 651.74 322.4
    晶屑凝灰岩 QZ003 11.74~14.45, 103.65~140 m 17 28.57 473.49 207.39
    火山角砾岩 QZ002 20.35~80 m 32 177.46 651.74 322.4
    下载: 导出CSV 
    | 显示表格

    M1磁异常为一椭圆状磁异常,长轴轴向为近东西向,长约1500 m,宽约1300 m。异常正值区基本上在南部,负值区在北部,ΔTmax为490 nT,ΔTmin为−275 nT(图6)。磁异常北部负值区地表岩性为鄂拉山组晶屑凝灰岩、二长花岗岩、黑云母花岗岩,正值区地表岩性为鄂拉山组安山岩、流纹岩,此两种岩性的磁性均较弱,均不足以引起如此规模异常,推测磁性体有一定的埋深和规模。

    图  6  各玛龙矿区M1磁异常ΔT等值线图
    Figure  6.  The ΔT isogram of M1 Magnetic Anomaly in Gemalong mining area

    根据已知磁性参数数据,对CP01磁法剖面进行二度半人机交互反演圈出6个磁性体(图7)。通过对1号和4号磁性体工程验证,1号反应Ⅴ-PbAuAg矿带,在黄铁绢英岩化带内圈定一铅矿体,含矿岩性及围岩为火山角砾岩(图8);4号反应Ⅶ-Cu矿带,在黄铁绢英岩化与泥化的混合带中圈定Cu矿化体,含矿岩性为花岗闪长斑岩,圈出5条铜(金)矿化体,总厚为14.4 m,Cu平均品位为0.15%,最高品位为0.36%(图9)。其余磁性体均未验证,其中2号磁性体规模、埋深较大,形态上呈一岩体状,截面积为66185 m2,长度为400 m,体积为26474082 m3,埋深>500 m,朝东陡倾,推测为隐伏岩体。

    图  7  CP01号磁测剖面拟合反演图
    1.流纹岩;2.晶屑凝灰岩;3.二长花岗岩;4.Pb矿化体;5.Cu矿化体;6.流纹质火山角砾岩;7.安山岩;8.磁异常;9.钻孔位置及编号
    Figure  7.  Magnetic profile fitting inversion map of CP01.
    图  8  Ⅴ-Pb含矿蚀变带QZ002钻孔(a)和Ⅶ-Cu含矿蚀变带QZ003钻孔(b)剖面图
    1.第四系残坡积物;2.流纹质火山角砾岩;3.流纹岩;4.晶屑凝灰岩;5.安山岩;6.石英斑岩;7.花岗闪长斑岩;8.Pb矿(化)体;9.破碎蚀变带;10.Cu矿(化)体;11.Au矿(化)体;12.钻孔位置及编号;13.实测及推测地质界线;14.平均品位/厚度;15.样品及采样位置
    Figure  8.  (a)The section of QZ002in the V-PbOre-bearing alteration zone and (b)the section of QZ003in the VII-Cu Ore-bearing alteration zone
    图  9  1∶5000激电中梯等值线图
    Figure  9.  The isogram of 1∶5000 Ip intermediate gradient

    从激电异常看,M1磁异常区内1∶5000激电异常整体呈为低阻高极化的面状异常特征,东侧未封闭,电阻率平均500 Ω·m,极化率平均6.8%,激电异常规模、形态均与磁异常套合,推测激电异常是隐伏岩体浅部的黄铁矿绢英岩化带的反映(图9)。

    从测深异常看,M1磁异常区经测量两条激电、可控源大地音频电磁测深综合剖面(DPM-1、DPM-2),激电异常均显示为低阻高极化,电阻率平均为500 Ω·m,极化率平均为6.8%,最高为11.2%;可控源大地音频电磁测深反应,深部隐伏一长为600 m,宽为400 m,深10001500 m的低阻体,形态上呈一岩体,现认为测深异常反应的是黄铁绢英岩化带,从低阻异常中心位置看,两条剖面低阻异常区对应较好,隐伏岩体向东埋深逐渐增大(图10)。

    图  10  DPM-1(a)、DPM-3地物(b)(激电中梯、可控源大地音频电磁测深)综合剖面图
    1.晶屑凝灰岩;2.流纹岩;3.安山岩;4.可控源大地音频电磁测深电阻率等值线图;5.激电中梯剖面视电阻率曲线;6.激电中梯剖面视极化率曲线;7.地磁剖面ΔT曲线
    Figure  10.  (a) The Composite profile of DPM-1 and (b) DPM-3 (Ip intermediate gradient\Controlled source magnetotelluric sounding)

    从化探异常看,围绕M1磁异常1∶10000土壤剖面具分带性,具体为Cu、Sn、Bi等中高温元素异常集中分布于M1磁异常区,Ag、Au、Pb等中低温元素异常扩散范围较远,推测M1磁异常深部隐伏岩体为本区成矿物质和热液来源(图11)。

    图  11  1∶5000土壤异常等值线图
    Figure  11.  The isogram of 1∶5000 Soil anomalies

    从地质特征看,围绕M1磁异常自外而内类似斑岩型的青盘岩化、黄铁绢英岩化、泥化蚀变分带;垂向上M1磁异常区地表为面状泥化,再往下为黄铁绢英岩化,深部测深高阻与低阻过度带推测是钾化,也具斑岩型蚀变分带。平面上成矿元素从从Ⅰ含矿蚀变带Pb、Ⅲ含矿蚀变带AgAu、Ⅳ含矿蚀变带PbZn、Ⅴ含矿蚀变带AgAuPb,再到CuⅡ含矿蚀变带、AuⅠ含矿蚀变带,围绕磁异常自外而内具低温-中高温的元素组合分带性(图12a),通过综合分析建立了M1磁异常预测找矿模型(图12b),磁异常区发现的Ⅶ-Cu矿带含矿岩性为花岗闪长斑岩,黄铜矿化呈稀疏浸染状分布,与斑岩型铜矿特征类似,现认为Ⅶ-Cu矿带为M1磁异常深部隐伏岩体的一个小岩脉,推测隐伏岩体岩性为花岗闪长斑岩,M1磁异常具备寻找斑岩型铜矿的潜力(李保平等,2011)。因此,综上认为各玛龙银多金属矿床可能发育浅部浅成低温热液矿化-深部斑岩型矿化的这样一套成矿系统,与我国西藏多龙矿集区、紫金山矿田类似(唐菊兴等,20142016张德全等,20032005李斌等,2015)。

    图  12  矿区成矿元素及蚀变分带平面(a)和剖面(b)示意图
    1.晚三叠世鄂拉山组晶屑凝灰岩;2.晚三叠世鄂拉山组流纹岩;3.晚三叠世鄂拉山组火山角砾岩;4.晚三叠世鄂拉山组安山岩;5.早三叠世二长花岗岩;6.晚三叠世花岗闪长岩;7.青磐岩化;8.黄铁绢英岩化;9.钾化;10.泥化;11.矿(化)体类型及编号;12.化探元素异常组合及分布范围;13.M1磁异常位置
    Figure  12.  (a)The plane and (b)sectionsketch map of Metallogenic elements and Alteration zoning in the mining area.

    根据综合分析,M1磁异常是由深部隐伏的花岗闪长斑岩体中发育磁黄铁矿化引起,地表面状的低阻高极化激电异常是浅部泥化带中发育黄铁矿化引起,测深异常是对深部黄铁绢英岩化带中发育金属硫化物的反应;再结合地表及浅部已经存在泥化、黄铁绢英岩化、钾化的面状蚀变分带,认为M1磁异常具备寻找斑岩型矿的较大潜力。经验证浅部已发现含铜花岗闪长斑岩脉,深部尚存在厚大的测深低阻体,认为是黄铁绢英岩化带中发育黄铜矿等金属硫化物的反应,存在较大的找矿潜力。

    (1)根据矿区地物化资料综合推测M1磁异常为深部隐伏的花岗闪长斑岩体引起;M1磁异常区域的地质显示其与斑岩型矿床的成矿地质条件相近,具备寻找斑岩型矿的较大潜力。

    (2)因为激电异常是隐伏岩体浅部的黄铁矿绢英岩化带的反映,测深异常反应的是黄铁绢英岩化带。下步应根据测深、磁法、激电综合异常,对M1磁异常应采用钻探工程进行深部验证,揭穿浅部的黄铁绢英岩化带,至深部钾化带附近寻找斑岩型矿。

  • 图  1   研究区及监测点位置图

    Figure  1.   Location of study area and monitoring wells

    图  2   GM(1,1)灰色模型流程图

    Figure  2.   Flowchart interpreting the procedures of GM (1,1) Grey Model setup

    图  3   BP神经网络原理图

    Figure  3.   Schematic diagram of BP neural network

    图  4   BP神经网络流程图

    Figure  4.   Flowchart showing the procedures of BP neural network establishment

    图  5   2010~2020年降水量、开采量与地下水位平均埋深

    Figure  5.   Precipitation, groundwater abstraction, and average groundwater level depth from 2010 to 2020

    图  6   不同年份降水量和地下水位月变化曲线

    a. 2011年 (丰水年);b. 2016年 (枯水年)

    Figure  6.   Monthly variation of precipitation and groundwater level in different years

    图  7   地下水埋深预测结果与变化趋势图

    Figure  7.   Prediction results and trend of groundwater depth

    表  1   2010~2020年西安市降水、开采和地下水位埋深情况表

    Table  1   Precipitation, groundwater extraction, and groundwater level depth in Xi'an City from 2010 to 2020

    年份 降水量(mm) 降水量变幅(mm) 开采量(亿m3 地下水位埋深(m)
    2010 819 33 6.24 15.15
    2011 1002 216 6.01 14.42
    2012 659 −127 6.23 15.05
    2013 656 −130 6.37 15.48
    2014 802 16 9.57 15.11
    2015 779 −7 10.01 15.36
    2016 656 −130 10.26 15.99
    2017 823 37 10.26 15.57
    2018 728 −58 10.37 15.56
    2019 910 124 10.05 15.44
    2020 809 23 9.49 15.23
    下载: 导出CSV

    表  2   降雨量变幅、开采量变幅与地下水位埋深相关性分析表

    Table  2   Correlation of rainfall, groundwater extraction, and mean groundwater level depth

    变量 降雨量 开采量 地下水位埋深
    降雨量 1
    开采量 −0.176 1
    地下水位埋深 −0.673* 0.843** 1
    下载: 导出CSV

    表  3   地下水开采量灰色预测模型检验

    Table  3   Verification of groundwater extraction estimation using grey model

    年份2011201220132014201520162017201820192020
    光滑检验0.960.510.340.380.290.230.180.150.130.11
    误差检验0.110.240.290.260.180.030.0200.040.05
    下载: 导出CSV

    表  4   2025~2035年地下水开采量预测结果(亿m3

    Table  4   Prediction results of groundwater extraction from 2025 to 2035 (108 m3)

    年份20252026202720282029203020312032203320342035
    预测结果5.715.625.535.455.365.285.215.125.044.964.89
    下载: 导出CSV

    表  5   BP网络模型预测精度

    Table  5   Accuracy of BP network model prediction

    年份 水位实测值(m) BP网络模型
    预测值(m) 相对误差(%)
    2010 15.15 14.7656 −1.903
    2011 14.42 15.2047 −1.807
    2012 15.05 15.1268 0.086
    2013 15.48 15.1425 −1.443
    2014 15.11 15.2049 −4.906
    2015 15.36 14.7576 −5.196
    2016 15.99 15.2013 −2.273
    2017 15.57 14.7323 −4.558
    2018 15.56 15.1060 −0.805
    2019 15.44 15.0458 −2.553
    2020 15.23 14.8329 −2.607
    下载: 导出CSV
  • 李培月, 李佳慧, 吴健华, 等. 黄土-古土壤互层对土壤水分运移及土体微结构的影响[J]. 水文地质工程地质, 2024, 51(3): 1−11.

    LI Peiyue, LI Jiahui, WU Jianhua, et al. Effects of loess-paleosol interbedding on soil moisture transport and soil microstructure[J]. Hydrogeology & Engineering Geology,2024,51(3):1−11.

    杨俊岩. 西安市平原区地下水位动态及监测网优化[D]. 西安: 长安大学, 2024.

    YANG Junyan. Groundwater Level Dynamics and Monitoring Network Optimization in the Plain Area of Xi'an City, China[D]. Xi’an: Chang’an University, 2024.

    曾发琛. 西安市水资源供需平衡分析及优化配置研究[D]. 西安: 长安大学, 2008.

    ZENG Fachen. A Research on Supply-Requirement Analysis and Optimized Allocation of Water Resources in Xi’an City[D]. Xi’an: Chang’an University, 2008.

    Du Xinqiang, Chang Kaiyang, Lu Xiangqin. Characteristics and causes of groundwater dynamic changes in Naoli River Plain, Northeast China[J]. Water Supply,2020,20:2603−2615. doi: 10.2166/ws.2020.157

    Ferede M, Haile A T, Walker D, et al. Multi-method groundwater recharge estimation at Eshito micro-watershed, Rift Valley Basin in Ethiopia[J]. Hydrological Sciences Journal,2020,65:1596−1605. doi: 10.1080/02626667.2020.1762887

    Hou M, Chen S, Chen X, et al. A Hybrid Coupled Model for Groundwater-Level Simulation and Prediction: A Case Study of Yancheng City in Eastern China[J]. Water,2023,15:1085. doi: 10.3390/w15061085

    Hui P, Qu W, Tang J, et al. Traffic Indexes Prediction Based on Grey Prediction Model[C]. Sixth International Symposium on Computational Intelligence and Design, 2013: 244−247.

    Kacimov A R. Dynamics of groundwater mounds: analytical solutions and integral characteristics[J]. Hydrological Sciences Journal,1997,42:329−342. doi: 10.1080/02626669709492032

    Li H, Hou. Groundwater dynamic response mechanism and quantity vulnerability assessment under the influence of human activities[J]. Environmental Science and Pollution Research,2020,27:22046−22064. doi: 10.1007/s11356-020-08645-z

    Li H, Ma C, Zhou W B, et al. Characterizing the Evolution of Groundwater Flow Field and Its Driving Forces in Xi’an, China[J]. Journal of Hydrologic Engineering,2018,23(8):05018017. doi: 10.1061/(ASCE)HE.1943-5584.0001668

    Liu D, Hu Y, Li T, et al. Complexity Diagnosis of Regional Groundwater Resources System Based on Symbolic Dynamics[J]. Journal of Hydrologic Engineering,2016,21:05015011. doi: 10.1061/(ASCE)HE.1943-5584.0001252

    Liu J, Gao Z, Wang M, et al. Study on the dynamic characteristics of groundwater in the valley plain of Lhasa City[J]. Environmental Earth Sciences,2018,77:646. doi: 10.1007/s12665-018-7833-4

    Ma D, Bai H. Groundwater inflow prediction model of karst collapse pillar: a case study for mining-induced groundwater inrush risk[J]. Natural Hazards,2015,76:1319−1334. doi: 10.1007/s11069-014-1551-3

    Mu E L, Yan L, Ding A Z, et al. Determination of controlled limit value of groundwater level depth and management practice in Xi’an, China[J]. Scientific Reports,2020,10:15505. doi: 10.1038/s41598-020-72523-4

    Pai T Y, Wu R S, Chen C H, et al. Prediction of Groundwater Quality Using Seven Types of First-Order Univariate Grey Model in the Chishan Basin, Taiwan[J]. Water, Air & Soil Pollution, 2022, 233: 481.

    Peng J B, Sun X H, Wang W, et al. Characteristics of land subsidence, earth fissures and related disaster chain effects with respect to urban hazards in Xi’an, China[J]. Environmental Earth Sciences,2016,75:1190. doi: 10.1007/s12665-016-5928-3

    Pradhan A M S, Kim Y T, Shrestha S, et al. Application of deep neural network to capture groundwater potential zone in mountainous terrain, Nepal Himalaya[J]. Environmental Science and Pollution Research,2021,28:18501−18517. doi: 10.1007/s11356-020-10646-x

    Qiao J, Zhu Y, Jia X, et al. Distributions of arsenic and other heavy metals, and health risk assessments for groundwater in the Guanzhong Plain region of China[J]. Environmental Research,2020,181:108957. doi: 10.1016/j.envres.2019.108957

    Raimohan N, Al-futaisi A, Jamrah A. Evaluation of long-term groundwater level data in regular monitoring wells, Barka, Sultanate of Oman[J]. Hydrological Processes,2007,21(24):3367−3379.

    Su Z, Wu J, He X, et al. Temporal Changes of Groundwater Quality within the Groundwater Depression Cone and Prediction of Confined Groundwater Salinity Using Grey Markov Model in Yinchuan Area of Northwest China[J]. Exposure and Health,2020,12:447−468. doi: 10.1007/s12403-020-00355-8

    Wang Y Q, Wang Z F, Cheng W C. A review on land subsidence caused by groundwater withdrawal in Xi’an, China[J]. Bulletin of Engineering Geology and the Environment,2019,78:2851−2863. doi: 10.1007/s10064-018-1278-6

    Wu J, Li Z, Zhu L, et al. Optimized BP neural network for Dissolved Oxygen prediction[J]. IFAC-PapersOnLine,2018,51:596−601.

    Xie L. The Heat load Prediction Model based on BP Neural Network-markov Model[J]. Procedia Computer Science,2017,107:296−300. doi: 10.1016/j.procs.2017.03.108

    Yang Y, Li P, Elumalai V, et al. Groundwater Quality Assessment Using EWQI With Updated Water Quality Classification Criteria: A Case Study in and Around Zhouzhi County, Guanzhong Basin (China)[J]. Exposure and Health,2023,15:825−840. doi: 10.1007/s12403-022-00526-9

    Yao Y, Zhang M, Deng Y, et al. Evaluation of environmental engineering geology issues caused by rising groundwater levels in Xi'an, China[J]. Engineering Geology,2021,294:106350. doi: 10.1016/j.enggeo.2021.106350

    Yu X, Ghasemizadeh R, Padilla I Y, et al. Patterns of temporal scaling of groundwater level fluctuation[J]. Journal of Hydrology,2016,536:485−495. doi: 10.1016/j.jhydrol.2016.03.018

    Zeng B, Li C, Chen G, et al. Equivalency and unbiasedness of grey prediction models[J]. Journal of Systems Engineering and Electronics,2015,26:110−118. doi: 10.1109/JSEE.2015.00015

    Zhang Q, Li P, Ren X, et al. A new real-time groundwater level forecasting strategy: Coupling hybrid data-driven models with remote sensing data[J]. Journal of Hydrology,2023,625:129962.

    Zhou P, Wang G, Duan R. Impacts of long-term climate change on the groundwater flow dynamics in a regional groundwater system: Case modeling study in Alashan, China[J]. Journal of Hydrology, 2020, 590. 125557.

    Zhou W, Chai J, Xu Z, et al. Historical evolution of urban water conservancy projects in Xi'an, China in the past 3, 000 years and its revelations[J]. Water Supply,2021,21:2173−2190. doi: 10.2166/ws.2021.043

    Zhou Y, Wu J, Gao X, et al. Hydrochemical Background Levels and Threshold Values of Phreatic Groundwater in the Greater Xi’an Region, China: Spatiotemporal Distribution, Influencing Factors and Implication to Water Quality Management[J]. Exposure and Health,2023,15:757−771. doi: 10.1007/s12403-022-00521-0

    Zhu C J, Hao Z C, Ju Q. A prediction of groundwater quality using grey system neural network united model[C]. Chinese Control and Decision Conference, 2009, 3216−3219.

图(7)  /  表(5)
计量
  • 文章访问数:  87
  • HTML全文浏览量:  12
  • PDF下载量:  14
  • 被引次数: 0
出版历程
  • 收稿日期:  2024-10-30
  • 修回日期:  2024-12-03
  • 录用日期:  2024-12-04
  • 网络出版日期:  2024-12-24

目录

/

返回文章
返回