针对火电机组SO_(2)排放质量浓度的影响因素众多,难以准确预测的问题,提出一种改进向量加权平均(weighted mean of vectors,INFO)算法与双向长短期记忆(bi-directional long short term memory,Bi-LSTM)神经网络相结合的预测模型(改进IN...针对火电机组SO_(2)排放质量浓度的影响因素众多,难以准确预测的问题,提出一种改进向量加权平均(weighted mean of vectors,INFO)算法与双向长短期记忆(bi-directional long short term memory,Bi-LSTM)神经网络相结合的预测模型(改进INFO-Bi-LSTM模型)。采用Circle混沌映射和反向学习产生高质量初始化种群,引入自适应t分布提升INFO算法跳出局部最优解和全局搜索的能力。选取改进INFO-Bi-LSTM模型和多种预测模型对炉内外联合脱硫过程中4种典型工况下的SO_(2)排放质量浓度进行预测,将预测结果进行验证对比。结果表明:改进INFO算法的寻优能力得到提升,并且改进INFO-Bi-LSTM模型精度更高,更加适用于SO_(2)排放质量浓度的预测,可为变工况下的脱硫控制提供控制理论支撑。展开更多
AIM:To evaluate effects of endogenous dopamine induced by low concentration atropine eye drops on choroidal neovascularization(CNV)in high myopia mice.METHODS:The C57BL/6J mice were deprived of the right eye for 4wk,a...AIM:To evaluate effects of endogenous dopamine induced by low concentration atropine eye drops on choroidal neovascularization(CNV)in high myopia mice.METHODS:The C57BL/6J mice were deprived of the right eye for 4wk,and the high myopia was diagnosed by optometry,the diopter was less than-6.00 D,and CNV was induced by 532 nm laser.The changes of dopamine D1 receptor(DRD1),dopamine D2 receptor(DRD2),and vascular endothelial growth factor A(VEGFA)were detected by Western blot technology at 0.5,1,2h,and 7d after 0.01%,0.05%,and 0.1%atropine eye drops,respectively,the area of CNV was measured.RESULTS:Significant increases were observed on the expression of DRD2 in mouse high myopia model at 0.5,1,2h,7d with 0.05%and 0.1%atropine eye drops(P<0.05).Significant decreases were observed on the expression of DRD1 and VEGFA in mouse high myopia model at 0.5,1,2h,7d with 0.05%and 0.1%atropine eye drops(P<0.05).The area of CNV induced by laser in the drug-treated group was significantly smaller than that in the control group,and the higher the concentration,the more significant the inhibitory effect(P<0.05).CONCLUSION:The 0.01%,0.05%,0.1%atropine eye drops can decrease the level of VEGFA and inhibit high myopia CNV indirectly by up-regulating the level of DRD2 and down-regulating the level of DRD1,and the effect of 0.05%and 0.1%atropine eye drops is more significant.展开更多
The increased concentration of CO_(2) due to continuous breathing and no discharge of human beings in the manned closed space,like spacecraft and submarines,can be a threat to health and safety.Effective removal of lo...The increased concentration of CO_(2) due to continuous breathing and no discharge of human beings in the manned closed space,like spacecraft and submarines,can be a threat to health and safety.Effective removal of low concentration CO_(2) from the manned closed space is essential to meet the requirements of long-term space or deep-sea exploration,which is an international frontier and trend.Ionic liquids(ILs),as a widespread and green solvent,already showed its excellent performance on CO_(2) capture and absorption,indicating its potential application in low concentration CO_(2) capture.In this review,we first summarized the current methods and strategies for direct capture from low concentration CO_(2) in both the atmosphere and manned closed spaces.Then,the multi-scale simulation methods of CO_(2) capture by ionic liquids are described in detail,including screening ionic liquids by COSMO-RS methods,capture mechanism by density functional theory and molecular dynamics simulation,and absorption process by computational fluid dynamics simulation.Lastly,some typical IL-based green technologies for low concentration CO_(2) capture,such as functionalized ILs,co-solvent systems with ILs,and supported materials based on ILs,are introduced,and analyzed the subtle possibility in manned closed spaces.Finally,we look forward to the technology and development of low concentration CO_(2) capture,which can meet the needs of human survival in closed space and proposed that supported materials with ionic liquids have great advantages and infinite possibilities in the vital area.展开更多
文摘针对火电机组SO_(2)排放质量浓度的影响因素众多,难以准确预测的问题,提出一种改进向量加权平均(weighted mean of vectors,INFO)算法与双向长短期记忆(bi-directional long short term memory,Bi-LSTM)神经网络相结合的预测模型(改进INFO-Bi-LSTM模型)。采用Circle混沌映射和反向学习产生高质量初始化种群,引入自适应t分布提升INFO算法跳出局部最优解和全局搜索的能力。选取改进INFO-Bi-LSTM模型和多种预测模型对炉内外联合脱硫过程中4种典型工况下的SO_(2)排放质量浓度进行预测,将预测结果进行验证对比。结果表明:改进INFO算法的寻优能力得到提升,并且改进INFO-Bi-LSTM模型精度更高,更加适用于SO_(2)排放质量浓度的预测,可为变工况下的脱硫控制提供控制理论支撑。
文摘AIM:To evaluate effects of endogenous dopamine induced by low concentration atropine eye drops on choroidal neovascularization(CNV)in high myopia mice.METHODS:The C57BL/6J mice were deprived of the right eye for 4wk,and the high myopia was diagnosed by optometry,the diopter was less than-6.00 D,and CNV was induced by 532 nm laser.The changes of dopamine D1 receptor(DRD1),dopamine D2 receptor(DRD2),and vascular endothelial growth factor A(VEGFA)were detected by Western blot technology at 0.5,1,2h,and 7d after 0.01%,0.05%,and 0.1%atropine eye drops,respectively,the area of CNV was measured.RESULTS:Significant increases were observed on the expression of DRD2 in mouse high myopia model at 0.5,1,2h,7d with 0.05%and 0.1%atropine eye drops(P<0.05).Significant decreases were observed on the expression of DRD1 and VEGFA in mouse high myopia model at 0.5,1,2h,7d with 0.05%and 0.1%atropine eye drops(P<0.05).The area of CNV induced by laser in the drug-treated group was significantly smaller than that in the control group,and the higher the concentration,the more significant the inhibitory effect(P<0.05).CONCLUSION:The 0.01%,0.05%,0.1%atropine eye drops can decrease the level of VEGFA and inhibit high myopia CNV indirectly by up-regulating the level of DRD2 and down-regulating the level of DRD1,and the effect of 0.05%and 0.1%atropine eye drops is more significant.
基金supported by the National Natural Science Foundation of China(21878295,22078024)the Natural Science Foundation of Beijing(2192052)the Project funded by Liaoning Provincial Department of Education(LQ2020001)。
文摘The increased concentration of CO_(2) due to continuous breathing and no discharge of human beings in the manned closed space,like spacecraft and submarines,can be a threat to health and safety.Effective removal of low concentration CO_(2) from the manned closed space is essential to meet the requirements of long-term space or deep-sea exploration,which is an international frontier and trend.Ionic liquids(ILs),as a widespread and green solvent,already showed its excellent performance on CO_(2) capture and absorption,indicating its potential application in low concentration CO_(2) capture.In this review,we first summarized the current methods and strategies for direct capture from low concentration CO_(2) in both the atmosphere and manned closed spaces.Then,the multi-scale simulation methods of CO_(2) capture by ionic liquids are described in detail,including screening ionic liquids by COSMO-RS methods,capture mechanism by density functional theory and molecular dynamics simulation,and absorption process by computational fluid dynamics simulation.Lastly,some typical IL-based green technologies for low concentration CO_(2) capture,such as functionalized ILs,co-solvent systems with ILs,and supported materials based on ILs,are introduced,and analyzed the subtle possibility in manned closed spaces.Finally,we look forward to the technology and development of low concentration CO_(2) capture,which can meet the needs of human survival in closed space and proposed that supported materials with ionic liquids have great advantages and infinite possibilities in the vital area.
文摘针对燃煤电厂在吹扫等过程中脱硫系统出口SO_(2)浓度的不能及时检测的问题,提出了一种基于特征选择的改进粒子群优化算法优化门控循环单元神经网络(IPSO-GRU)的脱硫系统出口SO_(2)浓度预测模型。通过最大相关最小冗余(minimum Redundancy and Maximum Relevance,mRMR)算法对采集的目标数据进行预处理,挑选出合适的变量,随后将选定的变量作为IPSO-GRU预测模型的输入。针对门控循环单元(Gated Recurrent Unit,GRU)模型关键超参数难以确定的问题,使用改进粒子群(Improved Particle Swarm Optimization,IPSO)算法对模型参数进行训练,以降低GRU的训练成本。最终实现对脱硫系统出口二氧化硫浓度的预测。实验结果表明,所提模型与传统循环神经网络相比预测精度更高,在工程实际中更具应用价值。
