期刊文献+

基于CNN-GRU-SSA组合模型的PM_(2.5)浓度预测

Prediction of PM_(2.5) Concentration Based on CNN-GRU-SSACombined Model
下载PDF
导出
摘要 为了解决门控循环单元(gated recurrent unit,GRU)超参数选取困难的问题,提出一种基于卷积神经网络(convolutional neural networks,CNN)、门控循环单元和麻雀搜索算法(sparrow search algorithm,SSA)的组合模型(CNN-GRU-SSA)。首先利用CNN对输入的多维数据集进行特征提取;然后将CNN提取到的特征输入GRU模型;最后使用SSA算法优化GRU模型的超参数,并将其应用于PM_(2.5)浓度预测。选取西部城市成都与东部城市杭州作为研究区域,使用2021年12月1日—2022年2月13日的大气污染物、气象因素、边界层高度(boundary layer height,BLH)以及大气可降水量(precipitable water vapor,PWV)的小时数据进行建模,分别预测两市2022年2月14日—2月28日PM_(2.5)浓度变化。实验结果表明,CNN-GRU-SSA模型预测精度与其他模型相比有明显提高,其中成都的预测值最接近实际值。 To address the challenge of difficulty in selecting hyperparameters for gated recurrent units(GRU),a combined model(CNN-GRU-SSA)was proposed,integrating convolutional neural networks(CNN),gated recurrent units,and the sparrow search algorithm(SSA).Firstly,CNN for feature extraction from the multidimensional input dataset was utilized.Subsequently,the features extracted by CNN into the GRU model are inputted.Lastly,the hyperparameters of the GRU model using the SSA algorithm were optimized and were applied to the model to predict PM_(2.5) concentrations.Chengdu in the western region and Hangzhou in the eastern region was selected as research areas.Hourly data of air pollutants,meteorological factors,boundary layer height(BLH),and precipitable water vapor(PWV)from December 1,2021,to February 13,2022,were utilized for modeling.The aim was to predict the variation in PM_(2.5) concentrations in both cities from February 14 to February 28,2022.Experimental results indicate that the CNN-GRU-SSA model demonstrates a noticeable improvement in prediction accuracy compared to other models,with Chengdu exhibiting the closest proximity between predicted and actual values.
作者 林买金 张露露 唐友兵 孟春阳 张茗斐 万梓康 谢劭峰 LIN Mai-jin;ZHANG Lu-lu;TANG You-bing;MENG Chun-yang;ZHANG Ming-fei;WAN Zi-kang;XIE Shao-feng(College of Geomatics and Geoinformation,Guilin University of Technology,Guilin 541006,China;School of Tourism and Landscape Architecture,Guilin University of Technology,Guilin 541006,China)
出处 《科学技术与工程》 北大核心 2024年第31期13269-13276,共8页 Science Technology and Engineering
基金 国家自然科学基金(41864002)。
关键词 PM_(2.5) 麻雀搜索算法 卷积神经网络 门控循环单元 PWV PM_(2.5) sparrow search algorithm convolutional neural network gated recurrent unit PWV
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部