A method for calculating spectral irradiance based on broadband optical measurement data

Abstract

The present invention relates to the technical field of optical radiation measurement, and discloses a method for calculating spectral irradiance based on broadband optical measurement data. First, energy ratio correction is adopted to correct the broadband data into narrowband data. The energy ratio is obtained from the standard lamp spectrum and the measured transmittance of the filter, without human intervention. Second, considering the difference in spectral transmittance, it is normalized and multiplied by a calibration coefficient to correct the influence caused by the difference in filter transmittance. Third, polarization correction is carried out to correct the additional transmittance generated by the polarization caused by the coating film, and it is incorporated into the environmental correction algorithm to eliminate the error generated by the polarization effect of the film. In addition, radiation heating correction is carried out. According to the change of radiation deviation with radiation intensity, a fitting curve is obtained to correct the data. The present invention accurately corrects the broadband optical measurement data for the case of using a filter for spectroscopy, improves the calculation accuracy of spectral irradiance, and achieves a high-precision measurement effect on the basis of a low-cost structure.

Description

Technical Field

[0001] The present invention relates to the technical field of optical radiation measurement, and particularly to a method for calculating spectral irradiance based on broadband optical measurement data. Background Art

[0002] Irradiance is an important optical parameter and is applied in many optical fields. In the field of atmospheric science, irradiance is mainly used to describe the radiation intensity of natural light, and its unit is the unit of energy flux, W m -2 , while spectral irradiance (also known as spectral irradiance) refers to the energy flux per unit wavelength, that is, the irradiance of a single spectral band, and its unit is mW m -2 nm -1 . The spectral irradiance referred to in the present invention is also the spectral irradiance defined conventionally in the art, and will not be elaborated hereinafter.

[0003] Natural light is directly or indirectly related to solar radiation and can generally be divided into direct light and scattered light. Direct light represents the radiation flux that directly reaches the instrument after being attenuated by the atmosphere, and scattered light is the downward radiation flux caused by the scattering of sunlight by atmospheric molecules and suspended particulate matter. The scattering caused by gas molecules in the atmosphere is called Rayleigh scattering, and its downward radiation flux appears as the blue color of the sky on sunny days. The scattering caused by larger particulate matter such as fog droplets and aerosols is described by Mie scattering. In addition to scattering, the action of atmospheric substances on light also includes absorption, that is, atmospheric substances absorb solar radiation energy and convert it into heat energy. The heat energy is then added to the atmospheric energy transfer in other forms of energy. The absorption of the atmosphere is not reflected in the irradiance, but is reflected in the attenuation of light in the atmosphere.

[0004] The instrument for measuring irradiance is an irradiance meter, which uses the photoelectric effect and the law of conservation of energy to obtain the signal of optical radiation intensity. The irradiance meter can measure spectral irradiance data of multi-spectral or hyper-spectral. The technology for measuring spectral irradiance has been very mature, and a variety of instruments can meet the measurement needs of spectral irradiance. The main difference in irradiance measurement technology lies in the spectral splitting method. Generally, a grating is required for spectral splitting in hyper-spectral instruments, while a filter can also be used for spectral splitting in instruments with fewer spectral bands.

[0005] The ideal narrow-bandwidth optical signal has a rectangular-shaped band-pass transmission, i.e., it is constant within the bandwidth of ±Δλ (unit: nm) and zero outside the bandwidth. The grating spectroscopy technique can output a rectangular signal, while although the filter is band-pass, the output signal is basically a peak-shaped signal, and there are significant differences between the signal intensity and bandwidth and the rectangular signal. The most important is the bandwidth signal. Measuring spectral irradiance requires a collector with a relatively narrow spectral width to ensure high measurement accuracy. The instrument bandwidth of grating spectroscopy can be achieved at ±3 nm, while the bandwidth of filter spectroscopy is much wider. Even with clear requirements for the bandwidth, the peak-shaped structure of its output signal still leads to the input of out-of-band signals and the broadening of the bandwidth. The wider the bandwidth of the filter, the greater the energy entering, and the higher the value of the output voltage. These deviations will be regarded as "signals" and are the main error sources of the instrument. Therefore, there will be significant errors in measuring irradiance with a wide-bandwidth filter, which is one of the unsolved problems.

[0006] Chinese Patent No. 201910302823.2 discloses a haze vertical structure detector, which belongs to a miniature irradiance meter. It is carried into the air by a radiosonde balloon to detect the vertical profile of atmospheric irradiance and invert the substance distribution in the atmosphere, constituting a fog visibility profiler (FVP). This patent mentions that "the method of using filters to obtain spectral data is adopted. The selection of the spectrum is divided into rough selection and fine selection. The rough selection is used to ensure effective coverage of the spectral range, with an even distribution in the range of 400 - 1000 nm. The fine selection is used to avoid the main spectral absorption bands. The optical sensor is located above the haze detector, and there are multiple filters above the optical sensor. Each filter obtains data by a phototube, and the number of phototubes is the same as the number of spectra. The data collected by the photovoltaic panel is collected, amplified, and output by a data amplification circuit. The filter is covered with a cosine diffuser above to achieve the function of uniform light". This kind of irradiance meter uses filters to provide band-pass data for each channel. Since this irradiance meter is a disposable instrument and cannot use components with high precision but high cost, it can only use filters with a relatively wide bandwidth and hopes to obtain the same measurement accuracy as a narrow-bandwidth instrument. However, it is difficult to accurately give the spectral irradiance value from the data obtained by using a wide-bandwidth collector, and sometimes the error can reach more than 1 time the actual value. Therefore, how to improve the spectral irradiance measurement accuracy of a wide-bandwidth collector using filter spectroscopy has become an urgent problem to be solved. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for calculating spectral irradiance based on wide-bandwidth optical measurement data. For an irradiance meter using filter spectroscopy, accurate correction is performed on the wide-bandwidth optical measurement data to improve the calculation accuracy of spectral irradiance and achieve a high-precision measurement effect on the basis of a low-cost structure.

[0008] To achieve the above object, the present invention provides the following solution:

[0009] A method for calculating spectral irradiance based on broadband optical measurement data, the method comprising the following steps:

[0010] (1) Energy ratio correction: Determine the energy ratio σ sd (λ n ) based on the standard lamp spectrum and the measured transmittance of the filter, where the energy ratio σ sd (λ n ) represents the proportion of energy in each spectral band, and the formula is expressed as:

[0011]

[0012] In the formula, E max is the maximum irradiance of the standard lamp in the visible spectral band, E sd (λ) is the irradiance of the standard lamp, γ(λ) is the measured transmittance of each spectral band of the filter, λ is the wavelength variable, λ n represents the nth specific central wavelength, and Δλ is the half bandwidth;

[0013] (2) Transmittance correction: Use Trios as the standard instrument to correct the transmittance of each spectral band and perform normalization, and the transmittance of each spectral band is normalized to η(λ n ); among them, the formula for correcting the transmittance of each spectral band is expressed as:

[0014]

[0015] In the formula, when Trios is at the set distance from the standard lamp, the measured irradiance is E γ0 (λ), and on this basis, adding the filter, cosine diffuser and coating, the measured irradiance is E(λ), γ film (λ) is the transmittance caused by the coating, γ cos (λ) is the transmittance caused by the cosine diffuser, γ filt (λ) is the transmittance caused by the filter;

[0016] (3) Polarization correction: Obtain the transmittance T(λ n ) caused by polarization according to the refractive index and transmittance of each spectral band of the coating material, and perform transmittance correction point by point. Among them, considering both direct light and scattered light, assume that T(λ n ) after correction consists of two parts, which is expressed as:

[0017]

[0018] In the formula, T max (λ n) is the maximum value of the transmittance caused by polarization; f0(λ n ) represents the influence of scattered light on irradiance; f1(λ n ) is the influence of direct light on irradiance, and the constraint condition between f0(λ n ) and f1(λ n ) is

[0019] f0(λ n ) + f1(λ n ) = 1 (15)

[0020] (4) Calculate the spectral irradiance: Based on the energy ratio σ sd (λ n ) and the transmittance η(λ n ), improve the calibration coefficient S(λ n ). Substitute the improved S(λ n ) and κ(λ n ) into the following formula to calculate the spectral irradiance, that is, the input energy E(λ n ) of the nth spectral band. The formula is expressed as:

[0021]

[0022] The improved S(λ n ) is used for the static calibration of the instrument, and κ(λ n ) is related to the solar zenith angle and belongs to environmental correction;

[0023] Among them, the improved calibration coefficient:

[0024]

[0025] In the formula, the output voltage corresponding to the nth spectral band is v(λ n ), and the output voltage v dark (λ n ) generated by the dark current of the instrument is measured when the incident radiation is zero; V sd (λ n ) is the broadband output voltage when using a filter for spectral measurement, and E sd (λ n ) is the irradiance of the standard lamp.

[0026] Furthermore, it also includes:

[0027] (5) Radiation heating correction: Considering the error caused by the heating of the instrument housing by solar radiation, calculate the spectral irradiance, that is, the input energy of the nth spectral band is E(λ n ) and is expressed as:

[0028]

[0029] Among them, the measured spectral irradiance affected by radiative heating is E0(λ n ), and the fitting coefficients a(λ n ), b(λ n ) and c(λ n ) are all functions of temperature.

[0030] Furthermore, in the radiative heating correction process of step (5), the derivation process of formula (20) is as follows:

[0031] Let the measured spectral irradiance affected by radiative heating be E0(λ n ), the irradiance change caused by temperature be ΔE0(λ n ), then the irradiance caused solely by the radiation intensity, E(λ n ), is:

[0032] E(λ n ) = E0(λ n ) - ΔE0(λ n ) (17)

[0033] According to the change curve of ΔE0(λ n ) with respect to E(λ n ), it is quadratic-fitted as:

[0034] ΔE0(λ n ) = a(λ n ) + b(λ n )E(λ n ) + c(λ n )E 2 (λ n ) (18)

[0035] Among them, E(λ n ) is the value provided by the comparison instrument Trios; the response of each spectral band to temperature is different, and a(λ n ), b(λ n ), c(λ n ) are related to the spectral band and are obtained by fitting the measured data;

[0036] Substituting equation (18) into equation (17), we get

[0037] E(λ n ) = E0(λ n ) - [a(λ n ) + b(λ n )E(λ n ) + c(λ n )E 2 (λ n )] (19)

[0038] In the formula, formula (20) is obtained by solving based on formula (19).

[0039] Further, in step (3), the transmittance T(λ n ) is a function related to the wavelength and is expressed as:

[0040]

[0041] where i1 and i2 are the incident angle and the refraction angle respectively, and are determined by the sine theorem:

[0042] n1sin i1 = n2sin i2 (10)

[0043] where n1 is the refractive index of the atmosphere and n2 is the refractive index of the film coating.

[0044] Further, in step (4), based on the energy ratio σ sd (λ n ) and the transmittance η(λ n ), the calibration coefficient S(λ n ) is improved, specifically including:

[0045] Based on the energy ratio and the transmittance, bandwidth correction and transmittance correction are performed on the wide-bandwidth voltage data, and the narrow-bandwidth voltage data obtained is:

[0046] v sd (λ n ) - v dark (λ n ) = [V sd (λ n ) - v dark (λ n )]σ sd (λ n )η(λ n ) (6)

[0047] In the formula, v sd (λ n ) is the narrow-bandwidth output voltage when measuring the standard lamp, V sd (λ n ) is the wide-bandwidth output voltage when measuring by means of filter spectroscopy, and formula (6) establishes the conversion relationship between the narrow-bandwidth output voltage and the wide-bandwidth output voltage;

[0048] The calculation formula of the calibration coefficient S(λ n ) before improvement is:

[0049]

[0050] In the formula, E sd (λ n) is the irradiance of the standard lamp;

[0051] Substitute formula (6) into the calculation formula (3) of the calibration coefficient S(λ before improvement n ) to obtain the improved calibration coefficient:

[0052]

[0053] The improved calibration coefficient of formula (8) increases the energy ratio and transmittance, and is used to convert broadband data into narrowband data.

[0054] Furthermore, in the step (4), the derivation process of formula (16) is as follows:

[0055] Use λ to represent the wavelength variable, and use λ n to represent the central wavelength of the nth spectral band, Δλ is the half bandwidth, the ideal narrowband optical signal is a rectangular wave, which is a constant within the bandwidth ±Δλ and zero outside the bandwidth. Let the measured input energy of the nth spectral band be E(λ n ), and the output voltage be v(λ n ). The relationship between the two is described by a linear function:

[0056] E(λ n ) = p(λ n )v(λ n ) + q(λ n ) (1)

[0057] where p(λ n ) and q(λ n ) are the slope and intercept of the linear fitting. When the incident radiation is zero, the output voltage v dark (λ n ) generated by the instrument dark current is measured, and equation (1) becomes:

[0058]

[0059] Since the transmittance caused by polarization is related to the incident angle and belongs to the environmental parameter, it cannot be given in the calibration coefficient and can only be corrected for the transmittance point by point, that is

[0060]

[0061] Correct T(λ n ) to κ(λ n ), and substitute the corrected S(λ n ) and κ(λ n ) into (13) to obtain formula (16).

[0062] According to the specific embodiments provided by the present invention, the method for calculating spectral irradiance based on broadband optical measurement data provided by the present invention discloses the following technical effects:

[0063] When developing a miniature irradiance meter that uses a filter for spectroscopy, due to the relatively wide bandwidth of the filter, the obtained irradiance error is very large. From an optical perspective, it is a difficult problem to invert the true physical quantity using the measurement results with a wide bandwidth. The present invention proposes a bandwidth correction algorithm, which realizes the accurate inversion of irradiance through the correction of multiple parameters. First, the energy ratio method is used for correction to correct the broadband data into narrowband data. The energy ratio is obtained from the standard lamp spectrum and the measured filter transmittance without any artificial intervention. Second, considering the difference in spectral transmittance, it is normalized and multiplied by a calibration coefficient to correct the influence caused by the difference in filter transmittance. The third step is called polarization correction, which corrects the additional transmittance caused by the polarization of the coating film and incorporates it into the environmental correction algorithm to eliminate the error caused by the polarization effect of the film. In addition, radiation heating correction can also be performed. According to the change of radiation deviation with radiation intensity, a fitting curve is obtained to correct the data. The results show that high-precision irradiance data can be obtained by using the method described in the present invention. Standardized tests show that after these corrections, the results are highly consistent with the standard instrument, achieving a good measurement effect, indicating that these correction steps constitute a reliable bandwidth correction algorithm. Therefore, when the present invention uses a filter for spectroscopy to save costs, it accurately corrects the broadband optical measurement data, improves the calculation accuracy of spectral irradiance, realizes a high-precision measurement effect on the basis of a low-cost structure, is more practical, and is applicable to practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0065] Figure 1 It is a cross-sectional view of the optical signal acquisition system components of a multi-spectral instrument that uses a filter for spectroscopy in an embodiment of the present invention. Among them, 1 is the coating film, 2 is the cosine diffuser, 3 is the filter, 4 is the optical sealing ring, 5 is the phototube, and 6 is the data acquisition circuit board. Figure 1 The cross-sections of all components are circular;

[0066] Figure 2Schematic diagram of the optical signal passing through the filter in the embodiment of the present invention. Among them, (a) is the transmittance of five spectral bands measured, (b) is the spectral distribution of the transmittance of the 535 nm filter, the green solid line is the transmission spectral curve of the filter, the black solid line is the position of the central spectrum, and the pink shadow represents the range of the set bandwidth;

[0067] Figure 3 Schematic diagram of the change of film transmittance with the solar altitude angle;

[0068] Figure 4 Schematic diagram of the selection result of empirically determining f1;

[0069] Figure 5 Error of irradiance inversion under strong radiation conditions;

[0070] Figure 6 For the cloudless weather on July 3, 2024, the variation of radiation deviation with radiation intensity and its fitting curves in five spectral bands of 427 nm wavelength, 474 nm wavelength, 535 nm wavelength, 606 nm wavelength, and 671 nm wavelength;

[0071] Figure 7 Schematic diagram of the functions played by various correction means;

[0072] Figure 8 Comparison chart of the FVP instrument (red line) and the standard instrument Trios (black line) in the standardized experiment at Baduan Mountain in Qingdao. Among them, the left column is the comparison chart for July 3, July 4, and July 5 corresponding to the 427 nm wavelength respectively; the right column is the comparison chart for July 3, July 4, and July 5 corresponding to the 606 nm wavelength;

[0073] Figure 9 Comparison chart of the synchronous observation results of the FVP instrument and the standard instrument Trios in the standardized experiment at Baduan Mountain in Qingdao. The left figure is the comparison chart of the results at the 427 nm wavelength, and the right figure is the comparison chart of the results at the 606 nm wavelength;

[0074] Figure 10 Average relative deviation chart of the FVP inversion data and Trios at each station. The numbers in the figure are the station numbers of each station;

[0075] Figure 11 Schematic diagram of the comparison between the FVP inversion result and the synchronous observation result of Trios;

[0076] Figure 12 Schematic diagram of the method flow for calculating spectral irradiance based on broadband optical measurement data in the embodiment of the present invention. Detailed implementation mode

[0077] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0078] The object of the present invention is to provide a method for calculating spectral irradiance based on broadband optical measurement data, so as to obtain accurate irradiance measurement results when developing a micro-irradiance meter using a filter for spectroscopy.

[0079] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0080] The measurement principle of the micro-irradiance meter is to utilize the photoelectric effect in physics, use a phototube as the data acquisition unit, and a front optical filter. The incident radiant energy is received by the acquisition unit, which changes the current of the phototube, and the generated output voltage is used as the output signal of the micro-irradiance meter. The optical signal acquisition system of the micro-irradiance meter is as Figure 1 shown: The system includes a filter 3 for obtaining light in a specific wavelength range; a cosine diffuser 2, also known as a Lambertian body, which converts light from various directions into cosine-scattered light; a film 1 plays a fixing role and can adjust the light transmittance; a phototube 5 and a data acquisition circuit board 6 carrying the phototube are used for collecting optical signals; the entire optical signal acquisition system achieves optical sealing through an optical sealing ring 4. The optical signal acquisition system with the filter as the main body has the characteristic of a relatively wide bandwidth, and it has a large error in obtaining irradiance data. Next, how to effectively correct the data to meet the needs of high-precision irradiance measurement will be studied.

[0081] As Figure 12 shown, the method for calculating spectral irradiance based on broadband optical measurement data provided by the present invention is as follows:

[0082] (1) Energy ratio correction: Determine the energy ratio σ sd (λ n ) based on the standard lamp spectrum and the measured transmittance of the filter;

[0083] (2) Transmittance correction: Use Trios as the standard instrument to correct the transmittance of each spectral band and perform normalization;

[0084] (3) Polarization correction: According to the refractive index and transmittance of each spectral band of the film material, obtain the transmittance T(λ n ) caused by polarization, and perform transmittance correction point by point. Among them, considering both direct light and scattered light, correct T(λ n) is κ(λ n );

[0085] (4) Calculate the spectral irradiance: Based on the energy ratio σ sd (λ n ) and the transmittance η(λ n ), improve the calibration coefficient S(λ n ). Substitute the improved S(λ n ) and κ(λ n ) into the following formula to calculate the spectral irradiance, that is, the input energy E(λ n ) of the nth spectral band. The formula is expressed as:

[0086]

[0087] Among them, the improved calibration coefficient:

[0088]

[0089] In the formula, the output voltage corresponding to the nth spectral band is v(λ n ), and the output voltage v dark (λ n ) generated by the instrument dark current is measured when the incident radiation is zero;

[0090] (5) Radiation heating correction: Considering the error caused by the heating of the instrument housing by solar radiation, calculate the spectral irradiance, that is, the input energy of the nth spectral band is E(λ n ) is expressed as:

[0091]

[0092] Among them, the measured spectral irradiance affected by radiation heating is E0(λ n ), and the fitting coefficients a(λ n ), b(λ n ) and c(λ n ) are all functions of temperature.

[0093] Example:

[0094] The present invention provides a method for calculating spectral irradiance based on broadband optical measurement data for the embodiment of the optical signal acquisition system of the micro-irradiance meter shown in Figure 1 . The specific steps are as follows:

[0095] (1) Standard calibration method of irradiance measurement instrument

[0096] For the sake of clarity, hereinafter, λ is used to represent the wavelength variable, and λ nrepresents the central wavelength of the nth spectral band, and Δλ is the half-bandwidth. The ideal narrow-bandwidth optical signal is a rectangular wave, which is a constant within the bandwidth of ±Δλ (unit: nm) and zero outside the bandwidth. Let the measured input energy of the nth spectral band be E(λ n )(unit: mW m -2 nm -1 ), and the output voltage be v(λ n )(unit: V). The relationship between the two is described by a linear function:

[0097] E(λ n ) = p(λ n )v(λ n ) + q(λ n ) (1)

[0098] where p(λ n ) and q(λ n ) are the slope and intercept of the linear fitting. When the incident radiation is zero, the output voltage v dark (λ n ) generated by the instrument dark current can be measured. Equation (1) becomes

[0099]

[0100] where S(λ n ) is the calibration coefficient, which needs to be obtained through optical calibration. In a darkroom, the instrument directly measures a standard lamp to obtain

[0101]

[0102] In the formula, E sd (λ n ) is the irradiance of the standard lamp, and v sd (λ n ) is the output voltage when measuring the standard lamp. As long as S(λ n ) is determined, the accurate spectral irradiance value can be calculated using Equation (2).

[0103] However, Equation (2) requires meeting the narrow-bandwidth requirement, that is, the narrower the half-bandwidth Δλ, the better. For example, the bandwidth of the optical instrument Trios using the grating spectroscopy principle is 3 nm, and the accurate irradiance value can be obtained using Equation (2). If the bandwidth is very large, the method of Equation (2) cannot be directly used because the measured output voltage contains the energy obtained within a larger bandwidth range, and the obtained irradiance is also the average value within a larger range, which cannot reflect the correct spectral irradiance. Therefore, obtaining irradiance using wide-bandwidth data is a difficult problem in optical measurement.

[0104] Our research shows that to obtain accurate irradiance with broadband data, the problems to be solved are to accurately understand the optical characteristics of the instrument and the signal, and to correct the output voltage signal of the broadband to convert it into an accurate irradiance value.

[0105] (2) Bandwidth and transmittance of the filter

[0106] The optical signal passing through the filter is very different from the rectangular peak and shows a peak-like structure, as Figure 2 shown in (a) below.

[0107] As Figure 2 shown in (b) below, taking the transmission spectral characteristics of the 535nm spectral band filter as an example, the transmittance distribution of the transmission spectrum is given. The measured voltage V sd (λ n ) actually reflects the result of integrating the incident radiation energy over the entire spectral range. The pink shaded area is the range of ±10nm, and the signal within this range is the signal we need to measure. It can be seen from the figure that outside the set bandwidth, there is still some light that can pass through the filter and enter the sensor. Therefore, it is necessary to accurately extract the narrow-bandwidth voltage data, v sd (λ n ) and v(λ sd ) and v(λ n ) from the broadband voltage data V n ) in order to calculate the irradiance using equations (2) and (3).

[0108] Considering that the irradiance of the standard lamp represents the radiation energy and is the known energy flux, the proportion σ sd (λ) of the energy in each spectral band can be calculated.

[0109]

[0110] In the formula, E max is the maximum value of the irradiance of the standard lamp in the visible spectral band. Equation (4) is equivalent to the weighted average and normalization of the radiation energy in each spectral band, and the weight function γ(λ) is the change of the transmittance with wavelength. Taking Δλ = 10nm to calculate σ sd (λ n ). The red light of the standard lamp is very strong, and the energy obtained within the set bandwidth is close to 80%, while the energy obtained by the violet light in the same bandwidth only accounts for 17%. In order to make the σ sd (λ n ) obtained from the standard lamp satisfy the measurement of natural light, it is necessary to ensure that the calculation result of equation (4) is as accurate as possible.

[0111] A careful study of equation (4) shows that σ sdActually reflects the distribution characteristics of the standard light spectrum, that is, the variation of spectral irradiance with wavelength, which is basically the average value of the spectrum of the standard lamp within the set bandwidth range. It is necessary to consider the actual transmittance to reflect what proportion of this energy actually enters the instrument. The spectral characteristics reflecting the transmittance are represented by the spectral transmittance γ(λ n ) Expression. In an ideal situation, the transmittance of each spectral band should be the same; however, due to inevitable deviations during the manufacturing process of the filter, the actual transmittances of each spectral band are not consistent. For example, the transmittance at 427nm is significantly less than that of other spectral bands, as shown in Figure 2 (a). Therefore, it is necessary to correct using the actual transmittance. Normalize the transmittance of each spectral band to

[0112]

[0113] In the formula, γ max is the maximum value of the transmittances of all spectral bands.

[0114] Perform bandwidth correction and transmittance correction on the broadband voltage data to obtain the narrowband voltage data as

[0115] v sd (λ n ) - v dark (λ n ) = [V sd (λ n ) - v dark (λ n )]σ sd (λ n )η(λ n ) (6)

[0116] (3) Determination of spectral transmittance

[0117] According to equation (4), the prerequisite for accurately calculating σ sd (λ n ) is to know the spectral distribution γ(λ) of the transmittance. Measuring the spectral distribution requires a hyperspectral instrument. Here, Trios is used as the standard instrument. When Trios is 50 cm away from the standard lamp, the measured irradiance is E γ0 (λ). On this basis, adding a filter, a cosine diffuser, and a fixed film, the received irradiance is E(λ). The overall spectral transmittance of the system is:

[0118]

[0119] In the formula, γ film (λ) is the transmittance caused by the thin film, γ cos (λ) is the transmittance caused by the cosine diffuser, γ filt(λ) is the transmittance caused by the filter. Note that to measure E(λ), the filter, cosine diffuser, and fixing film need to be closely attached to the trios lens without reducing the measurement distance and achieving optical sealing. If a light-shielding lens is used, which changes the measurement distance, the algorithm for obtaining γ(λ) will be much more complex.

[0120] As Figure 2 As shown in (a) of [], each filter does not produce an ideal rectangular transmittance. Instead, it has a peak-shaped transmittance centered on the central spectrum, decreasing on both sides, and there is also a certain degree of asymmetry. Therefore, regardless of the nominal spectral width of the filter, it is necessary to correct the voltage data to obtain the irradiance.

[0121] (4) Determination of the calibration coefficient

[0122] Substituting Equation (6) into Equation (3), we get:

[0123]

[0124] The physical meaning of Equation (8) is to obtain the energy within the narrow bandwidth by adjusting the calibration coefficient, so as to give an accurate irradiance value.

[0125] We expect to obtain an accurate irradiance using Equation (8) as the calibration coefficient. However, the actual results show that this method only has good results when the sunlight is close to perpendicular incidence, while there will be deviations when the sunlight is obliquely incident, and sometimes the deviations are large.

[0126] (5) Influence of the film on the transmittance

[0127] Since there is a film on the surface of the instrument to fix the cosine diffuser and the filter. It is this film that changes the light energy entering the instrument, bringing signal problems. Research shows that it is the effect of the film that changes the solar radiation flux entering the instrument, resulting in deviations in the measurement results during oblique incidence. In fact, the film can be not used to eliminate this problem, but because the film has the function of improving the light transmittance, therefore, the film is retained.

[0128] The film is an optical medium, and reflection and transmission will occur at its upper and lower interfaces. The influence on light has three aspects: (a) attenuation of the transmitted light; (b) spectral frequency shift, and (c) change in the transmittance of obliquely incident light. Among them, Equation (7) already includes the influence of the film on the transmittance; there is still an air layer between the film and the cosine diffuser, so the influence of frequency shift can be not considered. What needs to be solved is the influence of the film on obliquely incident light.

[0129] Reflection and refraction of light at the film interface will produce polarized light. According to the optical polarization theory, the transmittance considering polarization is a function related to the wavelength

[0130]

[0131] Among them, i1 and i2 are the incident angle and the refraction angle respectively, which are determined by the sine theorem:

[0132] n1sin i1 = n2sin i2 (10)

[0133] Among them, n1 is the refractive index of the atmosphere, and n2 is the refractive index of the thin film. According to the measurement standard of the refractive index, the nominal refractive index of the material is the refractive index of sodium yellow light. For oblique incidence, the refractive indices of different lights are different, so chromatic dispersion will occur. Actually, the refractive indices of lights of each wavelength are different, and they will be higher or lower than the nominal refractive index. Generally speaking, the longer the wavelength, the smaller the refractive index. The relationship between the refractive index and the wavelength is

[0134]

[0135] Assume that the wavelength λ0 of sodium yellow light is equal to 589.3 nm, and the nominal refractive index n0 of the medium is 1.73, then there is

[0136]

[0137] That is, the refractive index is inversely proportional to the wavelength. The results obtained by using equation (12) are shown in Table 1. It can be seen from the table that the actual refractive index of short waves is higher than the nominal refractive index, and the actual refractive index of long waves is lower than the nominal refractive index.

[0138] Table 1 Relationship between the transmittance and the refractive index of the thin film

[0139]

[0140] The transmittance calculated by using equation (11) is as Figure 3 shown. It can be seen from the figure that when the incident angle is within 40°, the difference is very small, but after 40°, the transmitted radiation decreases significantly. When the incident angle is greater than the Brewster angle (about equal to 62.2°), the transmittance of light decreases rapidly. This is the reason for the large error generated during oblique incidence.

[0141] The difference in refractive index caused by polarization leads to the difference in transmittance. Taking the light with an incident angle of 46° as an example, the transmittance of long waves is the largest, and the transmittance of short waves decreases. Under the condition of oblique incidence, the thin film weakens the transmitted light, making the energy received by the instrument less than the actual value. It is necessary to divide by the transmittance when calculating the irradiance to offset the influence of the thin film. Since the transmittance is related to the incident angle and belongs to the environmental parameter, it cannot be given in the calibration coefficient, and only the transmittance correction can be carried out point by point, that is

[0142]

[0143] The physical meaning of this correction is that the actually received voltage signal is affected by the thin film and weakened, which will lead to the inversion irradiance being lower than the actual value. The transmittance T(λ n ) represents the proportion of the weakened optical signal, so it will correct the voltage signal to obtain the accurate irradiance.

[0144] However, there is a problem with equation (13), that is, it is very accurate when the solar zenith angle is small, but when the solar zenith angle is close to horizontal, the error increases rapidly and even distortion occurs. We analyzed the reasons for the problem as follows: The transmittance given by equation (13) is only suitable for direct sunlight. In the case of thick fog with no direct sunlight at all, the instrument will still receive scattered light from the fog. Therefore, in the inversion algorithm, not only direct sunlight but also scattered light needs to be considered.

[0145] To reflect this feature, let T(λ n ) consist of two parts,

[0146]

[0147] where, T max (λ n ) is the maximum value of the transmittance; f0(λ n ) represents the influence of scattered light on the irradiance; f1 is the influence of direct sunlight, and the constraint condition between f0 and f1 is,

[0148] f0(λ n ) + f1(λ n ) = 1, (15)

[0149] That is, only one value of f0 and f1 needs to be determined. When the zenith angle is the smallest, the transmittance is the largest, and κ(λ n ) is equal to the maximum transmittance T max (λ n ), and has a good correction effect. When the zenith angle is very large, the transmittance is very small, and κ(λ n ) is equivalent to the influence of scattered light and tends to f0(λ n )T max (λ n ).

[0150] In fact, there is no available theory for the selection of f0 and f1, nor is there any mature experience. We can only determine it empirically, that is, when inverting all the comparison measurement data, determine the f1 with the smallest variance as the calibration coefficient. Figure 4It shows that the minimum value of the relative deviation occurs when f0:f1 is equal to 0.5:0.5. The physical meaning expressed by the ratio of the scattered light to the direct light given by equation (14) is that although the ratio of the direct light to the scattered light is constantly changing, a ratio can be selected for the entire observation process, and under this ratio, the relative deviation of the irradiance inversion is the smallest.

[0151] Therefore, when the zenith angle of the sun during measurement is known, equation (15) is adjusted to, that is

[0152]

[0153] The improved S(λ n ) is used for the static calibration of the instrument, and κ(λ n ) is related to the solar zenith angle and belongs to environmental correction.

[0154] This result shows that for the data obtained with a broadband filter, the standard calibration process can only obtain the irradiance of the vertically incident sunlight; once the sunlight is obliquely incident, further correction is required to eliminate the additional effects caused by the polarization generated by the thin film.

[0155] (6) Errors under strong radiation conditions

[0156] After the above corrections, the obtained data is in good agreement with the results of Trios. When the temperature is not very high, without considering the errors caused by the radiation intensity, the precise measurement can be achieved by using the aforementioned three corrections: energy ratio correction, transmittance correction, and polarization correction. However, the errors are still very large under strong radiation conditions ( Figure 5 blue curve), and further radiation heating correction is required. Our research shows that the errors under such strong radiation conditions are caused by the heating of the instrument housing by solar radiation. The plastic housing of the instrument will absorb solar radiation energy under sunlight irradiation, and the increase in the housing temperature will generate additional thermal radiation, which affects the measurement results. The influence of solar radiation can be weakened in two ways: one is to coat the outer surface of the instrument with a reflective material to weaken the solar energy entering the instrument; the other is to add a metal heat transfer layer on the outer surface to transfer the heat out as soon as possible so that it cannot enter the interior of the instrument. Since the reflected light generated by the reflective material will affect the scattered light intensity of the particulate matter in the overlying atmosphere, resulting in signal drift or noise, this method cannot be used. Installing a metal housing for the instrument is an effective method, and some precision optical instruments use metal housings. However, using a thick metal housing is heavy and not suitable for profiling detection; while using a thin metal housing cannot completely eliminate radiation heating.

[0157] We studied the influence of radiation heating, and the obtained results show that if the instrument housing does not use a reflective material or a high thermal conductivity material and is allowed to be heated by solar radiation, the temperature of the instrument will rise quickly, but it will no longer rise after reaching an equilibrium temperature.Figure 6 Using the data on July 3, 2024, the variation of radiation deviation with radiation intensity under cloudless weather conditions is presented. It can be clearly seen from the figure that under direct solar radiation, the radiation deviations in each spectral band are closely related to the radiation intensity, which gives us a hint that the error caused by instrument heating can be corrected through its relationship with the radiation intensity.

[0158] Let the measured spectral irradiance affected by radiation heating be E0(λ n ), and the irradiance change caused by temperature be ΔE0(λ n ), then the irradiance solely caused by radiation intensity is E(λ n ),

[0159] E(λ n ) = E0(λ n ) - ΔE0(λ n ) (17)

[0160] According to the variation curve of ΔE0(λ n ) with E(λ n ), its quadratic fitting is:

[0161] ΔE0(λ n ) = a(λ n ) + b(λ n )E(λ n ) + c(λ n )E 2 (λ n ) (18)

[0162] Among them, E(λ n ) is the value provided by the comparison instrument Trios. The response of each spectral band to temperature is different. Therefore, a(λ n ), b(λ n ), c(λ n ) are related to the spectral band and need to be obtained by fitting from the measured data. The fitting results are as Figure 6 shown by the red line, which can obviously well reflect the deviation caused by radiation heating.

[0163] Substituting equation (18) into equation (17), we get:

[0164] E(λ n ) = E0(λ n ) - [a(λ n ) + b(λ n )E(λ n ) + c(λ n )E 2 (λ n )] (19)

[0165] In the formula, E(λ n ) is the actually observed value, and solving for

[0166]

[0167] (Equation 20) is the final corrected result, which well eliminates the influence of radiation ( Figure 5 red line).

[0168] It should be noted that the influence of solar radiation on irradiance is not only related to the radiation intensity, but also related to the air temperature. According to the laws of thermodynamics, there is an equilibrium between radiation heating and heat conduction on the surface of the instrument. Therefore, the fitting coefficients a(λ n ), b(λ n ), and c(λ n ) are all functions of temperature.

[0169] Although the instrument temperature will change under the conditions of solar radiation heating and external air temperature, essentially these correction coefficients are still quantities related to the instrument structure, depending on the material, thickness of the instrument shell, and its response to heating. The fitting coefficients related to the instrument material can be obtained through on-site tests under various weather conditions to achieve the correction of instrument radiation heating.

[0170] (7) Functions of various corrections

[0171] The present invention proposes various correction algorithms to solve the problem of obtaining accurate irradiance by broadband sensors. Taking the 34th station of the Qianliyan Island observation as an example, the functions played by various correction methods are as Figure 7 shown. The measurement time of the 34th station is April 28, 2024, at 8:01 in the morning, the solar zenith angle is 57.35°, and thick fog appeared during the observation period, which has good representativeness for its correction.

[0172] In Figure 7 , the curve "standard" is the correction result obtained by standard calibration according to Equation (3), indicating that there is a quite large difference between the broadband data and the results of the comparison instrument, and the difference in irradiance is greater for shorter wavelengths. The curve "standard+band" is the result after adding the bandwidth correction σ sd , which converts the broadband data into narrowband data, and the result has been substantially improved, but the error in the short-wave part is still large. The curve "standard+band+trans" adds the correction of transmittance as shown in Equation (8), which greatly improves the transmittance of the spectral band with originally low transmittance. The curve "Total" means "standard+band+trans+film", adding the correction of the film on the basis of the existing corrections, that is, reflecting the additional influence of polarization on transmittance. Since Figure 7It represents the observation at 8:00 in the morning. At that time, the temperature was relatively low and the radiation intensity was also low, so no radiation heating correction was required.

[0173] After these corrections, the results showed a high degree of consistency with the standard comparison instrument. The retrieved irradiance was very close to that of the standard instrument, achieving an ideal measurement effect and becoming a reliable algorithm.

[0174] (8) Verification of the bandwidth correction algorithm

[0175] To prove the reliability of the algorithm, a standardization test was carried out in the field. Under natural conditions, for different weather conditions and different solar zenith angle situations, the FVP instrument using the method described in the present invention was bundled with the newly calibrated standard instrument for synchronous observation to determine the accuracy of the retrieved irradiance.

[0176] The verification test was conducted at the Badguanshan Meteorological Observatory of Ocean University of China from July 3 to 5, 2024. Among them, July 3 was a clear sky condition, and the following two days were cloudy. To test the adaptability of the algorithm to various solar zenith angles, the daily test was carried out from noon to 18:00 in the afternoon, and the measurement results satisfying the minimum and maximum zenith angles were given. Since the observation duration of the FVP instrument was within 1 hour, the instrument needed to be restarted after the observation ended. The instrument was observed more than 5 times a day, and each interruption was about 10 minutes.

[0177] Using the data with spectral irradiance greater than 100 mW m -2 nm -1 on July 3 to determine the fitting coefficients a(λ n ), b(λ n ), and c(λ n ) for the calibration of all test results. The obtained results are as Figure 8 shown. Due to space limitations, Figure 8 only the results at 427 nm and 606 nm are given. The results of other spectral bands are similarly close. As can be seen from the figure, the retrieved results are highly consistent with those of the standard instrument, and both the overall change trend and the change details caused by clouds match well. This result indicates that the bandwidth correction algorithm proposed in the present invention meets the needs of data correction in terms of accuracy.

[0178] The comparison diagram between FVP and Trios can be seen in Figure 9 It can be seen that whether it is strong radiation or weak radiation, the obtained irradiance is highly consistent with the observation results of Trios, achieving a satisfactory effect.

[0179] (9) Application examples of the bandwidth correction algorithm

[0180] We conducted a sea fog observation at the Qianliyan Island Marine Station in April 2024. A total of 38 balloons were released, and 3 thick fog profiles were successfully observed. At the same time, several atmospheric profiles without fog were also observed. Meanwhile, Trios observations were carried out. The FVP calculates irradiance using a bandwidth correction algorithm. Before the balloon release for some time, the measurement results of the FVP can be compared with the observation results of Trios to conduct an application experiment on the spectral irradiance inversion algorithm.

[0181] Under field conditions, it is impossible to use the true value for verification, and only a comparison measurement can be carried out. The field experiment is different from the standardized test, and there are some uncertain factors: (1) Limited by the narrow terrain and sunlight conditions on the island, the placement position of Trios is not exactly the same as the release position of the FVP, which affects the comparability between the two. (2) Since there are obvious spatial differences in irradiance when there is thick fog, the deviation may very well represent the actual situation, and it may not necessarily be a problem with the instrument algorithm. (3) During the comparison measurement, Trios was placed under unobstructed conditions, while there was a balloon hanging above the FVP instrument, which had a certain impact on the reception of direct sunlight.

[0182] The measurement results of the FVP are very close to the synchronous measurement results of Trios. The root mean square deviation of the 5 spectral bands is 67 mW m - 2 nm -1 ,and the absolute deviation is only 4.24 mW m -2 nm -1 ,which is an ideal result.

[0183] A total of 38 sets of data were obtained from the observations at Qianliyan. The relative deviations of the remaining Trios data are as shown in Figure 10 . It can be seen from the figure that there are 4 stations with relatively large relative errors: 2, 11, 29, and 30. At Station No. 2, it was 5 am and the solar altitude angle was too low. At Station No. 11, the position of the FVP was relatively low, at the upper edge of the fog, while the position of Trios was relatively high, and the two were not comparable. The inversion result at Station No. 29 was actually very good, but there was a problem with the phototube in the 4th spectral band, resulting in a very low irradiance. Station No. 30 was the only station measured in the rain. Since there is a lack of necessary research and tests on the influence of rain on irradiance, its inversion result is lower than that of Trios. Excluding the data of these 4 stations, there are still 34 stations of data available. Excluding these 4 stations, the relative deviations of the remaining stations are shown in Figure 10 . It can be seen that the average relative deviation of all stations is below 15%. Among them, the average relative deviation of the inversion at 28 stations is below 10%, meeting the practical requirements.

[0184] The results of the comparison between irradiance and Trios are as shown in Figure 11 . It can be seen that in the case of fog, the irradiance is generally lower than 800 mW m -2 nm-1 , all the measured values are near the diagonal, indicating that the algorithm of the present invention can give accurate inversion results. In the case of large irradiance, it belongs to the fog-free weather, and direct light dominates. The relative deviation is below 0.18, but the inversion results of Station 37 and Station 19 have relatively large deviations, which are related to the position differences of the two instruments.

[0185] When developing a micro-irradiance meter with filter spectroscopy, we hope to obtain accurate irradiance measurement results. However, due to the wide bandwidth of the filter and its peak-shaped light transmission structure, the acquired data contains energy in a relatively large wavelength range, making it impossible to accurately invert the irradiance. From an optical perspective, it is a difficult problem to invert the true physical quantity using the measurement results with a wide bandwidth. The present invention proposes a bandwidth correction algorithm, and through four-step correction, accurate inversion of the irradiance is achieved.

[0186] The first-step correction is called energy ratio correction. First, σ sd (λ n ) is introduced by the energy ratio method to correct the wide-bandwidth data. The physical meaning of σ sd (λ n ) is the proportion of the received light energy within the bandwidth of ±Δλ. Through this coefficient, the data in the wide spectral band is successfully converted into the measurement results in the narrow spectral band. Another important role of σ sd (λ n ) is to eliminate the significant difference between the standard lamp spectrum and the solar spectrum, and establish the applicability of the calibration coefficient to the solar spectrum. σ sd (λ n ) reflects the mutual influence between the standard lamp spectrum and the filter transmittance, thus establishing a common measurement system for light and the filter. This relationship has been proven to be applicable under various natural light conditions, and the difference in light illumination under foggy and fog-free conditions can be ignored. The unique feature of σ sd (λ n ) is that it is completely obtained from the standard lamp spectrum and the measured filter transmittance without any artificial intervention.

[0187] The second-step correction is called transmittance correction. The difference in the filter transmittance γ(λ n ) is mainly caused by the coating thickness control process during production. Since there are certain differences in the coating materials for different wavelengths, it is difficult to achieve the same transmittance. Therefore, we normalize γ(λ n ) and multiply it by the calibration coefficient, which actually corrects the influence caused by the difference in the filter transmittance and makes the measurement results in each spectral band more comparable.

[0188] The third step of calibration is called polarization calibration. The film covering the instrument acts as a polarizer for the reflected and transmitted light, generating polarized light. When the incident angle is large, the transmittance of the light passing through the film changes, causing significant errors and requiring additional correction to obtain accurate irradiance. This patent application proposes to obtain the transmittance caused by polarization based on the refractive index and transmittance of the film material in each spectral band and incorporate it into the calibration algorithm, eliminating the errors caused by the polarization effect of the film.

[0189] The fourth step is called radiative heating calibration. When the solar radiation is very strong, it directly heats the housing of the instrument, causing the temperature of the instrument housing to rise. Since the photoelectric tube that receives the optical signal receives additional energy due to the increase in the instrument temperature, measurement errors are caused. Our research shows that there are two characteristics of the radiative heating of the instrument housing: First, the outer shell of the instrument is very thin and heats up quickly; Second, the instrument quickly reaches the equilibrium temperature. These two characteristics enable us to obtain a fitting curve based on the change of the radiation deviation with the radiation intensity and correct the data. The results show that high-precision irradiance data can be obtained by using this correction method.

[0190] After these calibrations, the results are highly consistent with the standard instrument. From July 3rd to 5th, 2024, we conducted a standardized test on the effectiveness of the above algorithm at the Baduan Mountain Meteorological Observatory in Qingdao. The results show that whether there is fog or not, the obtained irradiance is highly consistent with the data of the standard instrument, achieving an ideal measurement effect, indicating that these correction steps constitute a reliable bandwidth correction algorithm.

[0191] We conducted a sea fog observation at the Qianliyan Island Marine Station in April 2024, releasing a total of 38 balloons. Before the balloons took off, the FVP data could be compared with the standard instrument Trios to conduct an application experiment on the spectral irradiance inversion algorithm. The results show that the data obtained by using the calculation method provided by the present invention is highly consistent with the comparison instrument.

[0192] Therefore, the research proves that there are large deviations in the broadband data obtained by the filter, and the algorithm provided by the present invention effectively corrects the broadband data to obtain accurate irradiance inversion results.

[0193] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for calculating spectral irradiance based on broadband optical measurement data as described above.

[0194] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0195] In this article, specific examples are used to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes.

[0196] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the industry should understand that the present invention is not limited by the above embodiments. The above embodiments and the description in the specification only illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for calculating spectral irradiance based on wide bandwidth optical measurement data, applied to a radiometer using filter spectroscopy, characterized in that: The following steps are involved: (1) Energy ratio correction: Determine the energy ratio σ based on the standard lamp spectrum and the measured filter transmittance sd (λ n ), the energy ratio σ sd (λ n ) represents the proportion of energy in each spectral band, and the formula is expressed as: In the formula, E max is the maximum irradiance of the standard lamp in the visible spectrum, E sd (λ) is the irradiance of the standard lamp, γ(λ) is the measured transmittance of each spectral band of the filter, λ is the wavelength variable, and λ n represents the nth specific central wavelength, Δλ is the half bandwidth; (2) Transmittance calibration: Using Trios as a standard instrument, calibrate the transmittance of each spectral band and normalize it. The transmittance of each spectral band is normalized to η(λ n );wherein, the transmittance of each spectral band is corrected, and the formula is expressed as: In the formula, when Trios is at a set distance from the standard lamp, the measured irradiance is E γ0 (λ), on this basis, add the filter, cosine scatterer and coating, and the measured irradiance is E(λ), γ film (λ) is the transmittance caused by the coating, γ cos (λ) is the transmittance caused by the cosine scatterer, γ filt (λ) is the transmittance caused by the filter; (3) Polarization correction: According to the refractive index and transmittance of each spectral band of the coating material, the transmittance T(λ) caused by polarization is obtained. n ), and the transmittance is corrected point by point. Considering both direct light and scattered light, let T(λ n ) after correction consists of two parts, expressed as: Where, T max (λ n ) is the maximum transmittance caused by polarization; f0(λ n ) represents the influence of scattered light on irradiance; f1(λ n ) is the effect of direct light on irradiance, f0(λ n ) and f1(λ n ) are: f0(λ n )+f1(λ n )=1 (15) (4) Calculation of spectral irradiance: Based on energy ratio σ sd (λ n ) and transmittance η(λ n ) Improved calibration coefficient S(λ n ), the improved S(λ n ) and κ(λ n ) into the following formula to calculate the spectral irradiance, that is, the input energy E(λ n ), the formula is: Among them, the improved calibration coefficient is: In the formula, the output voltage corresponding to the nth spectrum segment is v(λ n ), the output voltage v generated by the dark current of the instrument is measured when the incident radiation is zero dark (λ n );V sd (λ n ) is the wide bandwidth output voltage when using filter spectrometry, E sd (λ n ) is the irradiance of the standard lamp.

2. The method for calculating spectral irradiance based on wide-bandwidth optical measurement data according to claim 1, characterized in that: Also includes: (5) Radiation heating correction: Considering the error caused by solar radiation heating the instrument shell, the spectral irradiance is calculated, that is, the input energy of the nth spectral band is E(λ n ) is expressed as: Among them, the measured spectral irradiance affected by radiation heating is E0(λ n ), fitting coefficient a(λ n ), b(λ n ) and c(λ n ) are all functions of temperature.

3. The method for calculating spectral irradiance based on wide-bandwidth optical measurement data according to claim 2, characterized in that: In the radiation heating correction process of step (5), the derivation process of formula (20) is as follows: Assume that the measured spectral irradiance affected by radiation heating is E0(λ n ), the irradiance change caused by temperature is ΔE0(λ n ), then the irradiance caused by the radiation intensity alone is E(λ n )for: E(λ n )=E0(λ n )-ΔE0(λ n ) (17) According to ΔE0(λ n ) with E(λ n )’s changing curve, and its quadratic fitting is: ΔE0(λ n )=a(λ n )+b(λ n )E(λ n )+c(λ n )E 2 (l n ) (18) Among them, E(λ n ) is the value provided by the comparative instrument Trios; each spectrum segment has a different response to temperature, a(λ n ), b(λ n ), c(λ n ) is related to the spectrum and is obtained by fitting the measured data; Substituting (18) into (17), we obtain E(λ n )=E0(λ n )-[a(λ n )+b(λ n )E(λ n )+c(λ n )E 2 (l n )] (19) Wherein, formula (20) is obtained based on formula (19).

4. The method for calculating spectral irradiance based on wide-bandwidth optical measurement data according to claim 1, characterized in that: In the step (3), the polarized transmittance T(λ n ) is a function related to wavelength and is expressed as: Where i1 and i2 are the angle of incidence and angle of refraction, respectively, which are determined by the law of sines: n1sini1=n2sini2 (10) Where n1 is the refractive index of the atmosphere and n2 is the refractive index of the coating.

5. The method for calculating spectral irradiance based on wide-bandwidth optical measurement data according to claim 1, characterized in that: In step (4), based on the energy ratio σ sd (λ n ) and transmittance η(λ n ) Improved calibration coefficient S(λ n ), including: Based on the energy ratio and transmittance, the wide bandwidth voltage data is corrected for bandwidth and transmittance, and the narrow bandwidth voltage data is obtained as follows: v sd (l n )-v dark (l n )=[V sd (l n )-v dark (l n )]s sd (l n )η(λ n ) (6) In the formula, v sd (λ n ) is the narrow bandwidth output voltage when measuring standard lamps, V sd (λ n ) is the wide bandwidth output voltage when using filter spectrometry measurement, and formula (6) establishes the conversion relationship between the narrow bandwidth output voltage and the wide bandwidth output voltage; The calibration coefficient S(λ before improvement n ) is calculated as: In the formula, E sd (λ n ) is the irradiance of a standard lamp; Substitute formula (6) into the calibration coefficient S(λ n ) is calculated in formula (3), and the improved calibration coefficient is obtained: The improved calibration coefficients in formula (8) increase the energy ratio and transmittance and are used to convert wide bandwidth data into narrow bandwidth data.

6. The method for calculating spectral irradiance based on wide-bandwidth optical measurement data according to claim 1, characterized in that: In step (4), the derivation process of formula (16) is as follows: Let λ represent the wavelength variable, and λ n represents the central wavelength of the nth spectral band, Δλ is the half bandwidth, the ideal narrow bandwidth optical signal is a rectangular wave, which is constant within the bandwidth ±Δλ and zero outside the bandwidth. Let the measured input energy of the nth spectral band be E(λ n ), the output voltage is v(λ n ), the relationship between the two is described by a linear function: E(λ n )=p(λ n )v(λ n )+q(λ n ) (1) Among them, p(λ n ) and q(λ n ) are the slope and intercept of the linear fit. When the incident radiation is zero, the output voltage v generated by the dark current of the instrument is measured. dark (λ n ), (1) becomes: Since the transmittance caused by polarization is related to the incident angle and is an environmental parameter, it cannot be given in the calibration coefficient. The transmittance can only be corrected point by point, that is, T(λ n ) is corrected to κ(λ n ), the corrected S(λ n ) and κ(λ n ) into (13), we can get formula (16).

Images