In this review, first, we discussed the effect of phosphor features on optical properties by the software simulation in detail. A combination of these parameters: phosphor material, phosphor particle size and particle...In this review, first, we discussed the effect of phosphor features on optical properties by the software simulation in detail. A combination of these parameters: phosphor material, phosphor particle size and particle distribution, phosphor layer concentration, phosphor layer thickness, geometry, and location of the phosphor layer, will result in the final optical performance of the phosphor layer. Secondly, we introduced how to improve light extraction efficiency with various proposed methods. Thirdly, we summarized the thermal models to predict the phosphor temperature and the junction temperature. To stabilize the optical performance of phosphor-converted light emitting diodes (PC-LEDs), much effort has been made to reduce the junction temperature of the LED chips. The phosphor temperature, a critical reliability concern for PC-LEDs, should be attracted academic interest. Finally, we summed up optical-thermal coupled model for phosphors and summarized future optical- thermal issues exploring the light quality for LEDs. We foresee that optical-thermal coupled model for PC-LEDs should be paid more attention in the future.展开更多
针对变电站内各规格LED数码表数据实时监测出现的低分辨率、模糊、高光、反光小数点误识别问题,提出一种联合YOLOv5神经网络与LIG(local intensity and gradient)小数点识别算法的识别方法。采用YOLOv5识别LED数字部分,通过HSV颜色信息...针对变电站内各规格LED数码表数据实时监测出现的低分辨率、模糊、高光、反光小数点误识别问题,提出一种联合YOLOv5神经网络与LIG(local intensity and gradient)小数点识别算法的识别方法。采用YOLOv5识别LED数字部分,通过HSV颜色信息与灰度信息共同筛选LED部分,并与LIG算法确定小数点位置,完成LED字符数值读取。对采集到的LED数码表图像数据进行实验,结果表明,联合YOLOv5神经网络与LIG小数点识别算法小数点识别准确率平均提高了约8.5%。展开更多
Commercial phosphor-converted white LEDs(pc-WLEDs)face two inherent limitations,namely blue light hazard and low color rendering index,due to the use of blue LEDs as excitation source.To address these challenges,viole...Commercial phosphor-converted white LEDs(pc-WLEDs)face two inherent limitations,namely blue light hazard and low color rendering index,due to the use of blue LEDs as excitation source.To address these challenges,violet LEDs are proposed as an alternative solution.Currently,phosphors that can be efficiently excited by violet light(with wavelengths from 400 to 420 nm)remain under development still.In this study,we utilize large language models to construct a comprehensive database of Eu^(2+)and Ce^(3+)doped phosphors for discovering novel violet-excited phosphors.A total of 822 phosphor data entries,including elemental compositions,crystal structures and excitation/emission wavelengths,have been extracted and validated from 9551 research papers.Compared with Ce^(3+)doped phosphors,the Eu^(2+)are in general more suited for violet-excited phosphors,as well as red-emitting phosphors.In particular,Eu^(2+)doped nitrides and sulfides are worth of exploration for violet-excited phosphors.This database is expected to be useful in the future development of phosphors for pc-WLEDs based on artificial intelligence methods.The datasets in this article are listed in Science Data Bank at http://doi.org/10.57760/sciencedb.34314.展开更多
文摘In this review, first, we discussed the effect of phosphor features on optical properties by the software simulation in detail. A combination of these parameters: phosphor material, phosphor particle size and particle distribution, phosphor layer concentration, phosphor layer thickness, geometry, and location of the phosphor layer, will result in the final optical performance of the phosphor layer. Secondly, we introduced how to improve light extraction efficiency with various proposed methods. Thirdly, we summarized the thermal models to predict the phosphor temperature and the junction temperature. To stabilize the optical performance of phosphor-converted light emitting diodes (PC-LEDs), much effort has been made to reduce the junction temperature of the LED chips. The phosphor temperature, a critical reliability concern for PC-LEDs, should be attracted academic interest. Finally, we summed up optical-thermal coupled model for phosphors and summarized future optical- thermal issues exploring the light quality for LEDs. We foresee that optical-thermal coupled model for PC-LEDs should be paid more attention in the future.
文摘针对变电站内各规格LED数码表数据实时监测出现的低分辨率、模糊、高光、反光小数点误识别问题,提出一种联合YOLOv5神经网络与LIG(local intensity and gradient)小数点识别算法的识别方法。采用YOLOv5识别LED数字部分,通过HSV颜色信息与灰度信息共同筛选LED部分,并与LIG算法确定小数点位置,完成LED字符数值读取。对采集到的LED数码表图像数据进行实验,结果表明,联合YOLOv5神经网络与LIG小数点识别算法小数点识别准确率平均提高了约8.5%。
基金National Key Research and Development Program of China(2021YFB3500501)。
文摘Commercial phosphor-converted white LEDs(pc-WLEDs)face two inherent limitations,namely blue light hazard and low color rendering index,due to the use of blue LEDs as excitation source.To address these challenges,violet LEDs are proposed as an alternative solution.Currently,phosphors that can be efficiently excited by violet light(with wavelengths from 400 to 420 nm)remain under development still.In this study,we utilize large language models to construct a comprehensive database of Eu^(2+)and Ce^(3+)doped phosphors for discovering novel violet-excited phosphors.A total of 822 phosphor data entries,including elemental compositions,crystal structures and excitation/emission wavelengths,have been extracted and validated from 9551 research papers.Compared with Ce^(3+)doped phosphors,the Eu^(2+)are in general more suited for violet-excited phosphors,as well as red-emitting phosphors.In particular,Eu^(2+)doped nitrides and sulfides are worth of exploration for violet-excited phosphors.This database is expected to be useful in the future development of phosphors for pc-WLEDs based on artificial intelligence methods.The datasets in this article are listed in Science Data Bank at http://doi.org/10.57760/sciencedb.34314.
