基于集成機器學習的大理紅花大金元種植適宜性區(qū)劃

打開文本圖片集

中圖分類號：S572；S126 文獻標識碼：A文章編號：1007-5119（2025）02-0113-08

1，1*，2，3，4，1，1，1（1.河南農(nóng)業(yè)大學資源與環(huán)境學院，鄭州450046；2.云南省煙草公司大理州公司，云南大理67100；3.浙江萬里學院信息與智能學院，浙江寧波315000；4.河南農(nóng)業(yè)大學煙草學院，鄭州450046)

Planting Suitability Zones of Honghuadajinyuan in Dali Prefecture Based on Integrated Machine Learning

WENG Qianwen1, CHEN Weiqiang1*,WANG Dexun2, CHEN Yilin3, SHI Hongzhi4, MA Yuehong1, YU Ying1,YUAN Yuke1 (1.College ofResouresandEnviroment,Henan Agrcultural University,Zhengzhou45o46,China; 2.YuanTobaccoCompany Dali PrefectureCompany,Dali7o,Yunnan,Cina;3.ColegeofInformationand IntelligenceEnginering,Zhejng Wanli University,Ningbo350oo,Zhejang,China;4.TobacoColegeofHenanAgriculturalUniversityZhengzhou4546,China)

Abstract:Theflue-cured tobaccovariety“Honghuadajinyuan”exhibitspoorecologicaladaptability.Toestablishaninteligent Zoning methodforitscultivationsuitabilityinDali tobacogrowingarea,sevenecologicalenvironmentalovaratesereselected, includingaltitde,precipiationuringteaturitypriod,eantemperatueduringteaturitypriod,drotealit, soil pH,availablepotassumcontent,andwater-solublechlorieontent.Using752surveydatastsastraiingsamplesandthe ensembleofmachine learming algorithms，weinvestigatedthecultivationsuitabilityzoningof“Honghuadajinyuan”inDali Prefecture.Resultsdemonstratedasthefolows.(1）Thesuitabilityof“Honghuadajinyuan”showsmultidimensionalnonlinear relationshipwithenvironmentalcovariates,validatingtheuseofnonlinearmachneleamingmodels.2)Theoptimalmodelswere CHAIDdecisiontresitegratedwithaggingadbostingalgorits.Inthesuitabilitycassodel,thehydrotheralcoient, soilavailablepotas,ndmatuityperdprcipiatiwereteostiialinicatos;Iteabilityvelodelatity period precipitation,meantemperatureuringthematuritypriodandaltitudewereproritzed.(3)Among1458evalatioits, 471wereclasifiedas“mostsuitable”,456as“moderatelysuitable”and531as“unsuitable”.Suitablecultivationareas were concentratedinJianchanYunlong,Eryan,Dalirefecture,Weishan,Miduandanjiancountiesaswellasteastend western partsofBinchuanandthesouthwesternregionofYongping. (4)Validationusingsensory evaluationdata from48tobacco leaf samples confirmed the alignment of zoning results with actual quality.These findings provide a scientific basis for optimizing “Honghuadajinyuan” cultivation Zoning in Dali Prefecture.

Keywords: machine learning; Honghuadajinyuan; planting suitability assessment; Dali prefecture

紅花大金元品種煙葉清香型風格突出，香氣質(zhì)好，香氣量足，深受卷煙企業(yè)的青睞[1]。（剩余17257字）