[1]孙国祥,闫婷婷,汪小旵,等.基于小波变换和动态神经网络的温室黄瓜蒸腾速率预测[J].南京农业大学学报,2014,37(5):143-152.[doi:10.7685/j.issn.1000-2030.2014.05.023]
 SUN Guoxiang,YAN Tingting,WANG Xiaochan,et al.A method of cucumber transpiration rate forecast based on wavelet transform and dynamic neural network[J].Journal of Nanjing Agricultural University,2014,37(5):143-152.[doi:10.7685/j.issn.1000-2030.2014.05.023]
点击复制

基于小波变换和动态神经网络的温室黄瓜蒸腾速率预测()
分享到:

《南京农业大学学报》[ISSN:1000-2030/CN:32-1148/S]

卷:
37卷
期数:
2014年5期
页码:
143-152
栏目:
出版日期:
2014-09-25

文章信息/Info

Title:
A method of cucumber transpiration rate forecast based on wavelet transform and dynamic neural network
作者:
孙国祥 闫婷婷 汪小旵 陈满 张瑜 狄娇 施印炎 陈景波
南京农业大学工学院/江苏省现代设施农业技术与装备工程实验室, 江苏 南京 210031
Author(s):
SUN Guoxiang YAN Tingting WANG Xiaochan CHEN Man ZHANG Yu DI Jiao SHI Yinyan CHEN Jingbo
College of Engineering/Jiangsu Province Engineering Laboratory for Modern Facility Agriculture Technology and Equipment, Nanjing Agricultural University, Nanjing 210031, China
关键词:
温室黄瓜蒸腾小波变换动态神经网络时间序列预测
Keywords:
greenhousecucumbertranspirationwavelet transformdynamic neural networktime seriesforecast
分类号:
S625.5
DOI:
10.7685/j.issn.1000-2030.2014.05.023
摘要:
针对作物蒸腾速率与温室环境参数间非线性耦合时延性关系,以温室环境参数:空气温度、空气湿度、太阳辐射度、土壤温度、叶面温度、土壤含水量的时间序列为输入量,温室黄瓜蒸腾速率时间序列为输出量,采用小波分解重构方法,分别建立低频时间序列和高频时间序列的非线性自回归动态神经网络(NARX)子网络预测模型,以子网络的预测叠加值为蒸腾速率预测值。结果表明:1层小波分解重构的低频时间序列A1和高频时间序列D1的子网络预测值与蒸腾速率分解重构目标值间相关性决定系数R2分别为0.949和0.853,平均绝对误差(MAE)分别为5.36和2.00 g·h-1。2层小波分解重构的低频时间序列A2和高频时间序列D2的子网络预测值与蒸腾速率分解重构目标值间相关性决定系数R2分别为0.983和0.849,MAE分别为2.88和2.56 g·h-1。1层小波分解重构的时间序列的NARX子网络预测值合成值(A1+D1),2层小波分解重构的时间序列的NARX子网络预测值合成值(A2+D2+D1)和未小波分解重构的原时间序列的NARX预测值与蒸腾速率测量值间相关性决定系数R2分别为0.945、0.974和0.857,MAE分别为5.76、4.42和10.09 g·h-1。小波分解重构的高频和低频时间序列预测合成,能够提高时间序列的预测准确性。同时采用相同网络结构的BP神经网络和NAR动态神经网络预测蒸腾速率时间序列,其预测值与测量值间决定系数R2分别为0.596和0.839,MAE分别为19.55和9.45 g·h-1。NARX预测性能优于NAR和BP神经网络的预测性能,能够应用该方法预测温室黄瓜的蒸腾速率。该方法可推广至多变量非线性强耦合时延性系统中的变量预测。
Abstract:
In order to analyze the relationship of nonlinear, coupling, delay between crop transpiration rate and greenhouse environment parameters, the time series of greenhouse environment parameters:air temperature, air humidity, soil temperature, solar radiation, leaf temperature, and soil moisture were used as inputs, and the transpiration rate of cucumber was used as output. The low-frequency and high-frequency time series by the method of wavelet decomposition-reconstruction were used to establish the prediction models of nonlinear autoregressive with external input(NARX)dynamic neural sub networks respectively, and the predicted value of the subnets were accumulated as the predicted value of transpiration rate of cucumber. The results showed that:the regression coefficients(R2)between predicted values by the subnets based on the time series of low-frequency A1 and high-frequency D1 by one layer of wavelet decomposition-reconstruction and the target value of transpiration rate by decomposition-reconstruction were 0.949 and 0.853 respectively, and the mean absolute error(MAE)were 5.36 and 2.00 g·h-1 respectively. The regression coefficients between predicted values by the subnets based on the time series of low-frequency A2 and high-frequency D2 by two layers of wavelet decomposition-reconstruction and the target value of transpiration rate by decomposition-reconstruction were 0.983 and 0.849 respectively, and the MAE were 2.88 and 2.56 g·h-1 respectively. The regression coefficient between accumulated predictive values(A1+D1)by the NARX dynamic neural sub networks based on the time series of low-frequency A1 and high-frequency D1 by one layer of wavelet decomposition-reconstruction and the measured value of transpiration rate was 0.945, and the MAE was 5.76 g·h-1. The regression coefficient between accumulated predictive values(A2+D2+D1)by the NARX dynamic neural sub networks based on the time series of low-frequency A2 and high-frequency D2 and D1 by two layers of wavelet decomposition-reconstruction and the measured value of transpiration rate was 0.974, and the MAE was 4.42 g·h-1. The regression coefficients between predicted value of the original time series by the NARX dynamic neural network and the measured value of transpiration rate was 0.857, and the MAE was 10.09 g·h-1. Synthetic time series of low-frequency and high-frequency time series of forecast, can improve the accuracy of time series prediction. The regression coefficients between predicted value of the original time series by the BP neural network and NAR neural network under the same network structure and the measured value of transpiration rate were 0.596 and 0.839 respectively, and the MAE were 19.55 and 9.45 g·h-1 respectively. The forecast performance of NARX was better than NAR and BP neural network, which can forecast the transpiration rate of cucumber in the greenhouse accurately. This method can be extended up to predict the parameters for the multivariable, nonlinear, strong coupling and delay system.

参考文献/References:

[1] Harmanto S V M,Babel M S,Tantau H J. Water requirement of drip irrigated tomatoes grown in greenhouse in tropical environment[J]. Agricultural Water Management,2005,71(3):225-242
[2] Yuan B Z,Kang Y,Nishiyama S. Drip irrigation scheduling for tomatoes in unheated greenhouse[J]. Irrigation Science,2001,20(3):149-154
[3] Yuan B Z,Sun J,Nishiyama S. Effect of drip irrigation on strawberry growth and yield inside a plastic greenhouse[J]. Biosystems Engineering,2004,87(2):237-245
[4] 刘祖贵,段爱旺,吴海卿,等. 水肥调配施用对温室滴灌番茄产量及水分利用效率的影响[J]. 中国农村水利水电,2003(1):10-12
[Liu Z G,Duan A W,Wu H Q,et al. Impacts of water and fertilizer allocation on tomato yield and water use efficiency in drip irrigation greenhouse[J]. China Rural Water and Hydropower,2003(1):10-12(in Chinese with English abstract)]
[5] Hanson B R,May D M. Crop coefficients for drip-irrigated processing tomato[J]. Agricultural Water Management,2006,81(3):381-399
[6] Miranda F R,Gondim R S,Costa C A G. Evapotranspiration and crop coefficients for Tabasco pepper(Capsicum frutescens L.)[J]. Agricultural Water Management,2006,82(2):237-246
[7] Wang S,Boulard T. Greenhouse crop transpiration simulation from external climate conditions[J]. Agricultural and Forest Meteorology,2000,100(1):25-34
[8] Demrati H,Boulard T,Fatnassi H,et al. Microclimate and transpiration of a greenhouse banana crop[J]. Biosystems Engineering,2007,98(1):66-78
[9] 汪小旵,丁为民,罗卫红,等. 南方现代温室能耗预测模型的建立与分析[J]. 南京农业大学学报,2006,29(1):116-120. doi:10.7685/j.issn.1000-2030.2006.01.026
[Wang X C,Ding W M,Luo W H,et al. An energy prediction model for modern greenhouse in the south of China[J]. Journal of Nanjing Agricultural University,2006,29(1):116-120(in Chinese with English abstract)]
[10] 汪小旵,丁为民,罗卫红,等. 温室小气候测量试验设计及其夏季蒸腾研究[J]. 农业机械学报,2003,34(4):86-89
[Wang X C,Ding W M,Luo W H,et al. Design experiment to measure microclimate and study transpiration in summer in greenhouse[J]. Transactions of the Chinese Society for Agricultural Machinery,2003,34(4):86-89(in Chinese with English abstract)]
[11] 汪小旵,罗卫红,丁为民,等. 南方现代化温室黄瓜夏季蒸腾研究[J]. 中国农业科学,2002,35(11):1390-1395
[Wang X C,Luo W H,Ding W M,et al. Cucumber canopy transpiration in subtropical modern greenhouse under summer climate condition[J]. Scientia Agricultura Sinica,2002,35(11):1390-1395(in Chinese with English abstract)]
[12] 戴剑锋,金亮,罗卫红,等. 长江中下游Venlo型温室番茄蒸腾模拟研究[J]. 农业工程学报,2006,22(3):99-103
[Dai J F,Jin L,Luo W H,et al. Simulation of greenhouse tomato canopy transpiration in Yangtze River Delta[J]. Transactions of the CSAE,2006,22(3):99-103(in Chinese with English abstract)]
[13] 刘浩,段爱旺,孙景生,等. 基于Penman-Monteith方程的日光温室番茄蒸腾量估算模型[J]. 农业工程学报,2011,27(9):208-213
[Liu H,Duan A W,Sun J S,et al. Estimating model of transpiration for greenhouse tomato based on Penman-Monteith equation[J]. Transactions of the CSAE,2011,27(9):208-213(in Chinese with English abstract)]
[14] 彭致功. 日光温室滴管条件下小气候变化和植株蒸腾规律的研究[D]. 北京:中国农业科学院,2002
[Peng Z G. Study on the environmental factors and plant transpiration under drip irrigation solar greenhouse[D]. Beijing:Chinese Academy of Agricultural Sciences,2002(in Chinese with English abstract)]
[15] 姚勇哲,李建明,张荣,等. 温室番茄蒸腾量与其影响因子的相关分析及模型模拟[J]. 应用生态学报,2012,23(7):1869-1874
[Yao Y Z,Li J M,Zhang R,et al. Greenhouse tomato transpiration and its affecting factors:correlation analysis and model simulation[J]. Chinese Journal of Applied Ecology,2012,23(7):1869-1874(in Chinese with English abstract)]
[16] 牛勇,刘洪禄,吴文勇,等. 基于大型称重式蒸渗仪的日光温室黄瓜蒸腾规律研究[J]. 农业工程学报,2011,27(1):52-56
[Niu Y,Liu H L,Wu W Y,et al. Cucumber transpiration by large-scale weighting lysimeter in solar greenhouse[J]. Transactions of the CSAE,2011,27(1):52-56(in Chinese with English abstract)]
[17] 王世谦,苏娟,杜松怀. 基于小波变换和神经网络的短期风电功率预测方法[J]. 农业工程学报,2010,26(增刊2):125-129
[Wang S Q,Su J,Du S H. A method of short-term wind power forecast based on wavelet transform and neural network[J]. Transactions of the CSAE,2010,26(Suppl 2):125-129(in Chinese with English abstract)]
[18] 佟长福,史海滨,包小庆,等. 基于小波分析理论组合模型的农业需水量预测[J]. 农业工程学报,2011,27(5):93-98
[Tong C F,Shi H B,Bao X Q,et al. Application of a combined model based on wavelet analysis for predicting crop water requirement[J]. Transactions of the CSAE,2011,27(5):93-98(in Chinese with English abstract)]
[19] 柴琳娜,屈永华,张立新,等. 基于自回归神经网络的时间序列叶面积指数估算[J]. 地球科学进展,2009,24(7):756-768
[Chai L N,Qu Y H,Zhang L X,et al. Estimating time series leaf area index based on recurrent neural networks[J]. Advances in Earth Science,2009,24(7):756-768(in Chinese with English abstract)]

相似文献/References:

[1]薄凯亮,沈佳,钱春桃,等.‘ 北京截头’× 西双版纳黄瓜重组自交系群体重要农艺性状的遗传分析[J].南京农业大学学报,2011,34(3):20.[doi:10.7685/j.issn.1000-2030.2011.03.004]
 BO Kai-liang,SHEN Jia,QIAN Chun-tao,et al.Genetic analysis of the important agronomic traits on ‘Beijingjietou’× Xishuangbanna cucumber recombinant inbred lines[J].Journal of Nanjing Agricultural University,2011,34(5):20.[doi:10.7685/j.issn.1000-2030.2011.03.004]
[2]史建磊,娄群峰,钱春桃,等.黄瓜染色体片段导入系的构建与遗传评价[J].南京农业大学学报,2011,34(1):20.[doi:10.7685/j.issn.1000-2030.2011.01.004]
 SHI Jian-lei,LOU Qun-feng,QIAN Chun-tao,et al.Construction and genetic evaluation of chromosome segment introgression lines in cucumber[J].Journal of Nanjing Agricultural University,2011,34(5):20.[doi:10.7685/j.issn.1000-2030.2011.01.004]
[3]魏跃,陈啸寅,李振陆,等.黄瓜6磷酸葡萄糖酸脱氢酶基因cDNA片段的克隆及表达分析[J].南京农业大学学报,2010,33(1):37.[doi:10.7685/j.issn.1000-2030.2010.01.008]
 WEI Yue CHEN Xiao-yin,LI Zhen-lu,WANG Yong-ping,et al.Cloning of 6-phosphogluconate dehydrogenase gene cDNA fragments from cucumber and expression analysis[J].Journal of Nanjing Agricultural University,2010,33(5):37.[doi:10.7685/j.issn.1000-2030.2010.01.008]
[4]李为观,杨寅桂,魏跃,等.热胁迫下黄瓜幼苗生理生化指标变化及CSHSP70基因表达[J].南京农业大学学报,2010,33(3):47.[doi:10.7685/j.issn.1000-2030.2010.03.009]
 LI Wei-guan,YANG Yin-gui,WEI Yue,et al.Expression of CSHSP_{70} gene and changes of some physiological characters in cucumber seedlings during heat stress[J].Journal of Nanjing Agricultural University,2010,33(5):47.[doi:10.7685/j.issn.1000-2030.2010.03.009]
[5]娄丽娜,陈劲枫,钱春桃,等.利用胚培养诱导单性结实黄瓜果实形成单倍体植株的研究[J].南京农业大学学报,2009,32(2):30.[doi:10.7685/j.issn.1000-2030.2009.02.007]
 LOU Li-na,CHEN Jin-feng,QIAN Chun-tao,et al.Recovery of cucumber (Cucumis sativus L.)haploid plants through embryo culture in parthenocarpic cucumber fruits[J].Journal of Nanjing Agricultural University,2009,32(5):30.[doi:10.7685/j.issn.1000-2030.2009.02.007]
[6]耿贝贝,郭强,师慈,等.低浓度蔗糖对黄瓜下胚轴不定根发生有诱导作用[J].南京农业大学学报,2009,32(3):157.[doi:10.7685/j.issn.1000-2030.2009.03.029]
 GENG Bei-bei,GUO Qiang,SHI Ci,et al.Adventitious rooting of cucumbers(Cucumis sativus L.)induced by low concentration of sucrose[J].Journal of Nanjing Agricultural University,2009,32(5):157.[doi:10.7685/j.issn.1000-2030.2009.03.029]
[7]刁卫平,陈劲枫,雷春,等.影响黄瓜未授粉子房培养胚发生因素的研究[J].南京农业大学学报,2008,31(1):137.[doi:10.7685/j.issn.1000-2030.2008.01.028]
 DIAO Wei-ping,CHEN Jin-feng,LEI Chun,et al.Factors affecting embryo formation in unpollinated ovary culture of cucumber[J].Journal of Nanjing Agricultural University,2008,31(5):137.[doi:10.7685/j.issn.1000-2030.2008.01.028]
[8]张鹏,王飞,张列峰,等.黄瓜叶片生长期间内肽酶活性和同工酶变化及其生化特性[J].南京农业大学学报,2007,30(2):49.[doi:10.7685/j.issn.1000-2030.2007.02.010]
 ZHANG Peng,WANG Fei,ZHANG Lie-feng,et al.Study on endopeptidase isoenzymes and their characters in cucumber leaf development period[J].Journal of Nanjing Agricultural University,2007,30(5):49.[doi:10.7685/j.issn.1000-2030.2007.02.010]
[9]汪小旵,丁为民,罗为红,等.南方现代化温室能耗预测模型的建立与分析[J].南京农业大学学报,2006,29(1):116.[doi:10.7685/j.issn.1000-2030.2006.01.026]
 WANG Xiao-chan,DING Wei-min,LUO Wei-hong,et al.AnenergypredictionmodelformoderngreenhouseinthesouthofChina[J].Journal of Nanjing Agricultural University,2006,29(5):116.[doi:10.7685/j.issn.1000-2030.2006.01.026]
[10]钱春桃,陈劲枫,罗向东.黄瓜抗枯萎病异源易位植株AT-04的鉴定筛选[J].南京农业大学学报,2006,29(2):20.[doi:10.7685/j.issn.1000-2030.2006.02.005]
 QIAN Chun-tao,CHEN Jin-feng,LUO Xiang-dong.Identification and characterization of cucumber alien translocation plant AT-04 with resistance to fusarium wilt[J].Journal of Nanjing Agricultural University,2006,29(5):20.[doi:10.7685/j.issn.1000-2030.2006.02.005]

备注/Memo

备注/Memo:
收稿日期:2013-12-10。
基金项目:国家自然科学基金项目(61273227)
作者简介:孙国祥,博士研究生,主要从事农业智能化装备方面的研究,E-mail:sguoxiang@njau.edu.cn
通讯作者:汪小旵,教授,博导,主要从事农业生物环境模拟与调控方面的研究,E-mail:wangxiaochan@njau.edu.cn
更新日期/Last Update: 1900-01-01