基于自适应增强的BP模型的浙江省茶叶产量预测

doi:10.13305/j.cnki.jts.2021.04.009

摘要/Abstract

摘要： 本文采用1999—2018年浙江省59个县市的茶叶产量数据和地面气象要素驱动数据,提出了基于产量等级因子的自适应增强的反向传播（BP）神经网络模型的茶叶产量预测机制。首先分析提取了种植面积、年平均气温、3—7月的平均相对湿度、年平均相对湿度等11个影响因子,然后构建浙江省茶叶产量预测模型。试验结果表明,基于产量等级因子的自适应增强的BP模型算法相关系数达到0.893,相对误差的平均值和方差分别为0.187和0.136。在试验数据选取方面,相较于距离预测年份较远的数据,采用临近预测年份的数据,预测精度较高。根据本研究的茶叶产量预测机制,建立了浙江省茶叶产量预测误差空间分布图,其中1级优势区的平均误差为18.32%,2级次优势区为16.73%,3级一般产区为22.69%。预测模型能够实现浙江省各县市的茶叶产量预测,对茶叶生产的宏观管理具有一定指导意义。

关键词: 茶叶, 产量预测, 模型, 自适应增强, BP模型

Abstract: The study proposed the tea yield prediction mechanism using the adaboost BP network model with the tea yield level factor and China meteorological forcing dataset in 59 counties of Zhejiang in 1999-2018. We extracted 11 factors including the planting area, the yearly average temperature, the average relative humidity from March to July in the sensitivity analysis. The tea yield prediction model was established then. The result shows that the adaboost BP method with the yield level factor could reach the correlation coefficient as 0.893 and the average of the relative error as 0.187 and the variance of the relative error as 0.316. When selecting history data, the prediction error was lower when the data was closer to the prediction years. Based on the proposed method, the distribution of the prediction error was made. The average relative errors were 18.32%, 16.73% and 22.69% in level 1 high production area, level 2 medium area and level 3 general production area, respectively. The proposed model could realize the tea yield prediction in the counties of Zhejiang Province and could be used in the management of tea production process.

Key words: tea, yield prediction, model, adaboost, BP model

中图分类号:

陈冬梅, 韩文炎, 周贤锋, 吴开华, 张竞成. 基于自适应增强的BP模型的浙江省茶叶产量预测[J]. 茶叶科学, 2021, 41(4): 564-576. doi: 10.13305/j.cnki.jts.2021.04.009.

CHEN Dongmei, HAN Wenyan, ZHOU Xianfeng, WU Kaihua, ZHANG Jingcheng. Tea Yield Prediction in Zhejiang Province Based on Adaboost BP Model[J]. Journal of Tea Science, 2021, 41(4): 564-576. doi: 10.13305/j.cnki.jts.2021.04.009.

参考文献

[1] 毛祖法, 罗列万, 陆德彪, 等. 浙江茶叶产业转型升级基本方略研究[J]. 中国茶叶, 2010, 32(10): 6-9.
Mao Z F, Luo L W, Lu D B, et al.Study on basic strategy of transformation and upgrade of tea industry of Zhejiang Province[J]. China Tea, 2010, 32(10): 6-9.
[2] 胡克满, 胡海燕. 基于灰色神经网络的茶叶产量预测算法[J]. 浙江农业科学, 2019, 60(4): 577-579.
Hu K M, Hu H Y.Yield prediction algorithm of tea based on grey neural network[J]. Journal of Zhejiang Agricultural Sciences, 2019, 60(4): 577-579.
[3] 俞春芳. 中国茶叶生产布局特征及影响因素研究——基于全国408个茶叶生产县的调查[D]. 浙江: 浙江大学, 2018.
Yu C F.Study on the Characteristics and influencing factors of China’s tea production distribution: based on the survey of 408 counties [D]. Zhejiang: Zhejiang University, 2018.
[4] 张璠, 肖斌. 茶叶产量与气象因子的灰色关联度分析——以陕南茶区为例[J]. 西北农业学报, 2018, 27(5): 735-740.
Zhang P, Xiao B.Grey correlation analysis between tea yield and meteorological factors: case study of tea region in southern Shaanxi[J]. Acta Agriculturae Boreali-occidentalis Sinica, 2018, 27(5): 735-740.
[5] 孙智敏. 提高春茶产量的主要技术措施[J]. 福建茶叶, 2003, 25(3): 37.
Sun Z M.Main technical measures to improve the yield of spring tea[J]. Tea in Fujian, 2003, 25(3): 37.
[6] 朱秀红, 郑美琴, 姚文军, 等. 基于SPSS的日照市茶叶产量预测模型的建立[J]. 河南农业科学, 2010(7): 31-33.
Zhu X H, Zheng M Q, Yao W J, et al.The tea yield prediction model based on SPSS statistical software in Rizhao city[J]. Journal of Henan Agricultural Sciences, 2010(7): 31-33.
[7] 高洁煌. 基于GM(1,1)模型的武夷山市茶叶产量预测[J]. 科技视界, 2014(22): 21-22.
Gao J H.Forecast on the tea production of Wuyishan city by GM(1,1) model[J]. Science & Technology Vision, 2014(22): 21-22.
[8] 吕海侠, 赵景惠, 傅霞. 基于残差融合的ARMA-GM(1,1)模型茶叶产量预测[J]. 甘肃科学学报, 2018, 30(5): 24-28.
Lv H X, Zhao J H, Fu X.Tea production prediction under ARMA-GM (1,1) model based on residual fusion[J]. Journal of Gansu Sciences, 2018, 30(5): 24-28.
[9] 方孝荣, 丁希斌, 李晓丽. 基于灰色马尔柯夫链模型的浙江省名优茶产量预测[J]. 农机化研究, 2014, 36(7): 18-21.
Fang X R, Ding X B, Li X L.Yield prediction of famous green tea in Zhejiang Province based on Grey-Markov chain theory[J]. Journal of Agricultural Mechanization Research, 2014, 36(7): 18-21.
[10] 刘春涛, 魏明明, 郭丽娜. 气象要素对青岛崂山茶叶产量影响分析[J]. 中低纬山地气象, 2018, 42(1): 57-60.
Liu C T, Wei M M, Guo L N.The effect of meteorological factors on tea yield in Laoshan[J]. Mid-Low Latitude Mountain Meteorology, 2018, 42(1): 57-60.
[11] 金志凤, 黄敬峰, 李波, 等. 基于GIS及气候-土壤-地形因子的浙江省茶树栽培适宜性评价[J]. 农业工程学报, 2011, 27(3): 231-236.
Jin Z F, Huang J F, Li B, et al.Suitability evaluation of tea trees cultivation based on GIS in Zhejiang Province[J]. Transactions of the Chinese Society of Agricultural Engineering, 2011, 27(3): 231-236.
[12] 赵辉, 米鸿涛, 杜子璇, 等. 河南省茶树适宜种植气候区划研究[J]. 茶叶科学, 2016, 36(3): 330-336.
Zhao H, Mi H T, Du Z X, et al.Study on the climate regionalization of tea plant in Henan Province[J]. Journal of Tea Science, 2016, 36(3): 330-336.
[13] 阳坤, 何杰. 中国区域地面气象要素驱动数据集(1979-2018)[DS]. 国家青藏高原科学数据中心, 2019. doi: 10.11888/AtmosphericPhysics.tpe.249369.file.
Yang K, He J.China meteorological forcing dataset (1979-2018) [DS]. National Tibetan Plateau Data Center, 2019. doi: 10.11888/AtmosphericPhysics.tpe.249369.file.
[14] He J, Yang K, Tang W, et al.The first high-resolution meteorological forcing dataset for land process studies over China[J]. Scientific Data, 2020, 7: 25. doi: 10.1038/s41597-020-0369-y.
[15] Yang K, He J, Tang W, et al.On downward shortwave and longwave radiations over high altitude regions: observation and modeling in the Tibetan Plateau[J]. Agricultural & Forest Meteorology, 2010, 150(1): 38-46.
[16] Moniz N, Branco P, Torgo L.Evaluation of ensemble methods in imbalanced regression tasks[C]//International Workshop on Learning with Imbalanced Domains: Theory and Applications. Proceedings of Machine Learning Research 74, 2017: 129-140.
[17] 孙炜. 基于代价敏感的改进AdaBoost算法在不平衡数据中的应用[D]. 广州: 暨南大学, 2018.
Sun W.The application of improved adaBoost algorithm based on cost sensitive in imbalanced data [D]. Guangzhou: Jinan University, 2018.
[18] Przemysław S, Krawczyk B.Influence of minority class instance types on SMOTE imbalanced data oversampling[C]//International Workshop on Learning with Imbalanced Domains: Theory and Applications. Proceedings of Machine Learning Research 74, 2017: 7-21.
[19] 王来, 樊重俊, 杨云鹏, 等. 面向不平衡数据分类的KFDA-Boosting算法[J]. 计算机应用研究, 2019, 36(3): 807-811.
Wang L, Fan C J, Yang Y P, et al.KFDA-Boosting algorithm oriented to imbalanced data classification[J]. Application Research of Computers, 2019, 36(3): 807-811.
[20] Zhu T, Lin Y, Liu Y, et al.Minority oversampling for imbalanced ordinal regression[J]. Knowledge-Based Systems, 2019, 166: 140-155.
[21] Martin F M.A scaled conjugate gradient algorithm for fast supervised learning[J]. Neural Networks, 1993, 6(4): 525-533.
[22] Freund Y, Schapire R E.A decision-theoretic generalization of on-line learning and an application to boosting[J]. Journal of Computer and System Sciences, 1997, 55(1): 119-139.
[23] Michael K, Leslie G V.Cryptographic limitation on learning Boolean formulae and finite automata[C]//Johnson D S. STOC'89: Proceedings of the twenty-first annual ACM symposium on Theory of computing. New York: Association for Computing Machinery, 1989: 433-444.
[24] 金志凤, 封秀燕. 基于GIS的浙江省茶树栽培气候区划[J]. 茶叶, 2006, 32(1): 7-10.
Jin Z F, Feng X Y.Tea plant climate division in Zhejiang Province base on GIS technology[J]. Journal of Tea, 2006, 32(1): 7-10.
[25] Waldner F, Canto G S, Defourny P.Automated annual cropland mapping using knowledge-based temporal features[J]. Isprs Journal of Photogrammetry & Remote Sensing, 2015, 110: 1-13.
[26] Zhou Y, Xiao X, Qin Y, et al.Mapping paddy rice planting area in rice-wetland coexistent areas through analysis of Landsat 8 OLI and MODIS images[J]. International Journal of Applied Earth Observation and Geoinformation, 2016, 46: 1-12.
[27] Rajapakse R M S S, Tripathi N K, Honda K. Spectral characterization and LAI modelling for the tea (Camellia sinensis (L.) O. Kuntze) canopy[J]. International Journal of Remote Sensing, 2002, 23(18): 3569-3577.
[28] 杨艳魁, 陈芸芝, 吴波, 等. 基于高分二号影像结合纹理信息的茶园提取[J]. 江苏农业科学, 2019, 47(2): 210-214.
Yang Y K, Chen Y Z, Wu B, et al.Tea garden extraction based on gaofen-2 image with texture information[J]. Jiangsu Agricultural Sciences, 2019, 47(2): 210-214.
[29] 朱泽润. 基于高分辨率遥感影像的茶园场景提取方法[D]. 武汉: 武汉大学, 2018.
Zhu Z R.Tea garden scene extraction method based on high resolution remote sensing image [D]. Wuhan: Wuhan University, 2018.
[30] 刘峻明, 和晓彤, 王鹏新, 等. 长时间序列气象数据结合随机森林法早期预测冬小麦产量[J]. 农业工程学报, 2019, 35(6): 158-166.
Liu J M, He X T, Wang P X, et al.Early prediction of winter wheat yield with long time series meteorological data and random forest method[J]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(6): 158-166.
[31] 丑洁明, 叶笃正. 构建一个经济-气候新模型评价气候变化对粮食产量的影响[J]. 气候与环境研究, 2006, 11(3): 347-353.
Chou J M, Ye D Z.Assessing the effect of climate changes on grains yields with a new economy-climate model[J]. Climatic and Environmental Research, 2006, 11(3): 347-353.
[32] 齐邦宇, 方成刚, 王群, 等. 基于经济-气候耦合模型的昆明冬小麦产量评估[J]. 西南农业学报, 2013, 26(6): 2241-2246.
Qi B Y, Fang C G, Wang Q, et al.Study on assessing total winter-wheat yield based on an economy-climate coupling model in Kunming city[J]. Southwest China Journal of Agricultural Sciences, 2013, 26(6): 2241-2246.