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

详细信息    查看全文 | 推荐本文 |

英文篇名：Study on LSTM deep learning model-based prediction of reference crop evapotranspiration in North China 作者：邢立文 ; 崔宁博 ; 董娟 英文作者：XING Liwen;CUI Ningbo;DONG Juan;Shanxi Academy of Water Resources and Hydropower Sciences;College of Water Resources and Hydropower, Sichuan University;Shanxi Institute of Biology; 关键词：LSTM模型 ; 参考作物蒸散量ET_0 ; 华北地区 ; 深度学习模型 英文关键词：LSTM model;;reference crop evapotranspiration ET_0;;North China region;;deep learning model 中文刊名：SJWJ 英文刊名：Water Resources and Hydropower Engineering 机构：山西省水利水电科学研究院;四川大学水利水电学院;山西省生物研究所; 出版日期：2019-04-20 出版单位：水利水电技术 年：2019 期：v.50;No.546 基金：国家重点研发计划项目(2016YFC0400206);; 国家自然科学基金项目(51779161) 语种：中文; 页：SJWJ201904009 页数：9 CN：04 ISSN：11-1757/TV 分类号：67-75

摘要

为有效提高华北地下水漏斗区参考作物蒸散量ET_0的预报精度,本文以华北地区7个气象代表站1958—2010年ET_(0-PM)(Penman-Monteith,P-M)的历史时间序列为训练集构建LSTM模型,以2011—2017年ET_(0-PM)的时间序列为验证集将LSTM模型与其他4种经验模型进行对比分析。结果表明:LSTM在华北地区预测的整体评价指标Gpi(Global performance indicator)排名第一,该模型可以作为华北地区逐月ET_0预测的推荐模型,为我国精准农业灌溉预报提供科学的依据。
        In order to effectively improve the accuracy of prediction on the reference crop evapotranspiration(ET_0) in groundwater funnel area of North China, the LSTM(Long Short-Term Memory) model is constructed through taking the historical time series of ET_(0-PM)(Penman-Monteith, P-M) from 1958 to 2010 from seven meteorological stations in North China as the training set, and then a comparative analysis between LSTM model and the other four empirical models is made by taking the time series of ET_(0-PM) from 2011 to 2017 as the validation set. The result shows that the global performance indicator(Gpi) of LSTM prediction in North China ranks number one, thus the model can be used as the recommended model for ET_0 prediction in North China and provide a scientific basis for precise agricultural irrigation prediction in China.

引文

[1] 崔宁博,魏俊,赵璐,等.基于MEA-BPNN的西北旱区参考作物蒸散量预报模型[J].农业机械学报,2018,43(8):228- 236.
    [2] 李晨,崔宁博,魏新平,等.改进Hargreaves模型估算川中丘陵地区参考作物蒸散量[J].农业工程学报,2015,31(11):129- 135.
    [3] 李志.参考作物蒸散量简易估算模型在黄土高原的适用性[J].农业工程学报,2012,28(6):106- 111.
    [4] 徐俊增,彭世彰,丁加丽,等.基于蒸渗仪实测数据的日参考作物蒸发腾发量计算方法评价[J].水利学报,2010,41(12):1497- 1505.
    [5] 石弘一.机器学习综述[J].通讯世界,2018(10):253- 254.
    [6] KISI O.Applicability of Mamdani and Sugeno fuzzy genetic approaches for modeling reference evapotranspiration[J].Journal of Hydrology,2013,504:160- 170.
    [7] 陈晟,李淼,陈雷,等.基于气温和 DC-BO-NN 的河西走廊月度ET0估算模型[J].农业机械学报,2015,46 (12):140- 147.
    [8] PENG X R,YE T D,WANG Y S.Research and design of precision irrigation system based on artificial neural network[C].Chinese Control And Decision Conference (CCDC),2018:3865- 3870.
    [9] 曹国清,张晓明,陈亚峰.基于PCA-LSTM的多变量矿山排土场滑坡预警研究[J].计算机系统应用,2018,27(11):252- 258.
    [10] 谈苗苗.基于LSTM和灰色模型集成的短期交通流预测[D].南京:南京邮电大学,2017.
    [11] 黄婷婷,余磊.SDAE-LSTM模型在金融时间序列预测中的应用[J].计算机工程与应用,2019,55(1):142- 148.
    [12] 国家统计局.中国统计年鉴[M].北京:中国统计出版社,2017.
    [13] 赵旭,李毅,刘俊民.新疆旱区草地参考作物腾发量随机模拟及应用[J].水利学报,2008,39(11):1267- 1272.
    [14] B J,N A.Estimation of diffuse solar radiation in humid-subtropical climatic region of India:Comparison of diffuse fraction and diffusion coefficient models[J].Energy,2017,131:149- 164.