Guo Qingchun, He Zhenfang, Wang Zhaosheng
School of Geography and Environment, Liaocheng University, Liaocheng 252000, China.
Institute of Huanghe Studies, Liaocheng University, Liaocheng 252000, China.
Toxics. 2023 Jan 3;11(1):51. doi: 10.3390/toxics11010051.
Anthropogenic sources of fine particulate matter (PM) threaten ecosystem security, human health and sustainable development. The accuracy prediction of daily PM concentration can give important information for people to reduce their exposure. Artificial neural networks (ANNs) and wavelet-ANNs (WANNs) are used to predict daily PM concentration in Shanghai. The PM concentration in Shanghai from 2014 to 2020 decreased by 39.3%. The serious COVID-19 epidemic had an unprecedented effect on PM concentration in Shanghai. The PM concentration during the lockdown in 2020 of Shanghai is significantly reduced compared to the period before the lockdown. First, the correlation analysis is utilized to identify the associations between PM and meteorological elements in Shanghai. Second, by estimating twelve training algorithms and twenty-one network structures for these models, the results show that the optimal input elements for daily PM concentration predicting models were the PM from the 3 previous days and fourteen meteorological elements. Finally, the activation function (tansig-purelin) for ANNs and WANNs in Shanghai is better than others in the training, validation and forecasting stages. Considering the correlation coefficients (R) between the PM in the next day and the input influence factors, the PM showed the closest relation with the PM 1 day lag and closer relationships with minimum atmospheric temperature, maximum atmospheric pressure, maximum atmospheric temperature, and PM 2 days lag. When Bayesian regularization (trainbr) was used to train, the ANN and WANN models precisely simulated the daily PM concentration in Shanghai during the training, calibration and predicting stages. It is emphasized that the WANN1 model obtained optimal predicting results in terms of R (0.9316). These results prove that WANNs are adept in daily PM concentration prediction because they can identify relationships between the input and output factors. Therefore, our research can offer a theoretical basis for air pollution control.
人为来源的细颗粒物（PM）威胁着生态系统安全、人类健康和可持续发展。准确预测每日PM浓度可为人们减少暴露提供重要信息。人工神经网络（ANN）和小波人工神经网络（WANN）被用于预测上海的每日PM浓度。2014年至2020年上海的PM浓度下降了39.3%。严重的新冠疫情对上海的PM浓度产生了前所未有的影响。2020年上海封控期间的PM浓度与封控前相比显著降低。首先，利用相关性分析来确定上海PM与气象要素之间的关联。其次，通过估计这些模型的十二种训练算法和二十一种网络结构，结果表明，每日PM浓度预测模型的最佳输入要素是前三天的PM和十四种气象要素。最后，上海ANN和WANN的激活函数（tansig-purelin）在训练、验证和预测阶段比其他函数表现更好。考虑到次日PM与输入影响因素之间的相关系数（R），PM与滞后1天的PM关系最为密切，与最低气温、最高气压、最高气温以及滞后2天的PM关系也较为密切。当使用贝叶斯正则化（trainbr）进行训练时，ANN和WANN模型在训练、校准和预测阶段精确模拟了上海的每日PM浓度。需要强调的是，WANN1模型在R（0.9316）方面获得了最佳预测结果。这些结果证明WANN擅长每日PM浓度预测，因为它们能够识别输入和输出因素之间的关系。因此，我们的研究可为空气污染控制提供理论依据。
