基于机器学习的机械压实对大豆产量的影响预测研究

周修理; 秦娜; 王开宇; 孙浩; 王大维; 乔金友

基于机器学习的机械压实对大豆产量的影响预测研究

1. 东北农业大学电气与信息学院;
2. 东北农业大学工程学院

基金项目:

国家重点研发计划项目(2021YFD2000405-2)

财政部和农业农村部:国家现代农业产业技术体系项目(CARS-04-PS24)

详细信息

作者简介:
周修理(1966—),男,副教授,主要从事智能检测与控制研究,E-mail:xlzhou@neau.edu.cn

通讯作者:
乔金友(1969—),男,教授,主要从事农业机械化生产与管理研究,E-mail:qiaojinyou@163.com

中图分类号: S565.1
计量
- 文章访问数: 0
- HTML全文浏览量: 0
- PDF下载量: 0
出版历程
- 收稿日期: 2023-06-13
- 刊出日期: 2023-11-24

Effect of Mechanical Compaction on Soybean Yield Based on Machine Learning

1. College of Electrical and Information, Northeast Agricultural University;
2. College of Engineering, Northeast Agricultural University

摘要

摘要: 为评估农业机械作业对大豆产量的影响，本文开展不同机型、不同压实次数的拖拉机压实试验，获取不同压实环境中的土壤物理性质和大豆产量数据，分别从影响大豆产量的机械因素、土壤因素和复合因素出发，使用多元线性回归（Multiple linear regression, MLR）、随机森林（Random forest, RF）、自适应增强模型（Adaptive boosting, AdaBoost）、人工神经网络（Artificial neural network, ANN）4种机器学习算法建立大豆产量影响预测模型，对模型性能及模型特征重要性进行综合分析。研究结果表明，机械作业与大豆产量间关系复杂，集成学习算法（AdaBoost和RF）所建立的模型具有更好的拟合效果，模型决定系数更高；利用复合因素对大豆产量建立的模型拟合度最高，其次为机械因素和土壤因素，其中基于AdaBoost的复合因素对大豆产量影响模型其拟合程度最优，其R²为0.92,MAE为1.33%,RMSE为1.86%;机械因素、土壤因素都会影响大豆产量，其中机械压实次数以及表层和亚表层的土壤坚实度为影响大豆产量的重要因素，在实际生产中可通过减少机械作业次数、疏松表层及亚表层土壤来改善机械压实影响。
- 机械压实 /
- 大豆产量 /
- 机器学习 /
- 土壤坚实度 /
- 预测模型
Abstract: Aiming to find a more accurate method to assess the effect of agricultural machinery compaction on soybean yield, data of soil physical properties and soybean yield in different compaction environments were obtained by carrying out different numbers of compaction walks with different types of tractors. Soybean yield forecast models were developed from mechanical factors, soil factors, and composite factors which affected soybean growth, respectively. To find out the differences of models built by different types of machine learning algorithms, multiple linear regression（MLR）, random forest（RF）, adaptive boosting（AdaBoost）, and artificial neural network（ANN） were used in modeling. In addition, the importance of model features was comprehensively analyzed. The results showed that the relationship between mechanical operation and crop yield was complex, and the models built by integrated learning algorithms（AdaBoost and RF） had a better fit and higher coefficient of determination. Among the machine learning algorithms used, the best performance of the models built was AdaBoost, followed by random forest, artificial neural network and multiple linear regression. The model built using composite factors for soybean yield had the best fit, followed by mechanical and soil factors. The AdaBoost-based composite factor for soybean yield forecast model had the optimal fit with R~2 of 0.92, MAE of 1.33% and RMSE of 1.86%. Mechanical factors and soil factors all had an effect on the variation of soybean yield. The number of mechanical compaction, soil penetration resistance in the surface and subsurface layers were the important factors affecting soybean yield. Therefore, the effects from mechanical compaction can be relieved by reducing the number of mechanical operation and loosening soil penetration resistance of the surface and subsurface soils.
- mechanical compaction /
- soybean yield /
- machine learning /
- soil penetration resistance /
- prediction models

HTML全文

参考文献(38)

[1]	乔金友,张丹,张宏彬,等.大中小型拖拉机压实对土壤坚实度和大豆产量的影响[J].农业工程学报,2019,35(21):26-33.QIAO Jinyou,ZHANG Dan,ZHANG Hongbin,et al.Effects of soil compaction by various tractors on soil penetration resistance and soybean yields[J].Transactions of the CSAE,2019,35(21):26-33.(in Chinese)
[2]	SHAH A N,TANVEER M,SHAHZAD B,et al.Soil compaction effects on soil health and crop productivity an overview[J].Environmental Science and Pollution Research,2017,24(11):10056-10067.
[3]	GREGORICH E G,LAPEN D R,MA B L,et al.Soil and crop response to varying levels of compaction,nitrogen fertilization,and clay content[J].Soil Science Society of America Journal,2011,75(4):1483-1492.
[4]	张兴义,隋跃宇.土壤压实对农作物影响概述[J].农业机械学报,2005,36(10):161-164.ZHANG Xingyi,SUI Yueyu.Summarization on the effect of soil compaction on crops[J].Transactions of the Chinese Society for Agricultural Machinery,2005,36(10):161-164.(in Chinese)
[5]	NAWAZ M F,BOURRIE G,TROLARD F.Soil compaction impact and modelling.A review[J].Agronomy for Sustainable Development,2013,33(2):291-309.
[6]	HEYDARI L,BAYAT H,CASTRIGNANÒ A.Scale-dependent geostatistical modelling of crop-soil relationships in view of precision agriculture[J].Precision Agriculture,2023,24(4):1261-1287.
[7]	SIVARAJAN S,MAHARLOOEI M,BAJWA S G,et al.Impact of soil compaction due to wheel traffic on corn and soybean growth,development and yield[J].Soil and Tillage Research,2018,175:234-243.
[8]	OBOUR P B,UGARTE C M.A meta-analysis of the impact of traffic-induced compaction on soil physical properties and grain yield[J].Soil and Tillage Research,2021,211:105019.
[9]	ZHANG X Y,CRUSE R M,SUI Y Y,et al.Soil compaction induced by small tractor traffic in Northeast China[J].Soil Science Society of America Journal,2006,70(2):613-619.
[10]	CARTER M R.Penetration resistance to characterize the depth and persistence of soil loosening in tillage studies[J].Canadian Journal of Soil Science,1988,68(4):657-668.
[11]	LIMA C L R,REINERT D J,REICHERT J M,et al.Yield of crops and soil resistance to penetration of the Alfisol under different management[J].Pesquisa Agropecuária Brasileira,2010,45(1):89-98.
[12]	杨敏丽,彭健,金剑,等.不同农用轮胎对东北黑土区土壤压实的影响[J].农业机械学报,2023,54(6):85-93.YANG Minli,PENG Jian,JIN Jian,et al.Influence of agricultural tire technology innovation on soil compaction in black soil region Northeast China[J].Transactions of the Chinese Society for Agricultural Machinery,2023,54(6):85-93.(in Chinese)
[13]	DA LUZ F B,GONZAGA L C,CASTIONI G A F,et al.Controlled traffic farming maintains soil physical functionality in sugarcane fields[J].Geoderma,2023,432:116427.
[14]	ZHUO Z,XING A,CAO M,et al.Identifying the position of the compacted layer by measuring soil penetration resistance in a dryland farming region in Northeast China[J].Soil Use and Management,2020,36(3):494-506.
[15]	KELLER T,SANDIN M,COLOMBI T,et al.Historical increase in agricultural machinery weights enhanced soil stress levels and adversely affected soil functioning[J].Soil and Tillage Research,2019,194:104293.
[16]	王宪良,王庆杰,李洪文,等.免耕条件下轮胎压实对土壤物理特性和作物根系的影响[J].农业机械学报,2017,48(6):168-175.WANG Xianliang,WANG Qingjie,LI Hongwen,et al.Effect of type induced soil compaction on soil properties and crop root growth under no-tillage system[J].Transactions of the Chinese Society for Agricultural Machinery,2017,48(6):168-175.(in Chinese)
[17]	TASER O F,KARA O.Silage maize (Zea mays L.) seedlings emergence as influenced by soil compaction treatments and contact pressures[J].Plant Soil and Environment,2005,51(7):289-295.
[18]	CORREA J,POSTMA J A,WOJCIECHOWSKI T.Phenotypic response to soil compaction varies among genotypes and correlates with plant size in sorghum[J].Plant and Soil,2022,472(1-2):59-76.
[19]	DE MORAES M T,DEBIASI H,FRANCHINI J C,et al.Soil compaction impacts soybean root growth in an Oxisol from subtropical Brazil[J].Soil and Tillage Research,2020,200:104611.
[20]	REINTAM E,TRÜKMANN K,KUHT J,et al.Soil compaction effects on soil bulk density and penetration resistance and growth of spring barley (Hordeum vulgare L.)[J].Acta Agriculturae Scandinavica Section B-Soil and Plant Science,2009,59(3):265-272.
[21]	MULAZZANI R P,GUBIANI P I,ZANON A J,et al.Impact of soil compaction on 30-year soybean yield simulated with CROPGRO-DSSAT[J].Agricultural Systems,2022,203:103523.
[22]	ELAOUD A,HASSEN H B,SALAH N B,et al.Modeling of soil penetration resistance using multiple linear regression (MLR)[J].Arabian Journal of Geosciences,2017,10(20):1-8.
[23]	GHADERNEJAD K,SHAHGHOLI G,MARDANI A,et al.Prediction effect of farmyard manure,multiple passes and moisture content on clay soil compaction using adaptive neuro-fuzzy inference system[J].Journal of Terramechanics,2018,77:49-57.
[24]	BEN HASSEN H,ELAOUD A,MASMOUDI K.Modeling of agricultural soil compaction using discrete Bayesian networks[J].International Journal of Environmental Science and Technology,2020,17(5):2571-2582.
[25]	KHANAL S,KLOPFENSTEIN A,KUSHAL K C,et al.Assessing the impact of agricultural field traffic on corn grain yield using remote sensing and machine learning[J].Soil and Tillage Research,2021,208:104880.
[26]	李民赞,任新建,杨玮,等.基于树莓派的农田表土层土壤容重检测系统研究[J].农业机械学报,2021,52(增刊):329-335,376.LI Minzan,REN Xinjian,YANG Wei,et al.Soil bulk density detection system of farmland topsoil based on raspberry[J] Transactions of the Chinese Society for Agricultural Machinery,2021,52(Supp.):329-335,376.(in Chinese)
[27]	周修理,张萍萍,秦娜,等.基于GA-RF模型土壤坚实度对黑土区大豆产量的影响[J].东北农业大学学报,2022,53(10):67-75.ZHOU Xiuli,ZHANG Pingping,QIN Na,et al.Effects of soil penetration resistance on soybean yield in black soil region based on GA-RF model[J].Journal of Northeast Agricultural University,2022,53(10):67-75.(in Chinese)
[28]	GB 5009.3—2016 食品安全国家标准食品中水分的测定[S].
[29]	段丹丹,刘仲华,赵春江,等.基于优化植被指数的多生育期茶叶游离氨基酸含量估算[J].农业机械学报,2022,53(2):393-400,420.DUAN Dandan,LIU Zhonghua,ZHAO Chunjiang,et al.Estimation of free amino acid content in fresh tea leaves at multiple growth periods based on optimized vegetation index[J].Transactions of the Chinese Society for Agricultural Machinery,2022,53(2):393-400,420.(in Chinese)
[30]	张唯,伍霞,卢灿炯,等.基于多元回归的栅格水系阈值计算模型[J].农业机械学报,2016,47(10):131-138,107.ZHANG Wei,WU Xia,LU Canjiong,et al.Determination of flow accumulation threshold based on multiple regression model in raster river networks extraction[J].Transactions of the Chinese Society for Agricultural Machinery,2016,47(10):131-138,107.(in Chinese)
[31]	冯建英,苏允汇,龚劭齐,等.基于集成学习的农业生产技术效率评价方法[J].农业机械学报,2021,52(增刊):148-155.FENG Jianying,SU Yunhui,GONG Shaoqi,et al.Evaluation method of agricultural production technical efficiency based on ensemble learning[J].Transactions of the Chinese Society for Agricultural Machinery,2021,52(Supp.):148-155.(in Chinese)
[32]	吴静珠,张乐,李江波,等.基于高光谱与集成学习的单粒玉米种子水分检测模型[J].农业机械学报,2022,53(5):302-308.WU Jingzhu,ZHANG Le,LI Jiangbo,et al.Detection model of moisture content of single maize seed based on hyperspectral image and ensemble learning[J].Transactions of the Chinese Society for Agricultural Machinery,2022,53(5):302-308.(in Chinese)
[33]	齐建东,黄金泽,贾昕.基于XGBoost-ANN的城市绿地净碳交换模拟与特征响应[J].农业机械学报,2019,50(5):269-278.QI Jiandong,HUANG Jinze,JIA Xin.Simulation of NEE and characterization of urban green-land ecosystem responses to climatic controls based on XGBoost-ANN[J].Transactions of the Chinese Society for Agricultural Machinery,2019,50(5):269-278.(in Chinese)
[34]	HUANG Q,XUE H,LIU J,et al.A new method for spectral wavelength selection based on multiple linear regression combined with ant colony optimization and genetic algorithm[J].Journal of Spectroscopy,2022,2022:2440518.
[35]	LIAW A,WIENER M.Classification and regression by random forest[J].R news,2002,2(3):18-22.
[36]	GRÖMPING U.Variable importance assessment in regression:linear regression versus random forest[J].The American Statistician,2009,63(4):308-319.
[37]	GARSON G D.Interpreting neural-network connection weights[J].AI Expert,1991,6(4):46-51.
[38]	GOH A T C.Back-propagation neural networks for modeling complex systems[J].Artificial Intelligence in Engineering,1995,9(3):143-151.