Izaskun Azca a e, Jose Julio Gu ié ez, Andoni Lazkano, Pu i Saiz, Koldo Redondo, Luís Albe o
Le u iondo, “Expe imen al s udy o he esponse o e icien ligh ing echnologies o complex
ol age luc ua ions”, In e na ional Jou nal o Elec ical Powe and Ene gy Sys ems, Volume
63, 2014, Pages 499-506, ISSN 0142-0615, h ps://doi.o g/10.1016/j.ijepes.2014.06.039
(h ps://www.sciencedi ec .com/science/a icle/pii/S0142061514003883)
Abs ac : The eplacemen o incandescen lamps wi h mo e ene gy-e icien ligh ing
echnologies has a di ec in luence on he way licke is measu ed. The In e na ional
Elec o echnical Commission (IEC) es ablished in he 61000-4-15 s anda d he unc ional
speci ica ions o a licke me e , aking a s anda d incandescen lamp’s esponse o ol age
luc ua ions as he e e ence. Du ing he pas en yea s, di e en wo ks ha e s udied he
sensi i i y o mode n lamps o analy ical ol age luc ua ions o low complexi y. F om hese
s udies, he mos widesp ead conclusion is ha mode n lamps a e less sensi i e o licke han
a e incandescen lamps. Based on hese esul s, in e na ional s anda diza ion o ganiza ions a e
cu en ly s udying wo di e en possibili ies o upda ing he licke assessmen p ocedu e:
adjus ing he IEC licke me e acco ding o a new less sensi i e e e ence lamp, o inc easing he
es ablished compa ibili y le els o ol age luc ua ions. This wo k p esen s o he i s ime a
sensi i i y analysis o a se o mode n lamps subjec ed o eal ol age signals ha a e mo e
complex han analy ical ol age luc ua ions. The ob ained esul s lead o he ollowing
conclusions: no all e icien lamps ha e a lowe sensi i i y o luc ua ions han do incandescen
lamps; he esponse o some lamps depends on he complexi y o he inpu ol age luc ua ion;
and he esponse o some lamps in eal scena ios, i.e., mo e complex ol age luc ua ions, does
no co ela e wi h hei esponse o simple ol age luc ua ions.
Keywo ds: E icien Ligh ing, Vol age Fluc ua ions, Flicke , Powe Quali y, Complexi y.
Expe imen al S udy o he Response o E icien Ligh ing
Technologies o Complex Vol age Fluc ua ions.
I. Azca a e∗,a, J.J. Gu ie eza, A. Lazkanoa, P. Saiza, K. Redondoa, L.A. Le u iondoa
A ilia ions and add esses:
aCommunica ions Enginee ing Depa men .
Uni e si y o he Basque Coun y UPV/EHU.
Alameda U quijo S/N
48013 Bilbao, Spain
Co esponding au ho :
∗
Izaskun Azca a e
email: izaskun.azka a [email protected]
Tel. : +34946018209
Fax. : +34946014259
Abs ac 1
The eplacemen o incandescen lamps wi h mo e ene gy-e icien ligh ing echnologies2
has a di ec in luence on he way licke is measu ed. The In e na ional Elec o echnical3
Commission (IEC) es ablished in he 61000-4-15 s anda d he unc ional speci ica ions o 4
a licke me e , aking a s anda d incandescen lamp’s esponse o ol age luc ua ions as5
he e e ence. Du ing he pas en yea s, di e en wo ks ha e s udied he sensi i i y o 6
mode n lamps o analy ical ol age luc ua ions o low complexi y. F om hese s udies,7
1
he mos widesp ead conclusion is ha mode n lamps a e less sensi i e o licke han a e8
incandescen lamps. Based on hese esul s, in e na ional s anda diza ion o ganiza ions a e9
cu en ly s udying wo di e en possibili ies o upda ing he licke assessmen p ocedu e:10
adjus ing he IEC licke me e acco ding o a new less sensi i e e e ence lamp, o inc easing11
he es ablished compa ibili y le els o ol age luc ua ions. This wo k p esen s o he i s 12
ime a sensi i i y analysis o a se o mode n lamps subjec ed o eal ol age signals ha 13
a e mo e complex han analy ical ol age luc ua ions. The ob ained esul s lead o he14
ollowing conclusions: no all e icien lamps ha e a lowe sensi i i y o luc ua ions han15
do incandescen lamps; he esponse o some lamps depends on he complexi y o he inpu 16
ol age luc ua ion; and he esponse o some lamps in eal scena ios, i.e., mo e complex17
ol age luc ua ions, does no co ela e wi h hei esponse o simple ol age luc ua ions.18
Keywo ds19
E icien Ligh ing, Vol age Fluc ua ions, Flicke , Powe Quali y, Complexi y.20
2
1. In oduc ion21
In e na ional egula ions ha e p omp ed he mass eplacemen o incandescen lamps22
wi h ene gy-e icien ligh ing echnologies. The e is a ising conce n o e he ela ionship23
be ween he la ge-scale in oduc ion o such lamps and he Powe Quali y [1–3]. In his24
sense, he way licke is measu ed now akes special ele ance [4]. Flicke is de ined as25
he i i a ion su e ed by humans when exposed o illuminance luc ua ions p oduced by26
changes in he supply ol age. The s anda d IEC 61000-4-15 [5], whose p edecesso was IEC27
868 [6], es ablishes he unc ional speci ica ions o a licke me e —a de ice ha objec i ely28
quan i ies he le el o i i a ion by using he sho - e m (10 min) and long- e m (2 hou s)29
licke se e i y, Ps and Pl , espec i ely. The speci ica ions and e e ence alues, de ailed in30
he s anda d om i s i s publica ion in 1986, we e based on he leading ligh ing echnology31
a ha ime: he 60 W incandescen lamp. Compa ibili y le els o ol age luc ua ions we e32
speci ied based on he assump ion ha a alue o Ps = 1 would lead o complain s om a 33
leas 50% o he people exposed o he ligh luc ua ion p oduced by an incandescen lamp.34
The gain cu e is adi ionally used o cha ac e ize he sensi i i y o lamps subjec ed o35
sinusoidal ol age luc ua ions. This me hod calcula es he ela ionship be ween he ela i e36
ampli udes o he illuminance and ol age luc ua ions o a gi en luc ua ion equency.37
Mode n lamps’ gain cu es a e gene ally lowe han hose o incandescen lamps, so new38
ligh ing echnologies a e conside ed signi ican ly less sensi i e o ol age luc ua ions [7, 8].39
Taking in o accoun he p og essi e eplacemen o incandescen lamps and assuming he40
lowe sensi i i y o ene gy-e icien ligh ing echnologies o ol age luc ua ions, in e na ional41
o ganiza ions o he s anda diza ion and imp o emen o elec ic powe sys ems a e cu -42
en ly s udying al e na i e licke measu emen p o ocols adap ed o cu en echnologies:43
o ins ance, adjus ing he IEC licke me e acco ding o a new, mode n e e ence lamp, o 44
inc easing he es ablished compa ibili y le els o ol age luc ua ions [9].45
The speci ica ions o he IEC licke me e use he gain cu e o an incandescen lamp46
as a linea model, combined wi h he human eye’s esponse o ligh luc ua ions. The IEC47
licke me e ’s beha io is hus always linea , i espec i e o he complexi y o ol age luc-48
3
ua ions. Howe e , i has no been demons a ed ha his mi o s he beha io o mode n49
ligh ing echnologies [10, 11]. The cu en wo k analyzes he beha io o mode n lamps wi h50
complex ol age supplies, such as ec angula analy ical luc ua ions and eal ol age sig-51
nals. The main objec i e was o s udy whe he hei gain cu es, which we e ob ained om52
simple sinusoidal luc ua ions, we e alid o cha ac e izing hei esponse o mo e complex53
luc ua ions, in which case he assumed insensi i i y would be con incingly demons a ed54
and he a o emen ioned p oposals would go ahead. Howe e , ou wo k e ealed ha no all55
mode n lamps show a consis en ela ionship be ween hei gain cu e and hei beha io 56
in eal scena ios. Mo eo e , he p esen wo k clea ly challenges he assumed insensi i i y o 57
new ligh ing echnologies. The esul s poin o unce ain y in he ou come o inc easing he58
licke compa ibili y le els, and di icul ies in adap ing he IEC licke me e o a less-sensi i e59
e e ence lamp.60
2. Expe imen al Se up61
This sec ion desc ibes he se o lamps unde es (LUTs) and he sys em used o supply62
he lamps and o eco d he illuminance signals.63
2.1. Se o LUTs64
The main cha ac e is ics o he selec ed LUTs a e gi en in Table 1. We selec ed a se 65
o comme cially a ailable lamps om di e en manu ac u e s and using di e en ligh ing66
echnologies, including halogen, linea luo escen (LFL), compac luo escen (CFL; using67
elec onic o elec omagne ic ballas ), and ligh -emi ing diode (LED) lamps. The s udy68
also included a CFL dimmable lamp because his ea u e in ol es addi ional b igh ness-69
con ol me hods ha also a ec he licke [12]. The LUTs had di e en ene gy e iciency70
a ings; hese we e based on Eu opean Union ene gy labeling, which uses classes om he71
mos e icien (A) o he leas e icien (G).72
2.2. Vol age Gene a ion and Illuminance Reco ding73
Fig. 1 depic s he expe imen al se up used o gene a e he supply ol age o he LUTs74
and o eco d hei illuminance signals. The se up could gene a e eal o analy ical ol age75
4
signals. The eal ol age signals we e p e iously eco ded a a sampling a e o 6400 Hz by an76
acquisi ion sys em based on an analog- o-digi al (A/D) con e e (Na ional Ins umen s (NI)77
USB-6281) wi h 18-bi esolu ion. The digi ized analy ical o eal signal was con e ed in o78
an analog signal by a digi al- o-analog (D/A) con e e (NI USB-6211) a a a e o 6400 Hz.79
The analog signal was ampli ied o 230 V by a 7500 K ohn-Hi e Ampli ie (75 W, om DC80
o 1 MHz) and a 120/230 V ans o me . The ou pu o he ans o me was supplied o he81
LUT, which was enclosed in a whi e box oge he wi h he ligh senso , and was connec ed82
o a luxme e (E4-X Hagne Digi al Luxme e ). This p o ided he illuminance signal, l( ),83
which was digi ized by ano he A/D con e e (NI USB-6211) a a a e o 6400 Hz wi h84
16-bi esolu ion, and hen inally s o ed.85
3. Response o Analy ical Vol age Fluc ua ions86
The esponses o lamps o ol age luc ua ions ha e adi ionally been s udied by means87
o hei gain cu es [8, 10, 13, 14]. A lamp is mo e sensi i e o ol age luc ua ions when88
i s gain cu e is highe . Each da a poin on his cu e ep esen s he gain ac o o a gi en89
equency o he ol age luc ua ion, m. This pa ame e assesses he ela ionship be ween90
he ela i e ampli udes o he illuminance and ol age luc ua ions, ∆L
Land ∆V
V, espec i ely:91
G( m) = ∆L/L
∆V/V .(1)
The supply ol age o he LUT, when subjec ed o analy ical ol age luc ua ions, can92
be exp essed as ollows:93
u( ) = √2A1 + gm( )cos(ωo ),(2)
whe e ωo= 2π o ep esen s he equency o he mains supply, A ep esen s i s Roo 94
Main Squa e (RMS) alue, and gm( ) ep esen s he luc ua ion.95
5
Fo a sinusoidal ol age luc ua ion, gm( ) consis s o a single equency, m, o ela i e96
ampli ude ∆V
V. In his case, he illuminance luc ua ion also consis s o a single equency, m,97
o ela i e ampli ude ∆L
L. Fig. 2a depic s he gain cu e o each LUT, GLUT, no malized o98
he alues co esponding o he e e ence incandescen lamp, GI1. Each LUT was subjec ed99
o 32 sinusoidal luc ua ions wi h m alues anging om 1 o 32 Hz, wi h ∆V
V= 1%,100
acco ding o (2). The esul s e eal ha he sensi i i y o he halogen lamp (H1) is qui e101
close o ha o I1 o e he whole equency ange, while he es o he LUTs show educed102
sensi i i y ( ela i e o I1) o e a wide equency ange, om app oxima ely 3 o 25 Hz.103
3.1. Sensi i i y analysis o ec angula luc ua ions104
Fo ec angula ol age luc ua ions, which a e cha ac e ized by an unlimi ed bandwid h,105
i is no possible o calcula e ∆L
L o some o he LUTs. Fig. 3 depic s he wa e o ms o 106
he illuminance luc ua ions o he I1 and F1 lamps o a ec angula ol age luc ua ion107
wi h m= 8 Hz and ∆V
V= 1.5%. In he case o I1 (Fig. 3a), he illuminance en elope s ill108
consis s o a p edominan equency luc ua ion ha allows he calcula ion o he ela i e109
ampli ude, ∆L. Howe e , in he case o F1 (Fig. 3b), he complexi y o he wa e o m does110
no allow he di ec iden i ica ion o ∆Lo he accu a e calcula ion o i s gain ac o .111
An al e na i e me hod o s udying he esponses o he lamps o complex ol age luc-112
ua ions should conside all he equency componen s o he illuminance luc ua ion ha 113
a ec he licke [15]. Thus, i would conside he eal cha ac e is ics o he LUT and p o ide114
a close app oxima ion o he eal i i a ion i p oduces. Assessmen o he licke se e i y115
by means o an illuminance licke me e mee s hese equi emen s.116
The IEC 61000-4-15 s anda d de ines he design speci ica ions o a licke me e based117
on he ol age supply, acco ding o a physiological model o he lamp–eye–b ain chain.118
Some o he i e blocks o he model a e based on he pe o mance cha ac e is ics o he119
60 W incandescen lamp. An illuminance licke me e would be based on he illuminance120
a he han he ol age signal, bu should o he wise be based on he IEC licke me e , wi h121
modi ica ion o he blocks in he IEC model ha use he incandescen lamp’s esponse o122
ol age luc ua ions as a e e ence. An illuminance licke me e only includes ou blocks123
6
(Fig. 4) because he quad a ic demodula ion (Block 2 o he IEC s anda d) is elimina ed.124
Mo eo e , he weigh ing il e in Block 3 o he IEC s anda d, co esponding o he lamp–eye125
esponse, is modi ied so ha he equency cha ac e is ics o he human eye a e only used126
in Block B. The cu en wo k used a highly accu a e implemen a ion o an illuminance127
licke me e ha was desc ibed in de ail and alida ed in [16].128
The acquisi ion sys em depic ed in Fig. 1 egis e ed he illuminance o each LUT sub-129
jec ed o ec angula ol age luc ua ions, gene a ed acco ding o (2). Each LUT was sub-130
jec ed o 32 luc ua ions wi h m alues om 1 o 32 Hz, co esponding o ∆V
V alues ha 131
p oduced one uni o licke se e i y o an incandescen lamp, i.e., Ps ,I1 = 1. The eco ded132
illuminance signals we e p ocessed by he illuminance licke me e , and he co esponding133
licke se e i y alues, Ps ,LUT, we e ob ained.134
Fig. 2b depic s hese Ps ,LUT alues. The esul s e eal h ee di e en beha io s. Fi s ,135
he C1 and L1 lamps exhibi low sensi i i ies, clea ly below ha o he incandescen lamp,136
wi h esul s compa able o hei co esponding gain cu es gene a ed based on sinusoidal137
ol age luc ua ions (Fig. 2a). Second, he sensi i i y o lamp H1 is close o, o e en highe 138
han, ha o he incandescen lamp, wi h esul s compa able o he gain cu e. Thi d,139
he es o he LUTs show di e en esponses o sinusoidal and ec angula luc ua ions.140
The esponse o C2 o ec angula luc ua ions is highe han he gain cu e o he whole141
equency ange and shows mo e sensi i i y han I1 om 18 Hz onwa d. The esponse o 142
F1 o ec angula luc ua ions is simila o he gain cu e up o 25 Hz; om his equency143
onwa d, he sensi i i y o F1 is qui e close o ha o he incandescen lamp. The esponse144
o C3 is also qui e simila o he gain cu e up o 25 Hz, bu om his equency onwa d i 145
shows g ea e sensi i i y han I1.146
The analysis showed ha he beha io o some LUTs was di e en when supplied wi h147
ec angula e sus sinusoidal ol age luc ua ions. The highe complexi y o he ec angula 148
luc ua ions p oduced licke ing beha io in some LUTs ha di e ed om hei expec ed149
esponses acco ding o hei gain cu es. Hence, i is necessa y o analyze he esponses o 150
he LUTs when hey a e supplied wi h eal signals, i.e., wi h luc ua ions ha a e p esumably151
mo e complex because o hei lack o epe i i e cha ac e is ics.152
7
4. Beha io in Real Scena ios153
We analyzed he esponse o each LUT o ol age signals egis e ed a ou di e en 154
loca ions o he Low Vol age (LV) ne wo k (230 V/50 Hz) in he no h o Spain. The155
loca ions we e selec ed based on ea u es such as he popula ion, ype o dis u bing loads,156
and le el o licke se e i y.157
4.1. Desc ip ion o he si es158
Fig. 5 depic s he e olu ion o he Ps and Pl alues o he eal ol age signals a each159
si e o e app oxima ely one week, as well as hei 99 h pe cen iles, assessed by means o 160
he IEC licke me e [5]. Addi ionally, he igu es include he pe cen age o ime, TTH, o 161
which bo h pa ame e s exceed he i i abili y h eshold (Ps = 1, Pl = 1). The Pl alues162
we e calcula ed using a ime in e al o wo hou s, i.e., 12 sho ime in e als. A Pl alue163
was calcula ed as he cubic a e age o 12 consecu i e Ps alues. Each new Pl alue was164
ob ained wi h he 11 mos ecen Ps alues used in he calcula ion o he p e ious Pl alue165
and he nex new Ps alue. Following his p ocedu e, one Pl alue o each Ps alue was166
ob ained. A b ie desc ip ion o each si e is de ailed nex .167
•Si e a (Sa): A me opoli an a ea o 900,000 inhabi an s wi h ele an indus ial ac i i y,168
including s eel mills wo king wi h a c u naces. This si e did no p esen excessi e169
licke se e i y le els: Ps ,99 = 1.21 and Pl ,99 = 0.93, being close o he i i abili y170
h eshold, wi h only 3.8% o he Ps alues exceeding his limi and all he Pl alues171
being below i (Fig. 5a).172
•Si e b (Sb): A ou is des ina ion o 100,000 inhabi an s ( esiden s and isi o s) du ing173
he holiday pe iod. This si e is loca ed a om big indus ial loads. In his case, he174
licke se e i y alues emained abo e he i i abili y h eshold: Ps ,99 = 1.74 and175
Pl ,99 = 1.49. The licke se e i y alues exceeded he limi 67% o he ime o Ps and176
86.6% o he ime o Pl (Fig. 5b).177
•Si e c (Sc): A small own o 15,000 inhabi an s loca ed in a s eel indus y a ea, p e-178
dominan ly wi h ins alla ions equipped wi h a c u naces. The du y cycle o hese179
8
p o ocol o new lamps.349
Acknowledgmen s350
This wo k ecei ed inancial suppo om he Go e nmen o Basque Coun y h ough351
he PhD s uden ship BFI-2012-315 and om he Uni e si y o he Basque Coun y UPV/EHU352
h ough he p ojec UFI11/16 and he PhD s uden ship PIF2011/169.353
The au ho s would like o hank Ligh Labs depa men o Philips Ligh ing (Eindho en)354
o p o iding us wi h lamp samples and he necessa y in o ma ion ega ding he wo king355
p inciples o hese lamps.356
Re e ences357
[1] M. L. Schenne, A. K. Zia ani, T. H. O meye , A no el adap i e licke measu emen echnique, In . J.358
Elec . Powe . Ene gy Sys . 33 (10) (2011) 1686–1694. doi:10.1016/j.ijepes.2011.08.008.359
[2] S. Uddin, H. Sha ee , A. Mohamed, Powe quali y pe o mance o ene gy-e icien low-wa age LED360
lamps, Measu emen 46 (10) (2013) 3783–3795. doi:10.1016/j.measu emen .2013.07.022.361
[3] A. Gil-de Cas o, S. K. R¨onnbe g, M. H. J. Bollen, A. Mo eno-Mu˜noz, S udy on ha monic emission o 362
domes ic equipmen combined wi h di e en ypes o ligh ing, In . J. Elec . Powe . Ene gy Sys . 55363
(2014) 116–127. doi:10.1016/j.ijepes.2013.09.001.364
[4] A. Hooshya , E. El-Saadany, De elopmen o a licke me e o measu e non-incandescen lamps licke ,365
IEEE T ans. Powe Deli . 28 (4) (2013) 2103–2115. doi:10.1109/TPWRD.2013.2267153.366
[5] IEC-61000-4-15ed2.0, Elec omagne ic Compa ibili y (EMC) Pa 4: Tes ing and measu emen ech-367
niques - Sec ion 15: Flicke me e unc ional and design speci ica ions (2010).368
[6] IEC-868, Flicke me e . Fun ional and design speci ica ions (1986).369
[7] M. Halpin e al., Re iew o licke objec i es o HV, MV and LV sys ems, CIGRE/CIRED WG C4.108370
(2009).371
[8] R. Cai e al., Flicke esponses o di e en lamp ypes, IET Gene . T ansm. Dis. 3 (9) (2009) 816–824.372
[9] Te ms o e e ence, Tech. ep., Re iew o LV and MV Compa ibili y Le els o Vol age Fluc ua ions373
CIGRE WG C4.111 (2010).374
[10] I. Azca a e e al., Sensi i i y o licke o dimmable and non-dimmable lamps, in: P oc. o I2MTC375
Con e ence, 2012, pp. 344 –347.376
[11] R. Cai, Flicke in e ac ion s udies and licke me e imp o emen , Ph.D. hesis, Facul y o Elec ical377
Enginee ing, Eindho en Uni e si y o Technology (2009).378
15
[12] J. Smi h, J. Speakes, M. Rashid, An o e iew o he mode n ligh dimme : design, ope a ion, and379
applica ion, in: P oc. o he 37 h NAPS Con e ence, 2005, pp. 299 – 303.380
[13] K. Chmielowiec, Flicke e ec o di e en ypes o ligh sou ces, in: P oc. o 11 h EPQU Con e ence,381
IEEE, 2011, pp. 1–6. doi:10.1109/EPQU.2011.6128852.382
[14] J. D apela, P. Toman, In e ha monic - Flicke cu es o lamps and compa ibili y le el o in e ha monic383
ol ages, in: Powe Tech, Lausanne, IEEE, 2007, pp. 1552–1557.384
[15] I. Azca a e, J. Gu ie ez, P. Saiz, A. Lazkano, L. Le u iondo, K. Redondo, Flicke cha ac e is ics385
o e icien ligh ing assessed by he IEC licke me e , Elec . Powe Sys . Res. 107 (2014) 21–27.386
doi:10.1016/j.eps .2013.09.005.387
[16] I. Azca a e e al., Type es ing o a highly accu a e illuminance licke me e , in: P oc. o 15 h ICHQP388
Con e ence, 2012, pp. 897 –903.389
[17] S. Kay, J. Ma ple, S.L., Spec um analysis–A mode n pe spec i e, P oceedings o he IEEE 69 (11)390
(1981) 1380–1419. doi:10.1109/PROC.1981.12184.391
[18] I. Rezek, S. Robe s, S ochas ic complexi y measu es o physiological signal analysis, IEEE T ans.392
Biomed. Eng 45 (9) (1998) 1186–1191. doi:10.1109/10.709563.393
[19] C.E. Shannon, A ma hema ical heo y o communica ion, Bell Sys . Tech. J. 27 (1948) 379–423 and394
623–656.395
[20] L. F a e , N. Wa son, Ligh licke sensi i i y o high e iciency compac luo escen lamps, in: P oc. o 396
AUPEC Con e ence, IEEE, 2007, pp. 1–6. doi:10.1109/AUPEC.2007.4548067.397
[21] B. He e nan, L. F a e , N. Wa son, LED eplacemen o luo escen ube ligh ing, in: P oc. o AUPEC398
Con e ence, IEEE, 2007, pp. 1–6. doi:10.1109/AUPEC.2007.4548064.399
[22] K. Taekhyun e al., LED lamp licke caused by in e ha monics, in: P oc. o IMTC Con e ence, IEEE,400
2008, pp. 1920–1925. doi:10.1109/IMTC.2008.4547361.401
[23] D. A l , M. S a k, C. Ebe lein, Examples o in e na ional licke equi emen s in high ol age ne wo ks402
and eal wo ld measu emen s, in: P oc. o 9 h EPQU Con e ence, 2007, pp. 1 –4.403
[24] J. Ruiz e al., A e iew o licke se e i y assessmen by he IEC licke me e , IEEE T ans. Ins um.404
Meas. 59 (8) (2010) 2037–2047.405
[25] McKinsey & Company Inc., Ligh ing he way: Pe spec i es on he global ligh ing ma ke , h p://406
img.ledsmagazine.com/pd /Ligh ing heWay.pd , [Online] (2011).407
[26] J. Slezing e al., A new simpli ied model o compac luo escen lamps in he scena io o sma g ids,408
in: P oc. o he 15 h ICHQP Con e ence, 2012, pp. 835 –841.409
16
Figu e Cap ions410
Figu e 1 Diag am o he ol age gene a ion and illuminance eco ding411
p ocesses.412
Figu e 2 LUT sensi i i y cu es. (a) Gain cu es o sinusoidal ol age413
luc ua ions and (b) Ps alues o ec angula ol age luc ua-414
ions.415
Figu e 3 Wa e o m o he illuminance luc ua ion o a ec angula ol -416
age luc ua ion. (a) I1 lamp and (b) F1 lamp.417
Figu e 4 Func ional diag am o he illuminance licke me e .418
Figu e 5 Time e olu ion o Ps and Pl alues o he eal ol age signals419
a each si e; (a) Sa, (b) Sb, (c) Scand (d) Sd.420
Figu e 6 Real ol age signal a Scsi e. (a) Tempo al e olu ion o he421
ol age luc ua ion and (b) powe spec al densi y by Welch422
es ima o .423
Figu e 7 Pl alues o each LUT a he selec ed si es. (a,b,c,d) Time424
e olu ion and (e, ,g,h) box-plo s, co esponding o Sa, Sb, Sc
425
and Sd.426
Table Ti les427
Table 1 Se o lamps unde es (LUTs).428
Table 2 Spec al En opy o he analyzed signals.429
Table 3 TS: pe cen age o Pl ,LUT alues exceeding he uni and simul a-430
neously he 90% o he Pl ,I1 alues.431
17
Vol age gene a ion
Illuminance eco ding
Analy ical
signal gene a o
Real Signals
Reco d. Rep od.
D/A con .
NI USB-6211
K ohn-Hi e
ampli ie
120/230V
ans o me
u( )
Hagne
Luxme e
Ligh
senso
A/D con .
NI USB-6211
l( )
Da a
s o ing
Figu e 1
Fluc ua ion F equency (Hz)
(b)
Ps , LUT
Fluc ua ion F equency (Hz)
(a)
GLUT
GI1
0 5 10 15 20 25 30
0 5 10 15 20 25 30
0.2
0.6
1
1.8
0.2
0.6
1
1.8
Figu e 2
Lux
Time (s)
(a)
∆L
0 0.10.20.30.40.5
1450
1480
L
1510
1540
Lux
Time (s)
(b)
0 0.10.20.30.40.5
1445
1455
L
1475
1485
Figu e 3
l( )
BLOCK A
INPUT
ILLUMINANCE
ADAPTER
BLOCK B
0.05 35 10
DEMODUL. AND EYE-RESPONSE FILTERS
BLOCK C
SQUARING
MULTIPLIER
+
SLIDING
MEAN FILTER
Pins
BLOCK D
STATISTICAL
EVALUATION
Ps
Figu e 4
Pl
Ps
Flicke Se e i y
Time (h)
(a)
TTH,Pl =0.0%
TTH,Ps =3.8%
Pl ,99 =0.93
Ps ,99 =1.21
Thu s.Wed.Tues.Mon.
Sun.
Sa .
F i.
16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h 16h
0
0.5
1
1.5
2
2.5
Pl
Ps
Flicke Se e i y
Time (h)
(b)
TTH,Pl =86.6%
TTH,Ps =67.0%
Pl ,99 =1.49
Ps ,99 =1.74
Sa .F i.Thu s.Wed.
Tues.
Mon.
Sun.
17h 01h 09h 17h 01h 09h 17h 01h 09h 17h 01h 09h 17h 01h 09h 17h 01h 09h 17h
0
0.5
1
1.5
2
2.5
Pl
Ps
Flicke Se e i y
Time (h)
(c)
TTH,Pl =68.6%
TTH,Ps =50.5%
Pl ,99 =1.63
Ps ,99 =2.06
Mon.Sun.Sa .F i.
Thu s.
Wed.
Tues.
16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h 16h
0
0.5
1
1.5
2
2.5
Pl
Ps
Flicke Se e i y
Time (h)
(d)
TTH,Pl =95.3%
TTH,Ps =77.2%
Pl ,99 =1.92
Ps ,99 =2.29
F i.Thu s.Wed.
Tues.
Mon.
Sun.
16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h 16h 00h 08h
0
0.5
1
1.5
2
2.5
Figu e 5
PSD (dB/Hz)
F equency (Hz)
(b)
Time (min)
(a)
RMS (V)
50 150 250 350
13579
−100
−50
0
227
228
229
Figu e 6
Time (h)
(a)
Pl
16h 00h 08h 16h 00h 08h 16h 00h 08h 16h
0
0.3
0.6
1
Pl
LUT
(e)
L1
F1C3C2C1H1I1
0
0.3
0.6
1
Time (h)
(b)
Pl
17h 01h 09h 17h 01h 09h 17h 01h 09h 17h
0
0.5
1
1.5
Pl
LUT
( )
L1
F1C3C2C1H1I1
0
0.5
1
1.5
Time (h)
(c)
Pl
16h 00h 08h 16h 00h 08h 16h 00h 08h 16h
0
0.7
1.4
2.1
Pl
LUT
(g)
L1
F1C3C2C1H1I1
0
0.7
1.4
2.1
Time (h)
(d)
Pl
16h 00h 08h 16h 00h 08h 16h 00h 08h 16h
0
0.7
1.4
2.1
Pl
LUT
(h)
L1
F1C3C2C1H1I1
0
0.7
1.4
2.1
Figu e 7
