采用具有阻尼因子的函数模型,使用遗传算法(genetic algorithm,GA)辅助非线性最小二乘(nonlinear least squares,NLS)方法对相位参数进行求解。结果表明:1)相较于标准余弦函数模型,该方法的反演相位与土壤湿度的相关系数有较为明显的提...采用具有阻尼因子的函数模型,使用遗传算法(genetic algorithm,GA)辅助非线性最小二乘(nonlinear least squares,NLS)方法对相位参数进行求解。结果表明:1)相较于标准余弦函数模型,该方法的反演相位与土壤湿度的相关系数有较为明显的提升,反演结果也更加稳定,在5°~15°、5°~20°、5°~25°三个高度角范围内的相关系数均大于0.68,不同高度角之间的相关系数差值小于0.07;2)反演精度有不同程度提高,R 2提高5.72%~76.06%,RMSE减小6.12%~24.24%,MAE减小2.7%~28.3%,将该方案所求相位用于多星线性回归模型后平均RMSE减小10%。展开更多
为了利用大气可降水量(Precipitable Water Vapor,PWV)预报降雨,针对PWV在降雨过程中处于动态变化的现象,提出使用长短期记忆神经网络(LSTM)构建基于时序数据的多气象参数短临降雨预报模型。实验表明,当设置预报时间在1~6 h的情况下,模...为了利用大气可降水量(Precipitable Water Vapor,PWV)预报降雨,针对PWV在降雨过程中处于动态变化的现象,提出使用长短期记忆神经网络(LSTM)构建基于时序数据的多气象参数短临降雨预报模型。实验表明,当设置预报时间在1~6 h的情况下,模型短临预报降雨的整体性能优于其他时间尺度的预报,能预报目标区域约47%的降雨,模型的准确率为66%,略优于传统PWV分析方法;误报率为8%,优于传统PWV分析的预报方法。展开更多
To understand the retention mechanism of phenol in mainstream cigarette smoke by cellulose acetate(CA)filters,the phenol filtration efficiency,the phenol distribution in the gas phase and particulate phase during filt...To understand the retention mechanism of phenol in mainstream cigarette smoke by cellulose acetate(CA)filters,the phenol filtration efficiency,the phenol distribution in the gas phase and particulate phase during filtration and the retention characteristics of the filter for both the gas phase and particulate phenol species were investigated.The distribution of particulate bound phenol as a function of the particle size was examined.The ratio of phenol transferring from the particulate phase to the gas phase during smoke flowing through the CA filter were studied.Retention mechanisms for phenol in the gas phase as well as particulate phase phenol in the filter and the phase transition of phenol were proposed.Phenol filtration efficiency of the cigarette filter was calculated based on the proposed retention mechanisms and compared with available experimental data.The results showed that:1)Comparing with nicotine and the other particulate bound smoke constituents,CA filter exhibited a certain retention selectivity for phenol.2)Although phenol was mainly bound to the particulate phase of the cigarette mainstream smoke,about 35% of gas phase phenol was found at the tobacco end of the filter,whilst this dropped to approximately 0 at the mouth end of the filter,mostly by selective chemical adsorption.3)The weighted average smoke particle size and particulate phenol was 0.44 and 0.38 pm,respectively.The filtration efficiency of the filter to particulate phenol was near 40%.4)About 19% of particulate bound phenol transferred to gas phase when smoke passing the filter,and was selectively adsorbed.5)The overall filtration efficiency of the CA filter for phenol in mainstream cigarette smoke was 68.7%,with-7.7% variation compared with experimental data.The results demonstrated the phase transition of phenol during the filtration process and the selective retention with varying particle sizes.Our results showed that the application of CA filter in product designs reduces harmful constituents such as phenolic compounds in mainstream cigarette smoke.展开更多
文摘采用具有阻尼因子的函数模型,使用遗传算法(genetic algorithm,GA)辅助非线性最小二乘(nonlinear least squares,NLS)方法对相位参数进行求解。结果表明:1)相较于标准余弦函数模型,该方法的反演相位与土壤湿度的相关系数有较为明显的提升,反演结果也更加稳定,在5°~15°、5°~20°、5°~25°三个高度角范围内的相关系数均大于0.68,不同高度角之间的相关系数差值小于0.07;2)反演精度有不同程度提高,R 2提高5.72%~76.06%,RMSE减小6.12%~24.24%,MAE减小2.7%~28.3%,将该方案所求相位用于多星线性回归模型后平均RMSE减小10%。
文摘为了利用大气可降水量(Precipitable Water Vapor,PWV)预报降雨,针对PWV在降雨过程中处于动态变化的现象,提出使用长短期记忆神经网络(LSTM)构建基于时序数据的多气象参数短临降雨预报模型。实验表明,当设置预报时间在1~6 h的情况下,模型短临预报降雨的整体性能优于其他时间尺度的预报,能预报目标区域约47%的降雨,模型的准确率为66%,略优于传统PWV分析方法;误报率为8%,优于传统PWV分析的预报方法。
文摘To understand the retention mechanism of phenol in mainstream cigarette smoke by cellulose acetate(CA)filters,the phenol filtration efficiency,the phenol distribution in the gas phase and particulate phase during filtration and the retention characteristics of the filter for both the gas phase and particulate phenol species were investigated.The distribution of particulate bound phenol as a function of the particle size was examined.The ratio of phenol transferring from the particulate phase to the gas phase during smoke flowing through the CA filter were studied.Retention mechanisms for phenol in the gas phase as well as particulate phase phenol in the filter and the phase transition of phenol were proposed.Phenol filtration efficiency of the cigarette filter was calculated based on the proposed retention mechanisms and compared with available experimental data.The results showed that:1)Comparing with nicotine and the other particulate bound smoke constituents,CA filter exhibited a certain retention selectivity for phenol.2)Although phenol was mainly bound to the particulate phase of the cigarette mainstream smoke,about 35% of gas phase phenol was found at the tobacco end of the filter,whilst this dropped to approximately 0 at the mouth end of the filter,mostly by selective chemical adsorption.3)The weighted average smoke particle size and particulate phenol was 0.44 and 0.38 pm,respectively.The filtration efficiency of the filter to particulate phenol was near 40%.4)About 19% of particulate bound phenol transferred to gas phase when smoke passing the filter,and was selectively adsorbed.5)The overall filtration efficiency of the CA filter for phenol in mainstream cigarette smoke was 68.7%,with-7.7% variation compared with experimental data.The results demonstrated the phase transition of phenol during the filtration process and the selective retention with varying particle sizes.Our results showed that the application of CA filter in product designs reduces harmful constituents such as phenolic compounds in mainstream cigarette smoke.
