This paper proposes a novel locally linear backpropagation based contribution(LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder(AE), LLBBC can deal with the fau...This paper proposes a novel locally linear backpropagation based contribution(LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder(AE), LLBBC can deal with the fault diagnosis problem through extracting nonlinear features. When the on-line fault diagnosis task is in progress, a locally linear model is firstly built at the current fault sample. According to the basic idea of reconstruction based contribution(RBC), the propagation of fault information is described by using back-propagation(BP) algorithm. Then, a contribution index is established to measure the correlation between the variable and the fault, and the final diagnosis result is obtained by searching variables with large contributions. The smearing effect, which is an important factor affecting the performance of fault diagnosis, can be suppressed as well,and the theoretical analysis reveals that the correct diagnosis can be guaranteed by LLBBC. Finally, the feasibility and effectiveness of the proposed method are verified through a nonlinear numerical example and the Tennessee Eastman benchmark process.展开更多
The Tibetan Plateau(TP)is a prevalent region for convection systems due to its unique thermodynamic forcing.This study investigated isolated deep convections(IDCs),which have a smaller spatial and temporal size than m...The Tibetan Plateau(TP)is a prevalent region for convection systems due to its unique thermodynamic forcing.This study investigated isolated deep convections(IDCs),which have a smaller spatial and temporal size than mesoscale convective systems(MCSs),over the TP in the rainy season(June-September)during 2001–2020.The authors used satellite precipitation and brightness temperature observations from the Global Precipitation Measurement mission.Results show that IDCs mainly concentrate over the southern TP.The IDC number per rainy season decreases from around 140 over the southern TP to around 10 over the northern TP,with an average 54.2.The initiation time of IDCs exhibits an obvious diurnal cycle,with the peak at 1400–1500 LST and the valley at 0900–1000 LST.Most IDCs last less than five hours and more than half appear for only one hour.IDCs generally have a cold cloud area of 7422.9 km^(2),containing a precipitation area of approximately 65%.The larger the IDC,the larger the fraction of intense precipitation it contains.IDCs contribute approximately 20%–30%to total precipitation and approximately 30%–40%to extreme precipitation over the TP,with a larger percentage in July and August than in June and September.In terms of spatial distribution,IDCs contribute more to both total precipitation and extreme precipitation over the TP compared to the surrounding plain regions.IDCs over the TP account for a larger fraction than MCSs,indicating the important role of IDCs over the region.展开更多
Phosphorescent and thermally activated delayed fluorescence(TADF)emitters can break through the spin statistics rules and achieve great success in external quantum efficiency(over 5%).However,maintaining high efficien...Phosphorescent and thermally activated delayed fluorescence(TADF)emitters can break through the spin statistics rules and achieve great success in external quantum efficiency(over 5%).However,maintaining high efficiency at high brightness is a tremendous challenge for applications of organic light emitting diodes.Hence,we reported two phenanthroimidazole derivatives PPI-An-CN and PPI-An-TP and achieved extremely low efficiency roll-off with about 99%of the maximum external quantum efficiency(EQEmax)maintained even at a high luminance of 1000 cd/cm2 based non-doped devices.When doping the two materials in CBP(4,4'-bis(N-carbazolyl)-1,1'-biphenyl),the doped devices still exhibited excellent stability at high brightness with CIEy≈0.07 and low turn-on voltage of only 2.8 V.The state-ofthe-art low efficiency roll-off makes the new materials attractive for potential applications.It is the first time that the Fragment Contribution Analysis method has been used to analyze the excited state properties of the molecules in the field of OLEDs,which helps us understand the mechanism more intuitively and deeply.展开更多
系统异构性和统计异构性的存在使得通信开销和通信效率成为联邦学习的关键瓶颈之一,在众多参与方中只选取一部分客户端执行模型更新和聚合可以有效降低通信开销,但是选择偏差和客户端上的数据质量分布不平衡对客户端采样方法提出了额外...系统异构性和统计异构性的存在使得通信开销和通信效率成为联邦学习的关键瓶颈之一,在众多参与方中只选取一部分客户端执行模型更新和聚合可以有效降低通信开销,但是选择偏差和客户端上的数据质量分布不平衡对客户端采样方法提出了额外的挑战。为此,提出数据质量评估的高效强化联邦学习节点动态采样优化方法(client dynamic sampling optimization of efficient reinforcement federated learning based on data quality assessment,RQCS),该方法采用沙普利值的贡献指数评估客户端上的数据质量,基于深度强化学习模型,智能的动态选择具有高数据质量且能提高最终模型精度的客户端参与每一轮的联邦学习,以抵消数据质量分布不平衡引入的偏差,加速模型收敛并提高模型精度。在MNIST及CIFAR-10数据集上的实验表明,所提出算法与其他算法相比,在减少通信开销的同时进一步加快了收敛速度,同时在模型最终准确性上也有较好的性能。展开更多
基金supported by the Key Project of National Natural Science Foundation of China(61933013)Ningbo 13th Five-year Marine Economic Innovation and Development Demonstration Project(NBH Y-2017-Z1)。
文摘This paper proposes a novel locally linear backpropagation based contribution(LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder(AE), LLBBC can deal with the fault diagnosis problem through extracting nonlinear features. When the on-line fault diagnosis task is in progress, a locally linear model is firstly built at the current fault sample. According to the basic idea of reconstruction based contribution(RBC), the propagation of fault information is described by using back-propagation(BP) algorithm. Then, a contribution index is established to measure the correlation between the variable and the fault, and the final diagnosis result is obtained by searching variables with large contributions. The smearing effect, which is an important factor affecting the performance of fault diagnosis, can be suppressed as well,and the theoretical analysis reveals that the correct diagnosis can be guaranteed by LLBBC. Finally, the feasibility and effectiveness of the proposed method are verified through a nonlinear numerical example and the Tennessee Eastman benchmark process.
基金supported by the National Natural Science Foundation of China[grant number 42105064]the Second Tibetan Plateau Scientific Expedition and Research(STEP)program[grant number 2019QZKK0102]the special fund of the Yunnan University“double first-class”construction.
文摘The Tibetan Plateau(TP)is a prevalent region for convection systems due to its unique thermodynamic forcing.This study investigated isolated deep convections(IDCs),which have a smaller spatial and temporal size than mesoscale convective systems(MCSs),over the TP in the rainy season(June-September)during 2001–2020.The authors used satellite precipitation and brightness temperature observations from the Global Precipitation Measurement mission.Results show that IDCs mainly concentrate over the southern TP.The IDC number per rainy season decreases from around 140 over the southern TP to around 10 over the northern TP,with an average 54.2.The initiation time of IDCs exhibits an obvious diurnal cycle,with the peak at 1400–1500 LST and the valley at 0900–1000 LST.Most IDCs last less than five hours and more than half appear for only one hour.IDCs generally have a cold cloud area of 7422.9 km^(2),containing a precipitation area of approximately 65%.The larger the IDC,the larger the fraction of intense precipitation it contains.IDCs contribute approximately 20%–30%to total precipitation and approximately 30%–40%to extreme precipitation over the TP,with a larger percentage in July and August than in June and September.In terms of spatial distribution,IDCs contribute more to both total precipitation and extreme precipitation over the TP compared to the surrounding plain regions.IDCs over the TP account for a larger fraction than MCSs,indicating the important role of IDCs over the region.
基金supported by the National Natural Science Foundation of China(No.51673113)the Key Project of DEGP(No.2018KZDXM032)
文摘Phosphorescent and thermally activated delayed fluorescence(TADF)emitters can break through the spin statistics rules and achieve great success in external quantum efficiency(over 5%).However,maintaining high efficiency at high brightness is a tremendous challenge for applications of organic light emitting diodes.Hence,we reported two phenanthroimidazole derivatives PPI-An-CN and PPI-An-TP and achieved extremely low efficiency roll-off with about 99%of the maximum external quantum efficiency(EQEmax)maintained even at a high luminance of 1000 cd/cm2 based non-doped devices.When doping the two materials in CBP(4,4'-bis(N-carbazolyl)-1,1'-biphenyl),the doped devices still exhibited excellent stability at high brightness with CIEy≈0.07 and low turn-on voltage of only 2.8 V.The state-ofthe-art low efficiency roll-off makes the new materials attractive for potential applications.It is the first time that the Fragment Contribution Analysis method has been used to analyze the excited state properties of the molecules in the field of OLEDs,which helps us understand the mechanism more intuitively and deeply.
文摘系统异构性和统计异构性的存在使得通信开销和通信效率成为联邦学习的关键瓶颈之一,在众多参与方中只选取一部分客户端执行模型更新和聚合可以有效降低通信开销,但是选择偏差和客户端上的数据质量分布不平衡对客户端采样方法提出了额外的挑战。为此,提出数据质量评估的高效强化联邦学习节点动态采样优化方法(client dynamic sampling optimization of efficient reinforcement federated learning based on data quality assessment,RQCS),该方法采用沙普利值的贡献指数评估客户端上的数据质量,基于深度强化学习模型,智能的动态选择具有高数据质量且能提高最终模型精度的客户端参与每一轮的联邦学习,以抵消数据质量分布不平衡引入的偏差,加速模型收敛并提高模型精度。在MNIST及CIFAR-10数据集上的实验表明,所提出算法与其他算法相比,在减少通信开销的同时进一步加快了收敛速度,同时在模型最终准确性上也有较好的性能。
