● Used a double-stage attention mechanism model to predict ozone.● The model can autonomously select the appropriate time series for forecasting.● The model outperforms other machine learning models and WRF-CMAQ.● We used the model to analyze the driving factors of VOCs that cause ozone pollution. Ozone is becoming a significant air pollutant in some regions, and VOCs are essential for ozone prediction as necessary ozone precursors. In this study, we proposed a recurrent neural network based on a double-stage attention mechanism model to predict ozone, selected an appropriate time series for prediction through the input attention and temporal attention mechanisms, and analyzed the cause of ozone generation according to the contribution of feature parameters. The experimental data show that our model had an RMSE of 7.71 μg/m3 and a mean absolute error of 5.97 μg/m3 for 1-h predictions. The DA-RNN model predicted ozone closer to observations than the other models. Based on the importance of the characteristics, we found that the ozone pollution in the Jinshan Industrial Zone mainly comes from the emissions of petrochemical enterprises, and the good generalization performance of the model is proved through testing multiple stations. Our experimental results demonstrate the validity and promising application of the DA-RNN model in predicting atmospheric pollutants and investigating their causes. 相似文献
Environmental Science and Pollution Research - This study examines the association among the green energy production (GEP), green technological innovation (GTI), and green international trade (GIT)... 相似文献
Environmental Science and Pollution Research - Rapid industrialization is deteriorating water quality, and fluoride pollution in water is one of the most serious environmental pollution problems.... 相似文献
Environmental Science and Pollution Research - Owing to rapid socio-economic development in China, trace metal emissions have increased and lakes even in remote areas have experienced marked... 相似文献
The quantitative assessment of landfill gas emissions is essential to assess the performance of the landfill cover and gas collection system. The relative error of the measured surface emission of landfill gas may be induced by the static flux chamber technique. This study aims to quantify effects of the size of the chamber, the insertion depth, pressure differential on the relative errors by using an integrated approach of in situ tests, and numerical modeling. A field experiment study of landfill gas emission is conducted by using a static chamber at one landfill site in Xi’an, Northwest China. Additionally, a two-dimensional axisymmetric numerical model for multi-component gas transport in the soil and the static chamber is developed based on the dusty-gas model (DGM). The proposed model is validated by the field data obtained in this study and a set of experimental data in the literature. The results show that DGM model has a better capacity to predict gas transport under a wider range of permeability compared to Blanc’s method. This is due to the fact that DGM model can explain the interaction among gases (e.g., CH4, CO2, O2, and N2) and the Knudsen diffusion process while these mechanisms are not included in Blanc’s model. Increasing the size and the insertion depth of static chambers can reduce the relative error for the flux of CH4 and CO2. For example, increasing the height of chambers from 0.55 to 1.1 m can decrease relative errors of CH4 and CO2 flux by 17% and 18%, respectively. Moreover, we find that gas emission fluxes for the case with positive pressure differential (?Pin-out) are greater than that of the case without considering pressure fluctuations. The Monte Carlo method was adopted to carry out the statistical analysis for quantifying the range of relative errors. The agreement of the measured field data and predicted results demonstrated that the proposed model has the capacity to quantify the emission of landfill gas from the landfill cover systems.
Environmental Science and Pollution Research - Environment-friendly algaecides based on allelopathy have been widely used to control harmful algal blooms. In this research, micro and nano scale... 相似文献