欢迎访问作物学报，今天是
作物学报

作物学报 ›› 2009, Vol. 35 ›› Issue (2): 341-347.doi: 10.3724/SP.J.1006.2009.00341

• 耕作栽培·生理生化 • 上一篇    下一篇

应用神经网络和统计模型预测大豆生长发育阶段

张久权1,2;张凌霄2;张明华3;WATSON Clarence 4   

  1. 1中国农业科学院烟草研究所，青岛266101；2Delta Research and Extension Center, Mississippi State University,Stoneville 38776, MS,USA;3Department of Land, Air and Water Resources,University of California, Davis 95616,CA,USA;4 Oklahoma State University, Stillwater 74078-6009,OK,USA
  • 收稿日期:2008-01-07 修回日期:2008-09-10 出版日期:2009-02-12 网络出版日期:2008-12-12
  • 基金资助:

    本研究由密西西比大豆促进委员会研究基金（20021013）资助

Nrediction of Soybean Growth and Development Stages Using Artificial Neural Network and Statistical Models

ZHANG Jiu-Quan1,2,ZHANG Ling-Xiao2,ZHANG Ming-Hua3, WATSON Clarence4   

  1. 1Tobacco Research Institute, Chinese Academy of Agricultural Sciences, Qingdao 266101,China；2Delta Research and Extension Center, Mississippi State University,Stoneville 38776, MS,USA;3Department of Land, Air and Water Resources,University of California, Davis 95616,CA,USA;4 Oklahoma State University, Stillwater 74078-6009,OK,USA
  • Received:2008-01-07 Revised:2008-09-10 Published:2009-02-12 Published online:2008-12-12

PDF

2466

被引次数

8

摘要：

预测大豆的物候期对指导大豆生产、安排农事活动等具有重大意义。本研究构建了一种简单、但有效的预测模型，使大豆种植者能够较准确地预测大豆各生育阶段的具体日期。试验地位于美国密西西比Delta研究推广中心(经度: 90°55'W，纬度: 33°25'N)。试验进行了5年，以前4年(1998—2001)数据构建模型，第5年(2002)的数据验证模型。为简化模型，杂草、病虫害、干旱等干扰因素被优化排除。采用逐步回归(SR)、神经网络(ANN)以及内插法构建模型。营养(V-stage)和生殖(R-stage)生长阶段分别建模。结果表明，通过播期(PD)和从播种到某阶段的相对平均天数能很准确地预测营养生长各阶段的具体日期；可通过播种日期和成熟期组数值(MG)准确预测生殖生长各阶段的具体日期。3种方法中，神经网络所构建的模型准确度最高，具有较好的推广应用价值。

关键词: 大豆, 生长期, 成熟期组, 神经网络, 模拟, 播种日期, 物候学

Abstract:

The prediction of soybean phenology is important in many aspects of soybean production. The study objective was to develop predictive models, using a simple and effective modeling technique, which can allow producers to predict soybean growth and development stages in their fields. The experiments were conducted at the Delta Research and Extension Center in Stoneville, Mississippi (latitude: 33°25' N, longitude: 90°55' W) under irrigated conditions. The models were constructed using four-year field data (1998 to 2001), and validated with the fifth-year data (2002). Potential factors affecting stages of soybean growth and development were considered for developing the models. Affecting factors, such as weeds, insects, diseases, and drought stress, were controlled optimally to simplify the modeling procedures. In addition, stepwise regression (SR) analysis, artificial neural networks (ANN), and interpolation approaches were used to construct the models. The modeling of soybean growth and development processes was separated into two distinct periods: vegetative growth stage (V-stage) and reproductive growth stage (R-stage). The models included ten V-stages (up to V8) and eight R-stages. In the V-stage models, PD (planting date) and mean relative time-span from planting to a particular stage were the only significant parameters, whereas in R-stage models, PD and MG (maturity group) were significant. The models obtained accurate predictions when only using PD, MG, and mean relative time-span from planting to a particular stage. The ANN method provided the greatest accuracy in predicting phenological events, indicating that the ANN method can be effectively applied in crop modeling.

Key words: Soybean, Growth stage, Maturity group, Artificial neural networks, Modeling, Planting date, Phenology

[1]Bernard R L. Two major genes for time of flowering and maturity in soybeans. Crop Sci, 1971, 11: 242–244
[2]Buzzel R I, Voldeng H D. Inheritance of insensitivity to long day length. Soybean Genet News, 1980, 7: 26–29
[3]McBlain B A, Bernard R L. A new gene affecting the time of flowering in soybeans. J Hered, 1987, 78: 160–162
[4]Borthwick H A, Parker M W. Effective of photoperiodic treat-ments of plants of different age. Bot Gaz, 1938, 100: 245–249
[5]Cregan P B, Hartwig E E. Characterization of flowering response to photoperiod in diverse soybean genotypes. Crop Sci, 1984, 24: 657–660
[6]Johnson H W, Brothwick H A, Leffel R C, Effects of photoperiod and time of planting on rate of development of soybeans in vari-ous stages of life cycle. Bot Gaz, 1960, 122: 77–95
[7]Thomas F J, Raper C D Jr. Photoperiodic control of seed filling for soybeans. Crop Sci, 1976, 16: 667–672
[8]Cartter J L, Hartwig E E. The Management for Soybeans. In: Norman A G ed. The Soybean. New York: Academic Press, 1963, pp 161–226
[9]Acock B, Pachepsky Y A, Acock M C, Reddy V R, Whisler F D. Modeling soybean cultivars development rate, using field data from the Mississippi valley. Agron J, 1997, 89: 994–1002
[10]Foroun N, Mundel H H, Saindon G, Entz T. Effect of level and timing of moisture stress on soybean plant development and yield components. Irrig Sci, 1993, 13: 149–155
[11]Jones P G, Laing D R. Simulation of the phenology of soybeans. Agric Syst, 1978, 3: 295–311
[12]Specht J E, Elmore R W, Eisenhauer D E, Klocke N W. Growth stage scheduling criteria for sprinkler-irrigated soybeans. Irrig Sci, 1991, 10: 99–111
[13]Welch S M, Jones J W, Brennan M W, Reeder G, Jacobson B M. PC Yield: Model-based decision support for soybean production. Agric Syst, 2002, 74: 79–98
[14]Blackard J A, Dean D J. Comparative accuracies of artificial neural networks and discriminate analysis in predicting forest cover types from cartographic variables. Comput Electron Agric, 1999, 24: 131–151
[15]Elizondo D R, McClendon R W, Hoogenboom G. Neural network models for predicting flowering and physiological maturity of soybean. Transact ASAE, 1994, 37: 981–988
[16]Zhang L X, Zhang J, Kyei-Boahen S, Watson C E. Developing phenological prediction tables for soybean. Crop Management. doi: 10.1094/ CM-2004-1025-01-RS (2004-10-25)
[2008-01-04] http://www.plantmanagementnetwork.org/pub/cm/research/2004/tables/.
[17]Fehr W R, Caviness C E. Stage of soybean development. In: Iowa State University Cooperative Extension Service Special Report, 1977, p 80
[18]Bishop C M. Neural Networks for Pattern Recognition. Oxford, England: Clarendon Publishing, 1995
[19]Jain A, Mao J. Artificial Neural Networks: A tutorial. Computer, 1996, 29: 31–44
[20]Mitchell T M. Machine Learning. Boston, MA: McGraw-Hill Co, 1997
[1] 金昱何, 王雪菲, 徐张一娃, 缪怡宁, 蒋云杰, 伊莹, 缪德麟, 朱静仪, 钟一帆, 陈铭亨, 方芳, 刘鹏. 外源激素对低温胁迫下大豆叶片叶绿素荧光参数及抗氧化酶系统的影响[J]. 作物学报, 2026, 52(6): 1817-1829.
[2] 唐宽强, 李公允, 宋美毅, 赵雪, 常春玲. 大豆株高性状全基因组关联分析及预测模型构建[J]. 作物学报, 2026, 52(6): 1743-1756.
[3] 姚术, 郭凯悦, 翟慧慧, 姚佳慧, 邓文琪, 闫玲, 黄驰, 高阳, 俞嫣然, 赵振邦, 李英慧, 王晓波, 李佳佳. 大豆苗期耐低铁综合评价及优异种质筛选[J]. 作物学报, 2026, 52(5): 1373-1387.
[4] 张晴, 杨昱, 郭茜, 岳霈尧, 殷丛丛, 牛景萍, 赵晋忠, 杜维俊, 岳爱琴. 大豆GmARA6a的克隆及响应盐胁迫的功能分析[J]. 作物学报, 2026, 52(2): 480-493.
[5] 王克晶, 李向华. 我国珍稀的大豆属多年生烟豆和短绒野大豆物种遗传资源濒危性评估分析[J]. 作物学报, 2025, 51(8): 2009-2019.
[6] 孟然, 李赵嘉, 冯薇, 陈悦, 刘路平, 杨春燕, 鲁雪林, 王秀萍. 大豆不同生育时期耐盐性综合评价及耐盐种质筛选[J]. 作物学报, 2025, 51(8): 1991-2008.
[7] 贺红利, 张雨涵, 杨静, 程云清, 赵杨, 李星诺, 司洪亮, 张兴政, 杨向东. 大豆e1-as基因突变体的创制及生理分析[J]. 作物学报, 2025, 51(8): 2228-2239.
[8] 胡蒙, 沙丹, 张晟瑞, 谷勇哲, 张世碧, 李静, 孙君明, 邱丽娟, 李斌. 大豆分枝数QTL定位及候选基因筛选[J]. 作物学报, 2025, 51(7): 1747-1756.
[9] 王琼, 邹丹霞, 陈兴运, 张威, 张红梅, 刘晓庆, 贾倩茹, 魏利斌, 崔晓艳, 陈新, 王学军, 陈华涛. 大豆开花时间和成熟期性状全基因组关联分析与候选基因预测[J]. 作物学报, 2025, 51(6): 1558-1568.
[10] 殷丛丛, 李睿琦, 岳霈尧, 李晨, 牛景萍, 赵晋忠, 杜维俊, 岳爱琴. 基于闭合哑铃介导等温扩增可视化检测大豆花叶病毒SC15方法的建立及应用[J]. 作物学报, 2025, 51(5): 1248-1260.
[11] 许睿, 何妙华, 王昊, 李卫, 任杰, 夏志强. 基于空间转录组技术解析大豆种胚对X射线辐射的响应机制[J]. 作物学报, 2025, 51(12): 3121-3132.
[12] 林洋, 史晓蕾, 陈强, 刘兵强, 杨庆, 于慧娟, 闫龙, 武小霞, 杨春燕. 大豆蛋白质脂肪及脂肪酸组分相关QTL定位[J]. 作物学报, 2025, 51(11): 2899-2910.
[13] 王浩辰, 王克晶, 韩娟, 李向华. 东南沿海短绒野大豆两种代表性生境自然种群的空间遗传结构特征:种群内取样策略研究[J]. 作物学报, 2025, 51(11): 2875-2885.
[14] 李威, 朱玉鹏, 孙宾成, 温有祥, 吴宗声, 徐一帆, 宋雯雯, 徐彩龙, 吴存祥. 转基因大豆结合免耕平作实现东北地区大豆生产轻简化[J]. 作物学报, 2025, 51(10): 2738-2749.
[15] 陈敏, 贾蓉, 张金传, 张辰煜, 褚俊聪, 姚伟, 葛军勇, 王星宇, 杨亚东, 曾昭海, 臧华栋. 半干旱区燕麦与豆科作物带状复合种植的产量优势及氮素利用特征研究[J]. 作物学报, 2025, 51(10): 2727-2737.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
京ICP备15008789号-2