Volume 51 Issue 7
Aug.  2022
Turn off MathJax
Article Contents

Li Jianfei, Zhou Xiaoming. Research on lighting system based on LED current and temperature spectrum model[J]. Infrared and Laser Engineering, 2022, 51(7): 20210727. doi: 10.3788/IRLA20210727
Citation: Li Jianfei, Zhou Xiaoming. Research on lighting system based on LED current and temperature spectrum model[J]. Infrared and Laser Engineering, 2022, 51(7): 20210727. doi: 10.3788/IRLA20210727

Research on lighting system based on LED current and temperature spectrum model

doi: 10.3788/IRLA20210727
Funds:  Open Project of State Key Laboratory of Subtropical Building Science(2020 ZA05)
  • Received Date: 2022-01-12
  • Rev Recd Date: 2022-03-05
  • Publish Date: 2022-08-05
  • Changes in current and temperature can affect the spectral distribution of the LED, which in turn affects the visual and non-visual parameters of the light source. In this paper, starting from the spectral model based on the distribution of photon energy levels in the LED chip luminescent material, the current and temperature model was established for the RGBY four-color LED. The R-square of the spectral fitting could reach 0.99. On this basis, the genetic algorithm (GA) was used to optimize the circadian rhythm factor and luminous efficiency with multiple objectives. When the illuminance was 300 lx, 8 sets of visual parameters (color rendering index and blue light hazard efficiency) and non-visual parameters ( circadian rhythm stimulus) were designed to verify the feasibility of this model. Then, the relationship between two non visual parameters and temperature was explored. The results show that the circadian rhythm factors increased with the increase of temperature, but the circadian rhythm stimulus decreased with the increase of temperature. The reason for this situation was that the two parameters were differently affected by the illuminance. After compensating the illuminance of the light source, it was found that the two non-visual parameters both increased with the increase of temperature, and the two showed a certain positive correlation. This research started from the perspective of the light source spectrum and provided a reference for the consideration of non-visual effects in the design of LED lighting sources.
  • [1] Berson D M, Dunn F A, Takao M. Phototransduction by retinal ganglion cells that set the circadian clock [J]. Science, 2002, 295: 1070-1073. doi:  10.1126/science.1067262
    [2] Gall D, Bieske K. Definition and measurement of circadian radiometric quantities[C]//Vienna: Commission internationale de l'éclairage, 2004: 129-132.
    [3] Rea M S, Figueiro M G. Light as a circadian stimulus for architectural lighting [J]. Lighting Research & Technology, 2018, 50(4): 497-510.
    [4] 吕玫, 张丽, 张彦, 等. 量子点发光二极管稳定性提高策略[J]. 中国光学, 2021, 14(01): 117-134. doi:  10.37188/CO.2020-0184

    Lv M, Zhang L, Zhang Y, et al. Strategies for improving the stability of quantum dots light-emitting diodes [J]. Chinese Optics, 2021, 14(1): 117-134. (in Chinese) doi:  10.37188/CO.2020-0184
    [5] 郑莉莉, 郭自泉, 严威, 等. 三基色白光LED的司辰节律因子研究[J]. 发光学报, 2016, 37(011): 1384-1389. doi:  10.3788/fgxb20163711.1384

    Zheng L L, Guo Z Q, Yan W, et al. Investigation on the circadian action factor of RGB white LEDs [J]. Chinese Journal of Luminescence, 2016, 37(11): 1384-1389. (in Chinese) doi:  10.3788/fgxb20163711.1384
    [6] Jarboe C, Snyder J, Figueiro M G. The effectiveness of light-emitting diode lighting for providing circadian stimulus in office spaces while minimizing energy use [J]. Lighting Research & Technology, 2020, 52(2): 167-188.
    [7] Figueiro M G, Steverson B, Heerwagen J, et al. The impact of daytime light exposures on sleep and mood in office workers [J]. Sleep Health, 2017, 3(3): 204-215. doi:  10.1016/j.sleh.2017.03.005
    [8] 陈风, 袁银麟, 郑小兵, 等. LED 的光谱分布可调光源的设计[J]. 光学精密工程, 2008, 16(11): 2060-2064. doi:  10.3321/j.issn:1004-924X.2008.11.003

    Chen F, Yuan Y L, Zheng X B, et al. Design of spectral tunable LED light source [J]. Optics and Precision Engineering, 2008, 16(11): 2060-2064. (in Chinese) doi:  10.3321/j.issn:1004-924X.2008.11.003
    [9] 蒋腾, 卓宁泽, 谢琳艳, 等. 基于电流和双高斯模型的RGB LED光谱仿真与研究[J]. 照明工程学报, 2020, 031(001): 22-26.

    Jiang T, Zhuo N Z, Xie L Y, et al. Research and simulation of RGB LED spectrums based on currents and double Gaussian model [J]. China Illuminating Engineering Journal, 2020, 31(1): 22-26. (in Chinese)
    [10] 马乐, 陆威, 姜鹏, 等. 基于双高斯方差优化的Gm-APD三维距离像预处理方法研究[J]. 红外与激光工程, 2020, 49(S2): 131-136.

    Ma L, Lu W, Jiang P, et al. Research on Gm-APD 3 D range image preprocessing method based on double Gaussian variance optimization [J]. Infrared and Laser Engineering, 2020, 49(S2): 20200388. (in Chinese)
    [11] 王滨辉, 宋沙磊, 曹雄, 等. 多光谱激光雷达波形数据处理及应用[J]. 红外与激光工程, 2020, 49(S2): 96-102.

    Wang B H, Song S L, Cao X, et al. Processing and application of multispectral lidar waveform data [J]. Infrared and Laser Engineering, 2020, 49(S2): 20200368. (in Chinese)
    [12] 徐广强, 于慧嫒, 张竞辉, 等. 基于单色 LED 补偿白光 LED 技术的模拟太阳光谱研究[J]. 发光学报, 2017, 38(8): 1117-1124. doi:  10.3788/fgxb20173808.1117

    Xu G Q, Yu H M, Zhang J H, et al. Solar spectrum matching based on white LED compensated with monochromatic LEDs [J]. Chinese Journal of Luminescence, 2017, 38(8): 1117-1124. (in Chinese) doi:  10.3788/fgxb20173808.1117
    [13] Badgutdinov M L, Yunovich A É. Emission spectra of InGaN/AlGaN/GaN quantum well heterostructures: Model of the two-dimensional joint density of states [J]. Semiconductors, 2008, 42(4): 429-438. doi:  10.1134/S1063782608040106
    [14] 王雨思, 周贤建, 李青原, 等. 截止斜率可控的蓝光防护光学薄膜[J]. 中国光学, 2021, 14(03): 544-551. doi:  10.37188/CO.2020-0212

    Wang Y S, Zhou X J, Li Q Y, et al. Blue-blocking optical thin films with controllable cutoff slope [J]. Chinese Optics, 2021, 14(3): 544-551. (in Chinese) doi:  10.37188/CO.2020-0212
    [15] 陈刚, 刘言, 杨贺超, 等. 低照度彩色图像的自适应亮度增强[J]. 光学精密工程, 2021, 29(08): 1999-2007. doi:  10.37188/OPE.20212908.1999

    Chen G, Liu Y, Yang H C, et al. Adaptive brightness correction of dim-lightening color images [J]. Optics and Precision Engineering, 2021, 29(8): 1999-2007. (in Chinese) doi:  10.37188/OPE.20212908.1999
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(10)  / Tables(3)

Article Metrics

Article views(219) PDF downloads(45) Cited by()

Related
Proportional views

Research on lighting system based on LED current and temperature spectrum model

doi: 10.3788/IRLA20210727
  • State Key Laboratory of Subtropical Architecture, School of Physics and Optoelectronics, South China University of Technology, Guangzhou 510641, China
Fund Project:  Open Project of State Key Laboratory of Subtropical Building Science(2020 ZA05)

Abstract: Changes in current and temperature can affect the spectral distribution of the LED, which in turn affects the visual and non-visual parameters of the light source. In this paper, starting from the spectral model based on the distribution of photon energy levels in the LED chip luminescent material, the current and temperature model was established for the RGBY four-color LED. The R-square of the spectral fitting could reach 0.99. On this basis, the genetic algorithm (GA) was used to optimize the circadian rhythm factor and luminous efficiency with multiple objectives. When the illuminance was 300 lx, 8 sets of visual parameters (color rendering index and blue light hazard efficiency) and non-visual parameters ( circadian rhythm stimulus) were designed to verify the feasibility of this model. Then, the relationship between two non visual parameters and temperature was explored. The results show that the circadian rhythm factors increased with the increase of temperature, but the circadian rhythm stimulus decreased with the increase of temperature. The reason for this situation was that the two parameters were differently affected by the illuminance. After compensating the illuminance of the light source, it was found that the two non-visual parameters both increased with the increase of temperature, and the two showed a certain positive correlation. This research started from the perspective of the light source spectrum and provided a reference for the consideration of non-visual effects in the design of LED lighting sources.

    • 2002年,Berson等[1]发现了哺乳动物视网膜的第三类感光细胞(ipRGC),打破了人眼只存在视觉通路的观念,非视觉通道逐渐被大众熟悉。ipRGC能参与调节人体光生物效应,包括激素分泌、昼夜节律和瞳孔大小等,光不仅影响着人体生理节律,在视觉工作、主观情绪等方面也存在影响。因此,照明质量的评价应由原来单一视觉评价过渡到视觉和非视觉的双重评价。目前对非视觉的评估尚未形成统一标准,Gall等[2]通过非视觉光谱响应曲线和明视觉光谱响应曲线,提出用昼夜节律因子$ {C}_{AF} $ (Circadian Action Factor)表示非视觉效应的大小;Rea等[3]提出用昼夜节律刺激值CS (Circadian Stimulus)和昼夜节律照度$ {CL}_{A} $ (Circadian Light)描述非视觉效应的效果,此模型考虑了S型视锥细胞光谱灵敏度曲线以及人眼黑视素光谱灵敏度曲线的影响,昼夜节律刺激的阈值和饱和值分别为0.1和0.7,昼夜节律照度是光源在2856 K黑体辐射体标准化之下的辐射照度值。

      由于LED具有亮度大、可调节性强等特点,它的使用越来越广泛[4]。为了设计满足生活工作需要的照明系统,需要兼顾视觉参数(显色性指数、辐射发光效率、蓝光危害效率)和非视觉参数(昼夜节律因子、昼夜节律刺激值)。郑莉莉等[5]通过RGB照明系统,研究了昼夜节律因子和相关色温受各通道电流的影响,给出了昼夜节律因子、相关色温与电流之间的数学模型,并模拟了自然光在不同时段的光谱,不过该研究只探讨了电流对昼夜节律因子的影响,未涉及视觉参数。Jarboe等[6-7]评估了几种LED在办公室的使用效果,研究表明,白天时至少维持2 h超过0.3的CS值,有利于提高工作效率和警觉性,该研究着重分析了昼夜节律刺激参数,但缺少昼夜节律因子的分析。

      文中以LED芯片发光材料中光子能级分布规律的光谱模型为基础,通过改变LED所处的电流、温度来模拟光谱,确定RGBY各通道光源的绝对光谱和电流、温度之间的关系。通过遗传算法进行视觉和非视觉参数的多目标优化,对辐射发光效率和昼夜节律因子取Pareto最优解集,设置照度为300 lx,综合考量了昼夜节律刺激、蓝光危害效率、显色指数等指标,设计出8组适宜的照明系统,并探究8组方案非视觉参数与温度的关系。

    • 以往光谱拟合主要通过高斯函数或洛伦兹函数来实现,但拟合效果都不理想。高斯模型拟合尾部收敛较快的光谱效果不错,而洛伦兹模型更适合尾部收敛较慢的曲线拟合[8]。蒋腾等[9]通过双高斯模型[10-11]探究电流对RGB参数的影响,随着驱动电流增加,红光光谱向长波方向移动,绿光和蓝光光谱产生蓝移,拟合参数在0.9842以上。尽管以上研究综合了高斯和洛伦兹模型的优势,但拟合的效果依旧欠佳。徐广强等[12]采用光子在二维空间内联合态密度函数作为单色LED的光谱模型,实现了对目标光谱的匹配。以上研究分析了单一的非视觉参数和视觉参数的关系,未分析两种非视觉参数之间的关系。

      光谱产生本质上是光源在不同波长时能量分布不同[13]。基于LED芯片发光材料中光子能级分布规律,采用如下数学模型表征单色LED光谱:

      式中:$ S\left(\lambda \right) $表示光谱辐射强度;$ \lambda $表示波长;$ A、 {\lambda }_{c1}、 {\lambda }_{c2}、 {\omega }_{1}、{\omega }_{2} $均为拟合参数。

      利用RGBY四色光源来探究光源绝对光谱和电流、温度的关系。为防止外界光源对实验的影响,实验在密闭无光环境下进行,系统构建框图如图1所示。

      Figure 1.  Block diagram of RGBY four-color LED system

      光源驱动系统由微控制器、恒温加热台、驱动器、直流稳压电源(12 V)、直流降压模块(3.5 V)、RGBY灯珠组成。通过驱动模块DD313来驱动电路,驱动电流为350 mA。电流控制主要通过改变PWM通道占空比来实现,在每个通道串联一个0.1 Ω的电阻,通过蓝牙调节当前灯珠占空比进而改变电流,用万用表测量电阻的电压,这样就获得LED各通道电流值。建立光谱的电流模型时,将电流设置为100 mA、200 mA和300 mA获取数据。恒温加热台V-0505能灵活调节LED温度,范围设置在30°~120°内,以15°为间隔递增。使用SPIC-200光谱照度计测得不同电流、温度下的光谱。RGBY器件示意图以及数据测试实物图分别如图2图3所示。

      Figure 2.  Schematic diagram of RGBY device

      Figure 3.  Photos of data testing equipment

      利用1 stopt软件对RGBY绝对光谱分别拟合, 以红灯在30°、300 mA下为例,此时在635 nm处光功率最大,$ {P}_{max}\left(635\right) $为0.0552 ${\rm{W}}\cdot {\left({\rm{m}}^{2}\cdot {\rm{nm}}^{-1}\right)}^{-1}$,光谱见图4

      Figure 4.  Fitting effect of red light at 300 mA and 30°

      根据以上数据,文中建立了光谱与电流、温度的二阶方程。拟合参数$ {A、\lambda }_{c1}、{\lambda }_{c2}、{\omega }_{1}、{\omega }_{2} $与电流I和温度T之间的关系见公式(2)~(6):

      以上拟合参数R-square均大于0.95,拟合效果较好。根据公式(6)可知,当给出单色LED所处的电流、温度值时,可以计算出公式(1)中的拟合参数,以红光为例,其电流和温度的综合影响模型如公式(7)所示,同理可以得出绿光、蓝光和黄光的光谱功率分布。这样就可以计算出当前光源系统的光谱功率分布,进而得出光源的视觉和非视觉参数。

    • 视觉参数包括[14-15]辐射发光效率LER、蓝光危害效率$ {\eta }_{B} $、显色指数$ {R}_{a} $、相关色温CCT和照度E,非视觉参数包括昼夜节律因子$ {C}_{AF} $、昼夜节律刺激CS。昼夜节律因子$ {C}_{AF} $的计算见公式(8):

      式中:$ P(\lambda ,I,T) $为光源在电流I、温度T下光谱功率的叠加;$ C(\lambda ) $为非视觉光谱响应曲线;$ V(\lambda ) $为明视觉光谱响应曲线。昼夜节律刺激CS值可由公式(9)来计算:

      式中:$ C{L_A} $为生理节律光照,1548是归一化常数; $ {E_\lambda } $为光谱辐照度分布;$ {S_\lambda } $S型视锥细胞光谱灵敏度曲线;$ m{p_\lambda } $是黄斑色素透射率曲线;$ {V_\lambda } $$ {V_\lambda }^, $分别代表明视觉以及暗视觉的光谱响应曲线;$RodS at、 k、 {a}_{b-y}$$ {a}_{rod} $为常数,分别为6.5 W/${\rm{m}}^{2}$、0.2616、0.7和3.3。

      辐射发光效率LER和照度E的计算如下:

      $ {\mathrm{式}\mathrm{中}:K}_{m} $表示光源最大光通量,为683 lm/W。

      为了设计既有较高$ {C_{AF}} $又有较高LER的照明系统,需对两参数折衷处理。根据国际照明委员会CIE对办公室环境的参数要求,照度设为300 lx,显色指数$ {R}_{a} $大于90。考虑到现实环境因素会阻碍CS传递到人眼,将CS值设置为0.3~0.4之间,蓝光危害效率$ {\eta }_{B} $控制在0.3以下,在2500~8000 K色温下寻求满足上述条件的照明方案。采用遗传算法对$ {C_{AF}} $和LER进行多目标优化,为了消除量纲的影响,把$ {C_{AF}} $和LER归一化处理, 通过Matlab仿真得出$ {C_{AF}} $在限制下最大值和最小值分别为1.1647和0.1963;LER最大值和最小值分别为368 lm/W和265 lm/W。优化的方程如下,以四路占空比大小作为未知量$ {D_r},{D_g},{D_b},{D_y} $:

      $ {C_{AF}} $和LER的Pareto解集时,最优前端个体系数ParetoFraction取0.08,种群大小PopulationSize取100,最大遗传代数Generations设置为500代,停止代数StallGenLimit为500代,适应度函数偏差TolFun设为$ 1{e}^{-100} $,得到的8个照明系统方案如表1所示。

      表1可知,照度为300 lx时,照明方案的色温分布在2855~7628 K,节律刺激值CS随色温的变化曲线如图5所示, CS在2800~3500 K、4000~8000 K时随色温增加而变大,在3500~4000 K呈现急剧下降的趋势,这是因为$ {CL}_{A} $和色温有公式(9)的关系,而CS和$ {CL}_{A} $呈正相关。要想维持CS值在0.3~0.4之间,色温的可选空间较大。

      No.CCT/KCAFCSLER/lm·W−1RaηB
      128550.22230.3008360900.0483
      231310.25140.3107355900.1005
      335320.30630.3306347900.1432
      440000.38960.3016340910.1648
      549860.57830.3344326910.2097
      659320.76030.3661304910.2305
      766610.86410.3897283920.2482
      876281.01920.3986278920.2837

      Table 1.  Calculated values ​​of the parameters of the eight groups of optimal solutions

      Figure 5.  CS value of each scheme under different color temperature

    • 为验证该光谱模型的准确性,测试这8种方案下各路电流值(见表2),再改变灯珠温度,将各方案实测值和模型值进行比较。8种方案是按照色温的大小来排列的,即按照昼夜节律因子$ {C_{AF}} $的大小排列,方案8的色温最高,即方案8对人体的非视觉效应影响最大。由于是在2500~8000 K色温内的寻优,所以这8种方案可以代表限制条件下的最佳方案。

      No.Ir/mAIg/mAIb/mAIy/mA
      1227287324
      21834812317
      31408326309
      411913654301
      571169113231
      663170178233
      764174204218
      873106238212

      Table 2.  Measured four current values under each scheme

      由于目前常用的照明光源为D50和D65,而方案5和方案7的色温分别最接近D50和D65,以方案5和方案7为例,分析实测光谱和模型光谱的拟合情况,如图6所示。

      Figure 6.  Comparison of measured spectrum and model spectrum

      从拟合效果看,两个方案均能准确反映实际光谱。为了更精确地体现模型和实测光谱的差距,表3计算了8种方案各参数指标的误差,并标明了最大误差时的温度。由于SPIC-200可以直接测出相关色温CCT、显色指数$ {R}_{a} $、照度E,但辐射发光效率LER、昼夜节律因子$ {C}_{AF} $等参数无法直接测出,对于无法直接测得的参数,通过SPIC-200测量出实际的光谱,再根据公式(10)计算得出。

      No.CCT/KCAFCSLER/lm·W−1RaηB
      13.84% (45 ℃)2.98% (75 ℃)3.07% (120 ℃)5.03% (105 ℃)4.81% (105 ℃)2.91% (120 ℃)
      24.94% (90 ℃)3.61% (90 ℃)2.16% (105 ℃)4.78% (60 ℃)3.26% (120 ℃)4.18% (65 ℃)
      34.56% (105 ℃)3.98% (45 ℃)3.89% (105 ℃)5.21% (120 ℃)3.96% (60 ℃)5.06% (105 ℃)
      42.57% (75 ℃)3.85% (60 ℃)5.08% (90 ℃)6.89% (120 ℃)4.68% (75 ℃)5.68% (120 ℃)
      55.66% (120 ℃)5.76% (90 ℃)4.93% (105 ℃)5.82% (60 ℃)4.98% (105 ℃)4.82% (90 ℃)
      63.89% (105 ℃)2.68% (75 ℃)5.31% (60 ℃)4.78% (60 ℃)5.96% (75 ℃)5.41% (90 ℃)
      76.94% (120 ℃)5.33% (30 ℃)2.85% (75 ℃)6.74% (90 ℃)6.12% (105 ℃)3.56% (45 ℃)
      86.57% (30 ℃)7.06% (120 ℃)5.97% (105 ℃)5.87% (60 ℃)5.96% (30 ℃)6.78% (90 ℃)

      Table 3.  Comparison of measured values ​​and model values ​​of the parameters of each scheme

      表3中,模型和实测参数的绝对误差最大为7.06%,说明该模型能较好拟合实际光谱。初始照度为300 lx,但随着LED温度的升高,E逐渐衰减。图7给出了照度E和温度T的拟合曲线,两者呈线性负相关,$ {R}^{2} $为0.9869。不同方案参数的最大误差值有差别,影响参数误差的因素为LED的结温差、热耦合效应等。驱动LED芯片时,连续波电流会使LED因热效应进而结温上升,影响实验结果,在实验中采用脉冲电流驱动LED发光,并且仅在测量光谱的一瞬间闭合开关,最大限度减小LED的结温差对实验数据的影响。由于四路LED芯片之间最小的距离为19 mm左右,热耦合效应很微弱,对实验影响极小。

      整个实验的温度调节是通过恒温加热平台V-0505实现的,尽管将温度从30°~120°进行了等分,但温度随时间的变化并非是均匀的,因此需要考虑梯度温度的影响。该实验的温度仅受时间影响,所以在一维状态下,梯度温度可看成是温度的变化率。经过测量可知,温度从30°升到120°需要18 min,以1 min为间隔,通过红外线测温仪来测量实验平台的实时温度,绘制出温度随时间的变化曲线见图8

      Figure 7.  Straight-line fitting curve of illuminance E and temperature T

      Figure 8.  Temperature curve with time

      图8可以看出,加热过程中,温度的变化率逐渐衰减,12 min之后,温度几乎平稳。说明在刚开始时,光源的非视觉参数变化速率较快,随着温度的逐渐平稳,非视觉参数也增加到最大,趋于平稳。

      以方案1、5、8为例,两种非视觉参数随温度的变化见图9,可以看出:温度升高, $ {C}_{AF} $值逐渐增加,CS值逐渐减小。当从30 ℃上升到120 ℃时,方案1的$ {C}_{AF} $从0.2223增至0.2459,CS从0.3008减至0.2597;方案5的$ {C}_{AF} $由0.5783上升到0.6701,CS从0.3344减少到0.2998;方案8的$ {C}_{AF} $值由1.0192上升到1.0902,CS从0.3986减少到0.3657。

      Figure 9.  Variation curve of non-visual parameter with temperature for scheme 1, 5, 8

      随着温度变化,蓝光占比逐渐增加,非视觉效应增强,但$ {C}_{AF} $和CS的计算受照度影响不同,说明二者在描述非视觉效应时存在一些差异。为消除照度的影响,对光谱进行照度补偿,使照度恒定在300 lx,通过电流、温度模型计算当前光谱分布,补偿公式如下:

      式中:${D_{{r}}},{D_{{g}}},{D_{{b}}},{D_{{y}}}$为温度T下补偿占空比;${P_{{I,T}}}{\left( \lambda \right)}$${P_{{{{I}}_0},{{{T}}_0}}}{\left( \lambda \right) }$指电流I、温度T和初始电流I0;温度T0下光源的光谱; ${C_{{{{A}}{{F}}}_{{I,T}}}}$${C_{{{{A}}{{F}}}_{{{I}_0,{{T}}_0}}}}$为电流I、温度T和初始电流I0,温度T0下昼夜节律因子;${\text{CS}_{{I,T}}}$${\text{CS}_{{{{{I}}_0}},{{{T}_0}}}}$为电流I、温度T和初始电流${{{I}}_0}$,温度${{{T}}_0}$下昼夜节律刺激;$({x_{{I,T}}},{y_{{I,T}}})$$({x_{{{{I}}_0}{\text{,}}{{{T}}_0}}},{y_{{{{I}}_0},{{{T}}_0}}})$为电流I、温度T和初始电流I0,温度T0下的色坐标;$ {k_1},{k_2},{k_3},{k_4},{k_5} $为各项的权重。昼夜节律因子$ {C}_{AF} $和昼夜节律刺激CS作为两种衡量光的非视觉效应的变量,理应随温度的变化呈现相同的趋势。但在补偿前发现二者随温度的变化趋势截然相反,是因温度升高后有物理量的变化,导致二者趋势不同。经分析发现照度受温度的影响很大,于是决定用PWM调光补偿照度,其采用高频率的周期数字脉冲驱动光源,通过改变周期脉冲内有效电平脉宽的宽度来调整输出的有效电流,从而调节光源参数。PWM调光补偿的响应速度更快,光源控制精度更高,可控性更强,是照明调光领域中的优选方式。补偿后照度始终维持在300 lx,其余的光源参数尽可能地接近补偿前,从而得到合理的补偿方案。补偿后的非视觉参数随温度的变化趋势如图10所示。

      Figure 10.  Temperature-dependent variation curve of non-visual parameters of schemes 1, 5, and 8 after compensation

      补偿后,随着温度升高,CAF的增大趋势略微减缓,CS由下降变为上升。方案1的CAF从0.2223逐渐增加到0.2405,CS的值从0.3008增到0.3204;方案5的CAF从0.5783增加到0.6522,CS的值从0.3344增加到0.3637;方案8的CAF从0.5783逐渐增加到0.6522,CS的值从0.3344增加到0.3637。CAF和CS均随温度升高而增大,且二者表现出一定的正相关性。

    • 文中基于LED芯片发光材料中光子能级分布的规律,提出了光谱的电流、温度模型。通过该模型可以根据LED当前的电流、温度参数,模拟光源的实际光谱。在综合考虑非视觉参数和视觉参数的条件下,设计出8种符合照明要求的照明方案,并针对各个方案验证了电流、温度模型的可靠性。同时比较各方案下,昼夜节律因子和昼夜节律刺激随温度的变化趋势,发现昼夜节律因子随温度的升高而增大,但昼夜节律刺激随温度增加而减小,照度随温度的增大而线性衰减。补偿照度后,两种非视觉参数的值均随温度的升高而增大,且二者表现出一定的正相关性,这表明随着温度的升高,人体的非视觉效应的确在逐渐增强。该研究为LED照明光源设计中非视觉效应的考量提供了一定参考依据。

Reference (15)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return