高级检索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于BP神经网络的广东省针阔混交异龄林立地质量评价

沈剑波 王应宽 雷相东 雷渊才 汪求来 叶金盛

沈剑波, 王应宽, 雷相东, 雷渊才, 汪求来, 叶金盛. 基于BP神经网络的广东省针阔混交异龄林立地质量评价[J]. 北京林业大学学报, 2019, 41(5): 38-47. doi: 10.13332/j.1000-1522.20190028
引用本文: 沈剑波, 王应宽, 雷相东, 雷渊才, 汪求来, 叶金盛. 基于BP神经网络的广东省针阔混交异龄林立地质量评价[J]. 北京林业大学学报, 2019, 41(5): 38-47. doi: 10.13332/j.1000-1522.20190028
Shen Jianbo, Wang Yingkuan, Lei Xiangdong, Lei Yuancai, Wang Qiulai, Ye Jinsheng. Site quality evaluation of uneven-aged mixed coniferous and broadleaved stands in Guangdong Province of southern China based on BP neural network[J]. Journal of Beijing Forestry University, 2019, 41(5): 38-47. doi: 10.13332/j.1000-1522.20190028
Citation: Shen Jianbo, Wang Yingkuan, Lei Xiangdong, Lei Yuancai, Wang Qiulai, Ye Jinsheng. Site quality evaluation of uneven-aged mixed coniferous and broadleaved stands in Guangdong Province of southern China based on BP neural network[J]. Journal of Beijing Forestry University, 2019, 41(5): 38-47. doi: 10.13332/j.1000-1522.20190028

基于BP神经网络的广东省针阔混交异龄林立地质量评价

doi: 10.13332/j.1000-1522.20190028
基金项目: 国家林业公益性行业科研专项(201504303)
详细信息
    作者简介:

    沈剑波,博士。主要研究方向:生物数学模型及农业信息化。Email:lyshenjianbo@163.com 地址:100125 北京市朝阳区麦子店街41号

    责任作者:

    汪求来,博士生,高级工程师。主要研究方向:森林资源与生态监测及生物数学模型。Email:wangqiulai2014@126.com 地址:510520 广东省广州市广汕一路338号

Site quality evaluation of uneven-aged mixed coniferous and broadleaved stands in Guangdong Province of southern China based on BP neural network

  • 摘要: 目的针阔混交异龄林的地位指数计算一直是立地质量评价中的难点,国内外对针阔混交异龄林的地位指数模型的研究较少,为建立更精确的针阔混交异龄林地位指数模型,把神经网络模型引入针阔混交异龄林的立地质量评价。方法以广东省针阔混交异龄林为研究对象,建立基于神经网络方法的林分优势高模型以及针阔混交林地位指数模型,除年龄因子外,加入了海拔、坡度、坡向、坡位、土壤厚度、腐殖层厚度等立地因子,另外考虑针叶树种与阔叶树种的断面积比重对针阔混交异龄林样地地位指数的影响,并建立针阔混交异龄林的地位指数的计算模型。结果针阔混交林的地位指数的最大值为21.4 m,最小值为6.1 m,平均值为13.7 m,中位数为13.6 m,标准差为3.2 m,地位指数的最大值与最小值的差值为15.3 m。结论从结果中可以反映出,广东省地貌复杂且破碎,山多平地少,立地状况差异较大;另外由于广东省林地树种及树种比例具有复杂多样性的特征,导致基准年龄的差异较大。故地位指数的变异较大。本研究在计算针阔混交林的地位指数时,加入了海拔、坡度、坡向、坡位、土壤厚度、腐殖层厚度等立地因子,提高了针阔混交林地位指数的预估精度。研究结果为针阔混交异龄林地位指数的计算提供了精度更高的方法。

     

  • 图  1  基于优势高−年龄模型的BP神经网络结构

    A. 年龄;H. 优势高。下同。A, age; H, dominant height. The same below.

    Figure  1.  Structure of BP neural network based on stand dominant height-age model

    图  2  含立地因子的优势高−年龄模型的BP神经网络结构

    SA.坡向;SP.坡位;Sl.坡度;Al.海拔;ST.土层厚度;HT.腐殖层厚度。SA, slope aspect; SP, slope position; Sl, slope; Al, altitude; ST, soil thickness; HT, humus thickness.

    Figure  2.  BP neural network structure of stand dominant height-age model including site facors

    图  3  不同输入因子的林分优势高预测值散点图

        a. 输入因子为年龄,b. 输入因子为年龄以及立地因子。a, input factor is age; b, input factors are age and site factors.

    Figure  3.  Scatter diagram of predicted values of stand dominant height by different input factors

    图  4  基于神经网络模型的林分优势高残差图

        a. 输入因子为年龄,b. 输入因子为年龄以及立地因子。a, input factor is age; b, input factors are age and site factors.

    Figure  4.  Residual distribution of stand dominant height based on neural network

    表  1  林分优势高预测建模数据和检验数据统计

    Table  1.   Statistical characteristic of modeling and test data sets for predicting stand dominant height

    调查因子
    Surveying factor
    训练数据 Training data检验数据 Validation data
    最小值 Min.最大值 Max.平均值 Mean标准差 SD 最小值 Min.最大值 Max.平均值 Mean标准差 SD
    林分优势木平均年龄/a
    Stand dominant average age/year
    4.060.035.216.23.071.039.920.6
    坡度
    Slope/(°)
    0.044.027.38.015.038.027.56.6
    海拔
    Altitude/m
    25.01 458.0334.0219.245.0790.0360.5212.9
    土层厚度
    Soil thickness/mm
    40.0120.086.720.465.0100.096.39.3
    腐殖层厚度
    Humus thickness/cm
    0.025.08.25.94.020.09.14.3
    林分优势高
    Stand dominant height/m
    8.021.414.63.59.217.513.52.6
    下载: 导出CSV

    表  2  神经网络权值与阈值矩阵(输入:年龄)

    Table  2.   Weight martrix and threshold matrix of Bp neural network (input: age)

    类别 Category     权值矩阵 Weight matrix阈值矩阵 Threshold matrix
    输入层至隐含层 Input layer to hidden layer$ {\left[ \begin{array}{l}\;\;\;3.790\;2\\ - 3.975\;6\\\;\;\;4.225\;0\end{array} \right]}$$ {\left[ \begin{array}{l} - 3.904\;9\\\;\;\;1.490\;4\\\;\;\;3.355\;9\end{array} \right]}$
    隐含层至输出层 Hidden layer to output layer$\scriptstyle \left[ 0.107~3\ \ \ \ -0.204~2\ \ \ \ 0.184~1 \right]$[− 0.207 4]
    下载: 导出CSV

    表  3  神经网络各层之间传递函数

    Table  3.   Transfer function between each layer of neural network

    类别 Category 传递函数 Transfer function
    输入层至隐含层
    Input layer to hidden layer
    H1 = tansig(1.314 8A − 1.403 4Al − 1.393 7SA + 0.149 8SP − 1.110 8Sl + 0.923 3ST + 0.073 0HT − 1.436 7)
    H2 = tansig(1.050 6A − 0.239 7Al + 0.300 5SA + 0.190 1SP + 0.105 1Sl + 1.241 4ST + 1.390 8HT − 1.868 0)
    H3 = tansig(− 0.482 6A − 1.961 3Al − 0.692 0SA − 0.431 4SP + 0.115 2Sl + 0.188 9ST − 0.024 0HT − 1.355 7)
    H4 = tansig(− 0.656 3A − 0.478 0Al − 0.097 4SA + 0.444 5SP + 0.948 5Sl + 1.281 7ST − 1.283 0HT + 1.133 1)
    H5 = tansig(− 0.714 8A − 0.619 1Al − 1.123 9SA − 0.277 7SP + 0.819 4Sl + 0.779 3ST − 1.529 4HT + 0.858 1)
    H6 = tansig(− 0.047 2A + 0.575 2Al − 0.144 1SA + 1.034 8SP − 1.392 3Sl − 0.269 6ST + 0.371 2HT − 1.725 9)
    H7 = tansig(0.589 9A − 0.106 3Al + 1.498 2SA + 0.569 9SP + 0.987 0Sl + 0.625 0ST − 0.474 9HT + 0.471 3)
    H8 = tansig(− 0.485 6A − 0.874 3Al + 1.222 9SA − 0.037 9SP − 0.867 7Sl + 1.558 3ST − 0.899 0HT + 0.346 7)
    H9 = tansig(− 1.800 3A − 0.668 8Al − 0.853 7SA − 1.250 4SP + 0.577 5Sl − 0.044 1ST + 0.934 2HT + 0.092 1)
    H10 = tansig(0.203 2A + 1.406 4Al − 1.396 2SA − 0.931 7SP − 0.929 4Sl + 0.915 3ST − 2.187 4HT − 0.714 6)
    H11 = tansig(0.825 5A − 0.082 3Al − 0.529 0SA + 0.485 0SP − 1.846 2Sl − 0.136 0ST − 0.846 1HT + 0.929 9)
    H12 = tansig(− 0.759 4A + 0.553 8Al + 0.147 8SA − 1.040 3SP + 0.706 6Sl + 0.244 2ST + 1.312 3HT − 1.336 0)
    H13 = tansig(− 0.637 3A + 1.411 1Al − 0.521 9SA − 0.380 9SP + 0.480 8Sl + 0.542 9ST + 1.601 4HT − 1.005 8)
    H14 = tansig(0.343 5A − 1.982 6Al + 1.080 0SA + 1.395 8SP + 0.778 4Sl + 0.684 6ST + 0.337 6HT + 1.068 3)
    H15 = tansig(0.120 9A − 1.170 8Al − 0.813 6SA − 0.414 1SP + 1.126 2Sl + 1.656 5ST + 0.656 8HT − 1.741 5)
    隐含层至输出层
    Hidden layer to output layer
    H = purelin(0.610 8H1 − 0.548 2H2 + 0.421 0H3 − 0.760 9H4 + 0.086 1H5 + 0.356 3H6 + 0.517 9H7 − 0.039 1H8
    0.265 5H9 − 0.279 7H10 − 0.515 5H11 − 1.070 6H12 + 0.098 8H13 − 1.173 5H14 − 0.0155 1H15 + 0.716 2)
    下载: 导出CSV

    表  4  不同输入因子的林分优势高预测评价指标

    Table  4.   Comparisons of the results between different input factors (predicted stand dominant height)

    类别
    Category
    输入因子
    Input factor
    评价指标
    Evaluation index

    Value
    a年龄
    Age
    R20.478 3
    RMSE/m1.817 1
    MAE/m1.448 2
    RMAE/m0.102 4
    b年龄、海拔、坡向、坡位、坡度、土层厚度、腐殖层厚度
    Age, altitude, slope aspect, slope position, slope, soil layer thickness,humus layer thickness
    R20.532 7
    RMSE/m1.719 7
    MAE/m1.122 0
    RMAE/m0.075 6
    下载: 导出CSV

    表  5  针叶树种基准年龄

    Table  5.   Reference age of different needle tree species

    树种 Tree species    基准年龄/a Reference age/year
    加勒比松 Pinus caribaea30
    马尾松 Pinus massoniana30
    杉木 Cunninghamia lanceolata30
    湿地松 Pinus elliottii30
    下载: 导出CSV

    表  6  阔叶树种基准年龄

    Table  6.   Reference age of different broadleaf tree species

    树种
    Tree species
    基准年龄/a
    Reference age/year
    豺皮樟 Litsea rotundifolia var. oblongifolia30
    短序润楠 Machilus breviflora30
    红锥 Castanopsis hystrix30
    华润楠 Machilus chinensis 30
    黄樟 Cinnamomum porrectum30
    黧蒴栲(黎蒴) Castanopsis fissa30
    罗浮槭 Acer fabri30
    罗浮锥 Castanopsis faberi 30
    刨花楠 Machilus pauhoi30
    青冈 Quercus glauca30
    蕈树 Altingia chinensis30
    樟树 Cinnamomum camphora30
    中华锥 Castanopsis chinensis30
    醉香含笑(火力楠) Michelia macclurei30
    白楸 Mallotus paniculatus20
    赤杨叶 Alniphyllum fortunei20
    红木荷 Schima reinw20
    马占相思 Acacia mangium20
    尾叶桉 Eucalyptus urophylla20
    窿缘桉 Eucalyptus exserta20
    木莲 Manglietia fordiana20
    木荷 Schima superba 20
    柠檬桉 Eucalyptus citriodora20
    油桐 Vernicia fordii 20
    鸭脚木 Schefflera octophylla20
    台湾相思 Acacia confusa 20
    下载: 导出CSV

    表  7  神经网络权值与阈值矩阵(输入:年龄与立地因子)

    Table  7.   Weight martrix and threshold matrix of neural network (input: age and site factors)

    类别
    Category
    权值矩阵
    Weight matrix
    阈值矩阵
    Threshold matrix
    输入层至隐含层
    Input layer to hidden layer
    $ \left[ {\begin{array}{*{20}{c}} {- 0.647\;3} & {0.777\;1} & {- 0.437\;5} & {- 0.842\;8} & {0.867\;7} & {- 1.290\;2} & {0.498\;9}\\{ 0.303\;6} & {- 0.849\;2} & {- 1.193\;6} & {1.453\;8} & {- 0.309\;8} & {- 0.385\;4} & {0.249\;1}\\{ - 0.845\;2} & {- 0.914\;1} & {- 0.643\;3} & {0.493\;7} & {- 0.275\;9} & {- 0.510\;5} & {- 1.104\;3}\\{ - 0.005\;0} & {- 1.349\;1} & {0.860\;0} & {- 1.612\;7} & {- 0.476\;1} & {- 0.044\;2} & {0.992\;7}\\{ 0.204\;5} & {- 1.045\;9} & {0.443\;9} & {0.595\;3} & {0.260\;6} & {1.162\;4} & {1.034\;9}\\{ 0.872\;8} & {0.732\;4} & {- 0.283\;7} & {- 1.011\;3} & {- 0.012\;1} & {0.951\;4} & {1.316\;3}\\{ - 0.057\;3} & {- 0.382\;9} & {- 1.322\;4} & {0.911\;0} & {- 1.535\;5} & {- 0.043\;9} & {0.202\;4}\\{ - 0.680\;4} & {- 1.242\;4} & {0.397\;4} & {1.262\;7} & {- 0.788\;2} & {0.508\;3} & {- 0.058\;5}\\{ 0.255\;6} & {1.272\;7} & {- 1.610\;0} & {- 0.741\;7} & {- 0.224\;7} & {- 0.234\;5} & {- 1.520\;3}\\{ 0.866\;8} & {0.077\;7} & {0.693\;0} & {- 0.619\;2} & {- 1.344\;9} & {0.536\;8} & {1.160\;1}\\{ - 0.161\;4} & {1.410\;5} & {- 0.549\;0} & {- 0.535\;2} & {- 0.882\;9} & {0.976\;0} & {0.736\;7}\\{ - 0.972\;7} & {- 0.640\;7} & {0.172\;1} & {- 0.166\;7} & {1.174\;5} & {- 0.936\;7} & {1.059\;7}\\{ 0.508\;9} & {- 1.306\;7} & {- 0.464\;2} & {- 0.823\;3} & {0.524\;0} & {0.757\;7} & {- 0.605\;3}\\{ 0.535\;3} & {0.290\;1} & {- 1.324\;2} & {1.021\;0} & {0.717\;0} & {- 0.735\;0} & {0.249\;2}\\{ 0.007\;9} & {- 0.286\;1} & {- 0.279\;4} & {- 0.174\;6} & {- 1.443\;4} & {- 1.172\;9} & {0.976\;9} \end{array}} \right]$$ \left[ \begin{array}{*{20}{c}} 2.013\;8\\ - 2.300\;7\\ 1.606\;0\\ 1.054\;7\\ - 1.344\;1\\ - 1.056\;2\\ 0.350\;7\\ 0.355\;8\\ 0.138\;3\\ 0.481\;2\\ - 0.766\;9\\ - 1.418\;7\\ 1.554\;7\\ 1.695\;0\\ 2.098\;4 \end{array} \right]$
    隐含层至输出层
    Hidden layer to output layer
    $ \begin{array}{l}\left[ {\begin{array}{*{20}{c}}{ - 0.283\;3} & { - 0.703\;9} & { - 0.195\;3} & {\;0.205\;3} & {0.247\;8} & { - 0.290\;6} & {0.690\;9} & { - 0.976\;3}\end{array}} \right.\\\;\left. {\begin{array}{*{20}{c}}{ - 0.756\;1} & { - 0.287\;7} & { - 0.004\;8} & { - 0.640\;0} & {0.098\;7} & {0.253\;2} & { - 0.466\;5}\end{array}} \right]\end{array}$$\scriptstyle \left[ { - 0.551\;2} \right]$
    下载: 导出CSV

    表  8  神经网络各层之间传递函数

    Table  8.   Transfer function between each layer of neural network

    类别 Category传递函数 Transfer function
    输入层至隐含层
    Input layer to hidden layer
    H1 = tansig(− 0.647 3A + 0.777 1Al − 0.437 5SA − 0.842 8SP + 0.867 7Sl − 1.290 2ST + 0.498 9HT + 2.013 8)
    H2 = tansig(0.303 6A − 0.849 2Al − 1.193 6SA + 1.453 8SP − 0.309 8Sl − 0.385 4ST + 0.249 1HT − 2.300 7)
    H3 = tansig(− 0.845 2A − 0.914 1Al − 0.643 3SA + 0.493 7SP − 0.275 9Sl − 0.510 5ST − 1.104 3HT + 1.606 0)
    H4 = tansig(− 0.005 0A − 1.349 1Al + 0.860 0SA − 1.612 7SP − 0.476 1Sl − 0.044 2ST + 0.992 7HT + 1.054 7)
    H5 = tansig(0.204 5A − 1.045 9Al + 0.443 9SA + 0.595 3SP + 0.260 6Sl + 1.162 4ST + 1.034 9HT − 1.344 1)
    H6 = tansig(0.872 8A + 0.732 4Al − 0.283 7SA − 1.011 3SP − 0.012 1Sl + 0.951 4ST + 1.316 3HT − 1.056 2)
    H7 = tansig(− 0.057 3A − 0.382 9Al − 1.322 4SA + 0.911 0SP − 1.535 5Sl − 0.043 9ST + 0.202 4HT + 0.350 7)
    H8 = tansig(− 0.680 4A − 1.242 4Al + 0.397 4SA + 1.262 7SP − 0.788 2Sl + 0.508 3ST − 0.058 5HT + 0.355 8)
    H9 = tansig(0.255 6A + 1.272 7Al − 1.610 0SA − 0.741 7SP − 0.224 7Sl − 0.234 5ST − 1.520 3HT + 0.138 3)
    H10 = tansig(0.866 8A + 0.077 7Al + 0.693 0SA − 0.619 2SP − 1.344 9Sl + 0.536 8ST + 1.160 1HT + 0.481 2)
    H11 = tansig(− 0.161 4A + 1.410 5Al − 0.549 0SA − 0.535 2SP − 0.882 9Sl + 0.976 0ST + 0.736 7HT − 0.766 9)
    H12 = tansig(− 0.972 7A − 0.640 7Al + 0.172 1SA − 0.166 7SP + 1.174 5Sl − 0.936 7ST + 1.059 7HT − 1.418 7)
    H13 = tansig(0.508 9A − 1.306 7Al − 0.464 2SA − 0.823 3SP + 0.524 0Sl + 0.757 7ST − 0.605 3HT + 1.554 7)
    H14 = tansig(0.535 3A + 0.290 1Al − 1.324 2SA + 1.021 0SP + 0.717 0Sl − 0.735 0ST + 0.249 2HT + 1.695 0)
    H15 = tansig(0.007 9A − 0.286 1Al − 0.279 4SA − 0.174 6SP − 1.443 4Sl − 1.172 9ST + 0.976 9HT + 2.098 4)
    隐含层至输出层
    Hidden layer to output layer
    H = purelin(− 0.283 3H1 − 0.703 9H2 − 0.195 3H3 + 0.205 3H4 + 0.247 8H5 − 0.290 6H6 + 0.690 9H7 − 0.976 3H8
    0.756 1H9 − 0.287 7H10 − 0.004 8H11 − 0.640 0H12 + 0.098 7H13 + 0.253 2H14 − 0.466 5H15 − 0.551 2)
    下载: 导出CSV
  • [1] Kleinn C, Fernández B H, Campos J J. Site productivity estimation using height-diameter relationships in Costa Rican secondary forests[J]. Investigación Agraria Sistemas y Recursos Forestales, 2004, 13: 295−304.
    [2] 孟宪宇.测树学[M]. 第3版. 北京: 中国林业出版社, 2006.

    Meng X Y. Forest measurements[M]. Third edition. Beijing: China Forestry Publishing House, 2006.
    [3] 张超, 彭道黎, 黄国胜, 等. 基于森林清查数据的三峡库区林地立地质量评价[J]. 东北林业大学学报, 2015, 43(11):56−61. doi: 10.3969/j.issn.1000-5382.2015.11.012

    Zhang C, Peng D L, Huang G S, et al. Site quality evaluation in Three Gorges Reservoir region based on forest inventory data[J]. Journal of Northeast Forestry University, 2015, 43(11): 56−61. doi: 10.3969/j.issn.1000-5382.2015.11.012
    [4] 黄国胜, 马炜, 王雪军, 等. 基于一类清查数据的福建省立地质量评价技术[J]. 北京林业大学学报, 2014, 36(3):1−8.

    Huang G S, Ma W, Wang X J, et al. Forestland site quality evaluation of Fujian Province based on continuous forest inventory data[J]. Journal of Beijing Forestry University, 2014, 36(3): 1−8.
    [5] 李际平, 姚东和. BP模型在单木树高与胸径生长模拟中的应用[J]. 中南林业科技大学学报, 1996, 16(3):34−36.

    Li J P, Yao D H. Application of BP neural network model to the simulation of breast-height diameter and tree-height growth[J]. Journal of Central South Forestry University, 1996, 16(3): 34−36.
    [6] 马天晓, 赵晓峰, 黄家荣, 等. 基于人工神经网络的树高曲线模型研究[J]. 河南林业科技, 2006, 26(1):4−5. doi: 10.3969/j.issn.1003-2630.2006.01.002

    Ma T X, Zhao X F, Huang J R, et al. Research on the height-diameter model of tree with artificial neural network[J]. Journal of Henan Forestry Science and Technology, 2006, 26(1): 4−5. doi: 10.3969/j.issn.1003-2630.2006.01.002
    [7] 徐志扬. 基于BP神经网络的马尾松树高曲线模型[J]. 林业调查规划, 2015, 40(2):6−8. doi: 10.3969/j.issn.1671-3168.2015.02.002

    Xu Z Y. Height-diameter model for Pinus massoniana based on BP neural network[J]. Forest Inventory and Planning, 2015, 40(2): 6−8. doi: 10.3969/j.issn.1671-3168.2015.02.002
    [8] 董云飞, 孙玉军, 王轶夫, 等. 基于BP神经网络的杉木标准树高曲线[J]. 东北林业大学, 2014, 42(7):154−156. doi: 10.3969/j.issn.1000-5382.2014.07.036

    Dong Y F, Sun Y J, Wang Y F, et al. Generalized height-diameter model for Chinese fir based on BP neural network[J]. Journal of Northeast Forestry University, 2014, 42(7): 154−156. doi: 10.3969/j.issn.1000-5382.2014.07.036
    [9] 沈剑波, 雷相东, 李玉堂, 等. 基于BP神经网络的长白落叶松人工林林分平均高预测[J]. 南京林业大学学报(自然科学版), 2018, 42(2):147−154.

    Shen J B, Lei X D, Li Y T, et al. Prediction mean height for Larix olgensis plantation based on BP neural network[J]. Journal of Nanjing Forestry University (Natural Sciences Edition), 2018, 42(2): 147−154.
    [10] 张德丰. MATLAB神经网络应用设计[M]. 北京: 机械工业出版社, 2009.

    Zhang D F. MATLAB neural network application design[M]. Beijing: China Machine Press, 2009.
    [11] Sun J, Mao H, Liu J, et al. The research of paddy rice moisture lossless detection based on L-M BP neural network[J]. Computer & Computing Technologies in Agriculture, 2008, 120(1): 83−93.
    [12] Selvamuthu D, Kumar V, Mishra A. Indian stock market prediction using artificial neural networks ontick data[J]. Financial Innovation, 2019, 5(1): 1−12. doi: 10.1186/s40854-018-0118-9
    [13] 郭阳明, 冉从宝, 姬昕禹, 等. 基于组合优化BP神经网络的模拟电路故障诊断[J]. 西北工业大学学报, 2013, 31(1):44−48. doi: 10.3969/j.issn.1000-2758.2013.01.009

    Guo Y M, Ran C B, Ji X Y, et al. Fault diagnosis in analog circuits based on combined-optimization BP neural networks[J]. Journal of Northwestern Polytechnical University, 2013, 31(1): 44−48. doi: 10.3969/j.issn.1000-2758.2013.01.009
    [14] Nukala B T, Shibuya N, Rodriguez A, et al. An efficient and robust fall detection system using wireless gait analysis sensor with artificial neural network (ANN) and support vector machine (SVM) algorithms[J]. Open Journal of Applied Biosensor, 2015, 3(4): 29−39.
    [15] 陈永富, 杨彦臣, 张怀清, 等. 海南岛热带天然山地雨林立地质量评价研究[J]. 林业科学研究, 2000, 13(2):134−140. doi: 10.3321/j.issn:1001-1498.2000.02.005

    Chen Y F, Yang Y C, Zhang H Q, et al. A study on site quality evaluation of natural tropical mountainous rain forest in Hainan Island[J]. Forest Research, 2000, 13(2): 134−140. doi: 10.3321/j.issn:1001-1498.2000.02.005
    [16] 朱光玉, 康立. 森林立地生产力评价指标与方法[J]. 西北林学院学报, 2016, 31(6):275−281. doi: 10.3969/j.issn.1001-7461.2016.06.47

    Zhu G Y, Kang L. A review of forest site productivity evaluation indicators and methods[J]. Journal of Northwest Forestry University, 2016, 31(6): 275−281. doi: 10.3969/j.issn.1001-7461.2016.06.47
    [17] 翁国庆, 白卫国, 王维芳, 等. 地位指数表编制技术规程[S]. 北京: 中国标准出版社, 2015.

    Weng G Q, Bai W G, Wang W F, et al. Technical regulations for drafting of site index table[S]. Beijing: Standards Press of China, 2015.
    [18] 李际平,姚东和. BP模型在单木树高与胸径生长模拟中的应用[J]. 中南林学院学报, 1996, 16(3):34−36.

    Li J P, Yao D H. Application of BP neural network model to the simulation of breast-height diameter and tree-height growth[J]. Journal of Central South Forestry University, 1996, 16(3): 34−36.
    [19] 徐罗, 亢新刚, 郭韦韦, 等. 天然云冷杉针阔混交林立地质量评价[J]. 北京林业大学学报, 2016, 38(5):11−22.

    Xu L, Kang X G, Guo W W, et al. Site quality evaluation of natural spruce-fir and broadleaf mixed stands[J]. Journal of Beijing Forestry University, 2016, 38(5): 11−22.
    [20] 郭如意, 韦新良, 刘姗姗. 天目山区针阔混交林立地质量评价研究[J]. 西北林学院学报, 2016, 31(4):233−240. doi: 10.3969/j.issn.1001-7461.2016.04.39

    Guo R Y, Wei X L, Liu S S. Site quality evaluation of coniferous and broad-leaved mixed forest in Tianmu Mountain[J]. Journal of Northwest Forestry University, 2016, 31(4): 233−240. doi: 10.3969/j.issn.1001-7461.2016.04.39
    [21] 黄旭光, 周俊朝, 黄柏华, 等. 基于人工神经网络对栎树天然林地位指数模拟系统的研究[J]. 河南农业大学学报, 2015, 49(2):190−194.

    Huang X G, Zhou J C, Huang B H, et al. Study of oak growth dynamic simulation system based on artificial neural network[J]. Journal of Henan Agricultural University, 2015, 49(2): 190−194.
    [22] 阙泽胜. 广东省三维地形场景仿真研究[J]. 测绘地理信息, 2016, 41(3):100−102.

    Que Z S. Study on 3D terrain scene simulation in Guangdong Province[J]. Journal of Geomatic, 2016, 41(3): 100−102.
  • 加载中
图(4) / 表(8)
计量
  • 文章访问数:  1749
  • HTML全文浏览量:  399
  • PDF下载量:  81
  • 被引次数: 0
出版历程
  • 收稿日期:  2019-01-15
  • 修回日期:  2019-03-04
  • 网络出版日期:  2019-05-07
  • 刊出日期:  2019-05-01

目录

    /

    返回文章
    返回