主办单位：中国气象局沈阳大气环境研究所
国际刊号：ISSN 1673-503X
国内刊号：CN 21-1531/P
气象与环境学报

气象与环境学报 ›› 2025, Vol. 41 ›› Issue (2): 56-63.doi: 10.3969/j.issn.1673-503X.2025.02.007

• 论文 • 上一篇    

基于多种机器学习算法构建的呼和浩特市臭氧气象条件评估指数对比分析

王俊秀1, 张智2, 李二杰2, 姜学恭3, 杨泽华1, 王俊杰4, 兰劲青5   

  1. 1. 呼和浩特市气象局, 内蒙古呼和浩特 010020;
    2. 河北省气象灾害防御和环境气象中心, 河北石家庄 050021;
    3. 内蒙古自治区气象科学研究所, 内蒙古呼和浩特 010051;
    4. 遂宁市气象局, 四川遂宁 629000;
    5. 内蒙古自治区生态环境科学研究院, 内蒙古呼和浩特 010011
  • 收稿日期:2023-08-28 修回日期:2023-12-27 发布日期:2025-06-20
  • 通讯作者: 李二杰,男,高级工程师,E-mail:cafe-mate@163.com。 E-mail:cafe-mate@163.com
  • 作者简介:王俊秀,女,1990年生,工程师,主要从事天气预报、大气环境研究,E-mail:junxiuwang12@163.com。
  • 基金资助:
    国家自然科学基金项目(41965003)、内蒙古自治区气象局引导性创新基金项目(nmqxydcx202201,nmqxydcx202202)和内蒙古自治区气象局科技创新项目(nmqxkjcx202301)共同资助。

Comparative analysis of ozone meteorological condition assessment indices for Hohhot city based on multiple machine learning algorithms

WANG Junxiu1, ZAHNG Zhi2, LI Erjie2, JIANG Xuegong3, YANG Zehua1, WANG Junjie4, LAN Jingqing5   

  1. 1. Hohhot Meteorological Bureau, Hohhot 010020, China;
    2. Hebei Meteorological Disaster Prevention and Environmental Meteorology Center, Shijiazhuang 050021, China;
    3. Inner Mongolia Institute of Meteorological Sciences, Hohhot 010051, China;
    4. Suining Meteorological Bureau, Suining, 629000, China;
    5. Inner Mongolia Autonomous Region Academy of Eco-Environmental Sciences, Hohhot 010011, China
  • Received:2023-08-28 Revised:2023-12-27 Published:2025-06-20

PDF (PC)

7

摘要： 选用2018—2022年呼和浩特市O3监测资料和气象资料,分别建立基于轻量梯度提升机(LightGBM)、极端梯度提升树(XGBoost)、随机森林(RF)和长短期记忆网络(LSTM)的气象要素与日最大8 h平均O3浓度(ρ(O3-8h))模型,对比各模型的评价指标,选出最优模型并验证。结果表明:近5 a呼和浩特市ρ(O3-8h)＞160 g·m-3出现在4—10月,其中6月、7月天数最多,5月、8月次之;模型因子中的日最高气温对ρ(O3-8h)预测结果的贡献最大,占比为44%;模型模拟性能由高到低分别为:LightGBM＞LSTM＞XGBoost＞RF,LightGBM模型模拟结果最好,构建的臭氧气象条件评估指数与ρ(O3-8h)相关系数最高达0.86,较中国气象局臭氧气象条件评估指数提升了17.81%,可较好地评估呼和浩特市气象条件对O3浓度变化的影响。

关键词: 极端梯度提升树(XGBoost), 随机森林(RF), 长短期记忆网络(LSTM), 轻量级梯度提升机(LightGBM)

Abstract: Using ozone monitoring data and meteorological data in Hohhot city from 2018 to 2022,this study analyzes the O3 concentration variation characteristics.Models with relating meteorological factors to predict daily maximum 8-h average O3 concentration (ρ(O3-8h)) were established using Light Gradient Boosting Machine (Light-GBM),Extreme Gradient Boosting (XGBoost),Random Forest (RF),and Long Short-Term Memory networks (LSTM).A comparison of model performance metrics was conducted to identify the optimal model,which was validated.The results showed that over the past five years,O3 concentration in Hohhot City (ρ(O3-8h) exceeded 160 μg·m-3 from April to October,with the most exceedance days in June and July,followed by May and August.Among the model input factors,daily maximum temperature contributes the most to the prediction of (ρ(O3-8h),accounting for 44%.The LightGBM model provided the best simulation results,with overall model performance ranking from highest to lowest as follows: LightGBM> LSTM> XGBoost> RF.The locally constructed ozone meteorological condition assessment index showed a correlation coefficient up to 0.86 with (ρ(O3-8h),an improvement of 17.81% over the China Meteorological Administration's ozone meteorological condition assessment index.This demonstrates its effectiveness in assessing meteorological condition influences on O3 concentration variations in Hohhot City.

Key words: Extreme Gradient Boosting (XGBoost), Random Forest (RF), Long Short-Term Memory networks (LSTM), Light Gradient Boosting Machine (LightGBM)

中图分类号: