Prediction of watermelon transpiration in a greenhouse considering the stem flow rate based on the SMA-SVM model

doi:10.6048/j.issn.1001-4330.2024.10.011

Abstract

Abstract:

【Objective】 To accurately predict the water demand of greenhouse watermelons. 【Methods】 A greenhouse watermelon transpiration prediction model was proposed by using a combination of watermelon stem flow rate and meteorological factors as feature variables as model inputs, and a Support Vector Machine (Support Vector Machine, SVM) was established and optimized by slime mold algorithm (slime mold algorithm, SMA). 【Results】 The experimental results showed that the combined use of meteorological factors and stem flow rate as inputs resulted in higher accuracy in predicting transpiration than using meteorological factors alone as model inputs, and the SVM prediction model optimized by SMA had the best prediction performance. 【Conclusion】 The R² and RMSE of the SMA-SVM transpiration prediction model considering stem flow rate in watermelon at three stages are 0.83, 0.87, 0.92, and 0.38, 0.31, and 0.15, respectively and the predicted values of the model are close to the actual values, and the predicted results are reliable.

Key words: support vector machine; stem flow rate; transpiration rate; prediction

摘要：

【目的】 基于SMA-SVM模型预测温室西瓜需水量。【方法】 以西瓜茎流速率与气象因子结合的作为特征变量作为模型输入,建立黏菌算法(Slime mold algorithm,SMA)优化的支持向量机(Support vector machine, SVM)的温室西瓜蒸腾量预测模型。【结果】 气象因子与茎流速率共同作为输入要比气象因子单独作为模型输入的蒸腾量预测精度更高,且通过SMA优化后的SVM预测模型预测效果最好。【结论】 茎流速率的SMA-SVM蒸腾预测模型在西瓜三个时期的R²和RMSE分别为0.83、0.87、0.92和0.38、0.31和0.15;模型预测值与实际值接近,预测结果可靠。

关键词: 支持向量机, 茎流速率, 蒸腾量, 预测

CLC Number:

S627

ZHANG Jing, GUO Junxian, LIU Xiangjiang, CHAI Yangfan. Prediction of watermelon transpiration in a greenhouse considering the stem flow rate based on the SMA-SVM model[J]. Xinjiang Agricultural Sciences, 2024, 61(10): 2434-2443.

张静, 郭俊先, 刘湘江, 柴扬帆. 基于SMA-SVM模型的茎流速率预测温室西瓜蒸腾量[J]. 新疆农业科学, 2024, 61(10): 2434-2443.

Figures/Tables 9

References 28

[1]	李婷, 宋曙辉, 王正荣, 等. 北京地区不同品种西瓜果实的品质分析[J/OL]. 食品工业科技: 1-11 [2023-10-23].
	LI Ting, SONG Shuhui, WANG Zhengrong, et al. Quality analysis of watermelon fruits from different varieties in Beijing[J/OL]. Science and Technology of Food Industry: 1-11 [2023-10-23].
[2]	刘小刚, 符娜, 李闯, 等. 河南省主粮作物需水量变化趋势与成因分析[J]. 农业机械学报, 2015, 46(9): 188-197.
	LIU Xiaogang, FU Na, LI Chuang, et al. Trends and causes analysis of water requirement for main grain crops in Henan Province[J]. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(9): 188-197.
[3]	代志伟, 程曼, 袁洪波, 等. 温室灌溉控制策略研究进展[J]. 中国农机化学报, 2022, 43(9): 63-72.
	DAI Zhiwei, CHENG Man, YUAN Hongbo, et al. Review of irrigation control strategy for greenhouse[J]. Journal of Chinese Agricultural Mechanization, 2022, 43(9): 63-72.
[4]	周青云, 燕琪琦, 张宝忠, 等. 冬小麦不同冠层叶片光合蒸腾和水分利用效率变化特征及对灌溉的响应[J]. 麦类作物学报, 2021, 41(10): 1247-1255.
	ZHOU Qingyun, YAN Qiqi, ZHANG Baozhong, et al. Variation characteristics of photosynthesis, transpiration and WUE of winter wheat leaves in different canopy layers and their response to irrigation[J]. Journal of Triticeae Crops, 2021, 41(10): 1247-1255.
[5]	冯东雪, 刘海军, 高壮壮, 等. 夏玉米茎流速率变化规律及其影响因子研究[J]. 干旱地区农业研究, 2020, 38(2): 185-191, 198.
	FENG Dongxue, LIU Haijun, GAO Zhuangzhuang, et al. Study on the variation of sap flow rate and its influencing factors in summer maize growth season[J]. Agricultural Research in the Arid Areas, 2020, 38(2): 185-191, 198.
[6]	李建明, 任瑞丹, 范兵华, 等. 灌溉量对温室全有机营养液栽培甜瓜根际环境和茎流的影响[J]. 农业工程学报, 2020, 36(7): 217-224.
	LI Jianming, REN Ruidan, FAN Binghua, et al. Effects of irrigation amount on the rhizospheric environment and stem flow of melon cultivated with full organic nutrition in greenhouse[J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(7): 217-224.
[7]	王升, 陈洪松, 聂云鹏, 等. 基于基因表达式编程算法的参考作物腾发量模拟计算[J]. 农业机械学报, 2015, 46(4): 106-112.
	WANG Sheng, CHEN Hongsong, NIE Yunpeng, et al. Simulation of reference evapotranspiration based on gene-expression programming method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(4): 106-112.
[8]	惠瑞晗, 林丽, 曹伟, 等. 基于P-M模型的棉田智能灌溉系统的设计与试验[J]. 新疆农业科学, 2024, 61(2): 300-309. DOI
	HUI Ruihan, LIN Li, CAO Wei, et al. Design and experiment of intelligent irrigation system for cotton field based on P-M model[J]. Xinjiang Agricultural Sciences, 2024, 61(2): 300-309. DOI
[9]	王升, 付智勇, 陈洪松, 等. 基于随机森林算法的参考作物蒸发蒸腾量模拟计算[J]. 农业机械学报, 2017, 48(3): 302-309.
	WANG Sheng, FU Zhiyong, CHEN Hongsong, et al. Simulation of reference evapotranspiration based on random forest method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(3): 302-309.
[10]	颜俊, 孙美丽, 卢姿竹, 等. 灌溉方式对设施番茄蒸腾量和水分利用效率的影响[J]. 灌溉排水学报, 2021, 40(S2): 81-85.
	YAN Jun, SUN Meili, LU Zizhu, et al. Effects of irrigation methods on transpiration and water use efficiency of greenhouse tomato[J]. Journal of Irrigation and Drainage, 2021, 40(S2): 81-85.
[11]	郑健, 鲍婷婷, 杨赟, 等. 参考作物蒸发蒸腾量计算方法在甘肃省不同区域的适用性分析[J]. 灌溉排水学报, 2022, 41(10): 1-9.
	ZHENG Jian, BAO Tingting, YANG Yun, et al. Reference evapotranspiration in Gansu Province calculated using different methods[J]. Journal of Irrigation and Drainage, 2022, 41(10): 1-9.
[12]	何淑林, 刘慧敏, 金立强, 等. 基于神经网络算法的果树需水预测研究[J]. 灌溉排水学报, 2022, 41(1): 19-24.
	HE Shulin, LIU Huimin, JIN Liqiang, et al. Calculating demands of fruit trees for water using neural network algorithm[J]. Journal of Irrigation and Drainage, 2022, 41(1): 19-24.
[13]	李莉, 李伟, 耿磊, 等. 基于RF-GRU的温室番茄结果前期蒸腾量预测方法[J]. 农业机械学报, 2022, 53(3): 368-376.
	LI Li, LI Wei, GENG Lei, et al. Prediction method of greenhouse tomato transpiration in early fruiting stage based on RF-GRU[J]. Transactions of the Chinese Society for Agricultural Machinery, 2022, 53(3): 368-376.
[14]	薛冰, 王启杰, 马彬, 等. 基于MLR和人工神经网络的新疆杨日蒸腾量估算[J]. 华中农业大学学报, 2023, 42(5): 240-250.
	XUE Bing, WANG Qijie, MA Bin, et al. Estimating daily transpiration of Populus alba var. pyramidalis Bunge based on MLR and artificial neural network[J]. Journal of Huazhong Agricultural University, 2023, 42(5): 240-250.
[15]	毛亚萍, 房世峰. 基于机器学习的参考作物蒸散量估算研究[J]. 地球信息科学学报, 2020, 22(8): 1692-1701. DOI
	MAO Yaping, FANG Shifeng. Research of reference evapotranspiration’s simulation based on machine learning[J]. Journal of Geo-Information Science, 2020, 22(8): 1692-1701.
[16]	王辰璇, 陈莉. 小波-智能优化支持向量机模型及其对农业可持续发展的评价[J]. 农业工程学报, 2023, 39(4): 208-216.
	WANG Chenxuan, CHEN Li. Wavelet-intelligent optimization model using support vector machine and its application for evaluating sustainable agricultural development[J]. Transactions of the Chinese Society of Agricultural Engineering, 2023, 39(4): 208-216.
[17]	向友珍, 吴立峰, 张富仓, 等. 基于常规气象资料估算南方地区日辐射总量方法比较[J]. 农业机械学报, 2016, 47(10): 181-192, 155.
	XIANG Youzhen, WU Lifeng, ZHANG Fucang, et al. Comparison of total radiation estimation methods in south area based on conventional meteorological data[J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(10): 181-192, 155.
[18]	徐丽萍, 张朝晖. 基于Hydrus-1D的滴灌土壤水分运移数值模拟[J]. 节水灌溉, 2019,(2): 64-67.
	XU Liping, ZHANG Zhaohui. Numerical simulation of soil moisture movement under drip irrigation based on hydrus-1D[J]. Water Saving Irrigation, 2019,(2): 64-67.
[19]	周小红. 基于多元回归分析的农作物产量估测模型研究[J]. 中国农学通报, 2022, 38(8): 152-156. DOI
	ZHOU Xiaohong. The crop yield estimation model based on multiple regression analysis[J]. Chinese Agricultural Science Bulletin, 2022, 38(8): 152-156. DOI
[20]	Noble W S. What is a support vector machine?[J]. Nature Biotechnology, 2006, 24(12): 1565-1567. DOI PMID
[21]	马龙龙, 杜灵通, 丹杨, 等. 基于茎流-蒸渗仪法的荒漠草原带人工灌丛群落蒸散特征[J]. 植物生态学报, 2020, 44(8): 807-818.
	MA Longlong, DU Lingtong, DAN Yang, et al. Characteristics of evapotranspiration in planted shrub communities in desert steppe zone based on sap flow and lysimeter methods[J]. Chinese Journal of Plant Ecology, 2020, 44(8): 807-818.
[22]	冀健红, 刘新阳. 温室番茄植株蒸腾变化规律及主要影响因素研究[J]. 节水灌溉, 2020,(8): 40-43, 49.
	JI Jianhong, LIU Xinyang. A study on transpiration variation law and its main influencing factors of tomato grown in a solar greenhouse[J]. Water Saving Irrigation, 2020,(8): 40-43, 49.
[23]	冯倩, 尹义蕾, 王柳, 等. 基于作物表型的温室精准灌溉策略研究进展[J]. 节水灌溉, 2023,(6): 34-40. DOI
	FENG Qian, YIN Yilei, WANG Liu, et al. A review on precision irrigation strategy for greenhouse crops based on phenotype[J]. Water Saving Irrigation, 2023,(6): 34-40. DOI
[24]	梁自强, 胡笑涛, 冉辉, 等. 渭北旱塬不同水肥条件下气象因子对苹果树干茎流的影响[J]. 节水灌溉, 2020,(1): 10-14, 18.
	LIANG Ziqiang, HU Xiaotao, RAN Hui, et al. Effects of meteorological factors on stem flow of apple under different water and fertilizer conditions in the upland of Weibei River[J]. Water Saving Irrigation, 2020,(1): 10-14, 18.
[25]	张代俐, 汪廷华, 朱兴淋. SVM样本约简算法研究综述[J/OL]. 计算机科学: 1-17 [2023-10-23].
	ZHANG Daili, WANG Tinghua, ZHU Xinglin. Overview of Research on SVM Sample Reduction Algorithms[J/OL]. Computer Science: 1-17 [2023-10-23].
[26]	马宏忠, 王健, 杨启帆, 等. 基于声振特征区分的SMA优化SVM变压器机械松动识别方法[J]. 电机与控制学报, 2023, 27(10): 42-53.
	MA Hongzhong, WANG Jian, YANG Qifan, et al. SMA-optimized SVM transformer state identification method based on acoustic vibration feature differentiation[J]. Electric Machines and Control, 2023, 27(10): 42-53.
[27]	王卫强, 庄云铎, 闫帆, 等. 建筑安全投入的GWO-SVR模型构建与实证分析[J]. 大连交通大学学报, 2023, 44(3): 92-96.
	WANG Weiqiang, ZHUANG Yunduo, YAN Fan, et al. Model construction and empirical analysis of construction safety investment based on gray wolf optimization and support vector regression[J]. Journal of Dalian Jiaotong University, 2023, 44(3): 92-96.
[28]	张志政, 周威. 基于PSO-BP神经网络的参考作物蒸腾量预测[J]. 节水灌溉, 2014,(11): 87-90, 95.
	ZHANG Zhizheng, ZHOU Wei. Reference crop evapotranspiration prediction based on PSO-BP neural network[J]. Water Saving Irrigation, 2014,(11): 87-90, 95.

时期 Period	影响因子 Impact factor	回归方程 Regression equation	R²
幼苗期 Seedling stage	S	Y=0.06S+0.097	0.64
	T、H、V	Y=0.059T-0.032H-0.005V+1.208	0.58
	S、T、H、V	Y=0.058S+0.147T-0.053H-1.509V+1.630	0.82
伸蔓期 Stretching stage	S	Y=0.055S+0.037	0.65
	T、H、V	Y=0.054T+0.111H+0.016V-3.603	0.59
	S、T、H、V	Y=0.051S-0.222T+0.087H+2.385V-0.839	0.80
膨果期 Swelling stage	S	Y=0.62x-0.344	0.70
	T、H、V	Y=0.057T-0.024H-0.001V+0.323	0.61
	S、T、H、V	Y=0.064S-0.053T+0.006H+0.229V+0.273	0.83

时期 Period	影响因子 Impact factor	回归方程 Regression equation	R²
幼苗期 Seedling stage	S	Y=0.06S+0.097	0.64
	T、H、V	Y=0.059T-0.032H-0.005V+1.208	0.58
	S、T、H、V	Y=0.058S+0.147T-0.053H-1.509V+1.630	0.82
伸蔓期 Stretching stage	S	Y=0.055S+0.037	0.65
	T、H、V	Y=0.054T+0.111H+0.016V-3.603	0.59
	S、T、H、V	Y=0.051S-0.222T+0.087H+2.385V-0.839	0.80
膨果期 Swelling stage	S	Y=0.62x-0.344	0.70
	T、H、V	Y=0.057T-0.024H-0.001V+0.323	0.61
	S、T、H、V	Y=0.064S-0.053T+0.006H+0.229V+0.273	0.83

模型 Model	时期 Period	决定系数Coefficient of determination(R²)				均方根误差Root mean square error(RMSE)
模型 Model	时期 Period	训练集 Training set (h、t、v)	训练集 Training set (s、h、t、v)	测试集 Test set (h、t、v)	测试集 Test set (s、h、t、v)	训练集 Training set (h、t、v)	训练集 Training set (s、h、t、v)	测试集 Test set (h、t、v)	测试集 Test set (s、h、t、v)
SVM	幼苗	0.52	0.77	0.53	0.74	0.57	0.46	0.61	0.54
	伸蔓	0.61	0.82	0.58	0.76	0.63	0.43	0.56	0.48
	膨果	0.66	0.86	0.64	0.81	0.49	0.29	0.47	0.32
SMA-SVM	幼苗	0.59	0.88	0.62	0.83	0.54	0.34	0.52	0.38
	伸蔓	0.67	0.91	0.74	0.87	0.49	0.26	0.46	0.31
	膨果	0.72	0.94	0.76	0.92	0.44	0.13	0.34	0.15
GWO-SVM	幼苗	0.58	0.79	0.59	0.77	0.55	0.41	0.54	0.47
	伸蔓	0.75	0.86	0.64	0.83	0.46	0.36	0.55	0.41
	膨果	0.78	0.92	0.73	0.87	0.37	0.21	0.43	0.24
PSO-SVM	幼苗	0.61	0.80	0.67	0.77	0.52	0.37	0.51	0.42
	伸蔓	0.63	0.89	0.78	0.85	0.56	0.29	0.56	0.34
	膨果	0.69	0.87	0.66	0.83	0.48	0.26	0.44	0.33

模型 Model	时期 Period	决定系数Coefficient of determination(R²)				均方根误差Root mean square error(RMSE)
模型 Model	时期 Period	训练集 Training set (h、t、v)	训练集 Training set (s、h、t、v)	测试集 Test set (h、t、v)	测试集 Test set (s、h、t、v)	训练集 Training set (h、t、v)	训练集 Training set (s、h、t、v)	测试集 Test set (h、t、v)	测试集 Test set (s、h、t、v)
SVM	幼苗	0.52	0.77	0.53	0.74	0.57	0.46	0.61	0.54
	伸蔓	0.61	0.82	0.58	0.76	0.63	0.43	0.56	0.48
	膨果	0.66	0.86	0.64	0.81	0.49	0.29	0.47	0.32
SMA-SVM	幼苗	0.59	0.88	0.62	0.83	0.54	0.34	0.52	0.38
	伸蔓	0.67	0.91	0.74	0.87	0.49	0.26	0.46	0.31
	膨果	0.72	0.94	0.76	0.92	0.44	0.13	0.34	0.15
GWO-SVM	幼苗	0.58	0.79	0.59	0.77	0.55	0.41	0.54	0.47
	伸蔓	0.75	0.86	0.64	0.83	0.46	0.36	0.55	0.41
	膨果	0.78	0.92	0.73	0.87	0.37	0.21	0.43	0.24
PSO-SVM	幼苗	0.61	0.80	0.67	0.77	0.52	0.37	0.51	0.42
	伸蔓	0.63	0.89	0.78	0.85	0.56	0.29	0.56	0.34
	膨果	0.69	0.87	0.66	0.83	0.48	0.26	0.44	0.33