[1]马钊,任传栋,刘静,等.基于不同LSTM模型和Hargreaves模型估算鲁中地区参考作物蒸散量[J].江苏农业学报,2022,38(06):1559-1568.[doi:doi:10.3969/j.issn.1000-4440.2022.06.014]
　MA Zhao,REN Chuan-dong,LIU Jing,et al.Estimation of reference crop evapotranspiration in central Shandong by different LSTM models and Hargreaves models[J].,2022,38(06):1559-1568.[doi:doi:10.3969/j.issn.1000-4440.2022.06.014]
点击复制

基于不同LSTM模型和Hargreaves模型估算鲁中地区参考作物蒸散量()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
38
期数:
2022年06期
页码:
1559-1568
栏目:
农业信息工程
出版日期:
2022-12-31

文章信息/Info

Title:
Estimation of reference crop evapotranspiration in central Shandong by different LSTM models and Hargreaves models
作者:
马钊1; 任传栋1; 刘静1; 王志真2
（1.山东省水利勘测设计院有限公司，山东济南250013；2.山东省农业交流合作中心，山东济南250013）
Author(s):
MA Zhao1; REN Chuan-dong1; LIU Jing1; WANG Zhi-zhen2
(1.Shandong Survey and Design Institute of Water Conservancy Co., Ltd., Jinan 250013, China;2.Shandong Agricultural Exchange and Cooperation Center, Jinan 250013, China)
关键词:
鲁中地区; 参考作物蒸散量; 温度; 长短期记忆神经网络（LSTM）模型; 贝叶斯理论; Hargreaves模型
Keywords:
central region of Shandong; reference crop evapotranspiration; temperature; long short-term memory neural network (LSTM) model; Bayesian theory; Hargreaves model
分类号:
S274
DOI:
doi:10.3969/j.issn.1000-4440.2022.06.014
文献标志码:
A
摘要:
为找出在仅使用温度这一气象参数条件下适用于区域参考作物蒸散量（ET0）估算的简化模型，本研究以鲁中地区为研究区域，基于6个气象站点1961-2019年的逐日气象数据，以长短期记忆神经网络（LSTM）模型和Hargreaves（HS）模型为基础，利用粒子群优化LSTM模型（PSO-LSTM）、遗传算法优化LSTM模型（GA-LSTM）、贝叶斯理论优化LSTM模型（BA-LSTM）和5种HS改进模型估算ET0，并将估算结果与Penman-Monteith（PM）模型的ET0进行对比。结果表明，相同参数输入条件下，LSTM模型精度普遍优于HS模型，4种LSTM模型具有较强的适用性，其中BA-LSTM模型对ET0日值的估算效果最优，其均方根差（RMSE）、平均绝对误差（MAE）、决定系数（R2）、效率系数（Ens）的中位线分别为0.378 mm/d、0.276 mm/d、0.904和0.902，其综合性指标指数（GPI）的中位线为1.837。同时，BA-LSTM模型在全区的相对误差（RE）仅为0.01%～1.75%。因此，在仅有温度这一气象参数时，推荐使用BA-LSTM模型估算鲁中地区ET0。
Abstract:
To find out a simplified model suitable for regional reference crop evapotranspiration (ET0) estimation using only temperature data, we took the central region of Shandong as the research area in this study, and used the daily meteorological data of six meteorological stations from 1961 to 2019. Based on long short-term memory neural network (LSTM) model and Hargreaves (HS) model, three optimization models including particle swarm optimization LSTM (PSO-LSTM) model, genetic algorithm optimization LSTM (GA-LSTM) model and Bayesian theory optimization LSTM (BA-LSTM) model and five improved HS models were used to estimate ET0. The calculation results were compared with the Penman-Monteith (PM) model. The results showed that under the same parameter input conditions, the accuracy of the LSTM model was generally better than that of the HS model, and the four LSTM models had strong applicability. The BA-LSTM model had the best estimation effect on the daily value of ET0. The medians of root mean square error (RMSE), mean absolute error (MAE), determination coefficient (R2) and efficiency coefficient (Ens) were 0.378 mm/d, 0.276 mm/d, 0.904 and 0.902. The median of global performance index (GPI) was 1.837. The relative error (RE) of the BA-LSTM model in the whole area was only 0.01%-1.75%. Therefore, it is recommended to use BA-LSTM model to estimate ET0 in central Shandong when only temperature data are available.

参考文献/References:

[1]李可利，张鑫. 基于ANFIS的陕西省参考作物蒸散量计算[J]. 自然资源学报，2020，35(6)：1472-1483.
[2]关静，梁川，赵璐，等. 3种干旱指数在宁夏中部干旱带的适用性分析[J]. 水土保持研究，2018，25(3)：122-127.
[3]CHEN Z J, ZHU Z C, JIANG H, et al. Estimating daily reference evapotranspiration based on limited meteorological data using deep learning and classical machine learning methods[J]. Journal of Hydrology, 2020, 591:125286.
[4]褚荣浩，李萌，沙修竹，等. 13种典型参考作物蒸散量估算模型在安徽省的适用性评价[J]. 节水灌溉，2021(9)：61-70.
[5]YANG J Y, LIU Q, MEI X R, et al. Spatiotemporal characteristics of reference evapotranspiration and its sensitivity coefficients to climate factors in Huang-Huai-Hai Plain， China[J].Journal of Integrative Agriculture, 2013, 12(12): 2280-2291.
[6]张戴炜，王春颖，韩宇平，等. 基于温度的ET0模型改进及在海河流域的适用性[J].人民黄河，2021，43(3)：155-160.
[7]李晨. 川中丘陵区不同Hargreaves改进模型适应性评价[J]. 节水灌溉，2021(11)：88-96.
[8]HARGREAVES G H, ALLEN R G. History and evaluation of Hargreaves evapotranspiration equation[J]. Jounal of Irrigation and Drainage Engineering, 2003, 129(1): 53-63.
[9]胡庆芳，杨大文，王银堂，等. Hargreaves 公式的全局校正及适用性评价[J]. 水科学进展，2011，22(2)：160-167.
[10]李晨，崔宁博，魏新平，等. 改进Hargreaves模型估算川中丘陵区参考作物蒸散量[J]. 农业工程学报，2015， 31(11)：129-135.
[11]夏兴生，赵希振，汤童，等. Hargreaves公式在中国农业区的年内回归修正[J]. 排灌机械工程学报，2021，39(1)：81-88.
[12]TABARI H, KISI O, EZANI A, et al. SVM, ANFIS, regression and climate based models for reference evapotranspiration modeling using limited climatic data in a semi-arid highland environment[J]. Journal of Hydrology, 2012, 444: 78-89.
[13]武剑飞，康银红，宋鑫，等. 基于NARX模型的参考作物蒸散发预测[J]. 排灌机械工程学报，2021，39(5)：533-540.
[14]张薇，韦群，吴天傲，等. 基于GBDT算法的参考作物蒸散量模型在江苏省的预测[J]. 江苏农业学报，2020，36(5)：1169-1180.
[15]孙龙清，吴雨寒，孙希蓓，等. 基于IBAS和LSTM网络的池塘水溶解氧含量预测[J].农业机械学报，2021，52(增刊)：252-260.
[16]邢立文，崔宁博，董娟. 基于LSTM深度学习模型的华北地区参考作物蒸散量预测研究[J]. 水利水电技术，2019，50(4)：64-72.
[17]环海军，刘岩，孙燕玲，等. 鲁中地区近35a不同历时降水变化特征[J]. 沙漠与绿洲气象，2018，12(5)：83-89.
[18]ALLEN R G, PEREIRA L S, RAES D, et al. Crop evapotranspiration： guidelines for computing crop water requirements[M]. Rome: Food and Agriculture Organization of United Nation, 1998.
[19]ZHANG J, ZHU Y, ZHANG X, et al. Developing a Long Short-Term Memory (LSTM) based model for predicting water table depth in agricultural areas[J]. Journal of Hydrology, 2018, 561: 918-929.
[20]ZHU B, FENG Y, GONG D Z, et al. Hybrid particle swarm optimization with extreme learning machine for daily reference evapotranspiration prediction from limited climatic data[J]. Computers and Electronics in Agriculture, 2020, 173: 105430.
[21]任妮，鲍彤，刘杨，等. 基于粒子群优化算法和长短时记忆神经网络的蟹塘溶解氧预测[J]. 江苏农业学报，2021，37(2)：426-434.
[22]黄敏. 基于进化算法的LSTM降水量预测方法研究[D]. 阜新：辽宁工程技术大学，2020.
[23]HARGREAVES G H, ALLEN R G. History and evaluation of Hargreaves evapotranspiration equation[J]. Journal of Irrigation and Drainage Engineering, 2003, 129(1): 53-63.
[24]DROOGRES P, ALLEN R G. Estimating reference evapotranspiration under inaccurate data conditions[J]. Irrigation and Drainage Systems, 2002, 16(1): 33-45.
[25]胡庆芳，杨大文，王银堂，等. Hargreaves公式的全局校正及适用性评价[J]. 水科学进展，2011，22(2)：160-167.
[26]冯禹，崔宁博，龚道枝. 机器学习算法和 Hargreaves模型在四川盆地ET0计算中的比较[J]. 中国农业气象，2016，37(4)：415-421.
[27]赵凡萱，唐诗佳，张卓，等. 修正Hargreaves-Samani公式估算宁夏参考作物蒸散量适用性分析[J]. 中国防汛抗旱，2021，31(6)：57-63.
[28]蒋敏兰,吴沛伦,陈昊然. 基于多层LSTM的海兰褐蛋鸡产蛋率回归模型[J].江苏农业科学,2020,48(22):248-251.
[29]李莉，李文军，马德新，等. 基于LSTM的温室番茄蒸腾量预测模型研究[J]. 农业机械学报，2021，52(10)：369-376.
[30]谢家兴，高鹏，孙道宗，等. 基于长短期记忆的柑橘园蒸散量预测模型[J]. 农业机械学报，2020，51(增刊2)：351-356.

相似文献/References:

[1]张薇,韦群,吴天傲,等.基于GBDT算法的参考作物蒸散量模型在江苏省的预测[J].江苏农业学报,2020,(05):1169.[doi:doi:10.3969/j.issn.1000-4440.2020.05.014]
　ZHANG Wei,WEI Qun,WU Tian-ao,et al.Prediction models of reference crop evapotranspiration based on gradient boosting decision tree(GBDT) algorithm in Jiangsu province[J].,2020,(06):1169.[doi:doi:10.3969/j.issn.1000-4440.2020.05.014]
[2]尹起,周建平,许燕,等.基于粒子群优化（PSO）超限学习机预测新疆参考作物蒸散量[J].江苏农业学报,2021,(03):622.[doi:doi:10.3969/j.issn.1000-4440.2021.03.010]
　YIN Qi,ZHOU Jian-ping,XU Yan,et al.Prediction of reference crop evapotranspiration in Xinjiang based on particle swarm optimization(PSO) optimized extreme learning machine[J].,2021,(06):622.[doi:doi:10.3969/j.issn.1000-4440.2021.03.010]

备注/Memo

备注/Memo:
收稿日期：2022-03-01基金项目：河北省水利科研和推广项目（202065）作者简介：马钊（1980-），男，山东菏泽人，学士，高级工程师，主要从事水利工程规划设计研究。(E-mail)mazhao6842@126.com通讯作者：任传栋，(E-mail)253510814@qq.com
更新日期/Last Update: 2023-01-13