突变论尖角模型在树材积估算中的应用研究

doi:10.13209/j.0479-8023.2016.092

摘要/Abstract

摘要：

为了研究树木的生长是否也存在突变的问题, 将突变论的尖角模型理论应用于我国北方——内蒙古赤峰市旺业甸林场 10 个常见树种单木材积的测定和计算。选取 10 个北方常见树种, 对其树高(H)、胸径(D)、材积(V)、地径(D₀)等参数进行测定, 并以此作为真值, 利用改进差分进化算法对建立的“材积-树高-胸径”模型进行内符合精度检验, 利用边缘树种对模型进行外符合精度检验, 比较模型计算值与仪器测量值之间的差异。在经过一定的迭代之后, 结果表明: 模型的总体相对误差 RS 在[0.001, 0.05]范围, 平均相对误差 E 在 [-0.11, 0.02]范围, 总体预估精度 P 大于 80%, 说明模型拟合较好, 即突变论中的尖角模型理论适用于我国北方 10 个树种单木材积模型的建立, 同时说明随着树木的不断生长, 当树高达到一定高度时, 材积将遵循突变论的尖角模型理论。这是突变论的尖角模型理论首次应用到单木材积的测定, 并得到较好的拟合结果。同时, 在不连续的测量状态下建立的突变低维模型为树木材积的研究和各树种间的空间竞争研究提供了一定的理论和实践依据。

关键词: 突变论, 北方树种, 尖角模型, 改进差分进化算法, 树材积

Abstract:

In order to study whether the growth of trees existence mutations problems, measurement and calculation of the mutation based on the cusp model theory is applied to 10 common species single tree volume in Wang Ye Dian forest farm which is located in Chifeng City, Inner Mongolia, China. 10 northern common tree species were selected to carry out the experiment, the tree height (H), DBH (D), volume (V), diameter (D₀) were measured which as the true value. To test the Internal and external accord accuracy of the “V-H-D” model by using the improved differential evolution algorithm and edge species and then compare the difference between model calculation value and instrument measuring. After a certain iteration, the results showed that the overall relative error (RS) of the model is in the range of [0.001, 0.05], the average relative error (E) is in the range of [-0.11, 0.02], and the overall prediction accuracy (P) is greater than 80%. The outcome had shown the good fitness of the model, namely the cusp catastrophe theory model applies to northern China 10 single species tree volume. With the growing of trees, when the height of the tree reaches a certain height, the volume will follow the theory model of cusp catastrophe theory. This is the cusp catastrophe theory model was first applied to the determination of single tree volume and got a better fitting result. Meanwhile, a low-dimensional model of mutation was established in the discontinuous measurement, which provides a theoretical and practical basis for the study of tree volume and various species of spatial competition.

Key words: catastrophe theory, northern tree species, cusp model, improved differential evolution algorithm, trees volume

中图分类号:

S758

李伟, 陈秀万, 毛海颖, 李飞. 突变论尖角模型在树材积估算中的应用研究[J]. 北京大学学报自然科学版, 2017, 53(1): 189-196.

Wei LI, Xiuwan CHEN, Haiying MAO, Fei LI. Application and Researches for Trees Volume Estimation Based on Cusp Catastrophe Theory Model[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2017, 53(1): 189-196.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://xbna.pku.edu.cn/CN/10.13209/j.0479-8023.2016.092

https://xbna.pku.edu.cn/CN/Y2017/V53/I1/189

图/表 7

参考文献 24

[1]	王海波, 金沙. 突变论在种间生态学中的应用. 生态学杂志, 1988, 7(3): 41-45
[2]	张青. 突变论模型在生态系统中的应用. 北京林业大学学报, 1997, 19(4): 76-81
[3]	张宏蕃. 关于常微分方程数学模型的建立分析. 今日科苑, 2008(4): 275-276
[4]	Peter Y P, Mohamed A.A stochastic catastrophe model using two-fluid model parameters to investi-gate traffic safety on urban arterials. Accident Analysis & Prevention, 2011, 43(3): 1267-1278
[5]	Thom R.Atructural Stability and Morphogenesis. New York: Benjamin Press, 1972
[6]	李家贤. 奇点(含突变)理论的某些应用. 系统工程, 1991, 42(9): 1-7
[7]	Jones D D, Walters C J.Catastrophe theory and fisheries regulation. Journal of the Fisheries Research Board of Canada, 1976, 12: 2829-2833
[8]	Jones D D.Catastrophe theory applied to ecological systems. Simulation, 1977, 29: 1-15
[9]	Loehle C.Catastrophe theory in ecology: a critical review and an example of the butterfly catastrophe. Ecological Modelling, 1989, 49: 125-152
[10]	李勇, 戴勇, 吴大奎, 等. 基于突变论的储层边界尖点突变技术及应用研究. 成都理工大学学报: 自然科学版, 2010, 37(5): 546-549
[11]	尼庆伟, 王岩, 郑丙辉, 等. 基于突变理论模型的卧龙湖湿地退化判别研究. 湿地科学, 2014, 12(1): 1-6
[12]	孙文浩, 姚俭. 基于蜱虫效应的Sgn函数突变模型. 数学理论与应用, 2013, 33(4): 73-79
[13]	Hughes P, Brundrit G B.An index to assess South Africa vulnerability to sea-level rise. South African Journal of Science, 1992, 8: 308-311
[14]	王瑞燕, 赵庚星, 姜曙千, 等. 基于遥感及突变理论的生态环境脆弱性时空演变: 以黄河三角洲垦利县为例. 应用生态学报, 2008, 19(8): 1782-1788
[15]	孙玉荣, 张贵. 基于神经网络的森林火灾危害程度预测研究. 西北林学院学报, 2010, 25(3): 147-150
[16]	强明, 郭春喜, 周红宇. 基于神经网络的 GPS 高程拟合方法优选及精度分析. 重庆交通大学学报: 自然科学版, 2012, 31(4): 815-818
[17]	袁义福. 不同氮磷比条件对外来种火炬树与本地种麻栎、荆条的种间关系的影响[D]. 济南: 山东大学, 2014
[18]	梁影, 金铭, 乔晓林. 一种改进的遗传算法. 科学技术与工程, 2012, 15(12): 3636-3644
[19]	简静芳. 基于生物入侵思想的自适应遗传算法优化. 陕西理工学院学报: 自然科学版, 2014, 30(6): 50-54
[20]	Storn R, Price K.Differential evolution: a simple and efficient adaptive scheme for global optimization over continuous spaces. Journal of Global Optimization, 1995, 23(4): 341-359
[21]	王艳宜. 改进差分进化算法及其应用. 机械设计与研究, 2010, 26(5): 13-16
[22]	龙文, 徐松金, 焦建军. 一种改进的约束优化差分进化算法. 计算机工程与应用, 2012, 48(25): 34-38
[23]	赵娟. 遗传算法在最优化问题中的应用. 研究与开发, 2012(8): 17-20
[24]	王立国, 魏芳洁. 结合遗传算法和蚁群算法的高光谱图像波段选择. 中国图象图形学报, 2013, 18(2): 235-242

地区	年均气温/℃	年积温/℃	极端低温/℃	极端高温/℃	无霜期/d	年日照/h	年均风速/(m·s^-1)	年均降水/mm
赤峰市	0~7	2600~3800	-10	41	78~90	2700~3100	2.5	300~500
喀喇沁旗	3.5~7	2000~3200	-30.9	37.4	110~150	2913.3	2.8	591.2

地区	年均气温/℃	年积温/℃	极端低温/℃	极端高温/℃	无霜期/d	年日照/h	年均风速/(m·s^-1)	年均降水/mm
赤峰市	0~7	2600~3800	-10	41	78~90	2700~3100	2.5	300~500
喀喇沁旗	3.5~7	2000~3200	-30.9	37.4	110~150	2913.3	2.8	591.2

树种	模型	RMSE	SSE	R	R²	DC	C-S	F
油松 Pinus tabuliformis.	y=-40.671x₁⁴-2.975x₂² +2.742x₃+13.753	0.0200	0.0064	0.9969	0.9938	0.9938	0.00881	2232.1520
樟子松 Pinus sylvestnis var. mongolica Litv.	y=-5.048x₁⁴+9.821x₂² -0.0002x₃+0.0014	0.0044	0.0002	0.9995	0.9990	0.9990	0.0017	9134.0809
落叶松 Larix gmelinii (Ruprecht) Kuzeneva.	y=-1.927x₁⁴+9.067x₂² +2.754x₃-0.057	0.0520	0.0461	0.9919	0.9839	0.9839	0.1097	918.9520
红松 Pinus koraiensis Siebold et Zuccarini.	y=-7.958x₁⁴+0.0001x₂² +0.00032x₃-0.0429	0.0090	0.0007	0.9936	0.9873	0.9873	0.0041	467.1430
五角枫 Acer elegantulum Fang et P. L. Chiu.	y=-5.536x₁⁴+8.979x₂² -1.837x₃+4.467	0.0015	1.3536	0.9998	0.0006	0.9996	0.0002	9952.2379
榆树 Ulmus pumila L.	y=-4.199x₁⁴+0.0001x₂² -0.0001x₃+0.0003	0.0355	0.0151	0.9904	0.9809	0.9809	0.0289	514.8484
柞树 Xylosma racemosum.	y=-6.343x₁⁴+7.258x₂² +0.0005x₃-0.0421	0.0152	0.0032	0.9945	0.9891	0.9891	0.0400	1085.7894
黑桦 Betula dahurica Pall.	y=-3.544x₁⁴+9.346x₂² +2.177x₃-0.0183	0.0150	0.0033	0.9969	0.9939	0.9939	0.0135	1956.3780
白桦 Betula platyphylla Suk.	y=-5.166x₁⁴-9.459x₂² +0.0016x₃-0.1594	0.0498	0.0447	0.9829	0.9662	0.9662	0.1472	457.7420
山杨 Populus davidiana.	y=-2.507x₁⁴+7.583x₂² +0.00032x₃-0.0738	0.0700	0.0986	0.9913	0.9827	0.9827	-0.0509	1024.1950

树种	模型	RMSE	SSE	R	R²	DC	C-S	F
油松 Pinus tabuliformis.	y=-40.671x₁⁴-2.975x₂² +2.742x₃+13.753	0.0200	0.0064	0.9969	0.9938	0.9938	0.00881	2232.1520
樟子松 Pinus sylvestnis var. mongolica Litv.	y=-5.048x₁⁴+9.821x₂² -0.0002x₃+0.0014	0.0044	0.0002	0.9995	0.9990	0.9990	0.0017	9134.0809
落叶松 Larix gmelinii (Ruprecht) Kuzeneva.	y=-1.927x₁⁴+9.067x₂² +2.754x₃-0.057	0.0520	0.0461	0.9919	0.9839	0.9839	0.1097	918.9520
红松 Pinus koraiensis Siebold et Zuccarini.	y=-7.958x₁⁴+0.0001x₂² +0.00032x₃-0.0429	0.0090	0.0007	0.9936	0.9873	0.9873	0.0041	467.1430
五角枫 Acer elegantulum Fang et P. L. Chiu.	y=-5.536x₁⁴+8.979x₂² -1.837x₃+4.467	0.0015	1.3536	0.9998	0.0006	0.9996	0.0002	9952.2379
榆树 Ulmus pumila L.	y=-4.199x₁⁴+0.0001x₂² -0.0001x₃+0.0003	0.0355	0.0151	0.9904	0.9809	0.9809	0.0289	514.8484
柞树 Xylosma racemosum.	y=-6.343x₁⁴+7.258x₂² +0.0005x₃-0.0421	0.0152	0.0032	0.9945	0.9891	0.9891	0.0400	1085.7894
黑桦 Betula dahurica Pall.	y=-3.544x₁⁴+9.346x₂² +2.177x₃-0.0183	0.0150	0.0033	0.9969	0.9939	0.9939	0.0135	1956.3780
白桦 Betula platyphylla Suk.	y=-5.166x₁⁴-9.459x₂² +0.0016x₃-0.1594	0.0498	0.0447	0.9829	0.9662	0.9662	0.1472	457.7420
山杨 Populus davidiana.	y=-2.507x₁⁴+7.583x₂² +0.00032x₃-0.0738	0.0700	0.0986	0.9913	0.9827	0.9827	-0.0509	1024.1950

树种	迭代次数	RS	E	P/%
油松-Pinus tabuliformis.	317	0.0200	-0.027	90.13
樟子松- Pinus sylvestnis var. mongolica Litv.	309	0.0436	0.008	97.21
落叶松-Larix gmelinii (Ruprecht) Kuzeneva.	338	0.0520	0.015	89.10
红松-Pinus koraiensis Siebold et Zuccarini.	296	0.0092	-0.113	87.12
五角枫-Acer elegantulum Fang et P. L. Chiu.	314	0.0015	0.008	88.17
榆树-Ulmus pumila L.	314	0.0355	0.017	89.21
柞树-Xylosma racemosum.	246	0.0151	-0.021	90.97
黑桦-Betula dahurica Pall.	280	0.0154	0.001	93.26
白桦-Betula platyphylla Suk.	267	0.0498	0.019	94.55
山杨-Populus davidiana.	294	0.0702	-0.007	96.78