臭氧浓度的预测对于大气环境治理、空气质量改善等起到了重要的作用。本文提出了一种交互差分时空LSTM网络预测模型(ST-IDN)来挖掘臭氧浓度历史数据的时间相关性和空间相关性,并成功将其应用到网格化臭氧浓度数据预测上。在该模型中,首...臭氧浓度的预测对于大气环境治理、空气质量改善等起到了重要的作用。本文提出了一种交互差分时空LSTM网络预测模型(ST-IDN)来挖掘臭氧浓度历史数据的时间相关性和空间相关性,并成功将其应用到网格化臭氧浓度数据预测上。在该模型中,首先交互模块(IC)可以通过一系列的卷积操作来捕捉短期上下文信息,其次层融合模块(LF)可以融合不同层的空间信息来获得上一时刻丰富的空间信息,最后差分时空LSTM模块(DSTM)将捕捉到的时间信息和空间信息进行统一建模实现臭氧浓度预测。所构建模型分别与卷积LSTM网络(ConvLSTM)、预测循环神经网络(PredRNN)以及Memory in Memory网络(MIM)模型在河北省气象局提供的臭氧浓度数据上进行了对比分析,ST-IDN模型的平均绝对误差分别降低了19.836%、12.924%、7.506%。实验结果表明,所提出的模型能够提高臭氧浓度的预测精度。展开更多
The application of appropriate advanced treat- ment process in the municipal wastewater treatment plants （WWTPs） has become an important issue considering the elimination of emerging contaminants, such as pharma- ce...The application of appropriate advanced treat- ment process in the municipal wastewater treatment plants （WWTPs） has become an important issue considering the elimination of emerging contaminants, such as pharma- ceutical and personal care products （PPCPs）. In the present study, the removal of 13 PPCPs belonging to different therapeutic classes by the sequential ultraviolet （UV） and ozonation process in a full-scale WWTP in Beijing was investigated over the course of ten months. Most of the target PPCPs were effectively removed, and the median removal efficiencies of individual PPCPs, ranging from -13% to 89%, were dependent on their reaction rate constants with molecular ozone. Noticeable fluctuation in the removal efficiencies of the same PPCPs was observed in different sampling campaigns. Nevertheless, the sequential UV and ozonation process still made a significant contribution to the total elimination of most PPCPs in the full-scale WWTP, by compensating for the poor or fluctuant removal performance of PPCPs by biologic treatment process.展开更多
In the present study, the decomposition rates of carbon tetrachloride （CC14） and 2,4-dichlorophenol （2,4- DCP） in water by the ultraviolet （UV） light irradiation alone and H2O2AYV were experimentally investigate...In the present study, the decomposition rates of carbon tetrachloride （CC14） and 2,4-dichlorophenol （2,4- DCP） in water by the ultraviolet （UV） light irradiation alone and H2O2AYV were experimentally investigated. The detailed experimental studies have been conducted for examining treatment capacities of the two different ultraviolet light sources （low and medium pressure Hg arc） in H2O2/UV processes. The low or medium UV lamp alone resulted in a 60%-90% decomposition of 2,4-DCP while a slight addition of H2O2 resulted in a drastic enhancement of the 2,4-DCP decomposition rate. The decomposition rate of 2,4-DCP with the medium pressure UV lamp alone was about 3-6 times greater than the low pressure UV lamp alone. In the direct photolysis of aqueous CC14, the medium pressure UV lamp had advantage over the low pressure UV lamp because the molar extinction coefficient of CC14 at shorter wavelength （210-220 nm） is about 20 to 50 times higher than that at 254 nm. However, adding H202 to the medium pressure UV lamp system rendered a negative oxidation rate because H202 acted as a UV absorber being competitive with CC14 due to negligible reaction between CC14 and OH radicals. The results fi：om the present study indicated significant influence of the photochemical properties of the target contaminants on the photochemical treatment characteristics for designing cost-effective UV-based degradation of toxic contaminants.展开更多
文摘臭氧浓度的预测对于大气环境治理、空气质量改善等起到了重要的作用。本文提出了一种交互差分时空LSTM网络预测模型(ST-IDN)来挖掘臭氧浓度历史数据的时间相关性和空间相关性,并成功将其应用到网格化臭氧浓度数据预测上。在该模型中,首先交互模块(IC)可以通过一系列的卷积操作来捕捉短期上下文信息,其次层融合模块(LF)可以融合不同层的空间信息来获得上一时刻丰富的空间信息,最后差分时空LSTM模块(DSTM)将捕捉到的时间信息和空间信息进行统一建模实现臭氧浓度预测。所构建模型分别与卷积LSTM网络(ConvLSTM)、预测循环神经网络(PredRNN)以及Memory in Memory网络(MIM)模型在河北省气象局提供的臭氧浓度数据上进行了对比分析,ST-IDN模型的平均绝对误差分别降低了19.836%、12.924%、7.506%。实验结果表明,所提出的模型能够提高臭氧浓度的预测精度。
文摘The application of appropriate advanced treat- ment process in the municipal wastewater treatment plants （WWTPs） has become an important issue considering the elimination of emerging contaminants, such as pharma- ceutical and personal care products （PPCPs）. In the present study, the removal of 13 PPCPs belonging to different therapeutic classes by the sequential ultraviolet （UV） and ozonation process in a full-scale WWTP in Beijing was investigated over the course of ten months. Most of the target PPCPs were effectively removed, and the median removal efficiencies of individual PPCPs, ranging from -13% to 89%, were dependent on their reaction rate constants with molecular ozone. Noticeable fluctuation in the removal efficiencies of the same PPCPs was observed in different sampling campaigns. Nevertheless, the sequential UV and ozonation process still made a significant contribution to the total elimination of most PPCPs in the full-scale WWTP, by compensating for the poor or fluctuant removal performance of PPCPs by biologic treatment process.
文摘In the present study, the decomposition rates of carbon tetrachloride （CC14） and 2,4-dichlorophenol （2,4- DCP） in water by the ultraviolet （UV） light irradiation alone and H2O2AYV were experimentally investigated. The detailed experimental studies have been conducted for examining treatment capacities of the two different ultraviolet light sources （low and medium pressure Hg arc） in H2O2/UV processes. The low or medium UV lamp alone resulted in a 60%-90% decomposition of 2,4-DCP while a slight addition of H2O2 resulted in a drastic enhancement of the 2,4-DCP decomposition rate. The decomposition rate of 2,4-DCP with the medium pressure UV lamp alone was about 3-6 times greater than the low pressure UV lamp alone. In the direct photolysis of aqueous CC14, the medium pressure UV lamp had advantage over the low pressure UV lamp because the molar extinction coefficient of CC14 at shorter wavelength （210-220 nm） is about 20 to 50 times higher than that at 254 nm. However, adding H202 to the medium pressure UV lamp system rendered a negative oxidation rate because H202 acted as a UV absorber being competitive with CC14 due to negligible reaction between CC14 and OH radicals. The results fi：om the present study indicated significant influence of the photochemical properties of the target contaminants on the photochemical treatment characteristics for designing cost-effective UV-based degradation of toxic contaminants.
