Method and System for Generating Light Formulas Based on the Synthesis of Dendrobium officinale Polysaccharides

Abstract

The present invention discloses a method and system for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale, specifically relating to the technical field of light environment regulation in plant factories, and is used to solve the technical bottlenecks in the existing polarized light regulation, such as the instability of the polarization state caused by environmental interference and the insufficient uniformity of light signal penetration; by real-time sensing the polarization state distortion region based on Stokes parameters, extracting the physical interference factors of the reflection path and medium scattering, generating polarization state compensation parameters through cross-correlation modeling, synergistically optimizing the laser light source direction angle and phase delay amount, and using the colorimetric method to verify the penetration uniformity to trigger the recalibration of the distortion matrix; through the closed-loop logic of physical interference tracing and dynamic compensation, the stability of the light signal polarization state and the penetration of plant tissues are synchronously improved, ensuring the efficient coordination of the light response mechanism of Dendrobium officinale and the polysaccharide synthesis path.

Description

Technical Field

[0001] The present invention relates to the technical field of light environment regulation in plant factories, and more specifically, to a method and system for generating a light formula based on the synthesis of polysaccharides in Dendrobium officinale. Background Art

[0002] In a plant factory, crop yield and quality can be improved through light environment regulation; by regulating light signal parameters (such as intensity, wavelength, polarization state, etc.), the light response mechanism of plants can be directionally activated to promote the synthesis of metabolites (such as polysaccharides, secondary metabolites); research shows that the polarization characteristics of light signals have a significant impact on plant cell signal transduction. For example, light waves with a specific polarization state can enhance the activation efficiency of photoreceptor proteins, thereby regulating gene expression and material synthesis pathways; in the prior art, multi-dimensional light regulation means (such as spectral formula, dynamic lighting mode) have been widely used in plant factories, but the polarization state stability of light signals in complex environments (such as multi-path reflection, water mist scattering) is still a key technical bottleneck.

[0003] In the prior art, the light regulation system in a plant factory faces significant defects in the practical application of polarization state parameters: due to multi-source interference in the cultivation environment (such as equipment reflection, medium scattering), the polarization characteristics of incident light are prone to non-uniform distortion, resulting in blocked light signal transduction paths, directly weakening the regulation effect of polarized light on the photosensitive mechanism of plants, causing unstable synthesis efficiency of metabolites, and restricting the repeatability and large-scale application of high-quality production. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method and system for generating a light formula based on the synthesis of polysaccharides in Dendrobium officinale to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for generating a light formula based on the synthesis of polysaccharides in Dendrobium officinale includes the following steps:

[0007] S1. In a plant factory, the polarization state parameters of the light signal in the Dendrobium officinale cultivation area are obtained in real time, and the polarization state parameters of the light signal include Stokes parameters;

[0008] S2. Based on the Stokes parameters, a polarization state distortion matrix is constructed to identify the distortion area where the Stokes parameters deviate from the preset threshold;

[0009] S3. According to the polarization state distortion matrix and the distortion area, interference factors related to the physical characteristics of the cultivation environment are extracted;

[0010] S4. Generate polarization state compensation parameters by analyzing the cross-correlation characteristics between the reflection path distribution in the interference factor and the medium scattering coefficient through non-linear iteration;

[0011] S5. Adjust the emission direction angle and phase delay amount of the laser light source according to the polarization state compensation parameters to generate a target polarization state optical signal, including:

[0012] Map the polarization state compensation parameters to the azimuth angle coordinate system of the laser light source, and adjust the emission direction angle based on the proportional-integral-derivative control algorithm to align the light source axis with the propagation direction of the target polarization state optical signal;

[0013] According to the change gradient of the phase delay amount in the compensation parameters, dynamically adjust the refractive index of the laser crystal through a piezoelectric ceramic driver to generate a target phase delay amount;

[0014] Based on the deviation between the target polarization state optical signal and the preset polarization state parameters in real-time feedback, fine-tune the collaborative action weight of the emission direction angle and the phase delay amount, and the weight is dynamically allocated according to the contribution ratio of reflection and scattering in the compensation parameters;

[0015] Couple the adjusted emission direction angle and phase delay amount through a polarization beam combiner to generate a target polarization state optical signal, and verify the consistency of its Stokes parameters with the preset threshold;

[0016] S6. Verify the penetration uniformity of the target polarization state optical signal in the Dendrobium officinale leaf tissue through colorimetry. If the penetration uniformity does not meet the preset standard, recalibrate the polarization state distortion matrix.

[0017] In a preferred embodiment, the polarization state parameters of the optical signal in the Dendrobium officinale cultivation area are obtained in real-time in the plant factory. The polarization state parameters of the optical signal include Stokes parameters, including:

[0018] Arrange a polarization sensor array above and on the side of the Dendrobium officinale cultivation area;

[0019] Collect the incident optical signal within a preset wavelength range through the polarization sensor array;

[0020] Perform spectral splitting and filtering on the incident optical signal to separate the characteristic spectral bands related to the calculation of Stokes parameters;

[0021] Calculate the S0, S1, S2, and S3 components in the Stokes parameters based on the light intensity distribution and phase difference of the characteristic spectral bands.

[0022] In a preferred embodiment, construct a polarization state distortion matrix based on the Stokes parameters to identify the distortion regions where the Stokes parameters deviate from the preset threshold, including:

[0023] Based on the S0, S1, S2, and S3 components in the Stokes parameters, calculate the covariance matrix among the components. The covariance matrix characterizes the statistical distribution characteristics of the polarization state of the optical signal;

[0024] Perform principal component analysis on the covariance matrix, extract the first two principal component components, and construct a two-dimensional polarization state feature space;

[0025] In the two-dimensional polarization state feature space, mark the area where the Stokes parameters deviate from the preset threshold as the distortion area;

[0026] Based on the spatial distribution density and boundary continuity of the distortion area, merge adjacent distortion areas with a density higher than the critical value to generate the final set of distortion areas.

[0027] In a preferred embodiment, the preset threshold is set based on the light response sensitivity of Dendrobium officinale and the requirement for polarization state stability.

[0028] In a preferred embodiment, according to the polarization state distortion matrix and the distortion area, extract the interference factors related to the physical characteristics of the cultivation environment, including:

[0029] Decompose the polarization state distortion matrix into a reflection path component and a medium scattering component. The reflection path component is extracted through the correlation mapping between the high-variance eigenvector direction in the principal component analysis and the incident angle of the light source;

[0030] Based on the spatial distribution of the distortion area and the deviation direction of the Stokes parameters, calculate the reflection path distribution through geometric inversion;

[0031] According to the correlation between the medium scattering component and the S3 component of the Stokes parameters, combine with the light attenuation model to invert the medium scattering coefficient. The light attenuation model is established based on the exponential attenuation law of light intensity with the propagation distance;

[0032] Fuse the reflection path distribution and the medium scattering coefficient to generate a comprehensive interference factor. The fusion method is to perform normalized weighting on the two factors, and the weights are set according to the contribution ratio of reflection and scattering in the distortion area.

[0033] In a preferred embodiment, the reflection path distribution includes the position, inclination angle, and reflection optical path difference of the reflecting surface.

[0034] In a preferred embodiment, through non-linear iterative analysis of the cross-correlation characteristics between the reflection path distribution and the medium scattering coefficient in the interference factor, generate polarization state compensation parameters, including:

[0035] Construct a cross-correlation matrix based on the reflection path distribution and the medium scattering coefficient. The cross-correlation matrix characterizes the joint probability distribution of the reflection path optical path difference and the scattering coefficient;

[0036] According to the singular value decomposition result of the cross-correlation matrix, the coupling weights between the reflection path and the scattering medium are extracted. The coupling weights represent the contribution ratio of reflection and scattering to the polarization state distortion;

[0037] Through the non-linear least squares optimization algorithm, with the coupling weights as the constraint conditions, the objective function for iteratively solving the polarization state compensation parameters is minimized, where the objective function is the sum of squares of the residual polarization errors;

[0038] In each iteration, the optimization step size is dynamically adjusted according to the gradient of the objective function until the residual error is lower than the preset convergence threshold, and the final polarization state compensation parameters are output.

[0039] In a preferred embodiment, the penetration uniformity of the target polarization state optical signal in the Dendrobium officinale leaf tissue is verified by colorimetry. If the penetration uniformity does not meet the preset standard, the polarization state distortion matrix is recalibrated, including:

[0040] A multi-angle spectrophotometer is arranged on the surface of the Dendrobium officinale leaf to collect the transmission spectrum of the target polarization state optical signal in the leaf tissue;

[0041] Based on the light intensity distribution of the transmission spectrum, the penetration uniformity is calculated, where the penetration uniformity is the ratio of the mean square deviation of the light intensities at each wavelength to the average light intensity;

[0042] If the penetration uniformity exceeds the threshold, it is determined that the preset standard is not met, and the deviation direction of the Stokes parameters corresponding to the wavelength range with the worst penetration uniformity is extracted;

[0043] According to the deviation direction, the covariance weights of the polarization state distortion matrix are corrected in the reverse direction, and an updated set of distortion regions is generated and returned for recalibration.

[0044] On the other hand, the present invention provides a light recipe generation system based on the synthesis of Dendrobium officinale polysaccharides, including the following modules:

[0045] A polarization sensing module for real-time obtaining the polarization state parameters of the optical signal in the Dendrobium officinale cultivation area in a plant factory, where the polarization state parameters of the optical signal include Stokes parameters;

[0046] A distortion modeling module for constructing a polarization state distortion matrix based on the Stokes parameters and identifying the distortion regions where the Stokes parameters deviate from the preset threshold;

[0047] An interference analysis module for extracting the interference factors related to the physical characteristics of the cultivation environment according to the polarization state distortion matrix and the distortion regions;

[0048] An iterative compensation module for generating polarization state compensation parameters by non-linearly iteratively analyzing the cross-correlation characteristics between the reflection path distribution and the medium scattering coefficient in the interference factors;

[0049] A signal generation module, configured to adjust the emission direction angle and phase delay amount of a laser light source according to polarization state compensation parameters, and generate a target polarization state optical signal;

[0050] A feedback calibration module, configured to verify the penetration uniformity of the target polarization state optical signal in the Dendrobium officinale leaf tissue by colorimetry. If the penetration uniformity does not meet the preset standard, recalibrate the polarization state distortion matrix.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. Through the dynamic closed-loop regulation mechanism of polarization state parameters, the adaptive stability of optical signals in complex environments is achieved, significantly improving the efficiency and consistency of polysaccharide synthesis in Dendrobium officinale. The polarization state distortion matrix constructed based on Stokes parameters, combined with the multi-dimensional environmental interference factor extraction technology, can accurately locate the polarization state distortion regions caused by abnormal reflection paths and medium scattering effects. Through the integration of physical modeling and statistical analysis, the optical signal distortion problem is transformed into a quantifiable spatial distribution feature, breaking through the limitation of static adjustment of empirical parameters in traditional optical compensation methods. The non-linear iterative algorithm establishes a dynamic mapping relationship between polarization state compensation parameters and environmental interference by coupling the cross-correlation characteristics of the reflection path and the scattering medium, solving the parameter mismatch problem under the coupling action of multi-source interference, so that the compensated polarization state optical signal not only meets the preset polarization accuracy requirements but also can actively adapt to dynamic environmental changes such as equipment layout adjustment and temperature and humidity fluctuations.

[0053] 2. The collaborative optimization mechanism of polarization state compensation parameters and laser light source parameters realizes the physical interpretability and engineering controllability of optical signal regulation. By modeling the spatial distribution characteristics of environmental interference through the covariance matrix and combining the reverse feedback of colorimetric penetration uniformity verification, the system can establish a dynamic balance between polarization state stability and optical signal penetrability, forming a self-calibrating closed-loop control logic. Compared with traditional methods, this scheme quantitatively analyzes the polarization state distortion mechanism and directionally corrects the compensation parameters, making the optical formula parameters have both physical scene adaptability and biological effect directivity. It not only suppresses the distortion effect of environmental interferences such as equipment reflection and water mist scattering on the polarization state but also ensures the effective penetration and uniform distribution of optical signals in plant tissues. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flowchart of the method for generating an optical formula based on polysaccharide synthesis in Dendrobium officinale of the present invention;

[0055] Figure 2 is a schematic structural diagram of the system for generating an optical formula based on polysaccharide synthesis in Dendrobium officinale of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with 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 of 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.

[0057] Embodiment 1: Figure 1 A method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale is given, which includes the following steps:

[0058] S1. In a plant factory, the polarization state parameters of the light signal in the Dendrobium officinale cultivation area are obtained in real time. The polarization state parameters of the light signal include Stokes parameters;

[0059] S2. Based on the Stokes parameters, a polarization state distortion matrix is constructed to identify the distortion area where the Stokes parameters deviate from the preset threshold;

[0060] S3. According to the polarization state distortion matrix and the distortion area, interference factors related to the physical characteristics of the cultivation environment are extracted;

[0061] S4. Through non-linear iterative analysis of the cross-correlation characteristics between the reflection path distribution and the medium scattering coefficient in the interference factors, polarization state compensation parameters are generated;

[0062] S5. According to the polarization state compensation parameters, the emission direction angle and phase delay amount of the laser light source are adjusted to generate a target polarization state light signal;

[0063] S6. The penetration uniformity of the target polarization state light signal in the Dendrobium officinale leaf tissue is verified by colorimetry. If the penetration uniformity does not meet the preset standard, the polarization state distortion matrix is recalibrated.

[0064] When obtaining the polarization state parameters of the light signal in the Dendrobium officinale cultivation area in real time in a plant factory, first, a polarization sensor array is arranged above and on the side of the Dendrobium officinale cultivation area. The polarization sensor array consists of multiple linear polarization sensors and circular polarization sensors. The photosensitive surface of each sensor unit forms a specific angle with the growth direction of the main stem of Dendrobium officinale. The setting range of the angle is 30° to 60°, so as to adapt to the spatial distribution characteristics of the Dendrobium officinale leaves and ensure that the light signal collection direction is consistent with the normal direction of the leaf light-receiving surface. The setting basis of the angle is the morphological data of the Dendrobium officinale plant, including the inclination angle of the main stem and the leaf unfolding angle. The installation height of the polarization sensor array is 40 cm to 60 cm from the cultivation bed surface, and the horizontal spacing is 20 cm to 40 cm, covering the top to bottom area of the Dendrobium officinale canopy in the vertical direction. The setting basis of the installation height and spacing is the average height and canopy density of the Dendrobium officinale plant, where the average plant height is 50 cm and the canopy density is 10 to 15 plants per square meter.

[0065] The incident light signals within a preset wavelength range are collected by a polarization sensor array. The preset wavelength range is set from 280 nm to 730 nm, which covers the ultraviolet, visible, and near-infrared bands. The setting of the ultraviolet band from 280 nm to 400 nm is based on the activation threshold of the light defense metabolic pathway of Dendrobium officinale. The setting of the visible band from 400 nm to 700 nm is based on the light response curve of photosynthesis. The setting of the near-infrared band from 700 nm to 730 nm is based on the sensitive wavelength of photomorphogenesis regulation. During the collection process, each sensor unit synchronously records the light intensity, wavelength, and polarization angle data of the incident light signals at a sampling rate of 5 to 20 frames per second. The setting of the sampling rate is based on the refresh frequency of the lighting system in the plant factory and the signal noise level. The refresh frequency of the lighting system is 60 Hz, and the signal noise level is measured by a spectrum analyzer. The main noise frequencies are 50 Hz power frequency interference and high-frequency switching noise. The data is transmitted to the central control unit through the RS485 communication protocol. The data processing software of the central control unit is a signal processing program developed based on Python, with the version of Python 3.8, and the dependent libraries include NumPy 1.20 and SciPy 1.6.

[0066] When performing spectral splitting and filtering on the incident light signals, a band-pass filter is used to separate the characteristic spectral bands related to the calculation of Stokes parameters. The center wavelengths of the band-pass filter are set to 450 nm for the blue light band, 660 nm for the red light band, and 730 nm for the far-red light band, with a bandwidth of ±10 nm. The setting of the center wavelengths is based on the absorption peaks of the photoreceptor proteins of Dendrobium officinale. Among them, 450 nm corresponds to the absorption peak of phytochrome B, 660 nm corresponds to the absorption peak of phytochrome A, and 730 nm corresponds to the far-red light response peak of phytochrome A. The setting of the bandwidth is based on the full width at half maximum of the photoreceptor protein, and the full width at half maximum measured by an ultraviolet-visible spectrophotometer is 20 nm. After spectral splitting and filtering, the light signals are converted into electrical signals through a photoelectric conversion module. The response wavelength of the photoelectric conversion module is from 200 nm to 1000 nm, and the conversion efficiency is 0.5 A / W. A lock-in amplifier is used to extract the AC components of each band to eliminate the environmental background noise. The reference frequency of the lock-in amplifier is set from 1 kHz to 10 kHz, and the phase delay is adjusted from 0° to 90°. The setting of the reference frequency is based on the main frequency components of the environmental noise. After analyzing the noise spectrum through fast Fourier transform, the main interference frequencies are determined to be from 1 kHz to 5 kHz; the setting of the phase delay is based on the signal modulation method. When the light signal is modulated by a square wave, the phase delay is 0°, and when it is modulated by a sine wave, the phase delay is 90°.

[0067] Based on the light intensity distribution and phase difference in the characteristic spectral bands, calculate the S0, S1, S2, and S3 components in the Stokes parameters. The S0 component represents the total light intensity and is obtained by calculating the arithmetic sum of the light intensities in each band. The calculation formula is S0 = I0 + I90 + I45 + I135 + IR + IL, where I0 is the light intensity in the 0° linear polarization direction, I90 is the light intensity in the 90° linear polarization direction, I45 is the light intensity in the 45° linear polarization direction, I135 is the light intensity in the 135° linear polarization direction, IR is the light intensity in the right-handed circular polarization direction, and IL is the light intensity in the left-handed circular polarization direction. The S1 component represents the intensity difference between the linearly polarized light in the 0° direction and the 90° direction, and is calculated by the signal intensity difference between the 450 nm in the blue light band in the 0° polarization direction and the 90° polarization direction. The calculation formula is S1 = I0 - I90. The S2 component represents the intensity difference between the linearly polarized light in the 45° direction and the 135° direction, and is calculated by the signal intensity difference between the 660 nm in the red light band in the 45° polarization direction and the 135° polarization direction. The calculation formula is S2 = I45 - I135. The S3 component represents the intensity difference between the right-handed circular polarized light and the left-handed circular polarized light, and is calculated by the signal intensity difference between the 730 nm in the far red light band in the right-handed circular polarization direction and the left-handed circular polarization direction. The calculation formula is S3 = IR - IL. The unit of all light intensity parameters in the formula is W / m², and the calculation result is a dimensionless ratio or difference.

[0068] To verify the accuracy of the Stokes parameter calculation, a standard polarizer and a quarter-wave plate are combined to generate an optical signal with a known polarization state and input it into the polarization sensor array for calibration. During the calibration process, elliptically polarized light with a linear polarization angle from 0° to 180° and an ellipticity from 0 to 1 is input into the sensor. The linear polarization angle is controlled by rotating the angle of the polarizer, and the ellipticity is adjusted by the delay amount of the quarter-wave plate. The relative errors between the calculated S1 / S0, S2 / S0, S3 / S0 and the theoretical values are less than 5%, and the calibration results meet the accuracy requirements for light regulation in plant factories. For example, when elliptically polarized light with a linear polarization angle of 45° and an ellipticity of 0.5 is input, the theoretical values are S1 / S0 = 0.707, S2 / S0 = 0.707, S3 / S0 = 0.5, and the measured values are S1 / S0 = 0.698, S2 / S0 = 0.713, S3 / S0 = 0.485, and the relative errors are 1.3%, 0.8%, and 3.0% respectively.

[0069] In the spectral filtering process, if the optical intensity signal-to-noise ratio of a certain wavelength band is lower than the preset threshold of 20 dB, the gain adjustment mechanism is automatically triggered. The amplification factor of the photoelectric conversion module is adjusted to dynamically adapt the signal intensity to the range of the analog-to-digital converter. The adjustment range of the gain adjustment coefficient is from 1 times to 10 times, and the adjustment is based on the signal attenuation degree and the noise power spectral density. The signal attenuation degree is calculated by the ratio of the current optical intensity to the reference optical intensity, and the noise power spectral density is measured by a spectrum analyzer. For example, when the optical intensity of the 660 nm red light band is attenuated to 20% to 50% of the original value due to water mist scattering, the reference optical intensity is the optical intensity value of 100 W / m² under the condition of no interference, the current optical intensity is 20 W / m² to 50 W / m², and the attenuation degree is 20% to 50%. At this time, the gain coefficient is increased from 1 time to 3 times to 5 times, so that the input signal intensity of the analog-to-digital converter is maintained within the range of 1 V to 5 V.

[0070] When constructing the polarization state distortion matrix based on the Stokes parameters, first obtain the Stokes parameter components S0, S1, S2, and S3 of the Dendrobium officinale cultivation area from step S1. The Stokes parameter S0 represents the total optical intensity, with the unit of W / m² and the value range of 100 - 1000 W / m²; S1 represents the difference in optical intensity between the 0° linear polarization direction and the 90° linear polarization direction, with the unit of W / m² and the value range of -500 - 500 W / m²; S2 represents the difference in optical intensity between the 45° linear polarization direction and the 135° linear polarization direction, with the unit of W / m² and the value range of -500 - 500 W / m²; S3 represents the difference in optical intensity between the right-handed circular polarization direction and the left-handed circular polarization direction, with the unit of W / m² and the value range of -200 - 200 W / m².

[0071] Based on the S0, S1, S2, and S3 components in the Stokes parameters, calculate the covariance matrix between the components. The dimension of the covariance matrix is 4 rows and 4 columns, and each element in the matrix represents the covariance value between two Stokes parameter components. The calculation formula for the covariance value is: the covariance value is equal to the first parameter component minus its mean, multiplied by the second parameter component minus its mean, and then averaged over all sampling points. The mean is obtained by taking the arithmetic mean of the Stokes parameters at all sampling points in the Dendrobium officinale cultivation area. The number of sampling points is set to 10 to 20 per square meter, and the distribution method of the sampling points is a uniform grid distribution with a grid spacing of 10 cm. The calculation result of the covariance matrix is used to characterize the statistical distribution characteristics of the polarization state of the optical signal. For example, when the covariance value between S1 and S2 is positive and relatively high, it indicates that the change trends of the linearly polarized light in the 0° and 45° directions are consistent, which may be caused by the same interference source.

[0072] Perform principal component analysis on the covariance matrix and extract the first two principal component components. The implementation method of principal component analysis is as follows: calculate the eigenvalues and eigenvectors of the covariance matrix, sort them in descending order of eigenvalues, and select the eigenvectors corresponding to the first two largest eigenvalues as the principal component components. The direction of the eigenvector represents the main change direction of the Stokes parameters in the polarization state feature space, and the eigenvalue represents the variance contribution rate in each direction. The sum of the variance contribution rates of the first two principal component components should be greater than 80% to ensure that the two-dimensional feature space after dimensionality reduction can retain the core polarization state information of the original data. If the sum of the variance contribution rates is lower than 80%, then increase the number of principal component components to three or re-evaluate the interference factor extraction logic. The re-evaluation method is to check whether there are abnormal eigenvalues or data acquisition errors in the covariance matrix.

[0073] Based on the first two principal component components, construct a two-dimensional polarization state feature space. The horizontal axis of the two-dimensional space corresponds to the direction of the first principal component component, and the vertical axis corresponds to the direction of the second principal component component. The coordinates of each sampling point are obtained by projecting the four-dimensional Stokes parameter vector onto the two principal component directions. The calculation formula for the projection value is: the abscissa is equal to the dot product of the Stokes parameter vector and the first principal component eigenvector, and the ordinate is equal to the dot product of the Stokes parameter vector and the second principal component eigenvector. The unit of the projection value is dimensionless, and the numerical range is normalized to -100 to 100 according to the actual data. Through dimensionality reduction mapping, the complex four-dimensional polarization state data is simplified into a two-dimensional visualization space, which is convenient for subsequent distortion area identification.

[0074] In the two-dimensional polarization state feature space, mark the area where the Stokes parameter deviates from the preset threshold as the distortion area. The preset threshold is set based on the light response sensitivity and polarization state stability requirements of Dendrobium officinale, and is specifically calibrated through the following experiment: under the condition of no interference, measure the change curves of the photosynthetic rate and polysaccharide synthesis rate of Dendrobium officinale leaves with respect to the polarization state parameters, determine the polarization state parameter interval when the photosynthetic rate and polysaccharide synthesis rate are the largest, and use the boundary values of this interval as the preset threshold. For example, the threshold range of the first principal component component is -50 to 50, and the threshold range of the second principal component component is -30 to 30. If the abscissa of a sampling point exceeds -50 to 50 or the ordinate exceeds -30 to 30, it is marked as a distortion area. During the marking process, use the connected component analysis algorithm in image processing to identify discrete clusters of distortion points, and the neighborhood determination distance in the algorithm is 5 cm.

[0075] Based on the spatial distribution density and boundary continuity of the distortion regions, adjacent distortion regions with a density higher than the critical value are merged. The spatial distribution density is calculated by counting the number of distortion regions per square meter, and the density critical value is set at 5 per square meter. The boundary continuity is judged based on the Euclidean distance between the central points of adjacent distortion regions not exceeding 10 cm. The merging logic is as follows: if the distance between the central points of two distortion regions is less than 10 cm and the density of the merged region is higher than 5 per square meter, then the two are merged into a single continuous region. The final set of distortion regions is generated after merging and is used for subsequent extraction of interference factors. During the merging process, morphological dilation operations are used to fill the gaps between regions, and the kernel size of the dilation operation is 3 cm × 3 cm.

[0076] To verify the accuracy of the polarization state distortion matrix, known interference sources are artificially introduced in the plant factory and the marked results of the distortion regions are measured. Experiment 1: A metal reflector is placed 30 cm to the side of the cultivation area. The size of the reflector is 50 cm × 50 cm, and the reflective surface makes an angle of 30° with the incident light direction. The measurement results show that 95% of the distortion regions are distributed within the range of ±15 cm around the reflector, coinciding with the area covered by the reflected light of the reflector. Experiment 2: A water mist spraying device is arranged above the cultivation area, with a spraying intensity of 5 L / h and a water mist particle size of 10 - 50 μm. The measurement results show that 85% of the distortion regions coincide with the area covered by the water mist, and the unmarked regions are mainly distributed in positions with a lower water mist concentration.

[0077] In the case of extreme interference, if more than 80% of the sampling points are marked as distortion regions, the alarm mechanism is triggered and the global light environment calibration mode is started. In the calibration mode, the real-time compensation process is paused, and the reference Stokes parameters are re-acquired. The acquisition conditions for the reference parameters are to turn off all interference sources and maintain a basic light intensity of 500 W / m². After the acquisition is completed, the covariance matrix and the principal component components are updated to ensure that the system adapts to the dynamically changing interference environment. The eigenvalue distribution of the calibrated covariance matrix needs to satisfy that the variance contribution rates of the first two principal components are greater than 80%, otherwise a hardware failure or environmental anomaly is prompted.

[0078] When extracting the interference factors based on the polarization state distortion matrix and the distortion regions, first obtain the polarization state distortion matrix and the final set of distortion regions from step S2. The polarization state distortion matrix is a 4-row and 4-column covariance matrix, which characterizes the statistical distribution characteristics of the Stokes parameters S0, S1, S2, and S3, and the dimension of the matrix elements is W² / m 4 . The set of distortion regions is the region merged through spatial distribution density and boundary continuity. The region boundary coordinates are represented in the three-dimensional coordinate system of the cultivation area. The origin of the coordinate system is set at the center of the cultivation bed surface. The X-axis is parallel to the growth direction of the main stem of Dendrobium officinale, the Y-axis is perpendicular to the main stem growth direction, the Z-axis is vertically upward, and the unit of the coordinate system is meters.

[0079] Decompose the polarization state distortion matrix into a reflection path component and a medium scattering component. The extraction method of the reflection path component is as follows: perform principal component analysis on the polarization state distortion matrix, select the direction of the eigenvector with the highest variance contribution rate, and map it to the incident angle of the light source. The incident angle of the light source is calculated based on the geometric relationship between the installation position of the laser light source and the center point of the distortion area. The calculation formula is incident angle = arctan(light source height / horizontal distance). The light source height is the vertical distance from the center point of the laser light source to the surface of the culture bed, and the horizontal distance is the horizontal projection distance from the center point of the light source to the center point of the distortion area. If the angle deviation between the eigenvector direction and the incident angle direction is less than 5°, it is determined that this component is dominated by the reflection path; if the angle deviation is greater than 15°, it is determined to be dominated by scattering or other interferences. For example, the light source height is 1.5 m, the horizontal distance is 2.0 m, the incident angle is 36.87°, and if the angle deviation between the eigenvector direction and the incident angle is 3°, it is determined to be dominated by the reflection path.

[0080] Based on the spatial distribution of the distortion area and the deviation direction of the Stokes parameters, calculate the reflection path distribution through geometric inversion. The geometric inversion method is as follows: taking the center point of the distortion area as the reflection point, combining the incident angle of the light source and the law of reflection, calculate the position, inclination angle, and reflection optical path difference of the reflection surface. The formula of the law of reflection is that the incident angle is equal to the reflection angle, and the reflection angle is the angle between the normal of the reflection surface and the reflected light. The reflection optical path difference is the sum of the distance from the light source to the reflection surface and the distance from the reflection surface to the sampling point. The calculation formula is optical path difference = distance from the light source to the reflection surface + distance from the reflection surface to the sampling point, and the unit of distance is meter. For example, the distance from the light source to the reflection surface is 1.2 m, the distance from the reflection surface to the sampling point is 0.8 m, the optical path difference is 2.0 m, the inclination angle of the reflection surface is 30°, and the position coordinates are (0.5 m, 0.3 m, 0.2 m).

[0081] According to the correlation between the medium scattering component and the Stokes parameter S3 component, combine the light attenuation model to invert the medium scattering coefficient. The light attenuation model is based on the exponential attenuation law of light intensity with the propagation distance. The formula is that the attenuated light intensity is equal to the initial light intensity multiplied by the negative scattering coefficient of the natural constant e to the power of the propagation distance. The inversion method of the scattering coefficient is as follows: substitute the covariance value between the Stokes parameter S3 component and the medium scattering component into the attenuation model, and solve the scattering coefficient by least squares fitting. The objective function of the least squares method is to minimize the sum of the squares of the residuals between the actual measured light intensity and the model predicted light intensity. The unit of the scattering coefficient is m -1 and the numerical range is 0.1 m -1 to 1.0 m -1 and the specific value is related to the medium type and its concentration. For example, when the water mist concentration is 5 g / m³, the scattering coefficient is 0.5 m -1 ; when the dust concentration is 10 g / m³, the scattering coefficient is 0.8 m -1 .

[0082] Fuse the reflection path distribution and the medium scattering coefficient to generate a comprehensive interference factor. The fusion method is to perform normalized weighting on the reflection path distribution and the medium scattering coefficient, and the weights are set according to the contribution ratio of reflection and scattering in the distortion region. The contribution ratio is calculated by the variance ratio of the reflection path component and the medium scattering component in the principal component analysis. The variance ratio is the ratio of the variance value of each component to the total variance value, and the total variance value is the sum of the variance values of all components. For example, if the variance value of the reflection path component is 1200 W² / m 4 , and the variance value of the medium scattering component is 800 W² / m 4 , and the total variance value is 2000 W² / m 4 , then the weight of the reflection path is 1200 / 2000 = 0.6, and the weight of the medium scattering is 800 / 2000 = 0.4. The normalized weighting formula is that the comprehensive interference factor is equal to the reflection path distribution multiplied by the reflection path weight plus the medium scattering coefficient multiplied by the medium scattering weight, and the sum of the weights is 1, ensuring that the comprehensive interference factor is a dimensionless parameter.

[0083] In extreme cases, if the contribution ratio of reflection and scattering is close to 50% (such as the calculated weight values are 0.48 and 0.52), and the weight error exceeds ±10%, then the calibration mechanism is triggered. The calibration mechanism includes re-acquiring the polarization state distortion matrix and optimizing the principal component analysis parameters. When re-acquiring, increase the sampling point density to 30 per square meter, the sampling point distribution method is a uniform grid distribution, the grid spacing is 5 cm, and the data acquisition time is extended to 10 minutes to improve the statistical significance. The optimized principal component analysis needs to ensure that the sum of the variance contribution rates of the first two principal components is greater than 85%, and the weight error is controlled within ±5%. If the error requirement still cannot be met after calibration, it is prompted that the environmental interference exceeds the system compensation ability, and manual intervention is required to adjust the light source or environmental parameters.

[0084] When generating the polarization state compensation parameters by non-linearly iteratively analyzing the cross-correlation characteristics between the reflection path distribution and the medium scattering coefficient in the interference factor, first obtain the reflection path distribution and the medium scattering coefficient from step S3. The reflection path distribution includes the position, inclination angle of the reflection surface, and the reflection optical path difference, and the unit of the reflection optical path difference is meters, and the numerical range is from 0.5 meters to 3.0 meters; the unit of the medium scattering coefficient is reciprocal per meter (m -1 ), and the numerical range is from 0.1 m -1 to 1.0 m -1 , and the specific value is determined by the water mist or dust concentration.

[0085] Construct a cross-correlation matrix based on the reflection path distribution and the medium scattering coefficient. The dimension of the cross-correlation matrix is N rows by N columns, where N is the number of discrete sampling points of the optical path difference of the reflection path and the medium scattering coefficient. The number of sampling points is set to be 20 to 30 per square meter according to the area of the cultivation region. The value of the matrix element is the covariance between the optical path difference of the reflection path and the medium scattering coefficient. The calculation formula for the covariance is: the covariance value is equal to the optical path difference of the reflection path minus its mean, multiplied by the medium scattering coefficient minus its mean, and then taking the arithmetic mean of all sampling points. The mean is calculated by summing the optical path difference of the reflection path and the medium scattering coefficient respectively and then dividing by the total number of sampling points. For example, if the mean of the reflected optical path difference is 1.5 meters and the mean of the medium scattering coefficient is 0.5 m -1 , then the covariance value reflects the linear correlation between the two.

[0086] According to the singular value decomposition result of the cross-correlation matrix, extract the coupling weights of the reflection path and the scattering medium. The implementation method of singular value decomposition is: decompose the cross-correlation matrix into the product of three matrices, namely the left singular vector matrix, the singular value diagonal matrix, and the right singular vector matrix. The coupling weight is calculated by the product of the left and right singular vectors corresponding to the largest singular value. The calculation formula is: the coupling weight is equal to the sum of the products of the corresponding elements of the first column of the left singular vector and the first column of the right singular vector, and then multiplied by the largest singular value. The dimension of the coupling weight is dimensionless, and the numerical range is from 0 to 1, and the sum of the weights is 1. For example, if the first column of the left singular vector is [0.8, 0.2], the first column of the right singular vector is [0.7, 0.3], and the largest singular value is 1.5, then the coupling weight is (0.8×0.7 + 0.2×0.3) ×1.5 = 0.99. After normalization, the reflection path weight is 0.65 and the scattering medium weight is 0.35.

[0087] Through the nonlinear least squares optimization algorithm, with the coupling weight as the constraint condition, iteratively solve the objective function of the polarization state compensation parameter. The objective function is defined as minimizing the sum of the squares of the residual polarization errors. The residual polarization error is the difference between the actual measured polarization state and the compensated polarization state, and the unit of the difference is percentage. The nonlinear least squares optimization algorithm uses the Levenberg-Marquardt algorithm, and the iterative formula is: the parameter update amount is equal to the inverse matrix of the transpose of the Jacobian matrix multiplied by the Jacobian matrix plus the damping coefficient multiplied by the identity matrix, and then multiplied by the transpose of the Jacobian matrix multiplied by the residual error vector. The initial damping coefficient is set to 0.01, and the iteration termination condition is that the residual error is less than the preset convergence threshold 1×10⁻ 5 ⁶ or reaching the maximum number of iterations 100 times. Each element of the Jacobian matrix is the partial derivative of the residual error with respect to the polarization state compensation parameter, and the partial derivative is calculated by the numerical difference method with a step size of 0.001.

[0088] In each iteration, the optimization step size is dynamically adjusted according to the gradient of the objective function. The gradient is calculated through the Jacobian matrix, and the gradient direction is the direction in which the residual error decreases fastest. The step size adjustment logic is as follows: if the residual error of the current iteration is less than the error of the previous iteration, the damping coefficient is reduced to one-tenth of the original value to accelerate convergence; if the residual error increases, the damping coefficient is increased to ten times the original value to stabilize the search direction. For example, if the initial damping coefficient is 0.01, it is adjusted to 0.001 if the error decreases, and adjusted to 0.1 if the error increases.

[0089] To verify the accuracy of the polarization state compensation parameters, different interference scenarios are set in the plant factory and the compensation effects are measured. Experiment 1: A metal reflector is placed 50 cm to the side of the cultivation area, with an inclination angle of the reflector of 30 degrees, a reflected optical path difference of 2.0 m, and a medium scattering coefficient of 0.3 m -1 . After compensation, the polarization state error is reduced from 15% to 3%, and the sum of the squares of the residual errors is 0.09. Experiment 2: The water mist spraying device is turned on, with a water mist concentration of 5 g per cubic meter and a scattering coefficient of 0.5 m -1 , and the reflector is turned off. After compensation, the polarization state error is reduced from 20% to 4%, and the sum of the squares of the residual errors is 0.16. The experimental data show that the method can effectively suppress the polarization state distortion caused by environmental interference.

[0090] In extreme cases, if the number of iterations exceeds 100 times and still does not converge, an exception handling mechanism is triggered. The exception handling includes resetting the damping coefficient to the initial value of 0.01, and expanding the sampling range of the reflection path and the scattering coefficient to 50 points per square meter to enhance the statistical significance of the data. If convergence still cannot be achieved, a calibration failure is prompted and an alarm log is generated, and manual intervention is required to check the light source position or the distribution of environmental interference sources. The standard for calibration failure is that the residual error is greater than 5% or the weight calculation deviation exceeds ±10%.

[0091] In a complex optical environment, traditional optical compensation methods usually independently process reflection or scattering interference, resulting in a mismatch between the compensation parameters and the real physical scenario. By constructing the cross-correlation matrix of the reflection path and the scattering medium, quantifying the combined effect of the coupling of the two on the polarization state distortion, and extracting the dominant interference modes by combining singular value decomposition, the limitation of single-factor analysis is broken through; further, a non-linear optimization algorithm with dynamic step size adjustment is adopted, and the convergence speed and stability are adaptively balanced according to the error gradient, avoiding local optimality or oscillation caused by a fixed step size. Compared with the existing technology, by coupling weight constraints and joint probability modeling, the contribution ratio of the main interference sources is accurately identified, making the compensation parameters more in line with the actual environmental interference distribution, and significantly improving the accuracy and robustness of polarization state regulation.

[0092] When adjusting the emission direction angle and phase delay of the laser source according to the polarization state compensation parameters, first obtain the polarization state compensation parameters from step S4. The polarization state compensation parameters include the emission direction angle correction amount, the phase delay amount, and the contribution ratio weights of reflection and scattering. The unit of the emission direction angle correction amount is degrees, and the numerical range is -15° to +15°. The unit of the phase delay amount is radians, and the numerical range is 0 to π. The dimension of the weight is dimensionless, and the numerical range is 0 to 1.

[0093] Map the polarization state compensation parameters to the azimuth coordinate system of the laser source. The azimuth coordinate system takes the center of the laser source as the origin, the X-axis is parallel to the culture bed surface, the Y-axis is perpendicular to the culture bed surface, and the Z-axis is along the beam propagation direction. The mapping method is as follows: According to the direction angle correction amount in the compensation parameters, calculate the real-time adjustment angle of the laser source through the proportional-integral-derivative control algorithm. The proportional coefficient of the proportional-integral-derivative control algorithm is set to 0.5, the integral coefficient is 0.1, and the differential coefficient is 0.05. The parameter setting is based on the response speed and stability requirements of the light source mechanical structure. Specifically, determine the ratio of the proportional coefficient to the integral coefficient through a step response experiment, and optimize the differential coefficient through a damped oscillation test. For example, in the step response experiment, if the time required for the light source to adjust from 0° to 10° is 2 seconds and the overshoot is less than 5%, the proportional coefficient is set to 0.5 to balance speed and stability.

[0094] According to the change gradient of the phase delay amount in the compensation parameters, dynamically adjust the refractive index of the laser crystal through a piezoelectric ceramic actuator. The voltage input range of the piezoelectric ceramic actuator is 0 - 10V. The relationship between the voltage change amount and the phase delay amount is: phase delay amount = voltage × piezoelectric coefficient × crystal length, where the piezoelectric coefficient is 200 pm / V and the crystal length is 5 mm. The piezoelectric coefficient and the crystal length are obtained from the technical manual of the laser crystal supplier. For example, if a phase delay of π / 2 radians needs to be generated, the voltage value is calculated as (π / 2) / (200e-12×5e-3)=7.85V. During the adjustment process, the phase delay amount is monitored in real time through closed-loop feedback control. The sampling frequency is 1 kHz, and the monitoring error is less than 0.01 radians. The closed-loop feedback control uses a differential amplifier and an analog-to-digital converter. The gain of the differential amplifier is 100 times, and the resolution of the analog-to-digital converter is 16 bits.

[0095] Based on the deviation between the target polarization state optical signal and the preset polarization state parameters with real-time feedback, fine-tune the synergy weight of the emission direction angle and the phase delay amount. The weight is dynamically allocated according to the contribution ratio of reflection and scattering in the compensation parameters, and the allocation formula is: direction angle weight = reflection contribution ratio / (reflection contribution ratio + scattering contribution ratio), phase delay weight = scattering contribution ratio / (reflection contribution ratio + scattering contribution ratio). For example, if the reflection contribution ratio is 0.7 and the scattering contribution ratio is 0.3, then the direction angle weight is 0.7 and the phase delay weight is 0.3. The fine-tuning logic is: if the S1 or S2 component in the polarization state deviation of the real-time feedback exceeds the threshold, it is determined that the reflection interference is dominant, and the direction angle weight is increased by 10%; if the S3 component exceeds the threshold, it is determined that the scattering interference is dominant, and the phase delay weight is increased by 10%.

[0096] Couple the adjusted emission direction angle and phase delay amount through a polarization beam combiner to generate a target polarization state optical signal. The input ports of the polarization beam combiner receive linearly polarized light with the adjusted direction angle and circularly polarized light with the phase delay respectively, and the coupling method is polarization orthogonal superposition. The Stokes parameters of the superimposed optical signal are verified through a polarization beam splitter, and the verification method is: measure the S1, S2, and S3 components and compare them with the preset thresholds. The threshold ranges are S1 = ±10%, S2 = ±10%, S3 = ±5%. For example, if the measured S1 = 8%, S2 = 9%, and S3 = 4%, it is determined to meet the preset standard. During the verification process, the calibration accuracy of the polarization beam splitter is ±1%, and the calibration method is to generate a known polarization state through a combination of a standard polarizer and a quarter-wave plate for calibration.

[0097] To verify the adjustment effect, set different interference scenarios in the plant factory and measure the accuracy of the target optical signal. Experiment 1: Reflection-dominated scenario (reflection weight 0.8), after compensation, the direction angle error decreases from 3° to 0.5°, the phase delay error decreases from 0.1 radian to 0.02 radian, and the Stokes parameter deviation is less than 2%. Experiment 2: Scattering-dominated scenario (scattering weight 0.7), the phase delay error decreases from 0.15 radian to 0.03 radian, the influence of the direction angle error is negligible, and the Stokes parameter deviation is less than 3%. The experimental data is averaged by three repeated measurements, and the standard deviation is less than 0.5%.

[0098] In extreme cases, if the output angle of the proportional-integral-derivative control algorithm exceeds the mechanical structure limit of ±15°, the protection mechanism is triggered. The protection mechanism includes: locking the current angle and switching to the manual calibration mode. In the manual calibration mode, the light source angle is finely adjusted through a knob, with each adjustment step being 0.1°, and at the same time, a fault log is generated to indicate an abnormal hardware limit. If the deviation of the Stokes parameters of the output optical signal of the polarization beam combiner continuously exceeds the threshold of 5%, the redundant light source switching process is started, the standby laser light source is enabled, and the compensation parameters are re-initialized. The wavelength of the standby light source is the same as that of the main light source, the power deviation is less than 1%, and the switching time is 100 milliseconds.

[0099] When verifying the penetration uniformity of the target polarization state optical signal in the Dendrobium officinale leaf tissue by colorimetry, first, a multi-angle spectrophotometer is arranged on the surface of the Dendrobium officinale leaf. The fiber optic probes of the multi-angle spectrophotometer are installed on the upper and lower surfaces of the leaf in a uniform grid distribution manner. The grid spacing is set to 5 mm, the coverage area of each probe is a circular area with a diameter of 2 mm, and the installation direction of the probe is consistent with the normal direction of the leaf surface to ensure that the optical signal collection direction is perpendicular to the leaf tissue. The spectral wavelength range of the spectrophotometer is set to 400 nm to 800 nm, covering the visible to near-infrared bands. The spectral resolution is set to 1 nm, the light intensity measurement unit is watt per square meter (W / m²), the sampling interval is 5 nm, and the data collection time is set to 5 s to balance the requirements of signal-to-noise ratio and real-time performance.

[0100] The penetration uniformity is calculated based on the light intensity distribution of the transmission spectrum. The calculation formula for the penetration uniformity is: uniformity is equal to the standard deviation of the light intensity at each wavelength divided by the average light intensity and then multiplied by 100%. The mean square error threshold is set to 15% of the average light intensity. The threshold setting is based on the experimental data of the light absorption characteristics of Dendrobium officinale leaves and is determined specifically by the following method: Under interference-free conditions, the transmission spectra of 10 healthy Dendrobium officinale leaves are collected, the uniformity distribution is calculated, and the upper limit value of the 95% confidence interval is taken as the threshold. For example, if the average light intensity of a healthy leaf is 50 W / m² and the standard deviation is 5 W / m², then the uniformity is (5 / 50)×100% = 10%, and the threshold is conservatively set to 15% to tolerate environmental fluctuations.

[0101] If the penetration uniformity exceeds the threshold, it is determined that the preset standard is not met. Extract the deviation direction of the Stokes parameters corresponding to the wavelength range with the worst penetration uniformity. The extraction method is as follows: Divide the 400 - 800 nm spectrum into 8 sub - ranges at 50 nm intervals (400 - 450 nm, 450 - 500 nm... 750 - 800 nm), calculate the uniformity value of each sub - range, screen the sub - range with the largest uniformity value (such as 550 - 600 nm), and extract the deviation directions of the S1, S2, and S3 components of the Stokes parameters corresponding to this range. The deviation direction is determined by comparing the signs of the measured Stokes parameters and the preset values. For example, if the measured value of the S1 component is 10% higher than the preset value, the deviation direction is positive; if it is 15% lower, it is negative. The deviation direction determination rule is: When the absolute value of the deviation is greater than 5%, it is regarded as an effective direction, otherwise it is ignored.

[0102] Reverse - correct the covariance weight of the polarization state distortion matrix according to the deviation direction. The reverse - correction method is: Convert the deviation direction into a weight adjustment coefficient of the covariance matrix. The adjustment coefficient is the product of the reciprocal of the deviation direction and the current weight. For example, if the deviation direction of the S1 component is positive + 10%, the weight adjustment coefficient is - 0.1, and the weight of the elements related to S1 in the covariance matrix is updated to the original weight multiplied by 0.9. The corrected covariance matrix is used to generate an updated set of distortion regions. The update logic is: Recalculate the variance contribution rate of the principal component analysis, merge the distortion regions after weight adjustment, and eliminate the minor interference regions with weights lower than 5%. The elimination rule is: If the weight of a certain region is lower than 5% after three consecutive calibrations, it is permanently eliminated and marked as a low - interference region.

[0103] To verify the calibration effect of the penetration uniformity, two test scenarios are set in the plant factory. Experiment 1: Uniformity - up - to - standard scenario (mean square error 12%). The system determines that no re - calibration is required, the light formula parameters remain unchanged, the deviation of the Stokes parameters is stable within ±3%, and the average photosynthetic rate of the leaves remains at 20 μmol / m² / s.

[0104] Experiment 2: Uniformity - exceeding - standard scenario (mean square error 18%). The system extracts that the deviation direction in the 550 - 600 nm range is S1 positive + 12%. After reverse - correcting the covariance weight, the number of recalibrated distortion regions decreases by 30%, the uniformity drops to 13%, meeting the threshold requirement, and the photosynthetic rate recovers to 19 μmol / m² / s. The experimental data is averaged by five repeated measurements, and the standard deviation is less than 1%. The data reproducibility meets the industrial standard.

[0105] In extreme cases, if the penetration uniformity still exceeds the threshold after recalibration, the depth calibration mode is triggered. The depth calibration mode includes: increasing the sampling point density of the spectrophotometer to 50 points per square meter, shortening the wavelength interval to 1 nm, and extending the data acquisition time to 10 minutes to improve the signal-to-noise ratio. If the standard cannot be met after three depth calibrations (e.g., the mean square error remains higher than 15%), an alarm log is generated and it is recommended to manually check the health status of the leaves or the distribution of environmental interference sources. The manual inspection process includes: using a handheld spectrometer to verify the transmission uniformity of the leaves, checking whether the deviation of the light source installation angle exceeds ±2°, and detecting whether the humidity in the cultivation area exceeds 80% resulting in abnormal water mist scattering.

[0106] In a plant factory, traditional light regulation methods usually independently process the polarization state stability and light signal penetration, resulting in the disconnection between the light formula parameters and the actual environmental interference. In this embodiment, through the coupling of the polarization state distortion matrix and non-linear iterative analysis, the physical characteristics such as the reflection path distribution and the medium scattering coefficient are correlated and modeled with the statistical distribution of the Stokes parameters, breaking through the limitation of single-parameter independent compensation; based on the cross-correlation matrix and singular value decomposition, the coupling weights of the interference factors are extracted, and combined with the dynamic step size optimization algorithm, the accurate generation of the polarization state compensation parameters under complex interference scenarios is realized. Especially through the colorimetric penetration uniformity verification and feedback mechanism, the covariance weight is inversely corrected and the distorted area is iteratively calibrated to form a closed-loop control logic, solving the technical bottleneck of error accumulation in traditional open-loop calibration. Through the dynamic decomposition and reconstruction of the polarization state distortion matrix, combined with the physical model (reflection geometry, light attenuation law) and statistical methods (covariance analysis, principal component dimensionality reduction), the quantitative analysis of environmental interference and the directional correction of compensation parameters are realized. The polarization state regulation is transformed from a single optical problem into a cross-field collaboration of physical interference modeling and dynamic optimization, making the light formula generation process have both optical measurement accuracy and complex environment adaptability, and improving the reliability of large-scale light regulation in plant factories.

[0107] Embodiment 2: Figure 2 The structural schematic diagram of the light formula generation system based on the synthesis of Dendrobium officinale polysaccharide according to the present invention is given. The light formula generation system based on the synthesis of Dendrobium officinale polysaccharide includes the following modules:

[0108] A polarization perception module for real-time acquisition of the polarization state parameters of the light signal in the Dendrobium officinale cultivation area in the plant factory. The polarization state parameters of the light signal include Stokes parameters;

[0109] A distortion modeling module for constructing a polarization state distortion matrix based on the Stokes parameters and identifying the distorted areas where the Stokes parameters deviate from the preset threshold;

[0110] An interference analysis module for extracting interference factors related to the physical characteristics of the cultivation environment according to the polarization state distortion matrix and the distorted areas;

[0111] An iterative compensation module, which is used to generate polarization state compensation parameters by non-linearly iteratively analyzing the cross-correlation characteristics between the reflection path distribution in the interference factor and the medium scattering coefficient;

[0112] A signal generation module, which is used to adjust the emission direction angle and phase delay amount of a laser light source according to the polarization state compensation parameters to generate a target polarization state optical signal;

[0113] A feedback calibration module, which is used to verify the penetration uniformity of the target polarization state optical signal in the Dendrobium officinale leaf tissue by colorimetry. If the penetration uniformity does not meet the preset standard, the polarization state distortion matrix is recalibrated.

[0114] The calculations involved in the embodiments are all dimensionless numerical calculations. The preset parameters and threshold selections in the calculations are set by those skilled in the art according to the actual situation.

[0115] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0116] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application of the technical solution and the inventive constraints. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0117] In addition, in each embodiment of the present application, the various functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0118] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the shown or discussed mutual coupling or direct coupling or communication connection can be through some interfaces, and the indirect coupling or communication connection of the devices or modules can be in an electrical, mechanical or other form.

[0119] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the said claims.

[0120] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale, characterized in that, It includes the following steps: S1. In a plant factory, obtain the polarization state parameters of the light signal in the cultivation area of Dendrobium officinale in real time. The polarization state parameters of the light signal include Stokes parameters; S2. Based on the Stokes parameters, construct a polarization state distortion matrix and identify the distortion areas where the Stokes parameters deviate from the preset threshold; S3. According to the polarization state distortion matrix and the distortion areas, extract the interference factors related to the physical characteristics of the cultivation environment; S4. Through non-linear iterative analysis of the cross-correlation characteristics between the reflection path distribution and the medium scattering coefficient in the interference factors, generate polarization state compensation parameters; S5. According to the polarization state compensation parameters, adjust the emission direction angle and the phase delay amount of the laser light source to generate a target polarization state light signal, including: Map the polarization state compensation parameters to the azimuth coordinate system of the laser light source, and adjust the emission direction angle based on the proportional-integral-derivative control algorithm to align the light source axis with the propagation direction of the target polarization state light signal; According to the change gradient of the phase delay amount in the compensation parameters, dynamically adjust the refractive index of the laser crystal through a piezoelectric ceramic driver to generate a target phase delay amount; Based on the deviation between the target polarization state light signal and the preset polarization state parameters feedback in real time, fine-tune the cooperation weight of the emission direction angle and the phase delay amount, and the weight is dynamically allocated according to the contribution ratio of reflection and scattering in the compensation parameters; Couple the adjusted emission direction angle and the phase delay amount through a polarization beam combiner to generate a target polarization state light signal, and verify the consistency of its Stokes parameters with the preset threshold; S6. Verify the penetration uniformity of the target polarization state light signal in the leaf tissue of Dendrobium officinale by colorimetry. If the penetration uniformity does not meet the preset standard, recalibrate the polarization state distortion matrix.

2. The method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale as claimed in claim 1, wherein, In a plant factory, obtain the polarization state parameters of the light signal in the cultivation area of Dendrobium officinale in real time. The polarization state parameters of the light signal include Stokes parameters, including: Arrange a polarization sensor array above and on the side of the cultivation area of Dendrobium officinale; Collect the incident light signal within a preset wavelength range through the polarization sensor array; Perform spectral splitting and filtering processing on the incident light signal to separate the characteristic spectral bands related to the calculation of Stokes parameters; Based on the light intensity distribution and phase difference of the characteristic spectral bands, calculate the S0, S1, S2, and S3 components in the Stokes parameters.

3. The method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale as claimed in claim 2, wherein Based on the Stokes parameters, construct a polarization state distortion matrix and identify the distortion areas where the Stokes parameters deviate from the preset threshold, including: Based on the S0, S1, S2, and S3 components in the Stokes parameters, calculate the covariance matrix between the components. The covariance matrix characterizes the statistical distribution characteristics of the polarization state of the light signal; Perform principal component analysis on the covariance matrix, extract the first two principal component components, and construct a two-dimensional polarization state feature space; In the two-dimensional polarization state feature space, mark the areas where the Stokes parameters deviate from the preset threshold as distortion areas; Based on the spatial distribution density and boundary continuity of the distortion areas, merge adjacent distortion areas with a density higher than the critical value to generate a final set of distortion areas.

4. The method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale as claimed in claim 3, wherein, The setting basis of the preset threshold is the light response sensitivity and polarization state stability requirements of Dendrobium officinale.

5. The method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale as claimed in claim 4, wherein, According to the polarization state distortion matrix and the distortion areas, extract the interference factors related to the physical characteristics of the cultivation environment, including: Decompose the polarization state distortion matrix into a reflection path component and a medium scattering component. The reflection path component is extracted through the correlation mapping between the direction of the high-variance eigenvector in the principal component analysis and the incident angle of the light source. Based on the spatial distribution of the distortion region and the deviation direction of the Stokes parameters, calculate the reflection path distribution through geometric inversion. According to the correlation between the medium scattering component and the S3 component of the Stokes parameters, combined with the light attenuation model, invert the medium scattering coefficient. The light attenuation model is established based on the exponential attenuation law of light intensity with the propagation distance. Fuse the reflection path distribution and the medium scattering coefficient to generate a comprehensive interference factor. The fusion method is to perform normalized weighting on the two factors, and the weights are set according to the contribution ratio of reflection and scattering in the distortion region.

6. The method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale as claimed in claim 5, wherein, The reflection path distribution includes the position, inclination angle, and reflection optical path difference of the reflecting surface.

7. The method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale as claimed in claim 6, wherein, Generate polarization state compensation parameters through non-linear iterative analysis of the cross-correlation characteristics between the reflection path distribution and the medium scattering coefficient in the interference factor, including: Construct a cross-correlation matrix based on the reflection path distribution and the medium scattering coefficient. The cross-correlation matrix characterizes the joint probability distribution of the reflection path optical path difference and the scattering coefficient. According to the singular value decomposition result of the cross-correlation matrix, extract the coupling weights of the reflection path and the scattering medium. The coupling weights represent the contribution ratio of reflection and scattering to the polarization state distortion. Through the non-linear least squares optimization algorithm, with the coupling weights as the constraint conditions, iteratively solve the objective function of the polarization state compensation parameters. The objective function is to minimize the sum of the squares of the residual polarization errors. In each iteration, dynamically adjust the optimization step size according to the gradient of the objective function until the residual error is lower than the preset convergence threshold, and output the final polarization state compensation parameters.

8. The method for generating a light formula based on the synthesis of polysaccharides from Dendrobium officinale as claimed in claim 7, wherein, Verify the penetration uniformity of the target polarization state optical signal in the Dendrobium officinale leaf tissue by colorimetry. If the penetration uniformity does not meet the preset standard, recalibrate the polarization state distortion matrix, including: Deploy a multi-angle spectrophotometer on the surface of the Dendrobium officinale leaf to collect the transmission spectrum of the target polarization state optical signal in the leaf tissue. Calculate the penetration uniformity based on the light intensity distribution of the transmission spectrum. The penetration uniformity is the ratio of the mean square deviation of the light intensity at each wavelength to the average light intensity. If the penetration uniformity exceeds the threshold, it is determined that the preset standard is not met, and the deviation direction of the Stokes parameters corresponding to the wavelength range with the worst penetration uniformity is extracted. According to the deviation direction, reversely correct the covariance weight of the polarization state distortion matrix to generate an updated set of distortion regions and return for recalibration.

9. A light formula generation system based on the synthesis of polysaccharides from Dendrobium officinale, which is used to implement the light formula generation method based on the synthesis of polysaccharides from Dendrobium officinale according to any one of claims 1-8, characterized in that, Include the following modules: A polarization sensing module for real-time acquisition of the polarization state parameters of the optical signal in the Dendrobium officinale cultivation area in the plant factory. The polarization state parameters of the optical signal include Stokes parameters. A distortion modeling module for constructing a polarization state distortion matrix based on the Stokes parameters and identifying the distortion regions where the Stokes parameters deviate from the preset threshold. An interference analysis module for extracting the interference factors related to the physical characteristics of the cultivation environment according to the polarization state distortion matrix and the distortion regions. An iterative compensation module for generating polarization state compensation parameters through non-linear iterative analysis of the cross-correlation characteristics between the reflection path distribution and the medium scattering coefficient in the interference factor. A signal generation module, configured to adjust the emission direction angle and phase delay amount of a laser light source according to polarization state compensation parameters, and generate a target polarization state optical signal; A feedback calibration module, configured to verify the penetration uniformity of the target polarization state optical signal in Dendrobium officinale leaf tissues by colorimetry, and if the penetration uniformity does not meet the preset standard, recalibrate the polarization state distortion matrix.

Images