Shijie Qian, Tian Peng, Rui He, Jie Chen, Xuedong Zhang, Muhammad Shahzad Nazir, Chu Zhang
{"title":"A novel ensemble framework based on intelligent weight optimization and multi-model fusion for air quality index prediction","authors":"Shijie Qian, Tian Peng, Rui He, Jie Chen, Xuedong Zhang, Muhammad Shahzad Nazir, Chu Zhang","doi":"10.1016/j.uclim.2024.102233","DOIUrl":null,"url":null,"abstract":"The accuracy of air quality prediction is crucial for public health and environmental management. This paper proposes a hybrid deep learning model based on TimesNet, Crossformer and Modified Honey Badger Algorithm (MHBA) for air quality prediction. First, the original air quality index (AQI) series is decomposed using Seasonal-Trend decomposition based on Loess (STL). Then, the decomposed three components are predicted separately using TimesNet and Crossformer, while the hyperparameters of TimesNet and Crossformer are optimized using the Metis algorithm. In addition, half uniform initialization and Levy flight are added to the original HBA algorithm to make up for its shortcomings of slow optimization search speed and the tendency to fall into local optimal position, and the MHBA algorithm is obtained. Finally, the MHBA algorithm is used to weight the component prediction results of the two models, and compare the advantages and disadvantages of different weighting methods, and select the optimal weighting method to get the final AQI prediction results. The experimental results show that the STL-Metis-MHBA-TC model reduces RMSE, MAE, and MAPE by 19–34 %, 22–38 %, and 22–44 %, respectively, compared to the Transformer model. Therefore, the STL-Metis-MHBA-TC hybrid model proposed in this paper can effectively improve the AQI prediction accuracy.","PeriodicalId":48626,"journal":{"name":"Urban Climate","volume":"24 1","pages":""},"PeriodicalIF":6.0000,"publicationDate":"2024-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Urban Climate","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1016/j.uclim.2024.102233","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
The accuracy of air quality prediction is crucial for public health and environmental management. This paper proposes a hybrid deep learning model based on TimesNet, Crossformer and Modified Honey Badger Algorithm (MHBA) for air quality prediction. First, the original air quality index (AQI) series is decomposed using Seasonal-Trend decomposition based on Loess (STL). Then, the decomposed three components are predicted separately using TimesNet and Crossformer, while the hyperparameters of TimesNet and Crossformer are optimized using the Metis algorithm. In addition, half uniform initialization and Levy flight are added to the original HBA algorithm to make up for its shortcomings of slow optimization search speed and the tendency to fall into local optimal position, and the MHBA algorithm is obtained. Finally, the MHBA algorithm is used to weight the component prediction results of the two models, and compare the advantages and disadvantages of different weighting methods, and select the optimal weighting method to get the final AQI prediction results. The experimental results show that the STL-Metis-MHBA-TC model reduces RMSE, MAE, and MAPE by 19–34 %, 22–38 %, and 22–44 %, respectively, compared to the Transformer model. Therefore, the STL-Metis-MHBA-TC hybrid model proposed in this paper can effectively improve the AQI prediction accuracy.
期刊介绍:
Urban Climate serves the scientific and decision making communities with the publication of research on theory, science and applications relevant to understanding urban climatic conditions and change in relation to their geography and to demographic, socioeconomic, institutional, technological and environmental dynamics and global change. Targeted towards both disciplinary and interdisciplinary audiences, this journal publishes original research papers, comprehensive review articles, book reviews, and short communications on topics including, but not limited to, the following:
Urban meteorology and climate[...]
Urban environmental pollution[...]
Adaptation to global change[...]
Urban economic and social issues[...]
Research Approaches[...]