Method for detecting light transmittance of perovskite photovoltaic glass for building curtain wall

Abstract

The present application belongs to the technical field of glass transmittance detection, and provides a method for detecting the transmittance of perovskite photovoltaic glass for building curtain wall. Firstly, the present application extracts the advantage interference oscillation wavelength interval by constructing a zero-value crossing interval spectrum, and then combines the central wavelength to define the relative deviation and construct an empirical expression of dispersion compensation factor, and further establishes an equivalent interference optical path difference calculation model to convert the mathematical period into the film layer structure parameters. Then, the idealized interference oscillation sine function model is reconstructed in reverse to analyze the degree of coincidence between the ideal interference oscillation component and the measured interference oscillation component, and the interference confidence is defined to enable the system to evaluate the input quality. Finally, the three-level adaptive decoupling algorithm is constructed based on the confidence to extract the intrinsic transmission spectrum and complete the transmittance detection. The present method can stably extract the intrinsic transmittance, has a small error in visible light transmittance calculation, and can effectively improve the transmittance detection precision and robustness.

Description

Technical Field

[0001] This invention relates to the field of biomarker detection technology, and in particular to a method for detecting the light transmittance of perovskite photovoltaic glass used in building curtain walls. Background Technology

[0002] Perovskite photovoltaic glass, as a novel building curtain wall material, combines photovoltaic power generation and building lighting functions. Its transmittance is a key indicator for evaluating building aesthetics and photoelectric conversion efficiency. Currently, transmittance testing mainly employs spectrophotometry, obtaining spectral curves by measuring the transmitted light intensity at different wavelengths. However, perovskite photovoltaic glass has a multilayer film structure, and incident light is reflected and coherently superimposed at each interface, resulting in periodic interference oscillations superimposed in the measured spectrum, severely interfering with the accurate extraction of intrinsic transmittance. Existing testing methods typically use fixed-parameter Fourier filtering or moving averages for smoothing.

[0003] However, existing detection algorithms have the following shortcomings when processing spectral data of perovskite photovoltaic glass: First, they simply treat interference oscillations as noise and use general filtering algorithms for smoothing. However, oscillations are generated by thin-film interference and have a clear physical periodicity, not random noise. Simple filtering will destroy the structural information they carry. Second, the filtering parameters are fixed and cannot adapt to the changes in interference intensity caused by differences in film thickness and refractive index of different samples. In particular, excessively large parameters can smooth oscillations, but they are prone to damaging subtle features such as intrinsic absorption edges. Insufficiently small parameters can preserve details, but the noise reduction effect is insufficient, and residual oscillations interfere with the accuracy of transmittance calculation. Third, there is a lack of evaluation mechanism for the quality of the data itself. When there are multiple interference aliasing in the sample or the measurement noise is too large, the fixed processing procedure is still mechanically executed, resulting in unreliable output results. Summary of the Invention

[0004] To address the above technical problems, this invention provides a method for detecting the light transmittance of perovskite photovoltaic glass for building curtain walls.

[0005] According to the present invention, a method for testing the transmittance of perovskite photovoltaic glass for building curtain walls is provided, the method comprising: Collect raw spectral data of perovskite photovoltaic glass samples; Based on the original spectral data, the measured interference oscillation components are extracted, a zero-value crossing interval spectrum is constructed, and the dominant interference oscillation wavelength interval is determined. Based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, and by introducing the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained. Then, combined with the dominant interference oscillation wavelength interval, the relative deviation is defined, an empirical expression for the dispersion compensation factor is constructed, and an equivalent interference optical path difference solution model suitable for perovskite photovoltaic glass is established. Using the calculated equivalent interference optical path difference as a priori physical constraint, the idealized interference oscillation sinusoidal function model is reconstructed in reverse to obtain the ideal interference oscillation component. Then, the degree of agreement between the ideal interference oscillation component and the measured interference oscillation component is analyzed, and the interference confidence level is defined. Based on the original spectral data, the extraction path of the intrinsic transmission spectrum is selected according to the interference confidence level to obtain the intrinsic transmission spectrum and complete the transmittance detection.

[0006] In some embodiments of the present invention, the measured interference oscillation components are extracted from the original spectral data, including: The original spectral data is preprocessed with low-pass smoothing, and then the upper and lower envelopes are extracted using cubic spline interpolation to obtain the preliminary intrinsic trend lines of the original spectral data. Based on the original spectral data and the preliminary intrinsic trend line, the measured interference oscillation components are obtained.

[0007] In some embodiments of the present invention, constructing a zero-value crossover interval spectrum and determining the dominant interferometric oscillation wavelength interval includes: The measured interferometric oscillation components are traversed, and the sign change of the measured interferometric oscillation components is detected by linear interpolation to obtain a sequence of zero-value crossing points; Based on the zero-value crossing point sequence, the interference oscillation wavelength interval between the zero-value crossing points is calculated to obtain the interference oscillation wavelength interval sequence, and then the preliminary interval spectrum is obtained. Based on the preliminary interval spectrum, the consistency of the interference oscillation wavelength interval sequence is analyzed to obtain the enhanced interval spectrum, and the interference oscillation wavelength interval corresponding to the main peak position of the enhanced interval spectrum is determined as the dominant interference oscillation wavelength interval.

[0008] In some embodiments of the present invention, based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, and by introducing the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained, including: Based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, the interference oscillation wavelength interval between adjacent interference maxima is expressed as: the ratio of the product of the wavelengths corresponding to adjacent interference maxima to the optical path difference of a single dominant interference. If the wavelengths corresponding to adjacent interference maxima are all equal to the center wavelength of the analysis band, then the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained as follows: the interference oscillation wavelength interval of adjacent interference maxima is approximately equal to the ratio of the square of the center wavelength to the optical path difference of the single dominant interference.

[0009] In some embodiments of the present invention, the relative deviation is defined, including: Obtain the reference theoretical interference optical path difference of the perovskite photovoltaic glass sample, and construct the deviation from the benchmark based on the relationship; The relative deviation is defined based on the dominant interference oscillation wavelength interval and the deviation from the reference.

[0010] In some embodiments of the present invention, an empirical expression for the dispersion compensation factor is constructed, including: Based on the stable linear correlation between the relative deviation, the center wavelength, and the typical dispersion slope of perovskite materials, an empirical expression for the dispersion compensation factor is constructed.

[0011] In some embodiments of the present invention, the calculated equivalent interference optical path difference is used as a priori physical constraint to reconstruct the idealized interference oscillation sinusoidal function model in reverse, thereby obtaining the ideal interference oscillation components, including: Within the high transmission band of weak dispersion, the measured interference oscillation components are fitted to determine the amplitude coefficient and initial phase; Based on the calculated equivalent interference optical path difference, combined with the amplitude coefficient and the initial phase, an idealized sinusoidal function model between the interference oscillation and the wavelength is constructed to obtain the ideal interference oscillation component in the high transmission band with weak dispersion.

[0012] In some embodiments of the present invention, fitting the measured interference oscillation components within a weakly dispersed, high-transmission band to determine the amplitude coefficient and initial phase includes: Several local windows are uniformly selected in the high transmission band with weak dispersion. Within each local window, constrained by the dominant interference oscillation wavelength interval, the local amplitude and local phase within the local window are fitted using the least squares method, and the fitting residual for each local window is recorded. Analyze the distribution of the fitting residuals of all the local windows and filter candidate windows; Calculate the statistical distribution of local phases within all candidate windows, remove outlier candidate windows, and use the median of local amplitudes and the statistical mean of local phases within the remaining high-confidence candidate windows as the amplitude coefficient and initial phase of the global model, respectively.

[0013] In some embodiments of the present invention, the degree of agreement between the ideal interference oscillation component and the measured interference oscillation component is analyzed, and the interference confidence level is defined, including: In the high transmission band with weak dispersion, the sum of absolute residuals between the ideal interference oscillation component and the measured interference oscillation component is calculated. Combined with the total signal energy of the measured interference oscillation component, the degree of agreement between the ideal interference oscillation component and the measured interference oscillation component is obtained, and the interference confidence level is defined.

[0014] In some embodiments of the present invention, for the original spectral data, an extraction path for the intrinsic transmission spectrum is selected based on the interference confidence level to obtain the intrinsic transmission spectrum, including: Preset a lower threshold for high confidence and an upper threshold for low confidence; When the interference confidence level is greater than or equal to the high confidence lower limit threshold, the physical subtraction method is used to extract the intrinsic transmission spectrum from the original spectral data to obtain the intrinsic transmission spectrum. When the interference confidence level is less than the lower threshold of high confidence level but greater than the upper threshold of low confidence level, a physically guided Fourier filter method is used to extract the intrinsic transmission spectrum from the original spectral data to obtain the intrinsic transmission spectrum. When the interference confidence level is less than or equal to the low confidence upper limit threshold, the preliminary intrinsic trend line is output as a rough estimate of the intrinsic transmission spectrum and marked.

[0015] As can be seen from the above embodiments, the method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls provided by the present invention has the following beneficial effects: This invention first extracts measured interference oscillation components from the original spectral data, constructs a zero-value crossing interval spectrum, and determines the dominant interference oscillation wavelength interval, elevating the interference oscillation signal from the realm of random noise to an analytically deterministic physical characteristic, laying a data foundation for subsequent physical decoupling. Then, based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, the center wavelength of the analysis band is introduced. Combined with the dominant interference oscillation wavelength interval, the relative deviation is defined, and an empirical expression for the dispersion compensation factor is constructed. This leads to the establishment of an equivalent interference optical path difference solution model suitable for perovskite photovoltaic glass, i.e., establishing the equivalent optical path difference solution by introducing the dispersion compensation factor. The algorithm calculates the mathematical cycle and converts it into film structure parameters with clear physical meaning, enabling it to "recognize" the structural fingerprint of the sample. Then, using the calculated equivalent interference optical path difference as a priori physical constraint, it reconstructs the idealized interference oscillation sine function model to obtain the ideal interference oscillation components. The algorithm then analyzes the degree of agreement between the ideal and measured interference oscillation components, defines the interference confidence level, and quantifies the matching degree between the model and the data, enabling the system to assess its own input quality. Finally, based on the interference confidence level, a three-level adaptive decoupling algorithm is constructed for the original spectral data to select the extraction path of the intrinsic transmission spectrum and complete the transmittance detection. Experiments show that this method can stably extract the intrinsic transmittance of perovskite photovoltaic glass samples with different process fluctuations. The visible light transmittance calculation error is reduced by more than 50% compared with traditional filtering methods, effectively improving detection accuracy and robustness.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of the basic process for a method of detecting the transmittance of perovskite photovoltaic glass for building curtain walls provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the basic process of a method for determining the dominant interference oscillation wavelength interval provided in an embodiment of the present invention. Detailed Implementation

[0019] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for detecting the light transmittance of perovskite photovoltaic glass for building curtain walls proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device comprising a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of additional identical elements in the article or device that includes the element.

[0021] The following will describe in detail, with reference to the accompanying drawings, a method for detecting the light transmittance of perovskite photovoltaic glass for building curtain walls provided in this embodiment.

[0022] Please see Figure 1 This illustrates the basic process of a method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls, provided by an embodiment of the present invention.

[0023] like Figure 1 As shown, an embodiment of the present invention provides a method for detecting the light transmittance of perovskite photovoltaic glass for building curtain walls, which specifically includes the following steps: S100: Acquire raw spectral data of perovskite photovoltaic glass samples.

[0024] A UV-Vis-NIR spectrophotometer was used as the core measurement device, equipped with an integrating sphere to collect all transmitted light flux. Before measurement, the perovskite photovoltaic glass sample was cut or prepared to the size required by the equipment (typically 50mm × 50mm), and the surface was cleaned with anhydrous ethanol to ensure no contaminants interfered. During measurement, the perovskite photovoltaic glass sample was placed vertically in the optical path of the measurement device, with the incident light incident at a 0° angle, and the integrating sphere received all transmitted light. The scanning wavelength range covered 300nm to 1200nm, encompassing the main absorption bands of perovskite materials and the visible light region. The scanning step size was set to 1nm or 2nm, determined according to the required spectral resolution. Each perovskite photovoltaic glass sample was measured at least three times, and the average value was used as the raw spectral data. Simultaneously, background spectral data was collected as a reference to eliminate the influence of light source fluctuations and air absorption. The final output raw spectral data was a wavelength-transmittance sequence, which served as input for all subsequent analytical processing.

[0025] S200: For the original spectral data, extract the measured interference oscillation components, construct the zero-value crossing interval spectrum, and determine the dominant interference oscillation wavelength interval.

[0026] The raw spectral data is actually the result of the superposition of the intrinsic absorption trend of the glass and the interference oscillations. If the interference oscillations are analyzed directly, the intrinsic absorption trend often masks the details of the interference oscillations, especially near the absorption edge or in regions where the transmittance changes gently. The zero point and amplitude of the interference oscillations may be deviated. Therefore, before extracting the period of the interference oscillations, it is necessary to separate the interference oscillation components from the raw spectral data.

[0027] Based on the above analysis, in the embodiments of the present invention, the measured interference oscillation components are extracted from the original spectral data. For example... Figure 2 As shown, further includes: S201: Perform low-pass smoothing preprocessing on the original spectral data, and then use cubic spline interpolation to extract the upper and lower envelopes respectively to obtain the preliminary intrinsic trend lines of the original spectral data.

[0028] Specifically, considering that spectrophotometers inevitably introduce high-frequency white noise from the instrument's background during measurement, directly searching for extreme points based on the raw spectral data can easily lead to noise peaks being misidentified as local extrema, resulting in severe distortion of the extracted envelope. Therefore, before extracting the envelope, the raw spectral data undergoes low-pass smoothing preprocessing. This can be achieved using moving average filtering or Savitzky-Golay convolution smoothing algorithms. The smoothing window width is set to be less than half the minimum theoretical interference period to ensure that the original shape of the interference oscillation is effectively preserved while filtering out high-frequency noise. For the discretely sampled raw spectral data, cubic spline interpolation is then used to extract the upper and lower envelopes of the smoothed raw spectral data, denoted as […]. and Existing technology will not be elaborated here. Calculation of the upper envelope. and lower envelope The arithmetic mean of the original spectral data yields the preliminary intrinsic trend line:

[0029] In the formula, Preliminary intrinsic trend lines representing the raw spectral data; The upper envelope of the original spectral data after smoothing; This represents the lower envelope of the original spectral data after smoothing.

[0030] This step is equivalent to filtering out high-frequency oscillations, while retaining the macroscopic trend determined by the absorption characteristics of the glass itself.

[0031] S202: Based on the original spectral data and preliminary intrinsic trend lines, the measured interference oscillation components are obtained.

[0032] Specifically, by subtracting the preliminary intrinsic trend line point by point from the original spectral data, the measured interferometric oscillation components are obtained as follows:

[0033] In the formula, Indicates the first spectral data in the original spectral data The sampling point (the first) The wavelength of each sampling point is The measured values ​​of the interference oscillation components; Indicates the first The original spectral data values ​​of each sampling point; Indicates the first spectral data in the original spectral data Preliminary intrinsic trend line values ​​for each sampling point.

[0034] The advantage of obtaining the preliminary intrinsic trend line by the above method is that it is driven entirely by the extreme points of the original spectral data itself, without the need for a preset function form. It can adaptively track the macroscopic trend of the spectrum and avoid introducing artificial distortions into the interference oscillation signal.

[0035] After obtaining the measured interferometric oscillation components, the dominant interferometric oscillation wavelength interval is determined by constructing a zero-value crossing interval spectrum. For example... Figure 2 As shown, further includes: S203: Traverse the measured interference oscillation components, use linear interpolation to detect the sign change of the measured interference oscillation components, and obtain the zero-value crossing point sequence.

[0036] Specifically, after obtaining the measured interference oscillation components, it is necessary to lock the phase reversal position of the interference fringes, that is, to find the zero-value crossing point of the measured interference oscillation component curve. Because the measured data contains noise, it is directly obtained through... The zero-value crossing point is often unstable. Therefore, a linear interpolation method is used to detect sign changes and determine the zero-value crossing point. This involves traversing the entire measured interference oscillation component and finding the point where the product of the measured interference oscillation component values ​​of two adjacent sampling points satisfies the condition. When this occurs, it indicates that there is a zero-value crossing point between these two sampling points. At this point, the wavelength corresponding to this zero-value crossing point can be calculated using the linear interpolation formula:

[0037] In the formula, Indicates the first The sampling point and the first The wavelength corresponding to the zero-value crossing point between each sampling point; Indicates the first Wavelength of each sampling point; Indicates the first Wavelength of each sampling point; Indicates the first spectral data in the original spectral data The sampling point (the first) The wavelength of each sampling point is The measured values ​​of the interference oscillation components; Indicates the first spectral data in the original spectral data Measured interference oscillation component values ​​at each sampling point.

[0038] It should be noted that, in In actual calculations, to prevent the denominator from approaching zero and causing system collapse due to excessively small differences in the measured interference oscillation component values ​​between two adjacent sampling points, a zero-division prevention mechanism is introduced. Specifically, when the denominator ( The absolute value of ) is less than the preset minimum value (e.g. When the zero value crosses the point, the midpoint between the two sampling points is directly taken as the zero value crossing point, i.e., the first... The sampling point and the first The wavelength corresponding to the zero-value crossing point between sampling points is: .

[0039] After this operation, the continuous measured interferometric oscillation component waveforms are discretized into a sequence of zero-value crossing points. ,in They represent the first, the second, and the third, respectively. The and the first The wavelengths corresponding to the zero-value crossing points This represents the total number of zero-value crossing points. These zero-value crossing points correspond to the boundary lines of the alternating bright and dark fringes of the interference oscillation component.

[0040] S204: Based on the zero-value crossing point sequence, calculate the interference oscillation wavelength interval between the zero-value crossing points between phases to obtain the interference oscillation wavelength interval sequence, and then obtain the preliminary interval spectrum.

[0041] Specifically, the interval between two zero-value crossing points reflects the interference oscillation period of that local region. However, due to the influence of refractive index dispersion, the interference oscillation period is not strictly constant in the wavelength domain. Therefore, it is necessary to directly calculate the interference oscillation wavelength interval of all interphase elements in the zero-value crossing point sequence. ( Indicates the first Interference oscillation wavelength interval, Indicates the first (The wavelengths corresponding to the zero-value crossing points) will result in a fluctuating sequence of interference oscillation wavelength intervals. ,in They represent the first, the second, and the third, respectively. The and the first The interferometric oscillation wavelength intervals are calculated. To extract the most representative dominant interferometric oscillation period from this fluctuating sequence, the interferometric oscillation wavelength interval sequence is... Considering the sample as a whole, divide the wavelength range within a reasonable range of interferometric oscillation wavelengths (e.g., from the instrument resolution to one-tenth of the analysis band width) into several equally spaced wavelength intervals, and count the number of interferometric oscillation wavelength intervals falling into each wavelength interval, denoted as . (No. The number of interferometric oscillation wavelength intervals within each wavelength range is used to obtain a histogram of the interferometric oscillation wavelength interval distribution, which is the preliminary form of the interval spectrum, denoted as the preliminary interval spectrum.

[0042] S205: Based on the preliminary interval spectrum, analyze the consistency of the interference oscillation wavelength interval sequence to obtain the enhanced interval spectrum, and determine the interference oscillation wavelength interval corresponding to the main peak position of the enhanced interval spectrum as the dominant interference oscillation wavelength interval.

[0043] Specifically, firstly, to enhance the significance of the dominant interference oscillation period while suppressing pseudo-interference oscillation wavelength intervals caused by local fluctuations or noise, the idea of ​​enhancing the wavelength interval of nearest-neighbor interference oscillations is introduced into the statistics. The logic is that the wavelength interval sequence of interference oscillations generated by real interference oscillations should have continuity and consistency, and similar wavelength interval values ​​tend to appear densely. Therefore, the enhancement coefficient is defined as:

[0044] In the formula, Indicates the first... The center value of the interference oscillation wavelength interval of the square prisms The corresponding enhancement coefficient; Indicates the first... The center value of the interference oscillation wavelength interval of the square pillars; Indicates the first One interference oscillation wavelength interval; This represents the bandwidth parameter, which controls the smoothing range of the enhancement interval spectrum. The recommended value is set to 2 to 5 times the scanning step size of the spectrophotometer, so as to effectively suppress local jitter caused by sampling discretization while maintaining the identification of the dominant interference oscillation period. This represents the total number of zero-value crossing points; This represents the total number of wavelength intervals for the interference oscillations; Represented by natural constant An exponential function with base 0.

[0045] Then, by using the enhancement coefficient of each histogram, the number of interference oscillation wavelength intervals represented by the corresponding histogram is enhanced, i.e. ,in, Indicates the th in the enhanced histogram The number of interference oscillation wavelength intervals for each square prism is determined; ultimately, the enhanced interval spectrum is obtained, which will exhibit one or more peaks. For ideal interference oscillations caused by a single dominant interference, the spectrum will show a sharp main peak; if multiple interference optical path differences or strong noise exist, multiple broad peaks or an indistinct main peak may appear. Here, the interference oscillation wavelength interval value corresponding to the main peak position of the enhanced interval spectrum is considered to be... It carries the most stable and consistent periodic information of the interferometric oscillation within the measurement band, thus effectively avoiding the frequency ambiguity problem caused by directly performing Fourier transform under variable periods. Therefore, the interferometric oscillation wavelength interval corresponding to the main peak position of the enhanced interval spectrum is used. The dominant interference oscillation wavelength interval was determined to be the measured interference oscillation component.

[0046] S300: Based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, and introducing the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained. Then, combined with the dominant interference oscillation wavelength interval, the relative deviation is defined, and an empirical expression for the dispersion compensation factor is constructed. Finally, an equivalent interference optical path difference solution model suitable for perovskite photovoltaic glass is established.

[0047] Successfully extracted the dominant interference oscillation wavelength interval Next, it is necessary to convert these mathematical characteristics into film structure parameters with clear physical meaning. Simply relying on classical interferometry formulas for conversion introduces systematic errors because classical interferometry formulas implicitly assume that the refractive index is constant within the measurement wavelength range. This does not hold true for perovskite materials where the refractive index varies significantly with wavelength. For example, near the absorption edge, severe refractive index dispersion causes the actual interference oscillation period to deviate from the theoretical prediction. If this effect is ignored, the subsequent calculation of the interference optical path difference will retain significant phase distortion. Therefore, before calculating the interference optical path difference, it is necessary to compensate for the dispersion effect and correct the observed period to the equivalent optical path.

[0048] Based on the above analysis, in the embodiments of the present invention, based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, and introducing the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained. Then, combined with the dominant interference oscillation wavelength interval, the relative deviation is defined, and an empirical expression for the dispersion compensation factor is constructed. This leads to the establishment of an equivalent interference optical path difference calculation model suitable for perovskite photovoltaic glass. Wherein: Based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, and introducing the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained, further including: First, based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation component, the interference oscillation wavelength interval between adjacent interference maxima is obtained as: the ratio of the product of the wavelengths corresponding to adjacent interference maxima to the optical path difference of the single dominant interference. Specifically, for interference oscillation caused by a single dominant interference, the phase condition satisfied by adjacent interference maxima in the measured interference oscillation component is:

[0049] In the formula, Indicates the optical path difference of a single dominant interference; and Indicates the level of interference; Indicates the level of interference The wavelength corresponding to the time maximum; Indicates the level of interference The wavelength corresponding to the time maximum.

[0050] Subtracting the two equations and rearranging, we can obtain the wavelength interval of interference oscillation between adjacent interference maxima in the measured interference oscillation components. satisfy:

[0051] In the formula, This represents the wavelength interval between adjacent interference maxima in the measured interference oscillation components; Indicates the level of interference The wavelength corresponding to the time maximum; Indicates the level of interference The wavelength corresponding to the time maximum; This indicates the optical path difference of a single dominant interference.

[0052] Then, assuming that the wavelengths corresponding to adjacent interference maxima are all equal to the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained as follows: The interference oscillation wavelength interval of adjacent interference maxima is approximately equal to the ratio of the square of the center wavelength to the optical path difference of the single dominant interference. Specifically, when the interference oscillation wavelength interval of adjacent interference maxima is much smaller than its own wavelength, it can be approximated that the wavelengths corresponding to adjacent interference maxima are all equal to the center wavelength of the analysis band, thus obtaining a simplified relationship, namely, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is:

[0053] In the formula, This represents the wavelength interval between adjacent interference maxima in the measured interference oscillation components; Indicates the center wavelength of the analysis band; This indicates the optical path difference of a single dominant interference.

[0054] Among them, the center wavelength of the analysis band The selection of wavelength should be determined based on the subsequent application scenario. That is, if visible light transmittance is of interest, then select the wavelength around 550nm; if absorption edge characteristics in the near-infrared region are of interest, then select the center of the corresponding band.

[0055] However, this simplification ignores the difference between the wavelengths corresponding to adjacent interference maxima, which precisely implies the effect of dispersion. That is, when the refractive index changes with the wavelength, the effective interference optical path difference between adjacent interference orders is not strictly equal, causing the actual observed interference oscillation wavelength interval to shift relative to the predicted wavelength interval obtained through the center wavelength.

[0056] To capture the offset of the observed interferometric oscillation wavelength interval relative to the predicted wavelength interval obtained through the center wavelength, and for dispersion compensation, a relative deviation is defined based on the relationship, combined with the dominant interferometric oscillation wavelength interval. Further details include: First, the reference theoretical interference optical path difference of the perovskite photovoltaic glass sample is obtained. Then, based on the relevant formula, a deviation from the baseline is constructed. Specifically, to avoid the reference theoretical interference optical path difference... Due to the dominant interference oscillation wavelength interval The reverse reasoning leads to a logical deadlock where the relative deviation is always zero; this is based on the theoretical interference optical path difference. The parameters calculated from the prior structural parameters of the perovskite photovoltaic glass sample are as follows:

[0057] In the formula, This represents the reference theoretical interference optical path difference of the perovskite photovoltaic glass sample; This represents the preset reference refractive index value of the perovskite photovoltaic glass sample at the center wavelength (which can be estimated based on the typical dispersion model of perovskite materials). The nominal physical film thickness of the perovskite photovoltaic glass sample (derived from the manufacturing process parameters or design value).

[0058] Based on the reference theory of interference optical path difference, and combined with the relationship (the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength), the deviation from the reference is constructed as follows:

[0059] In the formula, This indicates the deviation from the reference (the predicted wavelength interval obtained by interferometric optical path difference between the center wavelength and the reference theory). Indicates the optical path difference in the reference theory interference; Indicates the center wavelength of the analysis band.

[0060] It should be understood that if the reference theoretical interference optical path difference of the perovskite photovoltaic glass sample... Then skip the dispersion compensation step and force the dispersion compensation factor to be set. .

[0061] Then, based on the dominant interference oscillation wavelength interval and the deviation from the reference, the relative deviation is defined as:

[0062] In the formula, This indicates the relative deviation between the observed interferometric oscillation wavelength interval and the deviation from the reference. This represents the dominant interferometric oscillation wavelength interval of the measured interferometric oscillation components; This indicates a deviation from the baseline.

[0063] relative deviation The deviation of the actual oscillation period caused by dispersion from the ideal oscillation period without dispersion was quantified. It should be noted that when... When the absolute value is less than the preset minimum value (e.g., 10^-6), directly set That is, the forced dispersion compensation factor .

[0064] After defining the relative deviation, an empirical expression for the dispersion compensation factor is constructed based on the relative deviation. Further steps include: Based on the stable linear correlation between relative deviation, center wavelength, and typical dispersion slope of perovskite materials, an empirical expression for the dispersion compensation factor is constructed. Specifically, through statistical analysis of historical spectral data from a large number of perovskite photovoltaic glass samples, a stable linear correlation was found between relative deviation and typical dispersion slope of perovskite materials. Based on this, the empirical expression for the dispersion compensation factor is constructed as follows:

[0065] In the formula, Indicates the dispersion compensation factor; The dimensionless normalization constant can be determined by fitting a set of standard perovskite photovoltaic glass samples with known film thicknesses (e.g., calibrated using an ellipsometry). In practical applications... The range of values ​​is usually 100. to The specific values ​​need to be calibrated using standard samples based on the instrument and material system. This indicates the relative deviation between the observed interferometric oscillation wavelength interval and the deviation from the reference.

[0066] Dispersion compensation factor Essentially, it describes the "compression" or "stretching" effect of the average rate of change of refractive index with wavelength on the interference period within a specific analytical band. Its dimension is consistent with the reciprocal of wavelength, and it is used to correct the systematic bias caused by neglecting dispersion in the classical interference formula.

[0067] Based on the empirical expression of the dispersion compensation factor, the equivalent interference optical path difference calculation model applicable to perovskite photovoltaic glass is established as follows:

[0068] In the formula, Indicates the equivalent interference optical path difference; Indicates the center wavelength of the analysis band; This represents the dominant interferometric oscillation wavelength interval of the measured interferometric oscillation components; This represents the dispersion compensation factor.

[0069] In practical calculations, to prevent the denominator from being negative or zero due to data anomalies or overcompensation, a conditional judgment needs to be added: if Then the dispersion compensation factor is forced. .

[0070] exist or the aforementioned , In this case, the formula for calculating the equivalent interference optical path difference degenerates into its classical form. It also indicates that the dispersion compensation for the current sample has failed and that the data quality needs to be manually reviewed.

[0071] The construction logic of the equivalent interference optical path difference solution model lies in: the denominator The term applies dispersion correction to the observation period. When dispersion is strong, the observation period is "compressed" or "stretched," requiring this term to be subtracted to restore the equivalent period under dispersion-free conditions. When dispersion is negligible, the formula degenerates into its classical form. In molecules It maintains the dimensional structure of the classical formula, ensuring It has the dimension of length and can be directly correlated with the physical thickness of the film. and average refractive index Related: (Consider the round-trip optical path when incident perpendicularly).

[0072] Here it is assumed that the equivalent interference optical path difference It is the first core physical parameter constructed. It is no longer a general signal frequency concept, but a "structural fingerprint" belonging to this specific perovskite photovoltaic glass sample. It is jointly determined by the data-driven interference oscillation wavelength interval (dominant period), the application-oriented center wavelength, and the dispersion compensation factor determined by the material properties. It directly reflects the equivalent optical thickness that dominates the interference effect in the multilayer film structure of perovskite photovoltaic glass.

[0073] S400: Using the calculated equivalent interference optical path difference as a priori physical constraint, the idealized interference oscillation sinusoidal function model is reconstructed in reverse to obtain the ideal interference oscillation component. Then, the degree of agreement between the ideal interference oscillation component and the measured interference oscillation component is analyzed, and the interference confidence level is defined.

[0074] After successfully calculating the equivalent interference optical path difference Next, its physical validity needs to be rigorously self-verified, and the proportion of components in the current spectral data that can be explained by a deterministic single-interference sinusoidal function model needs to be quantified. Simply obtaining a numerical equivalent interference optical path difference is insufficient. This is insufficient to determine its reliability. For example, when perovskite photovoltaic glass samples have multilayer film interference aliasing, interference ambiguity caused by uneven film thickness, or excessive measurement noise, relying on a single period (a single, determined interference oscillation wavelength interval) is not sufficient. Assuming the equivalent interference optical path difference is calculated in reverse... It may deviate significantly from the actual physical structure. If the subsequent intrinsic signal stripping is directly based on this, it will introduce artificial construction errors. Therefore, before performing the final intrinsic transmittance decoupling, it is necessary to construct a quantitative index that can measure the degree of "model-data" matching, namely the interference confidence level.

[0075] Based on the above analysis, in the embodiments of the present invention, the calculated equivalent interference optical path difference is used as a priori physical constraint to reconstruct the idealized interference oscillation sinusoidal function model in reverse, obtaining the ideal interference oscillation components. Then, the degree of agreement between the ideal interference oscillation components and the measured interference oscillation components is analyzed, and the interference confidence level is defined. Further steps include: First, within the high-transmission band of weak dispersion, the measured interferometric oscillation components are fitted to determine the amplitude coefficients and initial phase. Specifically, the amplitude coefficients and initial phase need to be determined by fitting the measured interferometric oscillation components... Fitting is performed to determine the amplitude coefficients and initial phase. However, directly performing global fitting on the entire waveband can introduce deviations due to local noise or interference from secondary interferometric modes. Therefore, a two-step strategy of "local optimal fitting - global consistency verification" is adopted here. A more specific implementation method is as follows: Within a weakly dispersed, high-transmission wavelength range (e.g., 800nm-1200nm), several local windows are uniformly selected. The window width needs to be much larger than the oscillation period but much smaller than the total width of the band, for example, around 50nm. Within each local window, the dominant interference oscillation wavelengths are spaced apart. To constrain this, the interferometric oscillation wavelength interval and the dominant interferometric oscillation wavelength interval are selected within each local window. Equal or similar range ( The oscillation period within the range is taken as the analysis object. The local amplitude and local phase within the local window are fitted by the least squares method, and the fitting residual of each local window is recorded. Here, it is assumed that in the pure region of a single dominant interference, the amplitude fitted by different local windows should change slowly, reflecting the amplitude modulation of intrinsic transmittance; the phase should remain basically constant, reflecting the phase transition characteristics of the reflection interface; and the fitting residual should be small.

[0076] To obtain a globally consistent reconstruction model, high-confidence candidate windows need to be selected from all local windows based on the fitting results. This involves analyzing the distribution of the fitting residuals of all local windows, setting a fitting residual threshold (which can be 0.2), and selecting local windows with fitting residuals less than the threshold as candidate windows. The statistical distribution of local phases within all candidate windows is then calculated, and candidate windows with outliers in their local phases are eliminated. The median of the local amplitude and the statistical mean of the local phases within the remaining high-confidence candidate windows are used as the amplitude coefficient and initial phase of the global model, respectively. This voting-based selection mechanism minimizes the impact of local noise or anomalous fluctuations on the model parameters, ensuring that the finally selected amplitude coefficients and initial phase truly represent the most prevalent and stable interferometric behavior characteristics across the entire band.

[0077] Then, based on the calculated equivalent interference optical path difference, combined with the amplitude coefficient and initial phase, an idealized sinusoidal function model of the interference oscillation and wavelength is constructed to obtain the ideal interference oscillation component in the low-dispersion, high-transmission band. Specifically, based on thin-film interference theory, for the interference oscillation caused by a single dominant interference, its normalized waveform can be approximately represented as a sinusoidal function model, thus restoring the calculated global physical quantities. Substituting the frequency position of the sine term, we can construct an idealized sine function model between the interference oscillation and the wavelength as follows:

[0078] In the formula, Represents the ideal interference oscillation component; Indicates the amplitude coefficient; Indicates the initial phase; Indicates wavelength; Indicates the equivalent interference optical path difference; This represents the sine function.

[0079] It should be noted that due to the strong intrinsic absorption and severe refractive index dispersion of perovskite materials in the short-wavelength region of 300-800 nm, their interference waveforms are severely distorted and cannot be accurately described by a single dominant sinusoidal function of the interference optical path difference. Therefore, this sinusoidal function model is mainly applicable to the weakly dispersed high-transmission band (e.g., the near-infrared region of 800 nm-1200 nm). Consequently, the initial local optimal fitting to obtain the amplitude coefficients and initial phase, as well as the subsequent interference confidence verification, are both limited to the weakly dispersed high-transmission band to ensure the consistency between the sinusoidal function model assumptions and physical reality.

[0080] Finally, in the high-transmission band with weak dispersion, the sum of the absolute residuals between the ideal and measured interference oscillation components is calculated. Combined with the total signal energy of the measured interference oscillation components, the degree of agreement between the ideal and measured components is obtained, and the interference confidence level is defined. Specifically, the formula for calculating the interference confidence level is:

[0081] In the formula, Indicates the confidence level of the interference; Represents the wavelength in the raw spectral data The measured interference oscillation component corresponding to the time; Indicates wavelength The corresponding ideal interference oscillation component; Indicates all wavelengths within the weak dispersion, high transmission band. The sum of absolute residuals between the corresponding ideal interference oscillation components and the measured interference oscillation components; Indicates all wavelengths within the weak dispersion, high transmission band. The total signal energy of the corresponding measured interference oscillation component; Indicates taking the absolute value; This represents the denominator correction parameter, which has the same dimensions as the denominator and takes the smallest value greater than 0. This is to prevent the denominator from being 0; for example, it can be set... This can prevent the denominator from being 0 without significantly interfering with the calculation of normal values.

[0082] It should be noted that the summation range here is limited to the weakly dispersive, high-transmission band used for fitting. The portion that cannot be explained by a single-interference sine function model was quantified. This represents the total signal energy that needs to be explained. The range of values ​​strictly falls within In the interval, when the sinusoidal function model of a single interference perfectly matches the measured interference oscillation component, the sum of the residuals is 0 ( ), When the sinusoidal function model of a single interference cannot fully explain the measured interference oscillation components, the residuals and Equal to the total signal energy , .

[0083] Interference confidence Essentially, it represents "the proportion of signal energy successfully interpreted within the weakly dispersive region," quantifying the portion of current spectral data that can be interpreted based on the equivalent interference optical path difference. The proportion of deterministic components described by the sinusoidal model of a single interference. Since the interference behavior in the weakly dispersive region is physically homologous to that across the entire wavelength range (both originating from the same film structure), this local interference confidence level can effectively characterize the overall effectiveness of the sinusoidal model of a single interference across the entire wavelength range. When When the value approaches 1, it indicates that the measured interferometric oscillation component in the weak dispersion region is almost entirely caused by a single dominant interference, suggesting that the data across the entire band is relatively pure; when At a moderate level, it indicates the presence of some minor interference or noise; when If the value is too low, it indicates that the current spectral data is not suitable for processing using a single interferometric sine function model.

[0084] S500: Based on the original spectral data, the intrinsic transmission spectrum extraction path is selected according to the interference confidence level to obtain the intrinsic transmission spectrum and complete the transmittance detection.

[0085] Interference confidence The extent to which the measured interferometric oscillation components can be explained by a sinusoidal function model of a single interference was quantified. Therefore, for the original spectral data, the extraction path of the intrinsic transmission spectrum was selected based on the interference confidence level to obtain the intrinsic transmission spectrum and complete the transmittance detection. Further steps include: First, preset a lower threshold for high confidence and an upper threshold for low confidence. Specifically, the lower threshold for high confidence... and low confidence upper limit threshold Pre-calibration was achieved through statistical learning on standard perovskite photovoltaic glass samples with known structures. Based on experimental statistics from a large number of perovskite photovoltaic glass samples, a high confidence lower limit was established. The value range is typically 0.80-0.90, with a lower confidence upper limit. The value range is usually 0.40-0.50, and the specific value can be fine-tuned according to the noise level of the actual measurement system and the accuracy requirements of the application scenario.

[0086] Then, the extraction path for the intrinsic transmission spectrum is selected based on the relationship between the interference confidence level and the lower and upper thresholds of high confidence. Wherein: When the interferometric confidence level is greater than or equal to the lower high confidence threshold, the physical subtraction method is used to extract the intrinsic transmission spectrum from the original spectral data, thus obtaining the intrinsic transmission spectrum. Specifically, when the interferometric confidence level is greater than or equal to the lower high confidence threshold, i.e. This indicates that the measured interference oscillation components are almost entirely based on the equivalent optical path difference. In cases where a single interference dominates, physical subtraction is employed to extract the intrinsic transmission spectrum from the original spectral data, i.e., using the amplitude coefficient. and initial phase Reconstruct the ideal interferometric oscillation components across the entire measurement band. Directly from raw spectral data Subtract the ideal interference oscillation component from the middle The intrinsic transmission spectrum is obtained as follows:

[0087] In the formula, Indicates the intrinsic transmission spectrum; Represents the raw spectral data; This represents the ideal interference oscillation component.

[0088] Physical subtraction angiography can preserve the fine structure of the spectrum, such as absorption edge features, to the greatest extent.

[0089] When the interference confidence level is less than the lower threshold of high confidence but greater than the upper threshold of low confidence, a physically guided Fourier filter is used to extract the intrinsic transmission spectrum from the original spectral data, thus obtaining the intrinsic transmission spectrum. Specifically, when the interference confidence level is less than the lower threshold of high confidence but greater than the upper threshold of low confidence, i.e. When the measured interference oscillation component is in the middle confidence region, the sinusoidal function model of a single interference can only explain part of the oscillation. Therefore, a physically guided Fourier filter is used to extract the intrinsic transmission spectrum from the original spectral data. Specifically, the original spectral data is first subjected to a Fourier transform to obtain the spectrum, and the eigenspectral components caused by the interference oscillation wavelength intervals are identified. The corresponding dominant frequency component is identified; then, a band-stop filter is designed centered on this dominant frequency. The stopband width is adaptively adjusted according to the interference confidence level, i.e., the lower the interference confidence level, the wider the stopband, to accommodate more uncertainty. The specific mapping relationship can be achieved through linear interpolation, i.e.:

[0090] In the formula, Indicates the stopband width; and These represent the preset maximum and minimum stopband widths, respectively. The maximum stopband width can be set to 20% of the dominant frequency, and the minimum stopband width can be set to 5% of the dominant frequency. Indicates the confidence level of the interference.

[0091] The filtered spectrum is then inversely transformed to obtain the processed spectrum. Then, compare it with the preliminary intrinsic trend line extracted earlier. The intrinsic transmission spectrum estimate is obtained by interferometric confidence-weighted fusion:

[0092] In the formula, Indicates the intrinsic transmission spectrum; Indicates the confidence level of the interference; This represents the spectrum after filtering the original spectral data; The initial intrinsic trend line represents the raw spectral data.

[0093] This weighted fusion strategy ensures that when the credibility of the single-interference sinusoidal function model is high, the filtered result takes precedence, while when the credibility of the single-interference sinusoidal function model is low, the robust trend line takes precedence.

[0094] When the interferometric confidence level is less than or equal to the lower confidence upper threshold, a preliminary intrinsic trend line is output as a rough estimate of the intrinsic transmission spectrum and labeled. Specifically, when the interferometric confidence level is less than or equal to the lower confidence upper threshold, i.e. when... When the measured interferometric oscillation component is in the low confidence region, the sinusoidal function model of a single interferometer fails, and the algorithm outputs a preliminary intrinsic trend line. As a rough estimate of the intrinsic transmission spectrum, the output intrinsic transmission spectrum is marked as "requires manual verification" in the test report and is not used for high-precision testing or automated quality assessment processes.

[0095] Finally, transmittance measurement is completed. Specifically, regardless of the mode used, an intrinsic transmission spectrum will ultimately be obtained. Based on the standard light source (such as the D65 light source) and the visual function specified in national standards (such as GB / T 2680-2021). By performing a weighted integral on the intrinsic transmission spectrum, the visible light transmittance is calculated as follows:

[0096] In the formula, Indicates visible light transmittance; Indicates the intrinsic transmission spectrum; This represents the spectral energy distribution of the D65 light source; This represents the view function.

[0097] In addition, indicators such as the absorption edge position (usually defined as the maximum point of the first derivative) and the average transmittance of a specific wavelength band can be extracted as a basis for evaluating the quality of perovskite photovoltaic glass.

[0098] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0099] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for testing the transmittance of perovskite photovoltaic glass for building curtain walls, characterized in that, The method includes: Collect raw spectral data of perovskite photovoltaic glass samples; Based on the original spectral data, the measured interference oscillation components are extracted, a zero-value crossing interval spectrum is constructed, and the dominant interference oscillation wavelength interval is determined. Based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, and by introducing the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained. Then, combined with the dominant interference oscillation wavelength interval, the relative deviation is defined, an empirical expression for the dispersion compensation factor is constructed, and an equivalent interference optical path difference solution model suitable for perovskite photovoltaic glass is established. Using the calculated equivalent interference optical path difference as a priori physical constraint, the idealized interference oscillation sinusoidal function model is reconstructed in reverse to obtain the ideal interference oscillation component. Then, the degree of agreement between the ideal interference oscillation component and the measured interference oscillation component is analyzed, and the interference confidence level is defined. Based on the original spectral data, the extraction path of the intrinsic transmission spectrum is selected according to the interference confidence level to obtain the intrinsic transmission spectrum and complete the transmittance detection.

2. The method for detecting the light transmittance of perovskite photovoltaic glass for building curtain walls according to claim 1, characterized in that, For the original spectral data, the measured interference oscillation components are extracted, including: The original spectral data is preprocessed with low-pass smoothing, and then the upper and lower envelopes are extracted using cubic spline interpolation to obtain the preliminary intrinsic trend lines of the original spectral data. Based on the original spectral data and the preliminary intrinsic trend line, the measured interference oscillation components are obtained.

3. The method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls according to claim 2, characterized in that, Constructing the zero-value crossing interval spectrum and determining the dominant interferometric oscillation wavelength interval, including: The measured interferometric oscillation components are traversed, and the sign change of the measured interferometric oscillation components is detected by linear interpolation to obtain a sequence of zero-value crossing points; Based on the zero-value crossing point sequence, the interference oscillation wavelength interval between the zero-value crossing points is calculated to obtain the interference oscillation wavelength interval sequence, and then the preliminary interval spectrum is obtained. Based on the preliminary interval spectrum, the consistency of the interference oscillation wavelength interval sequence is analyzed to obtain the enhanced interval spectrum, and the interference oscillation wavelength interval corresponding to the main peak position of the enhanced interval spectrum is determined as the dominant interference oscillation wavelength interval.

4. The method for detecting the light transmittance of perovskite photovoltaic glass for building curtain walls according to claim 1, characterized in that, Based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, and introducing the center wavelength of the analysis band, the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained, including: Based on the phase condition satisfied by adjacent interference maxima in the measured interference oscillation components, the interference oscillation wavelength interval between adjacent interference maxima is expressed as: the ratio of the product of the wavelengths corresponding to adjacent interference maxima to the optical path difference of a single dominant interference. If the wavelengths corresponding to adjacent interference maxima are all equal to the center wavelength of the analysis band, then the relationship between the interference oscillation wavelength interval of adjacent interference maxima and the center wavelength is obtained as follows: the interference oscillation wavelength interval of adjacent interference maxima is approximately equal to the ratio of the square of the center wavelength to the optical path difference of the single dominant interference.

5. The method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls according to claim 4, characterized in that, Define relative deviation, including: Obtain the reference theoretical interference optical path difference of the perovskite photovoltaic glass sample, and construct the deviation from the benchmark based on the relationship; The relative deviation is defined based on the dominant interference oscillation wavelength interval and the deviation from the reference.

6. The method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls according to claim 5, characterized in that, Constructing an empirical expression for the dispersion compensation factor, including: Based on the stable linear correlation between the relative deviation, the center wavelength, and the typical dispersion slope of perovskite materials, an empirical expression for the dispersion compensation factor is constructed.

7. The method for detecting the light transmittance of perovskite photovoltaic glass for building curtain walls according to claim 1, characterized in that, Using the calculated equivalent interference optical path difference as a priori physical constraint, the idealized interference oscillation sinusoidal function model is reconstructed in reverse to obtain the ideal interference oscillation components, including: Within the high transmission band of weak dispersion, the measured interference oscillation components are fitted to determine the amplitude coefficient and initial phase; Based on the calculated equivalent interference optical path difference, combined with the amplitude coefficient and the initial phase, an idealized sinusoidal function model between the interference oscillation and the wavelength is constructed to obtain the ideal interference oscillation component in the high transmission band with weak dispersion.

8. The method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls according to claim 7, characterized in that, Within the high transmission band of weak dispersion, the measured interference oscillation components are fitted to determine the amplitude coefficients and initial phase, including: Several local windows are uniformly selected in the high transmission band with weak dispersion. Within each local window, constrained by the dominant interference oscillation wavelength interval, the local amplitude and local phase within the local window are fitted using the least squares method, and the fitting residual for each local window is recorded. Analyze the distribution of the fitting residuals of all the local windows and filter candidate windows; Calculate the statistical distribution of local phases within all candidate windows, remove outlier candidate windows, and use the median of local amplitudes and the statistical mean of local phases within the remaining high-confidence candidate windows as the amplitude coefficient and initial phase of the global model, respectively.

9. The method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls according to claim 1, characterized in that, Analyze the degree of agreement between the ideal interference oscillation component and the measured interference oscillation component, and define the interference confidence level, including: In the high transmission band with weak dispersion, the sum of absolute residuals between the ideal interference oscillation component and the measured interference oscillation component is calculated. Combined with the total signal energy of the measured interference oscillation component, the degree of agreement between the ideal interference oscillation component and the measured interference oscillation component is obtained, and the interference confidence level is defined.

10. The method for detecting the transmittance of perovskite photovoltaic glass for building curtain walls according to claim 2, characterized in that, Based on the original spectral data and the interference confidence level, an extraction path for the intrinsic transmission spectrum is selected to obtain the intrinsic transmission spectrum, including: Preset a lower threshold for high confidence and an upper threshold for low confidence; When the interference confidence level is greater than or equal to the high confidence lower limit threshold, the physical subtraction method is used to extract the intrinsic transmission spectrum from the original spectral data to obtain the intrinsic transmission spectrum. When the interference confidence level is less than the lower threshold of high confidence level but greater than the upper threshold of low confidence level, a physically guided Fourier filter method is used to extract the intrinsic transmission spectrum from the original spectral data to obtain the intrinsic transmission spectrum. When the interference confidence level is less than or equal to the low confidence upper limit threshold, the preliminary intrinsic trend line is output as a rough estimate of the intrinsic transmission spectrum and marked.

Images