CN105934020B - A kind of method of multi-colored led match spectrum and illumination - Google Patents

Abstract

The invention discloses a kind of multi-colored led match spectrum and the method for illumination.In the method, first, the spectrum samples data of assorted single LED are obtained by sampling and measuring；Then, treat match spectrum and illumination is changed to obtain corresponding target optical spectrum radiance distribution；Finally, the driving value of LED is obtained by differential evolution algorithm Iterative matching, while additionally provides corresponding spectrum and illumination matching evaluation index to evaluate matching result.The present invention solves existing for current light spectrum matching process insufficient, can obtain the matching of illumination while match spectrum, and can significantly reduce because the maintenance of LED mixing light source and renewal and caused by calibration cost, it is easy to operate, it is easy to implement.In addition, also solved in matching process LED light source because fever and caused by spectrum change problem, can meet practical application at the same time to spectrum and the matched demand of illumination.

Description

A kind of method of multi-colored led match spectrum and illumination

Technical field

The present invention relates to field of LED illumination, more particularly to a kind of method of multi-colored led match spectrum and illumination.

Background technology

LED light source just gradually penetrates into each application field by the advantage such as environmentally friendly, energy saving, long-lived, small, such as aobvious Show, general illumination, plant culture, biologic medical, photovoltaic and radiancy, luminosity and colorimetry analysis calibration etc., and wherein Majority application is both needed to light source will also have certain spectral distribution property, therefore essence while certain brightness or illumination is capable of providing Really provide and meet that the light source light spectrum of application-specific demand and illumination have very significant actual application value.

The narrow-band characteristic of LED light source spectrum and its abundant distribution of color provide important base to realize that light source light spectrum is adjustable Plinth.At present, the method for realizing Spectral matching by LED light source mainly has：1) target light to be matched is realized using least square method To compose minimum the error between match spectrum, this method is simple and practicable, but for some matching situations, what least square method obtained Negative value occurs in the matching ratio of part LED, it is clear that negative value can not be achieved with actual LED, and simply discard appearance The LED light source of negative value can cause matching error to increase again；2) alternative manner declined using gradient, this method is to iterative parameter More sensitivity is set, once setting is unreasonable, often results in the drawbacks of iteration does not restrain；3) genetic algorithm, this method are used Though it can be not easy to implement to avoid above two method shortcoming.

The spectral characteristic of LED light source can produce change with its fever, and existing method often assumes the Spectral Properties of LED light source Property is constant, in turn results in matching error.The method that in addition, there will be do not refer to yet match LED light spectrum while how The problem of handling illumination matching.

The content of the invention

In order to overcome the shortcomings of existing Spectral matching method, the present invention provides a kind of multi-colored led match spectrum and illumination Method.

The technical solution adopted by the present invention is as follows：

A kind of method of multi-colored led match spectrum and illumination, includes the following steps：

S1：LED light composes the acquisition of sample data；

S2：The processing of spectrum to be matched and illumination data, obtains target optical spectrum spoke corresponding with spectrum to be matched and illumination Penetrate Luminance Distribution；

S3：Matching is iterated using differential evolution algorithm, is obtained corresponding all kinds of with the distribution of target optical spectrum radiance The driving value of LED；

S4：To measuring the spectrum of acquisition under above-mentioned driving value and illumination carries out matching evaluation.

Preferably, the acquisition process of LED light spectrum sample data specifically includes following steps in the step S1：

S101：LED drive data sample is established with a fixed sample interval in the driving value dynamic range of all kinds of LED, is remembered For d_i,j, represent j-th of driving data sample of the i-th class LED, the driving data sample of all kinds of LED be expressed as d with row vector_i =[d_i,1,d_i,2,...,d_i,max], wherein, d_i,maxRepresent the maximum sampling driving value of the i-th class LED, and all kinds of LED maximums are adopted The row vector that sample driving value is formed is denoted as D_max=[d_1,max,d_2,max,...,d_n,max], wherein n is the species number of LED；

S102：By the driving data sample in step S101, corresponding classification list LED is lighted in driving successively respectively, and is waited Its stable luminescence；

S103：Standard hawk or standard white plate (hereafter referred to collectively as on-gauge plate) are directly vertically below placed in all kinds of single LED, Measured successively using spectral radiometer and obtain the spectral radiance distribution through on-gauge plate reflection when all kinds of single LED light stable Data, are denoted as S_i,j(λ), represents the i-th class list LED in driving value d_i,jThe spectral radiance lighted through on-gauge plate reflection after stablizing is bright Degree distribution, unit is W/ (srm²Nm), λ is visible wavelength.

Preferably, the processing of spectrum to be matched and illumination data specifically includes following steps in the step S2：

S201：Spectrum S to be matched is set_t(λ), and the spectral reflectance of combined standard plate, according to light measurement theoretical calculation S_t The spectrum S that (λ) is reflected through on-gauge plate_t0(λ), the spectral reflectance of on-gauge plate are obtained by spectrophotometer measurement；

S202：According to S in light measurement theoretical calculation step S201_t0(λ) corresponding brightness L_t0；

S203：Illumination E to be matched is set_t, unit lx, and combine the brightness L in step S202_t0, according to light measurement Theoretical calculation spectrum to be matched and the corresponding target optical spectrum radiance distribution S of illumination_ta(λ)。

Preferably, spectrum and illumination matching are iterated matching using differential evolution algorithm in the step S3, it has Body step is：

S301：The equivalent amount of all kinds of LED is calculated according to illuminance principle of stacking, N is denoted as with row vector_max=[N_max,1, N_max,2,…,N_max,n], wherein n is the species number of LED；

S302：It is d that all kinds of LED are calculated according to the following formula in sampling driving value_i,jMixed spectra radiance point during opening Cloth S_i,j,M(λ)：

S_i,j,M(λ)=N_max,iS_i,j(λ),

N in formula_max,iFor the equivalent amount of the i-th class LED in step S301, S_i,j(λ) is that single LED is being adopted in step S103 Sample driving value is d_i,jWhen measure spectral radiance distribution；

S303：Predict all kinds of LED in its driving value dynamic range during any driving value by cubic spline interpolation algorithm Spectral radiance distribution S_i,M(λ), i.e.,

S_i,M(λ)=spline (d_i,S_i(λ),D_i),

Spline represents to perform cubic spline interpolation, d in formula_iSample what driving value was formed for the i-th class LED in step S101 Row vector, S_i(λ) is the row that the i-th class LED is formed in the spectral radiance of the different sampling driving values of af at wavelength lambda in step S302 Vector, D_iFor any driving values of the i-th class LED in its driving value dynamic range；

S304：Wavelength matched scope [λ is set₁,λ₂], i.e., in [λ₁,λ₂] the interior progress Spectral matching of wave-length coverage；

S305：Spectral matching, i.e., is converted to the minimum value for solving following formula by objective function f (D)：

Wherein

D=[D₁,D₂,…,D_n]

D is row vector to be solved in formula, each element D in vector_iRepresent that the i-th class LED appoints in its driving value dynamic range One driving value, n be LED species number, S_M(λ) represents multi-colored led mixed spectra radiance distribution, S_i,M(λ) is the i-th class LED in its driving value dynamic range at any driving value spectral radiance distribution, can be obtained by step S303, it is necessary to It is noted that calculated in the Wavelength matched scope that the object function need to be set in step s 304；

S306：Initialization operation is performed, i.e., in 0~D_maxBetween random Np equally distributed candidate's solution vector sets of generation, note For D_k,1=[D_1,k,1,D_2,k,1,…,D_n,k,1], wherein, k represents k-th of candidate's solution vector, and k={ 1,2 ..., Np }, and Np takes Certainly in the species number of LED, Np=10n under normal circumstances, 1 represents initial candidate solution vector set, when will equally pass through the G times iteration Np candidate's solution vector set be denoted as D_k,G=[D_1,k,G,D_2,k,G,…,D_n,k,G]；

S307：According to classical differential evolution algorithm principle to Np candidate's solution vector D in step S306_k,GPerform successively Variation, intersection and selection operation, Np candidate's solution vector of (i.e. the G+1 times) iteration next time is obtained through this sequence of maneuvers D_k,G+1；

S308：Optimal candidate solution vector during the G times iteration is obtained according to the following formula：

F (D in formula_k,G+1) it is the corresponding target function value of k-th of candidate's solution vector, D_O,GFor the G times iteration when optimal wait Solution vector is selected, Min represents calculated minimum；

S309：Step S307~S308 is repeated, until meeting the stopping criterion for iteration of setting, i.e. maximum iteration G_max, and the optimal solution vector finally obtained is denoted as D_O,end, as match the driving value of all kinds of LED of acquisition；

S310：The driving value for all kinds of LED for matching acquisition in step S309 is input to corresponding control software to light LED, and measure the spectral radiance distribution S matched respectively by spectral radiometer and illumination photometer_tm(λ) and illumination E_m。

Preferably, spectrum and illumination matching evaluation specifically include following steps in the step S4：

S401：Using spectral assessment indexs pair such as spectrum root-mean-square error (RMS), spectrum simulation goodness (GFC), p parameters Spectral matching result is evaluated, and the definition of these three indexs is respectively

λ in formula₁And λ₂Wavelength matched scope in corresponding step S304, N evaluate the quantity of wavelength, S for participation_ta(λ) is Target optical spectrum in step S203, S_tm(λ) is the spectra measurement in step S310；

S402：Using illumination percentage error ε_EIllumination matching result is evaluated, ε_EDefinition be

E in formula_tFor the illumination to be matched in step S203, E_mFor the illumination photometry value in step S310.

The beneficial effects of the invention are as follows：The present invention solves currently realizes deficiency existing for Spectral matching by LED light source, The method only measures the spectroscopic data of different type list LED, without measuring mixed light when similar LED is opened at the same time Spectrum, can significantly reduce because LED mixing light source safeguard and renewal and caused by calibration cost.In addition, pass through three in matching process Secondary spline interpolation consider LED light source because its fever and caused by spectrum change problem.The method is easy to operate, easy to implement, And illumination matching is taken into account while Spectral matching is realized, it disclosure satisfy that matched to light source light spectrum and illumination in practical application want Ask.

Brief description of the drawings

Fig. 1 is target optical spectrum corresponding with CIE standard illuminants A and matching measure spectrum；

Fig. 2 is target optical spectrum corresponding with CIE standard illuminants D65 and matching measure spectrum.

Embodiment

The method of multi-colored led match spectrum and illumination provided by the invention, mainly includes the following steps：

S1：LED light composes the acquisition of sample data；

S2：The processing of spectrum to be matched and illumination data, obtains target optical spectrum spoke corresponding with spectrum to be matched and illumination Penetrate Luminance Distribution；

S3：Matching is iterated using differential evolution algorithm, is obtained corresponding all kinds of with the distribution of target optical spectrum radiance The driving value of LED；

S4：To measuring the spectrum of acquisition under above-mentioned driving value and illumination carries out matching evaluation.

Above-mentioned matching process is described in detail with reference to the accompanying drawings and examples, to more fully understand the reality of the present invention Matter.

The multi-color LED light source that the present embodiment uses is a large scale being made of more than 1,000 monochromatic LED of 17 types LED matrix, and brightness adjustment control is carried out to it by the digital dimming mode based on PWM.It should be noted that not office of the invention The LED type and dimming mode being limited to employed in embodiment, as long as being suitable for the present invention using LED light source match spectrum.

The acquisition process of 1.LED spectrum samples data specifically includes following steps：

S101：LED drive data is established using 8 as driving value sampling interval in the driving value dynamic range of all kinds of LED Sample, is denoted as d_i,j, j-th of driving data sample of the i-th class LED is represented, with row vector by the driving data sample of all kinds of LED It is expressed as d_i=[d_i,1,d_i,2,...,d_i,max], wherein, d_i,maxRepresent the maximum sampling driving value of the i-th class LED, driving value takes It is 0~255 to be worth scope, and the row vector that the maximum sampling driving value of all kinds of LED is formed is denoted as D_max=[d_1,max, d_2,max,...,d_n,max], wherein n=17；

S102：By the driving data sample in step S101, corresponding classification list LED is lighted in driving successively respectively, and is waited Its stable luminescence；

S103：X-Rite ColorChecker Passport standard hawks are directly vertically below placed in all kinds of single LED, Measured and obtained when all kinds of single LED light stable through standard gray successively using spectral radiometer Konica Minolta CS-2000 The spectral radiance distributed data of plate reflection, is denoted as S_i,j(λ), represents the i-th class list LED in driving value d_i,jLight after stablizing Spectral radiance through the reflection of standard hawk is distributed, and unit is W/ (srm²Nm), λ is visible wavelength, wavelength measurement Scope is 380~780nm, and wavelength measurement is at intervals of 1nm.

2. the processing of spectrum to be matched and illumination data specifically includes following steps：

S201：CIE standard illuminants A and D65 is chosen as target optical spectrum S to be matched_t(λ), then in conjunction with standard hawk Spectral reflectance, according to light measurement theoretical calculation S_tThe spectrum S that (λ) is reflected through standard hawk_t0(λ), the spectrum of standard hawk Reflectivity is measured by spectrophotometer GretagMacBeth Color-Eye 7000A to be obtained, it is necessary to explanation, the present invention The spectrum of both types is not limited to, as long as being applicable in the present invention using LED matching light source light spectrums；

S202：According to S in light measurement theoretical calculation step S201_t0(λ) corresponding brightness L_t0；

S203：Illumination E to be matched is set_t=500lx is, it is necessary to explanation, the invention is not limited in this brightness value, As long as illumination to be matched is applicable in the present invention within multi-colored led achieved illumination range, and combines in step S202 Brightness L_t0, according to light measurement theoretical calculation spectrum to be matched and the corresponding target optical spectrum radiance distribution S of illumination_ta(λ), I.e. with CIE the standard illuminants A and D65 of selection, corresponding target optical spectrum radiance distribution, result of calculation are shown in Fig. 1 (A_ respectively Target) and Fig. 2 (D65_ targets).

3. spectrum and illumination matching are iterated matching using differential evolution algorithm, it is concretely comprised the following steps：

S301：The equivalent amount of all kinds of LED is calculated according to illuminance principle of stacking, N is denoted as with row vector_max=[N_max,1, N_max,2,…,N_max,n], wherein n be LED species number, i.e. n=17；

S302：It is d that all kinds of LED are calculated according to the following formula in sampling driving value_i,jMixed spectra radiance point during opening Cloth S_i,j,M(λ)：

S_i,j,M(λ)=N_max,iS_i,j(λ),

N in formula_max,iFor the equivalent amount of the i-th class LED in step S301, S_i,j(λ) is that single LED is being adopted in step S103 Sample driving value is d_i,jWhen measure spectral radiance distribution；

S303：Light of all kinds of LED in its driving value dynamic range during any driving value is predicted by cubic spline interpolation Compose radiance distribution S_i,M(λ), i.e.,

S_i,M(λ)=spline (d_i,S_i(λ),D_i),

Spline represents to perform cubic spline interpolation, d in formula_iSample what driving value was formed for the i-th class LED in step S101 Row vector, S_i(λ) is the row that the i-th class LED is formed in the spectral radiance of the different sampling driving values of af at wavelength lambda in step S302 Vector, D_iFor any driving values of the i-th class LED in its driving value dynamic range；

S304：Wavelength matched scope [λ is set₁,λ₂], i.e., in [λ₁,λ₂] the interior progress Spectral matching of wave-length coverage, it is real herein Apply and λ is set in example₁=430nm, λ₂=640nm is, it is necessary to which explanation, Wavelength matched scope are not limited to this scope, depend on In the spatial distribution of the LED actually used；

S305：Spectral matching, i.e., is converted to the minimum value for solving following formula by objective function：

Wherein

D=[D₁,D₂,…,D_n]

D is row vector to be solved in formula, each element D in vector_iRepresent that the i-th class LED appoints in its driving value dynamic range One driving value, n be LED species number, i.e. n=17, S_M(λ) represents multi-colored led mixed spectra radiance distribution, S_i,M(λ) For spectral radiance distributions of the i-th class LED in its driving value dynamic range at any driving value, can be obtained by step S303 , it should be noted that λ₁=430nm, λ₂The Wavelength matched scope that=640nm, i.e. step S304 are set；

S306：Initialization operation is performed, i.e., in 0~D_maxBetween random Np equally distributed candidate's solution vector sets of generation, note For D_k,1=[D_1,k,1,D_2,k,1,…,D_17,k,1], wherein, k represents k-th of candidate's solution vector, and k={ 1,2 ..., Np }, Np= 170,1 represent initial candidate solution vector set, and 170 candidate's solution vector sets when passing through the G times iteration equally are denoted as D_k,G= [D_1,k,G,D_2,k,G,…,D_17,k,G]；

S307：According to classical differential evolution algorithm principle to 170 candidate's solution vector D in step S306_k,GPerform successively Variation, intersection and selection operation, 170 candidate's solution vectors of (i.e. the G+1 times) iteration next time are obtained through this sequence of maneuvers D_k,G+1；

S308：Optimal candidate solution vector during the G times iteration is obtained according to the following formula：

F (D in formula_k,G+1) it is the corresponding target function value of k-th of candidate's solution vector, D_O,GFor the G times iteration when optimal wait Solution vector is selected, Min represents calculated minimum；

S309：Step S307~S308 is repeated, until meeting the stopping criterion for iteration of setting, i.e. maximum iteration G_max, G herein_max=200, bigger matching effect can be set better its appropriate adjusting, iterations according to actual match situation, But matching efficiency can be reduced, and the optimal solution vector finally obtained is denoted as D_O,end, as match the driving of all kinds of LED of acquisition Value；

S310：The driving value for all kinds of LED that acquisition is matched in step S309 is input in the control software write and is lighted LED matrix, and measure CIE the standard illuminants A and D65 matched respectively by the digital illumination photometers of CS-2000 and XYI-III Spectral radiance distribution S_tm(λ) and illumination E_m, specific matching result is shown in Fig. 1 (A_ matchings) and Fig. 2 (D65_ matchings) respectively, Illumination photometry value under two kinds of matching situations is respectively 476lx and 476lx.

4. spectrum and illumination matching evaluation specifically include following steps：

S401：Using spectral assessment indexs pair such as spectrum root-mean-square error (RMS), spectrum simulation goodness (GFC), p parameters Spectral matching result is evaluated, and the definition of these three indexs is respectively

λ in formula₁And λ₂Wavelength matched scope in corresponding step S304, is respectively that 430nm and 640nm, N evaluate for participation The quantity of wavelength, i.e., 211, S_ta(λ) be step S203 in target optical spectrum, S_tm(λ) is the spectra measurement in step S310；

Match spectrum in target optical spectrum and step S310 that CIE standard illuminants A and D65 are corresponded in step S203 is surveyed Value substitutes into above-mentioned formula respectively, can obtain Spectral matching evaluation result, as shown in table 1.

By table 1, Fig. 1 and Fig. 2 as it can be seen that Spectral matching method provided by the invention obtains in the Wavelength matched scope of setting Obtained preferable matching result, but the matching result of CIE standard illuminants D65 is worse than CIE standard illuminants A, this mainly by In D65 in subwave strong point there are spike and trough, the LED matrix that embodiment uses in addition is about in 540~580nm wavelength models Be in enclosing broadband LED and short wavelength regions LED light Spectral structure also not enough uniformly, as long as therefore can be in visible wavelength in putting into practice In the range of obtain LED as equally distributed as possible, Spectral matching method provided by the invention is obtained with the matching essence of satisfaction Degree.

1 Spectral matching evaluation result of table

S402：Using illumination percentage error ε_EIllumination matching result is evaluated, ε_EDefinition be

E in formula_tFor the illumination to be matched in step S203, i.e. 500lx, E_mIt is for the illumination photometry value in step S310, i.e., equal For 476lx.Measurement illumination in two kinds of situations is substituted into above formula respectively, it is 4.8% to obtain illumination matching error, the illumination With error within the acceptable range.Illumination matching error is mostly derived from A and D65 on the whole in 540~580nm wave-length coverages Measure spectrum radiance is below target optical spectrum radiance, in view of embodiment use LED light spectrum especially 540~ Limitation in 580nm wave-length coverages, so illumination matching process provided by the invention is effective.

Claims (5)

1. a kind of method of multi-colored led match spectrum and illumination, it is characterised in that include the following steps：

S1：LED light composes the acquisition of sample data；

S2：The processing of spectrum to be matched and illumination data, it is bright to obtain target optical spectrum radiation corresponding with spectrum to be matched and illumination Degree distribution；

S3：Matching is iterated using differential evolution algorithm, obtains all kinds of LEDs' corresponding with the distribution of target optical spectrum radiance Driving value；

S4：To measuring the spectrum of acquisition under above-mentioned driving value and illumination carries out matching evaluation.

2. the method for multi-colored led match spectrum and illumination as claimed in claim 1, it is characterised in that LED in the step S1 The acquisition process of spectrum samples data specifically includes following steps：

S101：LED drive data sample is established with a fixed sample interval in the driving value dynamic range of all kinds of LED, is denoted as d_i,j, represent j-th of driving data sample of the i-th class LED, the driving data sample of all kinds of LED be expressed as d with row vector_i= [d_i,1,d_i,2,...,d_i,max], wherein, d_i,maxRepresent the maximum sampling driving value of the i-th class LED, and all kinds of LED maximums are sampled The row vector that driving value is formed is denoted as D_max=[d_1,max,d_2,max,...,d_n,max], wherein n is the species number of LED；

S102：By the driving data sample in step S101, corresponding classification list LED is lighted in driving successively respectively, and waits its hair Light is stablized；

S103：On-gauge plate is directly vertically below placed in all kinds of single LED, is measured successively using spectral radiometer and obtains all kinds of single The spectral radiance distributed data reflected when LED lights stable through on-gauge plate, is denoted as S_i,j(λ), represents that the i-th class list LED exists Driving value d_i,jLight the spectral radiance through on-gauge plate reflection after stablizing to be distributed, unit is W/ (srm²Nm), λ is can See optical wavelength.

3. the method for multi-colored led match spectrum and illumination as claimed in claim 2, it is characterised in that treated in the step S2 The processing of match spectrum and illumination data specifically includes following steps：

S201：Spectrum S to be matched is set_t(λ), and the spectral reflectance of combined standard plate, according to light measurement theoretical calculation S_t(λ) The spectrum S reflected through on-gauge plate_t0(λ), the spectral reflectance of on-gauge plate are obtained by spectrophotometer measurement；

S202：According to S in light measurement theoretical calculation step S201_t0(λ) corresponding brightness L_t0；

S203：Illumination E to be matched is set_t, unit lx, and combine the brightness L in step S202_t0, according to luminosity theory Calculate spectrum to be matched and the corresponding target optical spectrum radiance distribution S of illumination_ta(λ)。

4. the method for multi-colored led match spectrum and illumination as claimed in claim 3, it is characterised in that light in the step S3 Spectrum and illumination matching are iterated matching using differential evolution algorithm, it is concretely comprised the following steps：

S301：The equivalent amount of all kinds of LED is calculated according to illuminance principle of stacking, N is denoted as with row vector_max=[N_max,1, N_max,2,…,N_max,n], wherein n is the species number of LED；

S302：It is d that all kinds of LED are calculated according to the following formula in sampling driving value_i,jMixed spectra radiance distribution during opening S_i,j,M(λ)：

S_i,j,M(λ)=N_max,iS_i,j(λ),

N in formula_max,iFor the equivalent amount of the i-th class LED in step S301, S_i,j(λ) is that single LED is driven in sampling in step S103 Dynamic value is d_i,jWhen measure spectral radiance distribution；

S303：Light of all kinds of LED in its driving value dynamic range during any driving value is predicted by cubic spline interpolation algorithm Compose radiance distribution S_i,M(λ), i.e.,

S_i,M(λ)=spline (d_i,S_i(λ),D_i),

Spline represents to perform cubic spline interpolation, d in formula_iFor the i-th class LED in step S101 sample row that driving value forms to Amount, S_i(λ) be the i-th class LED is formed in the spectral radiance of the different sampling driving value of af at wavelength lambda in step S302 row to Amount, D_iFor any driving values of the i-th class LED in its driving value dynamic range；

S304：Wavelength matched scope [λ is set₁,λ₂], i.e., in [λ₁,λ₂] the interior progress Spectral matching of wave-length coverage；

S305：Spectral matching, i.e., is converted to the minimum value for solving following formula by objective function f (D)：

<mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>D</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <mo>|</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>a</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>S</mi> <mi>M</mi> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> <mo>|</mo> </mrow> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>a</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>,</mo> </mrow>

Wherein

<mrow> <mtable> <mtr> <mtd> <mrow> <mi>D</mi> <mo>=</mo> <mo>&amp;lsqb;</mo> <msub> <mi>D</mi> <mn>1</mn> </msub> <mo>,</mo> <msub> <mi>D</mi> <mn>2</mn> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>D</mi> <mi>n</mi> </msub> <mo>&amp;rsqb;</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>S</mi> <mi>M</mi> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>S</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>M</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> </mtable> <mo>,</mo> </mrow>

D is row vector to be solved in formula, each element D in vector_iRepresent the i-th class LED any drivings in its driving value dynamic range Value, n be LED species number, S_M(λ) represents multi-colored led mixed spectra radiance distribution, S_i,M(λ) is the i-th class LED at it Spectral radiance distribution in driving value dynamic range at any driving value, can be obtained by step S303；

S306：Initialization operation is performed, i.e., in 0~D_maxBetween random Np equally distributed candidate's solution vector sets of generation, be denoted as D_k,1=[D_1,k,1,D_2,k,1,…,D_n,k,1], wherein, k represents k-th of candidate's solution vector, and k={ 1,2 ..., Np }, and 1 represents just Beginning candidate's solution vector set, is equally denoted as D by Np candidate's solution vector set when passing through the G times iteration_k,G=[D_1,k,G, D_2,k,G,…,D_n,k,G]；

S307：According to classical differential evolution algorithm principle to Np candidate's solution vector D in step S306_k,GPerform successively variation, Intersection and selection operation, Np candidate's solution vector D of next iteration is obtained through this sequence of maneuvers_k,G+1；

S308：Optimal candidate solution vector during the G times iteration is obtained according to the following formula：

<mrow> <msub> <mi>D</mi> <mrow> <mi>O</mi> <mo>,</mo> <mi>G</mi> </mrow> </msub> <mo>=</mo> <munder> <mrow> <mi>M</mi> <mi>i</mi> <mi>n</mi> </mrow> <mi>D</mi> </munder> <mo>{</mo> <mi>f</mi> <mrow> <mo>(</mo> <msub> <mi>D</mi> <mrow> <mn>1</mn> <mo>,</mo> <mi>G</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>)</mo> </mrow> <mo>,</mo> <mi>f</mi> <mrow> <mo>(</mo> <msub> <mi>D</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>G</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>)</mo> </mrow> <mo>,</mo> <mo>...</mo> <mo>,</mo> <mi>f</mi> <mrow> <mo>(</mo> <msub> <mi>D</mi> <mrow> <mi>N</mi> <mi>p</mi> <mo>,</mo> <mi>G</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>)</mo> </mrow> <mo>}</mo> <mo>,</mo> </mrow>

F (D in formula_k,G+1) it is the corresponding target function value of k-th of candidate's solution vector, D_O,GFor the G times iteration when optimal candidate solution Vector, Min represent calculated minimum；

S309：Step S307~S308 is repeated, until meeting the stopping criterion for iteration of setting, i.e. maximum iteration G_max, And the optimal solution vector finally obtained is denoted as D_O,end, as match the driving value of all kinds of LED of acquisition；

S310：The driving value for all kinds of LED that acquisition is matched in step S309 is input to corresponding control software and lights LED, and Measure the spectral radiance distribution S matched respectively by spectral radiometer and illumination photometer_tm(λ) and illumination E_m。

5. the method for multi-colored led match spectrum and illumination as claimed in claim 4, it is characterised in that light in the step S4 Spectrum and illumination matching evaluation specifically include following steps：

S401：Using spectrum root-mean-square error RMS, spectrum simulation goodness GFC, p parameter spectral assessment index to Spectral matching knot Fruit is evaluated, and the definition of these three indexs is respectively

<mrow> <mi>R</mi> <mi>M</mi> <mi>S</mi> <mo>=</mo> <msqrt> <mfrac> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>a</mi> </mrow> </msub> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> <mo>-</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>m</mi> </mrow> </msub> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> <mi>N</mi> </mfrac> </msqrt> <mo>,</mo> </mrow>

<mrow> <mi>G</mi> <mi>F</mi> <mi>C</mi> <mo>=</mo> <mfrac> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <mrow> <mo>(</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>a</mi> </mrow> </msub> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>m</mi> </mrow> </msub> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> <mo>)</mo> </mrow> </mrow> <mrow> <msqrt> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>a</mi> </mrow> </msub> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> </msqrt> <msqrt> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>m</mi> </mrow> </msub> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> </msqrt> </mrow> </mfrac> <mo>,</mo> </mrow>

<mrow> <mi>p</mi> <mo>=</mo> <mfrac> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <mo>|</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>a</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>m</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> <mo>|</mo> </mrow> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>&amp;lambda;</mi> <mo>=</mo> <msub> <mi>&amp;lambda;</mi> <mn>1</mn> </msub> </mrow> <msub> <mi>&amp;lambda;</mi> <mn>2</mn> </msub> </munderover> <msub> <mi>S</mi> <mrow> <mi>t</mi> <mi>a</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>&amp;lambda;</mi> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>,</mo> </mrow>

λ in formula₁And λ₂Wavelength matched scope in corresponding step S304, N evaluate the quantity of wavelength, S for participation_ta(λ) is step Target optical spectrum in S203, S_tm(λ) is the spectra measurement in step S310；

S402：Using illumination percentage error ε_EIllumination matching result is evaluated, ε_EDefinition be

<mrow> <msub> <mi>&amp;epsiv;</mi> <mi>E</mi> </msub> <mo>=</mo> <mfrac> <mrow> <mo>|</mo> <mrow> <msub> <mi>E</mi> <mi>t</mi> </msub> <mo>-</mo> <msub> <mi>E</mi> <mi>m</mi> </msub> </mrow> <mo>|</mo> </mrow> <msub> <mi>E</mi> <mi>t</mi> </msub> </mfrac> <mo>&amp;times;</mo> <mn>100</mn> <mi>%</mi> <mo>,</mo> </mrow>

E in formula_tFor the illumination to be matched in step S203, E_mFor the illumination photometry value in step S310.