本文已被:浏览 1294次 下载 769次 |
![本文二维码信息](/download_upload_file.aspx?file_name=/uploadfile/two_dimensional_code_img/zgnydxxb/2022/8/20220819.png) 码上扫一扫! |
基于CART算法的茶园精准灌溉方法 |
苗犇犇1,张武1,2*,王志鸿1,3,冯金磊1,王瑞卿1,江朝晖1,2,饶元1,2,金秀1,2
|
|
(1.安徽农业大学 信息与计算机学院, 合肥 230036;2.智慧农业技术与装备安徽省重点实验室, 合肥 230036;3.四创电子股份有限公司, 合肥 230036) |
|
摘要: |
针对茶园灌溉系统中灌溉时机与灌溉量难以确定的问题,以黄山太平地区猴魁茶园为研究对象,先利用CART算法对环境因子与灌溉之间的相关性进行分析,并与逻辑回归、支持向量机等预测模型进行对比,后采用Penman-Monteith公式等对灌溉量进行精确计算,再通过田间试验对本研究获得的灌溉方法进行验证。结果表明:1)地下20 cm、40 cm和地下60 cm处土壤湿度与灌溉之间的相关性最大,分别达到40.66%、12.74%、12.25%,因此,土壤湿度是判断灌溉时机最关键的环境因子;2)使用CART算法预测猴魁茶园的灌溉时机,其准确率、精确率、召回率、F1值较逻辑回归模型(Logistic Regression model)、支持向量机模型(Support vector machine model)分别提升了12%~16%、1.3%~1.4%、12%~16%、7%~9%;3)Penman-Monteith公式对灌溉量的计算结果表明,黄山太平地区猴魁茶园年灌溉量为200 ~500 mm,其中,夏梢期灌溉量最多,占全年灌溉量70%以上。田间试验结果表明该灌溉方法节水率达到30%~35%。因此,使用CART算法及Penman-Monteith公式所获取的灌溉策略对于该地区茶园精准灌溉有一定的借鉴意义。 |
关键词: 灌溉时机 灌溉量 CART算法 精准灌溉 |
DOI:10.11841/j.issn.1007-4333.2022.08.19 |
投稿时间:2021-12-28 |
基金项目:2022年安徽省重点研究和开发计划(202204C06020022);智慧农业技术与装备安徽省重点实验室开放基金(APKLSATE2019X001);2016年农业部农业物联网技术集成与应用重点实验室开放基金(2016KL05);2019年安徽省重点研发计划面上攻关项目(201904a06020056) |
|
Precision irrigation method for tea plantation based on CART algorithm |
MIAO Benben1,ZHANG Wu1,2*,WANG Zhihong1,3,FENG Jinlei1,WANG Ruiqing1,JIANG Zhaohui1,2,RAO Yuan1,2,JIN Xiu1,2
|
(1.School of Information and Computer, Anhui Agricultural University, Hefei 230036, China;2.Anhui Key Laboratory of Intelligent Agricultural Technology and Equipment, Hefei 230036, China;3.Sichuang Electronics Co., Ltd., Hefei 230036, China) |
Abstract: |
Aiming at the problem that it is difficult to determine the irrigation timing and amount in the tea plantation irrigation system, Houkui tea plantation in Taiping area of Huangshan is taken as the research object. The CART algorithm is used to analyze the correlation between environmental factors and irrigation and compared with logistic regression and support vector machines prediction models. The Penman-Monteith formula was used to calculate the irrigation amount accurately and the irrigation method obtained in this study is then verified by field experiments. The results showe that: 1)The correlation between soil moisture and irrigation is the largest at 20 cm, 40 cm and 60 cm underground, they are 40. 66%, 12. 74% and 12. 25%, respectively. Therefore, soil moisture is the most critical environment factor in judging irrigation timing; 2)Using the CART algorithm of Taiping Houkui, The accuracy, precision, recall and F1 value by using the CART algorithm to predict the irrigation timing areincreased by 12%-16%, 1. 3%-1. 4%, 12%-16% and 7%-9% than those of the Logistic Regression and the support vector machine models, respectively; 3)The the irrigation amount calculated by Penman-Monteith formula on shows that the annual irrigation amount of Houkui tea tree in Taiping area of Huangshan is 200-500 mm, and the summer shoot irrigation amount is the largest, accounting for more than 70% of the annual irrigation amount. Field experiments verifies that the water-saving rate of this irrigation method reaches 30% to 35%. Therefore, the irrigation strategy obtained by the CART algorithm and the Penman-Monteith formula has certain references for the precise irrigation of tea plantations in this area. |
Key words: irrigation timing irrigation quantity CART algorithm precision irrigation |