Page 189 - 《高原气象》2025年第6期
P. 189
6 期 周秋雪等:一种适用于复杂地形下最高气温订正的机器学习方法 1597
the Royal Meteorological Society, 142(694): 287-303. 10. 19517/j. 1671-6345. 20210092. He S S, Lan Y, Qi Y F,
Friedman J H, 2001. Greedy function approximation: a gradient boost‐ 2021. Moving average of optimal time-window method for 2m
ing machine[J]. Annals of Statistics, 29(5): 1189-1232. temperature forecast correction of GRAPES-GFS[J]. Meteorolog‐
IPCC, 2018. Special report on global warming of 1. 5 ℃[M]. UK: ical Science Technology, 49(5): 746-753. DOI: 10. 19517/j.
Cambridge University Press. 1671-6345. 20210092.
Yasser K, Hemayed E, 2017. Novelty detection for location predic‐ 康军, 张凡, 段宗涛, 等, 2020. 基于 LightGBM 的乘客候车路段推
tion problems using boosting trees[M]. Gervasi O, Murgante B, 荐方法[J]. 测控技术, 39(2): 56-62. DOI: 10. 19708/j. ckjs.
Misra S, et al, Lecture Notes in Computer Science, 3: 173-182. 2020. 02. 010. Kang J, Zhang F, Duan Z T, et al, 2020. Recom‐
曹渝昆, 朱萌, 2019. 基于主成分分析和 LightGBM 的风电场发电 mendation method of passengers' boarding sections based on
功率超短期预测[J]. 上海电力学院学报, 35(6): 562-566. LightGBM[J]. Measurement and Control Technology, 39(2):
DOI: 10. 3969/j. issn. 1006-4729. 2019. 06. 009. Cao Y K, Zhu 56-62. DOI: 10. 19708/j. ckjs. 2020. 02. 010.
M, 2019. Ultra-short-term prediction of wind farm power genera‐ 刘杰, 刘高平, 安晶晶, 等, 2024. 基于机器学习的模式温度预报订
tion based on principal component analysis and LightGBM[J]. 正方法[J]. 沙漠与绿洲气象, 18(3): 96-104. DOI: 10. 12057/
Journal of Shanghai University of Electric Power, 35(6): 562- jissn. 1002-0799. 2024. 03. 013. Liu J, Liu G P, An J J, et al,
566. DOI: 10. 3969/j. issn. 1006-4729. 2019. 06. 009. 2024. Correction method of model temperature forecast based on
陈昱文, 黄小猛, 李熠, 等, 2020. 基于 ECMWF 产品的站点气温预 machine learning[J]. Desert and Oasis Meteorology, 18(3): 96-
报集成学习误差订正[J]. 应用气象学报, 31(4): 494-503. 104. DOI: 10. 12057/jissn. 1002-0799. 2024. 03. 013.
DOI: 10. 11898/1001-7313. 20200411. Chen Y W, Huang X M, 刘军中, 2021. 基于机器学习的最大风速预测系统研究与实现[D].
Li Y, et al, 2020. Ensemble learning for bias correction of station 南昌: 南昌大学 . Liu J Z, 2021. Research and implementation of
temperature forecast based on ECMWF products[J]. Journal of maximum wind speed prediction system based on machine learn‐
Applied Meteorological Science, 31(4): 494-503. DOI: 10. ing[D]. Nanchang: Nanchang University.
11898/1001-7313. 20200411. 刘新伟, 黄武斌, 蒋盈沙, 等, 2021. 基于 LightGBM 算法的强对流
冯良敏, 周秋雪, 曹萍萍, 等, 2023. 四川地区气温转折过程 2m 温 天气分类识别研究[J]. 高原气象, 40(4): 909-918. DOI: 10.
度变化订正研究[J]. 干旱气象, 41(1): 164-172. DOI: 10. 7522/j. issn. 1000-0534. 2020. 00075. Liu X W, Huang W B, Ji‐
11755/j. issn. 1006-7639(2023)-01-0164. Feng L M, Zhou Q ang Y S, et al, 2021. Study of the classified identification of the
X, Cao P P, et al, 2023. Study of 2 m temperature variation cor‐ strong convective weathers based on the LightGBM algorithm
rection during transitional processes of temperature in Sichuan [J]. Plateau Meteorology, 40(4): 909-918. DOI: 10. 7522/j.
[J]. Journal of Arid Meteorological, 41(1): 164-172. DOI: 10. issn. 1000-0534. 2020. 00075.
11755/j. issn. 1006-7639(2023)-01-0164. 南东亮, 王维庆, 王海云, 2019. 基于消息队列的 LightGBM 超参数
冯良敏, 周秋雪, 康岚, 2019b. 四川地区 EC 细网格模式 2m 温度偏 优化[J]. 计算机工程与科学, 41(8): 1360-1365. DOI: CNKI:
差 订 正 研 究[J]. 高 原 山 地 气 象 研 究 , 39(1): 34-40. DOI: SUN: JSJK. 0. 2019-08-004. Nan D L, Wang W Q, Wang H Y,
10. 3969/j. issn. 1674-2184. 2019. 01. 006. Feng L M, Zhou Q 2019. LightGBM hyperparameter optimization based on message
X, Kang L, 2019. Study on 2m temperature bias correction of queue[J]. Computer Engineering and Science, 41(8): 1360-
EC model in Sichuan Province[J]. Plateau and Mountain Meteo‐ 1365. DOI: CNKI: SUN: JSJK. 0. 2019-08-004.
rology Research, 39(1): 34-40. DOI: 10. 3969/j. issn. 1674- 宁晓菊, 秦耀辰, 崔耀平, 等, 1999. 60年来中国农业水热气候条件
2184. 2019. 01. 006. 的时空变化[J]. 地理学报, 70(3): 364-379. Ning X J, Qin Y
冯良敏, 周秋雪, 陈朝平, 等, 2019a. 四川地区多模式 2m 温度预报 C, Cui Y P, et al, 1999. The Spatio-temporal change of agricul‐
性能分析及集成方法研究[J]. 高原山地气象研究, 39(2): 30- tural hydrothermal conditions in China from 1951 to 2010[J]. Ac‐
36. DOI: 10. 3969/j. issn. 1674-2184. 2019. 02. 005. Feng L M, ta Geographica Sinica, 70(3): 364-379.
Zhou Q X, Chen C P, et al, 2019. Performance analysis and 孙敏, 袁慧玲, 杜予罡, 2018. 上海地区春季最高气温预报失败案
study on integrated method of multi-model 2m temperature in Si‐ 例分析[J]. 气象, 44(1): 65-79. DOI: 10. 7519/j. issn. 1000-
chuan Provinc[J]. Plateau and Mountain Meteorology Research, 0526. 2018. 01. 006. Sun M, Yuan H L, Du Y G, 2018. Analysis
39(2): 30-36. DOI: 10. 3969/j. issn. 1674-2184. 2019. 02. 005. of the spring maximum 2 m temperature forecast failure in Shang‐
符娇兰, 代刊, 2016. 基于 CRA 空间检验技术的西南地区东部强降 hai[J]. Meteorological Monthly, 44(1): 65-79. DOI: 10. 7519/
水 EC 模式预报误差分析[J]. 气象, 42(12): 1456-1464. DOI: j. issn. 1000-0526. 2018. 01. 006.
10. 7519/j. issn. 1000-0526. 2016. 12. 003. Fu J L, Dai K, 2016. 谭江红, 陈伟亮, 王珊珊, 2018. 一种机器学习方法在湖北定时气
The ECMWF model precipitation systematic error in the east of 温预报中的应用试验[J]. 气象科技进展, 8(5): 46-50. DOI:
southwest China based on the contiguou[J]. Meteorological 10. 3969/j. issn. 2095-1973. 2018. 05. 006. Tan J H, Chen W L,
Monthly, 42(12): 1456-1464. DOI: 10. 7519/j. issn. 1000- Wang S S, 2018, Using a machine learning method for tempera‐
0526. 2016. 12. 003. ture forecast in Hubei Province[J]. Advances in Meteorological
何珊珊, 蓝盈, 戚云枫, 2021. GRAPES-GFS 模式 2m 温度预报的最 Science Technology, 8(5): 46-50. DOI: 10. 3969/j. issn. 2095-
优时窗滑动订正方法[J]. 气象科技, 49(5): 746-753. DOI: 1973. 2018. 05. 006.

