基于注意力机制和高斯概率估计的PM2.5浓度预测
作者:

PM2.5 Concentration Prediction Based on Attention Mechanism and Gaussian Probability Estimation
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [17]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    作为衡量空气污染物浓度的重要指标, 对PM2.5浓度进行监控预测, 能够有效地保护大气环境, 进一步地减少空气污染带来的危害. 随着空气质量自动监测站的大范围建立, 由传统的机器学习搭建的空气质量预测模型已经不能满足当今的需求. 本文提出了一种基于多头注意力机制和高斯概率估计的高斯-注意力预测模型, 并对沈阳市某监测站点的数据进行了训练和测试. 该模型考虑了PM2.5浓度受到其他空气质量数据的影响, 将空气质量数据的分层时间戳(周、日、小时)的信息对齐作为输入, 使用多头注意力机制对于不同子空间的时间序列关联特征进行提取, 能够获得更加完善有效的特征信息, 再经过高斯似然估计得到预测结果. 通过与多种基准模型进行对比, 相较于性能较优的DeepAR, 高斯-注意力预测模型的MSE、MAE分别下降了21%、15%, 有效地提高了预测准确率, 能够较准确地预测出PM2.5浓度.

    Abstract:

    PM2.5 is an important indicator for measuring the concentration of air pollutants, and monitoring and predicting its concentration can effectively protect the atmospheric environment and further reduce the harm caused by air pollution. As automatic air quality monitoring stations are constructed on a large scale, the air quality prediction model built by traditional machine learning can no longer meet the current needs. This study proposes a Gaussian-attention prediction model based on the multi-head attention mechanism and Gaussian probability estimation and utilizes the data from a monitoring station in Shenyang for training and tests. Because PM2.5 concentration is affected by other air quality data, this model uses the information alignment of hierarchical time stamps (week, day, and hour) of air quality data as input and extracts the time-series correlation features of different subspaces with the multi-head attention mechanism. More complete and effective feature information is thereby obtained, and prediction results are then acquired by Gaussian likelihood estimation. A comparison with multiple benchmark models is conducted, and the mean squared error (MSE) and mean absolute error (MAE) of the proposed Gaussian-attention prediction model are respectively 21% and 15% lower than that of the DeepAR model. Effectively improving prediction accuracy, the proposed model can accurately predict PM2.5 concentration.

    参考文献
    [1] Pui DYH, Chen SC, Zuo ZL. PM2.5 in China: Measurements, sources, visibility and health effects, and mitigation. Particuology, 2014, 13: 1–26. [doi: 10.1016/j.partic.2013.11.001
    [2] Chen PF, Bi XH, Zhang JQ, et al. Assessment of heavy metal pollution characteristics and human health risk of exposure to ambient PM2.5 in Tianjin, China. Particuology, 2015, 20: 104–109. [doi: 10.1016/j.partic.2014.04.020
    [3] 严宙宁, 牟敬锋, 赵星, 等. 基于ARIMA模型的深圳市大气PM2.5浓度时间序列预测分析. 现代预防医学, 2018, 45(2): 220–223, 242
    [4] 石峰, 楼文高, 张博. 基于灰狼群智能最优化的神经网络PM2.5浓度预测. 计算机应用, 2017, 37(10): 2854–2860. [doi: 10.11772/j.issn.1001-9081.2017.10.2854
    [5] 尹建光, 彭飞, 谢连科, 等. 基于小波分解与自适应多级残差修正的最小二乘支持向量回归预测模型的PM2.5浓度预测. 环境科学学报, 2018, 38(5): 2090–2098
    [6] 蒋洪迅, 石晓文, 孙彩虹, 等. 基于DLENN模型的沈阳地区PM2.5浓度预测. 系统工程, 2021, 39(2): 13–21
    [7] Mirzaei M, Amanollahi J, Tzanis CG. Evaluation of linear, nonlinear, and hybrid models for predicting PM2.5 based on a GTWR model and MODIS AOD data. Air Quality, Atmosphere & Health, 2019, 12(10): 1215–1224
    [8] Lim B, Zohren S. Time-series forecasting with deep learning: A survey. Philosophical Transactions of the Royal Society A, 2021, 379(2194): 20200209. [doi: 10.1098/rsta.2020.0209
    [9] 刘旭林, 赵文芳, 唐伟. 应用CNN-Seq2seq的PM2.5未来一小时浓度预测模型. 小型微型计算机系统, 2020, 41(5): 1000–1006. [doi: 10.3969/j.issn.1000-1220.2020.05.017
    [10] 邓丽, 邬群勇, 杨水荣. 融合SSAE深度特征学习和LSTM网络的PM2.5小时浓度预测. 环境科学学报, 2020, 40(9): 3422–3434
    [11] 黄伟建, 李丹阳, 黄远. 基于深度学习的PM2.5浓度长期预测. 计算机应用研究, 2021, 38(6): 1809–1814
    [12] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc., 2017. 6000–6010.
    [13] Zhou HY, Zhang SH, Peng JQ, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(12): 11106–11115
    [14] 傅颖颖, 张丰, 杜震洪, 等. 融合图卷积神经网络和注意力机制的PM2.5小时浓度多步预测. 浙江大学学报(理学版), 2021, 48(1): 74–83
    [15] 张莹, 张婕, 王式功, 等. 成都市PM2.5浓度变化的影响因素交互作用研究. 中国环境科学, 2021, 41(10): 4518–4528. [doi: 10.3969/j.issn.1000-6923.2021.10.006
    [16] Bahdanau D, Cho K, Bengio Y. Neural machine translation by jointly learning to align and translate. arXiv:1409.0473, 2014.
    [17] Salinas D, Flunkert V, Gasthaus J, et al. DeepAR: Probabilistic forecasting with autoregressive recurrent networks. International Journal of Forecasting, 2020, 36(3): 1181–1191. [doi: 10.1016/j.ijforecast.2019.07.001
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

杨柳,张镝,李建熹,康丽荥,赵思彤.基于注意力机制和高斯概率估计的PM2.5浓度预测.计算机系统应用,2023,32(7):305-311

复制
分享
文章指标
  • 点击次数:726
  • 下载次数: 1746
  • HTML阅读次数: 1033
  • 引用次数: 0
历史
  • 收稿日期:2022-01-14
  • 最后修改日期:2022-02-15
  • 在线发布日期: 2022-09-01
文章二维码
您是第11205190位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号