Wavelength Power Calibration Method and Device for Wavelength Tunable Light Source

Abstract

The present application relates to the technical field of wavelength-tunable light sources, and discloses a wavelength power calibration method and device for a wavelength-tunable light source. The method includes: collecting the spectral and power of the output optical signals of lasers in the S band, C band, and L band to generate a wavelength and power joint calibration parameter matrix; performing scoring calculations to obtain a parameter importance scoring matrix; constructing a first adaptive control unit and a second adaptive control unit, generating a wavelength compensation value through the first adaptive control unit, and generating a power compensation value through the second adaptive control unit; inputting the wavelength compensation value into a wavelength tuning mechanism, and inputting the power compensation value into a power adjustment mechanism for joint compensation control to obtain a calibrated output optical signal; calculating calibration accuracy data; generating a wavelength power calibration report based on the calibration accuracy data, the parameter importance scoring matrix, and the wavelength and power joint calibration parameter matrix. Furthermore, the calibration of wavelength accuracy and power stability is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of wavelength tunable light sources, and particularly to a wavelength power calibration method and device for a wavelength tunable light source. Background Art

[0002] As a key device in an optical fiber communication system, the wavelength and power stability of a wavelength tunable light source directly affect the transmission performance of the system. With the development of dense wavelength division multiplexing technology, higher requirements are put forward for the wavelength accuracy and power stability of wavelength tunable light sources in the S band, C band, and L band. Traditional wavelength tunable light source calibration methods usually adopt a single parameter calibration method, independently calibrating the wavelength and power respectively, ignoring the power fluctuation during the wavelength tuning process and the influence of power adjustment on the wavelength, resulting in unsatisfactory calibration effects.

[0003] The existing calibration methods mainly have the following problems: lack of in-depth analysis of the coupling relationship between wavelength drift and power fluctuation, resulting in a lack of scientific basis for the selection of calibration parameters; secondly, the data processing method during the calibration process is relatively simple, and the multi-dimensional measurement data characteristics are not fully utilized; the evaluation system for calibration results is not perfect, lacking a comprehensive evaluation of short-term and long-term stability. Summary of the Invention

[0004] The present application provides a wavelength power calibration method and device for a wavelength tunable light source, thereby realizing the calibration of wavelength accuracy and power stability.

[0005] In the first aspect of the present application, a wavelength power calibration method for a wavelength tunable light source is provided. The wavelength power calibration method for the wavelength tunable light source includes:

[0006] Collecting the spectral and power of the output optical signals of the lasers in the S band, C band, and L band to generate a wavelength and power joint calibration parameter matrix;

[0007] Calculating scores for the wavelength compensation coefficient, power compensation coefficient, and coupling influence coefficient in the wavelength and power joint calibration parameter matrix to obtain a parameter importance score matrix;

[0008] Constructing a first adaptive control unit and a second adaptive control unit based on the parameter importance score matrix, generating a wavelength compensation value through the first adaptive control unit, and generating a power compensation value through the second adaptive control unit;

[0009] Inputting the wavelength compensation value into a wavelength tuning mechanism, inputting the power compensation value into a power adjustment mechanism, and performing joint compensation control on the laser to obtain a calibrated output optical signal;

[0010] Perform wavelength and power tracking measurements on the calibrated output optical signal, and calculate the calibration accuracy data;

[0011] Generate a wavelength-power calibration report based on the calibration accuracy data, the parameter importance scoring matrix, and the wavelength and power joint calibration parameter matrix.

[0012] The second aspect of the present application provides a wavelength-power calibration device for a wavelength-tunable light source, and the wavelength-power calibration device for the wavelength-tunable light source includes:

[0013] An acquisition module, configured to perform spectrum and power acquisition on the output optical signals of lasers in the S band, C band, and L band, and generate a wavelength and power joint calibration parameter matrix;

[0014] A calculation module, configured to perform scoring calculations on the wavelength compensation coefficient, power compensation coefficient, and coupling influence coefficient in the wavelength and power joint calibration parameter matrix to obtain a parameter importance scoring matrix;

[0015] A processing module, configured to construct a first adaptive control unit and a second adaptive control unit based on the parameter importance scoring matrix, generate a wavelength compensation value through the first adaptive control unit, and generate a power compensation value through the second adaptive control unit;

[0016] A compensation control module, configured to input the wavelength compensation value into a wavelength tuning mechanism, input the power compensation value into a power adjustment mechanism, perform joint compensation control on the laser, and obtain a calibrated output optical signal;

[0017] A measurement module, configured to perform wavelength and power tracking measurements on the calibrated output optical signal, and calculate the calibration accuracy data;

[0018] A generation module, configured to generate a wavelength-power calibration report based on the calibration accuracy data, the parameter importance scoring matrix, and the wavelength and power joint calibration parameter matrix.

[0019] The third aspect of the present application provides an electronic device, including: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor calls the instructions in the memory to enable the electronic device to execute the above-mentioned wavelength-power calibration method for a wavelength-tunable light source.

[0020] The fourth aspect of the present application provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when the instructions are run on a computer, the computer is enabled to execute the above-mentioned wavelength-power calibration method for a wavelength-tunable light source.

[0021] Compared with the prior art, the present application has the following beneficial effects: By constructing a joint calibration parameter matrix of wavelength and power and combining with the density peak clustering algorithm, an accurate description of the coupling characteristics of wavelength drift and power fluctuation is achieved. The multi-level adaptive control unit architecture, including functional modules such as data preprocessing, feature extraction, compensation calculation, and output regulation, improves the robustness and adaptability of the calibration system. A parameter importance scoring mechanism is introduced, and through algorithms such as matrix analysis and singular value decomposition, a reasonable configuration and dynamic optimization of calibration parameters are realized. A dual closed-loop control system is designed, and through the coordinated action of the wavelength outer loop and the power inner loop, mutual interference is effectively suppressed, and the dynamic response performance of the system is improved. A perfect calibration effect evaluation system is established, including multiple dimensions such as wavelength accuracy, power stability, and short-term and long-term stability, ensuring the reliability of the calibration results. A professional data visualization system is developed, and through multi-dimensional displays such as parameter distribution, performance indicators, and stability evaluation, an intuitive calibration quality evaluation method is provided. Joint calibration of the entire wavelength band (S, C, L bands) is achieved, and the wavelength accuracy and power stability are significantly better than traditional single-parameter calibration methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0023] The structures, ratios, sizes, etc. shown in the drawings of this specification are only used to cooperate with the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the limiting conditions under which the present invention can be implemented. Therefore, they do not have a substantial technical meaning. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope that can be covered by the technical content disclosed in the present invention.

[0024] Figure 1 is a schematic flowchart of the wavelength-power calibration method for a wavelength-tunable light source provided by an embodiment of the present invention;

[0025] Figure 2 is a schematic block diagram of the structure of a wavelength-power calibration device for a wavelength-tunable light source provided by an embodiment of the present invention;

[0026] Figure 3 is a schematic block diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] 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 part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] The flowchart shown in the accompanying drawings is only an example illustration, and does not necessarily include all the content and operations / steps, nor does it necessarily need to be executed in the described order. For example, some operations / steps can also be decomposed, combined, or partially merged. Therefore, the actual execution order may change according to the actual situation.

[0029] It should also be understood that the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. As used in the specification of this application and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0030] It should be further understood that the term "and / or" used in the specification of this application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. Please refer to Figure 1 , an embodiment of the wavelength power calibration method of the wavelength tunable light source in the embodiments of this application includes:

[0031] Step 100: Collect the spectral and power of the laser output optical signals in the S band, C band, and L band to generate a wavelength and power joint calibration parameter matrix;

[0032] It can be understood that the execution subject of this application can be a wavelength power calibration device of a wavelength tunable light source, or a terminal or a server. Specifically, it is not limited here. The embodiments of this application are described by taking the server as the execution subject as an example.

[0033] Specifically, N wavelength sampling points are set in the S-band, C-band, and L-band respectively. Each wavelength sampling point is measured M times to obtain the original sampling data. Based on the original sampling data, the real-time wavelength value of each wavelength sampling point is recorded by a spectrum analyzer, and at the same time, the real-time power value of each wavelength sampling point is recorded by a power meter to obtain a real-time measurement data set, which reflects the actual output characteristics of the laser at different bands and different wavelengths. The difference between the wavelength value in the real-time measurement data set and the target wavelength value is calculated to obtain the wavelength drift amount, and the difference between the power value in the real-time measurement data set and the target power value is calculated to obtain the power fluctuation amount. The difference data can reflect the stability and accuracy of the wavelength and power of the laser at different sampling points. To systematically analyze the fluctuations of the wavelength and power, the wavelength drift amount and the power fluctuation amount are arranged according to the measurement time sequence respectively to generate a wavelength drift amount matrix and a power fluctuation amount matrix. These matrices record the fluctuation conditions of each wavelength sampling point and show the change trend of these fluctuations over time. For feature analysis, the wavelength drift amount matrix and the power fluctuation amount matrix are normalized to generate a normalized wavelength feature matrix and a normalized power feature matrix. The normalized wavelength feature matrix is analyzed to calculate its wavelength local density and wavelength distance. The local density is used to describe the density of points near each data point, and the distance is used to measure the relative distance between different data points. Similarly, a similar analysis is performed on the normalized power feature matrix to calculate the power local density and power distance. The calculation of the local density and distance can help identify the mutual relationship between data points and can identify representative data points for subsequent compensation calculations. Based on the first product of the wavelength local density and the wavelength distance, the data points in the normalized wavelength feature matrix are sorted in descending order to select the data point with the largest first product as the first clustering center point. Similarly, based on the second product of the power local density and the power distance, the data points in the normalized power feature matrix are sorted in descending order to select the data point with the largest second product as the second clustering center point. Based on the first clustering center point and the second clustering center point, the wavelength compensation coefficient, the power compensation coefficient, and the coupling influence coefficient are calculated. These coefficients are used to describe the coupling relationship between the wavelength and the power and their respective deviation compensations. Furthermore, the wavelength compensation coefficient, the power compensation coefficient, and the coupling influence coefficient are combined to form a wavelength and power joint calibration parameter matrix, which contains comprehensive compensation information for the output optical signal of the laser in the S-band, C-band, and L-band.

[0034] Step 200: Calculate the score of the wavelength compensation coefficient, the power compensation coefficient, and the coupling influence coefficient in the wavelength and power joint calibration parameter matrix to obtain a parameter importance score matrix;

[0035] Specifically, a wavelength scoring unit, a power scoring unit, and a coupling scoring unit are constructed based on the wavelength and power joint calibration parameter matrix. These three scoring units respectively process the wavelength compensation coefficient, the power compensation coefficient, and the coupling influence coefficient. Among them, the wavelength scoring unit is used to process the wavelength compensation coefficient, the power scoring unit is used to process the power compensation coefficient, and the coupling scoring unit is used to process the coupling influence coefficient. By constructing the scoring units, independent processing of different types of calibration parameters is realized, and the scores of various parameters are accurately calculated. Statistical characteristic parameters of the wavelength drift matrix are calculated, where the wavelength statistical characteristic parameters include the standard deviation of the wavelength data distribution and the mean of the wavelength data center. These statistical characteristic parameters effectively reflect the distribution and central tendency of the wavelength data. Based on the statistical characteristic parameters, a wavelength weight distribution vector is calculated through the normal distribution, providing a weight basis for scoring the wavelength compensation coefficient. Similarly, statistical characteristic parameters of the power fluctuation matrix are calculated, and the power statistical characteristic parameters include the standard deviation of the power data distribution and the mean of the power data center. By calculating these parameters, the distribution characteristics of the power data are obtained. A power weight distribution vector is calculated through the normal distribution for scoring the power compensation coefficient. The wavelength drift matrix and the power fluctuation matrix are subjected to a cross-correlation operation to obtain a correlation matrix, which reflects the correlation between the wavelength and the power. When there is a mutual influence between the two, the calculation of the correlation matrix can reveal the intensity of this coupling effect. Based on the correlation matrix, eigenvalue decomposition is performed to obtain a coupling weight distribution vector, which describes the coupling degree between the wavelength and the power. The wavelength weight distribution vector, the power weight distribution vector, and the coupling weight distribution vector are standardized to generate a standardized weight matrix. The purpose of the standardization process is to eliminate the differences between different dimensions, so that all weights can be compared on the same scale, ensuring the fairness and consistency of scoring. The standardized weight matrix and the wavelength and power joint calibration parameter matrix are subjected to a dot product operation to obtain a parameter scoring matrix. Each element of the parameter scoring matrix represents the score of the corresponding parameter in the calibration process, reflecting the influence degree of the parameter on the calibration accuracy. Each score value in the parameter scoring matrix is mapped to the interval from 0 to 100 according to a linear mapping relationship to generate a quantization scoring matrix. The quantization scoring matrix is subjected to a singular value decomposition operation to extract the singular values and the corresponding singular vectors of the matrix. Through the analysis of the singular values, the right singular vector corresponding to the largest singular value is selected as the parameter importance scoring matrix, ensuring that the extracted features have the maximum explanatory power and enabling the parameter importance scoring matrix to reflect the most important parameter information in the entire calibration process.

[0036] Step 300: Based on the parameter importance scoring matrix, a first adaptive control unit and a second adaptive control unit are constructed. A wavelength compensation value is generated through the first adaptive control unit, and a power compensation value is generated through the second adaptive control unit.

[0037] It should be noted that a first adaptive control unit for generating wavelength compensation values and a second adaptive control unit for generating power compensation values are respectively constructed. The first adaptive control unit includes a first data preprocessing layer, a first feature extraction layer, a first compensation calculation layer, and a first output adjustment layer. The second adaptive control unit includes a second data preprocessing layer, a second feature extraction layer, a second compensation calculation layer, and a second output adjustment layer. The spectrogram data collected in real time is input into the first data preprocessing layer in the first adaptive control unit. This layer consists of a 5-layer wavelet transform network structure. Each layer of the wavelet network includes a decomposition unit and a reconstruction unit. Among them, the decomposition unit uses the db4 wavelet basis function, and the reconstruction unit uses the soft threshold denoising method. By processing the spectrogram data layer by layer, the denoised spectrogram data is obtained. The denoised spectrogram data is input into the first feature extraction layer. The first feature extraction layer includes a spectral main peak extraction sublayer, an edge feature extraction sublayer, and a shape feature extraction sublayer. The spectral main peak extraction sublayer uses a 9-point quadratic polynomial fitting algorithm to extract the main peak position and intensity information in the spectrogram; the edge feature extraction sublayer uses the Sobel edge detection operator to extract the edge information of the spectrogram data; and the shape feature extraction sublayer uses 15 shape descriptors to describe the shape features of the spectrogram signal. The multi-dimensional feature extraction method can comprehensively capture the features of the spectrogram data from different angles and generate a spectrogram feature vector containing rich information. The spectrogram feature vector and the parameter importance scoring matrix are input into the first compensation calculation layer. This layer consists of 3 fully connected layers. The number of nodes in the first fully connected layer is 128, the number of nodes in the second fully connected layer is 64, and the number of nodes in the third fully connected layer is 32. The ReLU activation function is used between each layer to increase the nonlinear ability of the model. At the same time, the Dropout layer is used to prevent overfitting, thereby enhancing its generalization ability during the training and use of the model. Through calculation, the wavelength compensation parameter is obtained. The wavelength compensation parameter is input into the first output adjustment layer. This layer includes a proportional-integral loop and a feedforward compensation loop. Among them, the proportional coefficient of the proportional-integral loop is 0.8, and the integral coefficient is 0.2. The feedforward compensation loop uses a 5th-order FIR filter. By combining proportional-integral adjustment and feedforward compensation, the wavelength compensation value is generated to ensure that the output of the wavelength remains stable and accurately reaches the target value. The power data collected in real time and the generated wavelength compensation value are input into the second data preprocessing layer in the second adaptive control unit. The second data preprocessing layer includes a data synchronization unit and a normalization unit. Among them, the data synchronization unit uses the nearest neighbor interpolation algorithm to ensure the time alignment of the power data and the wavelength data; and the normalization unit uses the Z-score normalization algorithm to normalize the power data, eliminating the dimensional difference between different data and making the subsequent calculation more stable and accurate. The preprocessed power data is input into the second feature extraction layer. This layer includes a 7-layer time-domain feature extraction network and a 5-layer frequency-domain feature extraction network.In time-domain feature extraction, statistical moment features and trend features are adopted to capture the statistical variations and time trends of power data; in frequency-domain feature extraction, wavelet packet decomposition and Hilbert transform are used to obtain the frequency-domain characteristics of power data. Through this step, an information vector that comprehensively reflects power features, namely the power feature vector, is generated. Based on the power feature vector, wavelength compensation value, and parameter importance scoring matrix, the second compensation calculation layer and the second output adjustment layer adopt a cascade compensation control method to generate the final power compensation value. In the second compensation calculation layer, the dynamic matrix control algorithm is used to dynamically adjust the compensation calculation parameters according to the real-time input signal, thereby improving the response speed and accuracy of the system. In the second output adjustment layer, a composite control structure combining forward-channel compensation and feedback-channel compensation is adopted. Fast adjustment is achieved through forward compensation, while the steady-state error of the system is eliminated through feedback compensation to generate an accurate power compensation value.

[0038] Step 400: Input the wavelength compensation value into the wavelength tuning mechanism and input the power compensation value into the power adjustment mechanism to perform combined compensation control on the laser to obtain the calibrated output optical signal;

[0039] Specifically, the wavelength compensation value is converted into a voltage through the first digital-to-analog conversion unit to generate a wavelength tuning control voltage. The wavelength tuning control voltage is input into the wavelength tuning mechanism through the first drive amplification circuit. The first drive amplification circuit consists of three-stage amplification units. The first stage is a differential amplifier, which is used to perform preliminary differential amplification on the input signal to improve the signal-to-noise ratio. The second stage is a voltage follower, which is used to provide a high input impedance and a low output impedance to enhance the driving ability of the signal. The third stage is a power operational amplifier, which further increases the power of the signal to ensure that the wavelength tuning control voltage can effectively drive the wavelength tuning mechanism. At the same time, the power compensation value is converted into a power regulation control current through the second digital-to-analog conversion unit. The power regulation control current is input into the power regulation mechanism through the second drive amplification circuit. The second drive amplification circuit includes a constant current source unit and a current detection unit. The constant current source unit is used to provide a stable current signal to ensure the accuracy and stability of power regulation. The current detection unit is used to monitor the change of the current in real time so as to adjust the power regulation control in a timely manner. Through current conversion and drive amplification processing, the power regulation control current can be accurately transmitted to the power regulation mechanism, thereby effectively regulating the output power of the laser. After the input of the wavelength and power compensation values is completed, the compensated initial calibration optical signal is obtained. In order to analyze and feedback control this optical signal, the initial calibration optical signal is converted into an electrical signal through the optoelectronic conversion unit to obtain an optoelectronic conversion signal. This optoelectronic conversion process can convert the intensity and wavelength information of the optical signal into corresponding electrical signal forms, which is convenient for subsequent signal processing and analysis. The optoelectronic conversion signal is digitally sampled through the third digital-to-analog conversion unit to obtain digital feedback data. The digital feedback data is input into the double closed-loop control system to perform precise closed-loop control on the wavelength and power. The double closed-loop control system includes a wavelength outer loop controller and a power inner loop controller. The wavelength outer loop controller adopts a feedforward-feedback composite control structure. This control structure combines the fast response of feedforward control and the steady-state accuracy of feedback control. The feedforward part compensates for the disturbance of the system in advance, while the feedback part can dynamically adjust the error of the system to achieve precise control of the wavelength. The power inner loop controller adopts an incremental PID algorithm. This algorithm calculates and adjusts the increment of the power error to ensure the smoothness and accuracy of power regulation. Compared with the traditional PID control, the incremental PID algorithm has better anti-disturbance performance and can effectively avoid the influence of cumulative error when dealing with system errors. The double closed-loop control system outputs corresponding control quantities, which are respectively connected to the wavelength tuning mechanism and the power regulation mechanism, thereby jointly and precisely controlling the wavelength and power of the laser. Through double closed-loop control, on the basis of real-time dynamic adjustment, the output states of the wavelength and power are continuously optimized to ensure that the output optical signal of the laser meets the calibrated accuracy requirements.

[0040] Step 500: Perform wavelength and power tracking measurements on the calibrated output optical signal, and calculate the calibration accuracy data;

[0041] Specifically, pass the calibrated output optical signal through a high-precision wavelength measurement unit and a power measurement unit for sampling respectively to obtain the original wavelength-power sampling data. Perform time synchronization and data alignment processing on the original wavelength-power sampling data to generate a synchronized measurement data stream, ensuring that the wavelength and power measurement data are compared and analyzed under the same time reference. Divide the synchronized measurement data stream into bands according to the S band, C band, and L band to generate a sub-band measurement matrix. Analyze and process the data of different bands respectively to facilitate the identification and solution of specific problems in each band. Perform wavelength reference correction on the wavelength data in the sub-band measurement matrix to obtain the corrected wavelength measurement data. The wavelength reference correction process adjusts the actually measured wavelength data according to a precise reference signal to eliminate the systematic error in the measurement process and ensure the accuracy of the wavelength data. Calculate the wavelength deviation parameters based on the corrected wavelength measurement data. The wavelength deviation parameters include instantaneous deviation, root mean square deviation, and maximum deviation. Among them, the instantaneous deviation is used to describe the difference between the measured value and the target value at a specific moment, and can reflect the response performance of the system under transient conditions; the root mean square deviation is used to measure the fluctuation in the entire measurement process and characterize the overall level of wavelength stability; the maximum deviation represents the maximum amplitude of deviation from the target value among all sampling points and is a measure of the system accuracy in the worst case. By calculating the deviation parameters, obtain the wavelength accuracy evaluation result, reflecting the overall accuracy and stability of the wavelength measurement. Similarly, calculate the power deviation parameters based on the power data in the synchronized measurement data stream. The power deviation parameters include instantaneous deviation, root mean square deviation, and maximum deviation. By calculating the power deviation, obtain the power accuracy evaluation result, reflecting the accuracy of the power measurement, the changes at different moments, and the stability of the system in power output. Perform band correlation analysis on the wavelength accuracy evaluation result and the power accuracy evaluation result to construct a performance evaluation matrix for the S band, C band, and L band. Identify the mutual influence and correlation between wavelength and power in different bands to help reveal the overall performance of the system under different working conditions. Through the correlation analysis of wavelength and power, it is found that there is a strong coupling effect in some bands, and this coupling effect affects the overall performance of the system. Characterize the performance of each band systematically through the performance evaluation matrix. Perform index normalization processing on the performance evaluation matrix to generate the calibration accuracy data. Convert the performance indicators in different bands to the same scale for direct comparison and analysis.

[0042] Step 600: Generate a wavelength-power calibration report based on the calibration accuracy data, the parameter importance scoring matrix, and the wavelength and power joint calibration parameter matrix.

[0043] Specifically, the wavelength and power joint calibration parameter matrix is sorted and analyzed according to wavelength bands to generate a parameter distribution characteristic table, recording the parameter distribution of each band to help understand the characteristic performance of the system under different working conditions. The parameter distribution characteristic table and the parameter importance scoring matrix are input into the data association processing unit for parameter importance ranking to generate a parameter optimization weight table, reflecting the importance of each calibration parameter in the overall calibration process, which helps to determine the parameters that should be focused on and optimized in subsequent calibrations, improving the efficiency and effectiveness of calibration. Statistical analysis is performed on the calibration accuracy data, and key indicators such as the root mean square error of wavelength, the maximum deviation of wavelength, the root mean square error of power, and the maximum deviation of power are statistically processed to generate performance index analysis curves, reflecting the calibration accuracy of the system in terms of wavelength and power. Through these analysis curves, the performance performance of the system during different measurement periods and the stability of wavelength and power can be observed. The performance index analysis curves are input into the trend analysis unit to generate stability evaluation data. Through trend analysis, the stability change characteristics of the system during different time periods are identified, and then the long-term performance performance of the calibration system is evaluated. Based on the stability evaluation data obtained from trend analysis, it is divided according to the time dimension to generate stability data at multiple time scales. The stability data at multiple time scales is compared and analyzed with the preset control indicators to generate a calibration control effect evaluation table. The calibration control effect evaluation table can show the performance of the system after calibration in achieving the expected performance target, clearly indicating whether the system meets the design requirements and helping to identify improvement directions. A comprehensive evaluation matrix is constructed based on the parameter optimization weight table, the performance index analysis curves, and the calibration control effect evaluation table. The analytic hierarchy process is used to calculate the normalized weights of each evaluation index to generate a comprehensive score of calibration quality. The analytic hierarchy process compares the relative importance of different indicators to obtain the weight of each indicator in the overall evaluation, thus obtaining a more scientific and reasonable comprehensive evaluation result. The comprehensive score of calibration quality, the parameter distribution characteristic table, the stability data at multiple time scales, and the calibration control effect evaluation table are input into the data visualization engine for chart conversion to generate a wavelength power calibration report.

[0044] In the embodiments of the present application, by constructing a wavelength and power joint calibration parameter matrix and combining with the density peak clustering algorithm, an accurate description of the coupling characteristics of wavelength drift and power fluctuation is achieved. The multi-level adaptive control unit architecture is adopted, which includes functional modules such as data preprocessing, feature extraction, compensation calculation, and output regulation, improving the robustness and adaptability of the calibration system. A parameter importance scoring mechanism is introduced, and through algorithms such as matrix analysis and singular value decomposition, a reasonable configuration and dynamic optimization of the calibration parameters are realized. A double closed-loop control system is designed, and through the coordinated action of the wavelength outer loop and the power inner loop, mutual interference is effectively suppressed, and the dynamic response performance of the system is improved. A perfect calibration effect evaluation system is established, including multiple dimensions such as wavelength accuracy, power stability, and short-term and long-term stability, ensuring the reliability of the calibration results. A professional data visualization system is developed, and through multi-dimensional displays such as parameter distribution, performance indicators, and stability evaluation, an intuitive calibration quality evaluation method is provided. The joint calibration of the entire wavelength band (S, C, and L bands) is realized, making the wavelength accuracy and power stability significantly better than the traditional single-parameter calibration method.

[0045] In a specific embodiment, the process of executing step 100 may specifically include the following steps:

[0046] Set N wavelength sampling points in the S band, C band, and L band respectively, and perform M measurements on each wavelength sampling point to obtain the original sampling data;

[0047] According to the original sampling data, record the real-time wavelength value through a spectrum analyzer and record the real-time power value through a power meter to obtain a real-time measurement data set;

[0048] Calculate the difference between the wavelength value in the real-time measurement data set and the target wavelength value to obtain the wavelength drift amount, and calculate the difference between the power value in the real-time measurement data set and the target power value to obtain the power fluctuation amount;

[0049] Arrange the wavelength drift amount and the power fluctuation amount in the measurement time sequence respectively to generate a wavelength drift amount matrix and a power fluctuation amount matrix, and perform normalization processing on the wavelength drift amount matrix and the power fluctuation amount matrix to generate a normalized wavelength feature matrix and a normalized power feature matrix;

[0050] Calculate the wavelength local density and wavelength distance of the normalized wavelength feature matrix, and calculate the power local density and power distance of the normalized power feature matrix;

[0051] Sort the data points in the normalized wavelength feature matrix in descending order based on the first product of the local wavelength density and the wavelength distance, and select the data point with the largest first product as the first clustering center point. Then, sort the data points in the normalized power feature matrix in descending order based on the second product of the local power density and the power distance, and select the data point with the largest second product as the second clustering center point;

[0052] Based on the first and second clustering center points, calculate the wavelength compensation coefficient, the power compensation coefficient, and the coupling influence coefficient, and form a wavelength and power joint calibration parameter matrix with the wavelength compensation coefficient, the power compensation coefficient, and the coupling influence coefficient.

[0053] Specifically, set N wavelength sampling points in the S band, C band, and L band respectively. For each wavelength sampling point, perform M measurements to obtain the original sampling data. Assume that the wavelength sampling point is denoted as , where represents the band (S, C, L bands), represents the sampling point number, . The result of each measurement forms the original sampling data set, including the power and wavelength values recorded at each wavelength sampling point during M measurements. By analyzing the original sampling data set, use a spectrum analyzer to record the real-time wavelength value of each wavelength sampling point, and a power meter to record the real-time power value to obtain the real-time measurement data set. Let the wavelength measurement value be and the power measurement value be , where represents the number of measurements at each sampling point. Calculate the difference between the wavelength value in the real-time measurement data set and the target wavelength value to obtain the wavelength drift , and the formula is:

[0054] ;

[0055] where is the target value of the wavelength sampling point. Similarly, calculate the difference between the power measurement value and the target power value to obtain the power fluctuation :

[0056] ;

[0057] Through the above formula calculations, obtain the wavelength drift and power fluctuation in each measurement. Arrange the drift and fluctuation in the order of measurement time to form a wavelength drift matrix and a power fluctuation matrix, denoted as and respectively. Where is a matrix with a dimension of The matrix, where each row represents the change of wavelength drift amount of a wavelength sampling point over time; similarly, is also a matrix with dimensions of , representing the change of power fluctuation amount over time. The wavelength drift amount matrix and the power fluctuation amount matrix are normalized to obtain a normalized wavelength feature matrix and a normalized power feature matrix. Let the normalized wavelength feature matrix be , and the normalized power feature matrix be , and their normalization is carried out in the following way:

[0058] ;

[0059] where, represents the element value in the matrix, is the mean value of all elements in the matrix, is the standard deviation. Through normalization, the influence of the data dimension is eliminated to ensure that data in different bands can be compared and analyzed on the same scale. Analyze the normalized wavelength feature matrix and calculate the wavelength local density and the wavelength distance . The wavelength local density is defined as the inverse function of the distance between each data point and other data points in its neighborhood, expressed as:

[0060] ;

[0061] where, represents the Euclidean distance, is a very small positive number used to prevent division by zero errors. Similarly, calculate the wavelength distance , which represents the minimum distance between a certain data point and a data point with higher local density. A similar process is applied to the normalized power feature matrix to obtain the power local density and the power distance . Based on the product of the wavelength local density and the wavelength distance , the data points in the normalized wavelength feature matrix are sorted in descending order, and the point with the largest product is selected as the first clustering center point, denoted as . Similarly, based on the product of the power local density and the power distance , the data points in the normalized power feature matrix are sorted in descending order, and the point with the largest product is selected as the second clustering center point, denoted as . Taking the first clustering center point and the second clustering center point as a reference, calculate the wavelength compensation coefficient , power compensation coefficient and coupling influence coefficient . The calculation of these coefficients is based on the deviation between their respective eigenvalues and the cluster centers, so as to quantify the relative deviation degrees of wavelength and power at each sampling point. For example, the wavelength compensation coefficient is expressed as:

[0062] ;

[0063] where is a compensation function, and its specific form is defined according to the characteristics of the system. Usually, a linear or non-linear function is selected to effectively reduce the deviation. Similarly, the power compensation coefficient and the coupling influence coefficient are calculated in a similar way. The calculated wavelength compensation coefficient, power compensation coefficient and coupling influence coefficient form a wavelength and power joint calibration parameter matrix. The matrix contains the calibration information of all wavelength sampling points and can reflect the coupling characteristics between wavelength and power, which is used for subsequent compensation control and performance optimization.

[0064] In a specific embodiment, the process of executing step 200 may specifically include the following steps:

[0065] Based on the wavelength and power joint calibration parameter matrix, a wavelength scoring unit, a power scoring unit and a coupling scoring unit are constructed. The wavelength scoring unit is used to process the wavelength compensation coefficient, the power scoring unit is used to process the power compensation coefficient, and the coupling scoring unit is used to process the coupling influence coefficient;

[0066] Calculate the wavelength statistical characteristic parameters of the wavelength drift matrix. The wavelength statistical characteristic parameters include the standard deviation of wavelength data distribution and the mean value of wavelength data center. Perform a normal distribution calculation to obtain the wavelength weight distribution vector, and calculate the power statistical characteristic parameters of the power fluctuation matrix. The power statistical characteristic parameters include the standard deviation of power data distribution and the mean value of power data center. Perform a normal distribution calculation to obtain the power weight distribution vector;

[0067] Perform a cross-correlation operation on the wavelength drift matrix and the power fluctuation matrix to obtain a correlation matrix, and perform eigenvalue decomposition based on the correlation matrix to obtain a coupling weight distribution vector;

[0068] Perform a normalization process on the wavelength weight distribution vector, the power weight distribution vector and the coupling weight distribution vector to generate a normalized weight matrix, and perform a dot product operation on the normalized weight matrix and the wavelength and power joint calibration parameter matrix to obtain a parameter scoring matrix;

[0069] Map each scoring value in the parameter scoring matrix to the 0-100 interval according to a linear mapping relationship to generate a quantization scoring matrix, and perform a singular value decomposition operation on the quantization scoring matrix. Select the right singular vector corresponding to the largest singular value as the parameter importance scoring matrix.

[0070] Specifically, the mutual influences among wavelength, power, and coupling are analyzed and processed. The wavelength scoring unit is used to process the wavelength compensation coefficient, the power scoring unit is used to process the power compensation coefficient, and the coupling scoring unit is used to process the coupling influence coefficient. By separately processing different compensation coefficients, the contribution and influence of each calibration parameter on the overall system performance are evaluated. Calculate the wavelength statistical characteristic parameters of the wavelength drift matrix, including the standard deviation of the wavelength data distribution and the central mean value of the wavelength data. The standard deviation reflects the degree of dispersion of the wavelength offset data, while the mean value is the central tendency of the wavelength data during the measurement process. By calculating the mean value and the standard deviation, a normal distribution calculation is performed on the wavelength drift amount to generate a wavelength weight distribution vector, which is used to represent the statistical importance of different wavelength sampling points. Calculate the power statistical characteristic parameters of the power fluctuation matrix. The power statistical characteristic parameters include the standard deviation of the power data distribution and the central mean value of the power data. A normal distribution calculation is performed to obtain the power weight distribution vector. Perform a cross-correlation operation on the wavelength drift matrix and the power fluctuation matrix to obtain their correlation matrix . The cross-correlation calculation reflects the mutual relationship between wavelength and power and reveals the degree of coupling between them. Each element of the correlation matrix represents the degree of correlation between the wavelength drift amount and the power fluctuation amount, and is calculated by the following formula:

[0071] ;

[0072] where and are the mean values of wavelength and power respectively, and are the standard deviations. By calculating the correlation matrix, the coupling relationship between wavelength and power is quantified. Based on the correlation matrix , perform eigenvalue decomposition on it to obtain the coupling weight distribution vector , which is used to describe the intensity of the coupling influence between wavelength and power. Normalize the wavelength weight distribution vector , the power weight distribution vector , and the coupling weight distribution vector to generate a normalized weight matrix . The normalized weight matrix is expressed as:

[0073] ;

[0074] where represents the norm of the vector. The weight matrix obtained by normalization ensures that the wavelength, power, and coupling weights are within the same range, making the subsequent scoring calculation more reasonable. The normalized weight matrix Wavelength and power combined calibration parameter matrix Perform a dot product operation to obtain a parameter scoring matrix S:

[0075] ;

[0076] Wherein, is the wavelength and power combined calibration parameter matrix, which contains compensation information for each wavelength and power sampling point, is the parameter scoring matrix, and each element represents the contribution value of the corresponding parameter in the scoring. The parameter scoring matrix is used to evaluate the importance of different calibration parameters. Map each scoring value in the parameter scoring matrix to the interval from 0 to 100 according to a linear mapping relationship to generate a quantization scoring matrix . Perform singular value decomposition on the quantization scoring matrix Q to obtain the singular values of the matrix and the corresponding singular vectors. Singular value decomposition decomposes the matrix into three parts: the left singular vector matrix, the singular value diagonal matrix, and the right singular vector matrix. By analyzing the singular values, select the right singular vector corresponding to the largest singular value as the parameter importance scoring matrix. The singular value corresponding to the right singular vector is the largest, which means it has the largest energy or explanatory power and is used to represent the importance of each parameter in the entire calibration process.

[0077] In a specific embodiment, the process of executing step 300 may specifically include the following steps:

[0078] Construct a first adaptive control unit and a second adaptive control unit. The first adaptive control unit includes a first data preprocessing layer, a first feature extraction layer, a first compensation calculation layer, and a first output adjustment layer. The second adaptive control unit includes a second data preprocessing layer, a second feature extraction layer, a second compensation calculation layer, and a second output adjustment layer;

[0079] Input the real-time spectral data into the first data preprocessing layer. The first data preprocessing layer contains a 5-layer wavelet transform network structure. Each layer of the wavelet network includes a decomposition unit and a reconstruction unit. The decomposition unit uses the db4 wavelet basis function, and the reconstruction unit uses the soft threshold denoising method to obtain the denoised spectral data through layer-by-layer processing;

[0080] Input the denoised spectral data into the first feature extraction layer. The first feature extraction layer includes a spectral main peak extraction sublayer, an edge feature extraction sublayer, and a shape feature extraction sublayer. The spectral main peak extraction sublayer uses a 9-point quadratic polynomial fitting algorithm, the edge feature extraction sublayer uses the Sobel edge detection operator, and the shape feature extraction sublayer uses 15 shape descriptors to obtain a spectral feature vector;

[0081] Input the spectral feature vector and the parameter importance scoring matrix into the first compensation calculation layer. The first compensation calculation layer contains 3 fully connected layers. The number of nodes in the first fully connected layer is 128, the number of nodes in the second fully connected layer is 64, and the number of nodes in the third fully connected layer is 32. The ReLU activation function and the Dropout layer are used between each layer to obtain the wavelength compensation parameter;

[0082] Input the wavelength compensation parameter into the first output adjustment layer. The first output adjustment layer contains a proportional-integral loop and a feedforward compensation loop. The proportional coefficient of the proportional-integral loop is 0.8, and the integral coefficient is 0.2. The feedforward compensation loop uses a 5th-order FIR filter to generate the wavelength compensation value;

[0083] Input the real-time power data and the wavelength compensation value into the second data preprocessing layer. The second data preprocessing layer contains a data synchronization unit and a normalization unit. The data synchronization unit uses the nearest neighbor interpolation algorithm, and the normalization unit uses the Z-score normalization algorithm to obtain the preprocessed power data;

[0084] Input the preprocessed power data into the second feature extraction layer. The second feature extraction layer contains a 7-layer time-domain feature extraction network and a 5-layer frequency-domain feature extraction network. The time-domain feature extraction network uses statistical moment features and trend features, and the frequency-domain feature extraction network uses wavelet packet decomposition and Hilbert transform to obtain the power feature vector;

[0085] Based on the power feature vector, the wavelength compensation value, and the parameter importance scoring matrix, cascade compensation control is performed in the second compensation calculation layer and the second output adjustment layer. The second compensation calculation layer uses the dynamic matrix control algorithm, and the second output adjustment layer uses a composite control structure that combines forward channel compensation and feedback channel compensation to generate the power compensation value.

[0086] Specifically, the real-time spectral data is input into the first data preprocessing layer, which is processed using a 5-layer wavelet transform network structure. In each layer of the wavelet network, there are decomposition units and reconstruction units. The decomposition units use the db4 wavelet basis function to perform multi-resolution analysis on the signal, separating different frequency components of the signal; the reconstruction units then use the soft threshold denoising method to reconstruct the signal, effectively removing noise. Through layer-by-layer wavelet decomposition and reconstruction, high-frequency noise is gradually filtered out, important spectral features are retained, and denoised spectral data is obtained. The denoised spectral data is input into the first feature extraction layer for feature extraction. The first feature extraction layer consists of three sub-layers: the spectral main peak extraction sub-layer, the edge feature extraction sub-layer, and the shape feature extraction sub-layer. The spectral main peak extraction sub-layer extracts the main peak information in the spectrum through a 9-point quadratic polynomial fitting algorithm, thereby identifying the most significant peak positions and intensities in the spectrum; the edge feature extraction sub-layer uses the Sobel edge detection operator to extract the edge features in the spectral signal, and these edge features reflect the change trend of the spectrum; the shape feature extraction sub-layer uses 15 shape descriptors to describe the overall shape features of the spectrum. These extracted features are combined into a spectral feature vector , which is used to represent the main features of the spectral signal. The spectral feature vector and the parameter importance scoring matrix are input into the first compensation calculation layer. The first compensation calculation layer contains three fully connected layers. The number of nodes in the first fully connected layer is 128, the number of nodes in the second fully connected layer is 64, and the number of nodes in the third fully connected layer is 32. The ReLU activation function is used between each layer to increase the non-linearity of the model, and a Dropout layer is added to prevent overfitting and improve the generalization ability of the model. Finally, through the step-by-step calculation of these three fully connected networks, the wavelength compensation parameter is output. This parameter is used for subsequent compensation adjustment to help the wavelength reach the target value. The wavelength compensation parameter is input into the first output adjustment layer. The first output adjustment layer contains a proportional-integral loop and a feed-forward compensation loop. The proportional coefficient of the proportional-integral loop is 0.8, and the integral coefficient is 0.2. The proportional-integral loop (PI control) is used to reduce the wavelength error through real-time adjustment, and its output is expressed as:

[0087] ;

[0088] where, is the wavelength error at the current moment, is the proportional coefficient, is the integral coefficient. The feed-forward compensation loop uses a 5th-order FIR filter to quickly respond to external disturbances and changes, thereby improving the dynamic performance of the system. The first output adjustment layer generates the wavelength compensation value , adjust the wavelength to reach the target. Input the real-time power data and wavelength compensation value into the second data preprocessing layer. The second data preprocessing layer includes a data synchronization unit and a normalization unit. The data synchronization unit uses the nearest neighbor interpolation algorithm to align the power data and wavelength data, ensuring that they can be compared and analyzed at the same time point; the normalization unit uses the Z-score normalization algorithm to normalize the power data, and the formula is:

[0089] ;

[0090] where, represents the original power data, is the mean of the power data, is the standard deviation of the power data. By normalization, the dimensional differences between different data are eliminated, enabling the data to be processed on the same scale, and obtaining the preprocessed power data . Input the preprocessed power data into the second feature extraction layer. The second feature extraction layer includes a 7-layer time-domain feature extraction network and a 5-layer frequency-domain feature extraction network. In time-domain feature extraction, statistical moment features and trend features are used to describe the changes of power data over time, such as statistical features like mean, variance, skewness, and kurtosis; in frequency-domain feature extraction, wavelet packet decomposition and Hilbert transform are used to capture the frequency characteristics and phase characteristics of the power signal, obtaining a more comprehensive description of the power features. Through feature extraction, the power feature vector is obtained. Based on the power feature vector , wavelength compensation value and parameter importance scoring matrix , input them into the second compensation calculation layer and the second output adjustment layer for cascade compensation control. The second compensation calculation layer uses the dynamic matrix control algorithm. This algorithm is based on the model prediction of the system, and by predicting and optimizing the future output, the power compensation parameter is obtained. The control law of the dynamic matrix control algorithm is expressed as:

[0091] ;

[0092] where, is the dynamic matrix of the system, is the regularization parameter, is the target output, is the current power output. Through this control law, the appropriate control input To achieve power control. The second output regulation layer adopts a composite control structure that combines forward-channel compensation and feedback-channel compensation. The forward-channel compensation quickly responds to input changes and improves the dynamic performance of the system, while the feedback-channel compensation can continuously adjust the error of the system to ensure the accuracy of the system under steady state. The second output regulation layer generates a power compensation value , which is used to precisely adjust the power of the laser.

[0093] In a specific embodiment, the process of executing step 400 may specifically include the following steps:

[0094] The wavelength compensation value is voltage-converted by the first digital-to-analog conversion unit to obtain a wavelength tuning control voltage, and the wavelength tuning control voltage is input into the wavelength tuning mechanism through the first drive amplification circuit. The first drive amplification circuit includes three-stage amplification units. The first stage is a differential amplifier, the second stage is a voltage follower, and the third stage is a power operational amplifier;

[0095] The power compensation value is current-converted by the second digital-to-analog conversion unit to obtain a power regulation control current, and the power regulation control current is input into the power regulation mechanism through the second drive amplification circuit. The second drive amplification circuit includes a constant current source unit and a current detection unit;

[0096] The compensated initial calibration optical signal is obtained, and the initial calibration optical signal is electrically signal-converted through the photoelectric conversion unit to obtain a photoelectric conversion signal, and the photoelectric conversion signal is digitally sampled through the third digital-to-analog conversion unit to obtain digital feedback data;

[0097] The digital feedback data is input into a double closed-loop control system. The double closed-loop control system includes a wavelength outer-loop controller and a power inner-loop controller. The wavelength outer-loop controller adopts a feedforward-feedback composite control structure, and the power inner-loop controller adopts an incremental PID algorithm to obtain the control quantity output by the double closed-loop control system, and is respectively connected to the wavelength tuning mechanism and the power regulation mechanism to obtain a calibrated output optical signal.

[0098] Specifically, the wavelength compensation value is voltage-converted by the first digital-to-analog conversion unit to obtain a control voltage for wavelength tuning. The role of the digital-to-analog conversion unit is to convert the digitized compensation value into an analog voltage signal, so as to drive the subsequent analog circuit part. Assume the wavelength compensation value is , and the corresponding wavelength tuning control voltage is obtained through the first digital-to-analog conversion unit, and its relationship is described by the following formula:

[0099] ;

[0100] Among them, is the digital-to-analog conversion ratio coefficient, which determines the sensitivity of the digital-to-analog conversion and the dimension conversion. represents the wavelength tuning control voltage. After obtaining the control voltage, it is amplified by the first drive amplifier circuit and input into the wavelength tuning mechanism. The first drive amplifier circuit includes three-stage amplification units: the first stage is a differential amplifier, mainly used to suppress common-mode noise and enhance the anti-interference ability of the signal; the second stage is a voltage follower, used to provide a high input impedance and a low output impedance, thereby reducing the load effect of the signal source and maintaining the signal stability; the third stage is a power operational amplifier, used to increase the power of the signal to ensure that the control voltage can effectively drive the wavelength tuning mechanism. Through these three-stage amplification processes, the signal can be fully amplified and enhanced to meet the input requirements of the wavelength tuning mechanism and ensure the precise adjustment of the wavelength. In terms of power adjustment, the power compensation value is converted into a current through the second digital-to-analog conversion unit to obtain the power adjustment control current . The function of the digital-to-analog conversion unit is similar to that of the wavelength part. By converting the digital compensation value into a corresponding analog signal, it drives the subsequent power control system. The converted power adjustment control current is expressed as:

[0101] ;

[0102] where, is the digital-to-analog conversion ratio coefficient, is the power adjustment control current. This current is then transmitted to the power adjustment mechanism through the second drive amplifier circuit. The second drive amplifier circuit includes a constant current source unit and a current detection unit. The constant current source unit is used to ensure the stability of the current signal so that the power adjustment mechanism can obtain a consistent and accurate input, while the current detection unit is used to monitor the output current in real time to ensure the adjustment accuracy and the closed-loop nature of the feedback control. The initial calibrated optical signal after compensation is obtained and converted into an electrical signal through the optoelectronic conversion unit to obtain the optoelectronic conversion signal. The optoelectronic conversion process is realized by an optoelectronic detector, which converts the intensity information of the optical signal into a corresponding current or voltage signal for subsequent signal processing and feedback control. Assuming the optoelectronic conversion signal is , this signal is digitally sampled through the third digital-to-analog conversion unit to obtain digital feedback data. The function of the third digital-to-analog conversion unit is to convert the analog optoelectronic signal into data available for digital calculation, facilitating the real-time digital feedback control of the system. The digital feedback data is input into a double closed-loop control system, which includes a wavelength outer loop controller and a power inner loop controller. The wavelength outer loop controller adopts a feedforward-feedback composite control structure, where the feedforward part is used to actively compensate for external disturbances, thereby improving the response speed of the system; the feedback part is used to correct the error between the wavelength and the target value to ensure the stability of the wavelength output. The control quantity of the wavelength outer loop controller is expressed as:

[0103] ;

[0104] Wherein, is a reference signal representing the target wavelength, is the wavelength error, , and are the coefficients of feedforward, proportional, and integral control respectively. Through the combination of feedforward and feedback, the system can quickly respond and correct the wavelength deviation to ensure the accuracy of the wavelength. The power inner-loop controller adopts an incremental PID control algorithm for real-time adjustment of the power output. The incremental PID algorithm adjusts the control quantity according to the increment of the power error, and its control law is expressed as:

[0105] ;

[0106] Wherein, is the power adjustment increment at the current moment, is the increment of the power error, , , are the proportional, integral, and differential coefficients respectively, is the sampling time interval. Through incremental PID control, the power inner-loop controller dynamically adjusts the output power to ensure its stability near the target value. After the control quantity of the double-loop control system is output, it is respectively connected to the wavelength tuning mechanism and the power adjustment mechanism for joint compensation control of the wavelength and power. The output of the wavelength outer-loop controller is used to adjust the input voltage of the wavelength tuning mechanism so that the wavelength can be accurately tuned to the target value; while the output of the power inner-loop controller is used to adjust the input current of the power adjustment mechanism so that the power output remains within the target range. Finally, the calibrated output optical signal is obtained.

[0107] In a specific embodiment, the process of executing step 500 may specifically include the following steps:

[0108] The calibrated output optical signal is respectively sampled for wavelength and power through a high-precision wavelength measurement unit and a power measurement unit to obtain the original wavelength power sampling data, and the original wavelength power sampling data is subjected to time synchronization and data alignment processing to generate a synchronized measurement data stream;

[0109] The synchronized measurement data stream is divided into bands according to the S band, C band, and L band to generate a sub-band measurement matrix, and the wavelength data in the sub-band measurement matrix is corrected for the wavelength reference to obtain the corrected wavelength measurement data;

[0110] Calculate the wavelength deviation parameters based on the calibrated wavelength measurement data. The wavelength deviation parameters include instantaneous deviation, root mean square deviation, and maximum deviation, and obtain the wavelength accuracy evaluation result. Calculate the power deviation parameters based on the power data in the synchronous measurement data stream. The power deviation parameters include instantaneous deviation, root mean square deviation, and maximum deviation, and obtain the power accuracy evaluation result;

[0111] Perform band correlation analysis on the wavelength accuracy evaluation result and the power accuracy evaluation result, construct the performance evaluation matrices for the S band, C band, and L band, and perform index normalization processing on the performance evaluation matrices to generate calibration accuracy data.

[0112] Specifically, accurately measure the calibrated output optical signal. Through the high-precision wavelength measurement unit and power measurement unit, sample the wavelength and power of the optical signal respectively to obtain the original wavelength-power sampling data. Assume that the wavelength value collected by the wavelength measurement unit is and the power value collected by the power measurement unit is where represents the sampling time point, and the sampling frequency should be high enough to capture the dynamic changes of the wavelength and power. Perform time synchronization and data alignment processing on the original wavelength and power sampling data to generate a synchronous measurement data stream. For example, through methods such as linear interpolation or nearest neighbor interpolation, align the wavelength and power sampling data on the same time basis to obtain the synchronized wavelength and power data, denoted as and where represents the time point. Divide the synchronous measurement data stream into bands according to the S band, C band, and L band. By judging the wavelength data intervals, classify the data in the synchronous measurement data stream into different bands to generate the sub-band measurement matrices. Assume the wavelength data matrix is and the power data matrix is to obtain three sub-band measurement matrices, namely the S band matrix 、 , the C band matrix and the L band matrix . Perform wavelength reference calibration on the wavelength data in each sub-band measurement matrix to eliminate the system measurement error and make the wavelength data more consistent with the actual target. Assume the calibrated wavelength measurement data is , and its calculation method is implemented through the following formula:

[0113] ;

[0114] where is the wavelength reference correction value, which is given by the reference equipment of the system and is used to compensate for the errors caused by equipment or environment during the measurement process. Based on the corrected wavelength measurement data, the wavelength deviation parameters are calculated. The wavelength deviation parameters include instantaneous deviation, root mean square deviation, and maximum deviation. The instantaneous deviation is defined as the difference between the corrected wavelength and the target wavelength:

[0115] ;

[0116] where is the target wavelength. The root mean square deviation (RMS deviation) represents the degree of deviation of the wavelength from the target value over a period of time, and its calculation formula is:

[0117] ;

[0118] where is the observation time interval. The maximum deviation is the maximum deviation value of the wavelength during the observation period:

[0119] ;

[0120] Through the above parameter calculations, the evaluation results of wavelength accuracy are obtained, which reflect the stability and accuracy of the wavelength during the calibration process. Similarly, the power data in the synchronous measurement data stream is analyzed to calculate the power deviation parameters. The power deviation parameters include instantaneous deviation, root mean square deviation, and maximum deviation. The instantaneous deviation represents the difference between the corrected power and the target power:

[0121] ;

[0122] The root mean square deviation represents the average degree of deviation of the power from the target value over a period of time, and the calculation method is:

[0123] ;

[0124] The maximum deviation is the maximum deviation of the power during the observation time:

[0125] ;

[0126] Through the calculation of these power deviation parameters, the evaluation results of power accuracy are obtained, which can help to judge whether the power meets the required accuracy standard during the entire calibration process. The wavelength and power accuracy evaluation results are subjected to band correlation analysis to construct a performance evaluation matrix for the S-band, C-band, and L-band. The performance evaluation matrix is used to comprehensively analyze the performance of the wavelength and power in different bands, evaluate their mutual correlation and their comprehensive impact on the system performance. Let the performance evaluation matrix be , whose elements are represented as deviation parameters of wavelength and power in each band. For example, matrix element represents the sum of the root mean square deviations of wavelength and power in the S band, and is used to characterize the overall calibration accuracy within this band. Perform index normalization on the performance evaluation matrix to generate calibration accuracy data. Standardize the evaluation results of different bands to the same scale for easy comparison and analysis. The normalization uses the following formula:

[0127] ;

[0128] where is an element in the performance evaluation matrix, and are the minimum and maximum values in the matrix respectively. The matrix obtained through normalization, and the values of all elements are between 0 and 1, enabling effective comparison between different bands and indicators.

[0129] In a specific embodiment, the process of executing step 600 may specifically include the following steps:

[0130] Organize and analyze the wavelength and power joint calibration parameter matrix according to the band, generate a parameter distribution characteristic table, and input the parameter distribution characteristic table and the parameter importance scoring matrix into the data association processing unit for parameter importance ranking to generate a parameter optimization weight table;

[0131] Perform statistical analysis on the root mean square error of wavelength, the maximum deviation of wavelength, the root mean square error of power, and the maximum deviation of power in the calibration accuracy data to generate a performance index analysis curve;

[0132] Input the performance index analysis curve into the trend analysis unit to generate stability evaluation data, and divide the stability evaluation data according to the time dimension to generate multi-time scale stability data;

[0133] Compare and analyze the multi-time scale stability data with the preset control indicators to generate a calibration control effect evaluation table;

[0134] Construct a comprehensive evaluation matrix based on the parameter optimization weight table, the performance index analysis curve, and the calibration control effect evaluation table, and use the analytic hierarchy process to calculate the normalized weights of each evaluation index to generate a comprehensive calibration quality score;

[0135] Convert the comprehensive calibration quality score, the parameter distribution characteristic table, the multi-time scale stability data, and the calibration control effect evaluation table through a data visualization engine to generate a wavelength and power calibration report.

[0136] Specifically, the wavelength and power joint calibration parameter matrix is sorted and analyzed according to the S-band, C-band, and L-band to generate a parameter distribution characteristic table. The generation of the parameter distribution characteristic table is based on the statistics of the wavelength compensation coefficient, power compensation coefficient, and coupling influence coefficient in each band, including characteristic information such as mean, variance, and kurtosis. Through statistical analysis, the parameter distribution characteristics of each band are obtained, forming a parameter distribution characteristic table. The generated parameter distribution characteristic table and the parameter importance scoring matrix are input into the data correlation processing unit for parameter importance ranking. The parameter importance scoring matrix contains the relative importance of each parameter in the overall calibration process. Through data correlation processing, considering the characteristics within the band and the importance information in the scoring matrix, a parameter optimization weight table is generated. This weight table is used to clarify the priority of different parameters in calibration. Statistical analysis is performed on the root mean square error of wavelength, maximum wavelength deviation, root mean square error of power, and maximum power deviation in the calibration accuracy data to generate a performance index analysis curve. For example, the root mean square error of wavelength and the root mean square error of power respectively represent the fluctuation degrees of wavelength and power during the calibration process, and their calculation methods are as follows:

[0137] ;

[0138] ;

[0139] where and are the wavelength and power at the current time respectively, and are the target wavelength and power, is the observed time interval. By statistically analyzing these metrics, performance metric analysis curves of wavelength and power are plotted to reflect the calibration accuracy and stability of the system at different time periods. The performance metric analysis curves are input into the trend analysis unit to generate stability assessment data. The purpose of trend analysis is to identify the changing trends of wavelength and power during long-term operation and evaluate the stability of the system. The stability assessment data is divided by time dimension to generate multi-time scale stability data. The multi-time scale stability data is compared and analyzed with preset control metrics to generate a calibration control effect assessment table. The preset control metrics are performance targets during the calibration process. By comparing the multi-time scale stability data with these targets, it can be intuitively judged whether the system achieves the expected calibration effect at each time scale. If some metrics fail to meet the control requirements, the assessment table will show the specific situation of the deviation, providing guidance for subsequent improvement. Based on the parameter optimization weight table, performance metric analysis curves, and calibration control effect assessment table, a comprehensive evaluation matrix is constructed. The analytic hierarchy process is used to calculate the normalized weights of each evaluation metric to generate a comprehensive score for calibration quality. The process of the analytic hierarchy process includes steps such as constructing a judgment matrix, calculating a weight vector, and consistency checking. The elements in the comprehensive evaluation matrix represent the relative importance of the th metric to the th target. Through the method of eigenvalue decomposition, the normalized weight vector w is obtained:

[0140] ;

[0141] where is a unit vector and is the maximum eigenvalue of the judgment matrix. The normalized weight vector is used to comprehensively score each evaluation metric, thereby obtaining a comprehensive score for calibration quality, reflecting the performance of the system during the overall calibration process. The comprehensive score for calibration quality, parameter distribution characteristic table, multi-time scale stability data, and calibration control effect assessment table are converted into charts through a data visualization engine to generate a wavelength power calibration report. The data visualization engine converts complex data information into an intuitive chart form, making the calibration report easier to understand. For example, a pie chart of parameter distribution, a line chart of calibration accuracy, a bar chart of multi-time scale stability, etc. are generated to help users intuitively understand the performance of each part during the calibration process.

[0142] The wavelength power calibration method of the wavelength tunable light source in the embodiment of the present application is described above. Next, the wavelength power calibration device 10 of the wavelength tunable light source in the embodiment of the present application will be described. Please refer to Figure 2 , an embodiment of the wavelength power calibration device 10 of the wavelength tunable light source in the embodiment of the present application includes:

[0143] The acquisition module 11 is used to collect the spectral and power of the output optical signals of the lasers in the S-band, C-band, and L-band, and generate a wavelength and power joint calibration parameter matrix.

[0144] The calculation module 12 is used to calculate the scores of the wavelength compensation coefficient, power compensation coefficient, and coupling influence coefficient in the wavelength and power joint calibration parameter matrix, and obtain a parameter importance score matrix.

[0145] The processing module 13 is used to construct a first adaptive control unit and a second adaptive control unit based on the parameter importance score matrix, generate a wavelength compensation value through the first adaptive control unit, and generate a power compensation value through the second adaptive control unit.

[0146] The compensation control module 14 is used to input the wavelength compensation value into the wavelength tuning mechanism and input the power compensation value into the power adjustment mechanism, perform joint compensation control on the laser, and obtain a calibrated output optical signal.

[0147] The measurement module 15 is used to perform wavelength and power tracking measurements on the calibrated output optical signal, and calculate and obtain calibration accuracy data.

[0148] The generation module 16 is used to generate a wavelength and power calibration report based on the calibration accuracy data, parameter importance score matrix, and wavelength and power joint calibration parameter matrix.

[0149] Through the collaborative cooperation of the above-mentioned various components, by constructing a wavelength and power joint calibration parameter matrix and combining with the density peak clustering algorithm, an accurate description of the coupling characteristics of wavelength drift and power fluctuation is achieved. Adopting a multi-level adaptive control unit architecture, including functional modules such as data preprocessing, feature extraction, compensation calculation, and output regulation, improves the robustness and adaptability of the calibration system. Introducing a parameter importance scoring mechanism, through algorithms such as matrix analysis and singular value decomposition, the reasonable configuration and dynamic optimization of calibration parameters are realized. A dual-loop control system is designed, and through the synergistic effect of the wavelength outer loop and the power inner loop, mutual interference is effectively suppressed, and the dynamic response performance of the system is improved. A perfect calibration effect evaluation system is established, including multiple dimensions such as wavelength accuracy, power stability, and short-term and long-term stability, ensuring the reliability of the calibration results. A professional data visualization system is developed, and through multi-dimensional displays such as parameter distribution, performance indicators, and stability evaluation, an intuitive calibration quality evaluation method is provided. The joint calibration of the entire band (S, C, L bands) is realized, making the wavelength accuracy and power stability significantly better than the traditional single-parameter calibration method.

[0150] Please refer to Figure 3 , Figure 3Schematic block diagram of the structure of the electronic device 300 provided by the embodiments of the present application. The electronic device 300 includes a processor 301 and a memory 302. The processor 301 and the memory 302 are connected through a device bus 203. Among them, the memory 302 may include a non-volatile storage medium and an internal memory.

[0151] The non-volatile storage medium can store a computer program. The computer program includes program instructions. When the program instructions are executed by the processor 301, the processor 301 can be made to execute any of the above wavelength power calibration methods of the wavelength tunable light source.

[0152] The processor 301 is used to provide computing and control capabilities to support the operation of the entire electronic device 300.

[0153] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor 301, the processor 301 can be made to execute any of the above wavelength power calibration methods of the wavelength tunable light source.

[0154] Those skilled in the art can understand that Figure 3 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the electronic device 300 involved in the solution of the present application. The specific electronic device 300 may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0155] It should be understood that the processor 301 may be a central processing unit (CPU), and the processor 301 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0156] It should be noted that those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described electronic device 300 can refer to the corresponding process of the above wavelength power calibration method of the wavelength tunable light source, and will not be elaborated here.

[0157] The embodiments of the present application further provide a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by one or more processors, the one or more processors are caused to implement the wavelength power calibration method of the wavelength tunable light source provided by the embodiments of the present application.

[0158] Among them, the computer-readable storage medium may be an internal storage unit of the electronic device 300 in the foregoing embodiment, such as the hard disk or memory of the electronic device 300. The computer-readable storage medium may also be an external storage device of the electronic device 300, such as a plug-in hard disk equipped with the electronic device 300, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc.

[0159] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described in detail here.

[0160] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0161] The above is the case. The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A wavelength power calibration method for a wavelength tunable light source, characterized in that: The method comprises: The spectrum and power of the laser output optical signals of the S-band, C-band and L-band are collected to generate the wavelength and power joint calibration parameter matrix; Scoring and calculating the wavelength compensation coefficient, the power compensation coefficient and the coupling influence coefficient in the wavelength and power joint calibration parameter matrix to obtain a parameter importance scoring matrix; Constructing a first adaptive control unit and a second adaptive control unit based on the parameter importance scoring matrix, generating a wavelength compensation value through the first adaptive control unit, and generating a power compensation value through the second adaptive control unit; Inputting the wavelength compensation value into a wavelength tuning mechanism, inputting the power compensation value into a power adjustment mechanism, performing joint compensation control on the laser, and obtaining a calibrated output optical signal; Performing wavelength and power tracking measurement on the calibrated output optical signal to calculate calibration accuracy data; A wavelength power calibration report is generated based on the calibration accuracy data, the parameter importance score matrix and the wavelength and power joint calibration parameter matrix.

2. The wavelength power calibration method of a wavelength tunable light source according to claim 1, characterized in that: The spectrum and power of the laser output optical signals in the S band, C band and L band are collected to generate a wavelength and power joint calibration parameter matrix, including: N wavelength sampling points are set in the S band, C band and L band respectively, and each wavelength sampling point is measured M times to obtain the original sampling data; According to the original sampling data, a real-time wavelength value is recorded by a spectrum analyzer, and a real-time power value is recorded by a power meter to obtain a real-time measurement data set; Performing a difference calculation between the wavelength value in the real-time measurement data set and the target wavelength value to obtain a wavelength drift amount, and performing a difference calculation between the power value in the real-time measurement data set and the target power value to obtain a power fluctuation amount; Arranging the wavelength drift amount and the power fluctuation amount according to a measurement time sequence respectively to generate a wavelength drift amount matrix and a power fluctuation amount matrix, and normalizing the wavelength drift amount matrix and the power fluctuation amount matrix to generate a normalized wavelength characteristic matrix and a normalized power characteristic matrix; Calculating the wavelength local density and wavelength distance of the normalized wavelength characteristic matrix, and calculating the power local density and power distance of the normalized power characteristic matrix; Based on the first product of the wavelength local density and the wavelength distance, the data points in the normalized wavelength characteristic matrix are sorted in descending order, and the data point with the largest first product is selected as the first cluster center point; and based on the second product of the power local density and the power distance, the data points in the normalized power characteristic matrix are sorted in descending order, and the data point with the largest second product is selected as the second cluster center point; Taking the first cluster center point and the second cluster center point as a reference, a wavelength compensation coefficient, a power compensation coefficient and a coupling influence coefficient are calculated, and the wavelength compensation coefficient, the power compensation coefficient and the coupling influence coefficient are combined to form a wavelength and power joint calibration parameter matrix.

3. The wavelength power calibration method of a wavelength tunable light source according to claim 2, characterized in that: The wavelength compensation coefficient, power compensation coefficient and coupling influence coefficient in the wavelength and power joint calibration parameter matrix are scored and calculated to obtain a parameter importance scoring matrix, including: Constructing a wavelength scoring unit, a power scoring unit and a coupling scoring unit based on the wavelength and power joint calibration parameter matrix, wherein the wavelength scoring unit is used to process the wavelength compensation coefficient, the power scoring unit is used to process the power compensation coefficient, and the coupling scoring unit is used to process the coupling influence coefficient; Calculate the wavelength statistical characteristic parameters of the wavelength drift matrix, the wavelength statistical characteristic parameters include the wavelength data distribution standard deviation and the wavelength data center mean, perform normal distribution calculation to obtain the wavelength weight distribution vector, and calculate the power statistical characteristic parameters of the power fluctuation matrix, the power statistical characteristic parameters include the power data distribution standard deviation and the power data center mean, perform normal distribution calculation to obtain the power weight distribution vector; Performing a cross-correlation operation on the wavelength drift matrix and the power fluctuation matrix to obtain a correlation matrix, and performing eigenvalue decomposition based on the correlation matrix to obtain a coupling weight distribution vector; The wavelength weight distribution vector, the power weight distribution vector and the coupling weight distribution vector are normalized to generate a normalized weight matrix, and a dot product operation is performed on the normalized weight matrix and the wavelength and power joint calibration parameter matrix to obtain a parameter scoring matrix; Each score value in the parameter score matrix is ​​mapped to the interval of 0-100 according to a linear mapping relationship to generate a quantitative score matrix, and a singular value decomposition operation is performed on the quantitative score matrix, and the right singular vector corresponding to the maximum singular value is selected as the parameter importance score matrix.

4. The wavelength power calibration method of a wavelength tunable light source according to claim 3, characterized in that: The step of constructing a first adaptive control unit and a second adaptive control unit based on the parameter importance score matrix, generating a wavelength compensation value by the first adaptive control unit, and generating a power compensation value by the second adaptive control unit includes: Constructing a first adaptive control unit and a second adaptive control unit, wherein the first adaptive control unit includes a first data preprocessing layer, a first feature extraction layer, a first compensation calculation layer and a first output adjustment layer, and the second adaptive control unit includes a second data preprocessing layer, a second feature extraction layer, a second compensation calculation layer and a second output adjustment layer; Input the real-time spectral data into the first data preprocessing layer, the first data preprocessing layer comprises a 5-layer wavelet transform network structure, each layer of the wavelet network comprises a decomposition unit and a reconstruction unit, the decomposition unit adopts the db4 wavelet basis function, the reconstruction unit adopts the soft threshold denoising method, and the denoised spectral data is obtained by layer-by-layer processing; Input the de-noised spectrum data into the first feature extraction layer, the first feature extraction layer comprises a spectrum main peak extraction sublayer, an edge feature extraction sublayer and a shape feature extraction sublayer, the spectrum main peak extraction sublayer adopts a 9-point quadratic polynomial fitting algorithm, the edge feature extraction sublayer adopts a Sobel edge detection operator, and the shape feature extraction sublayer adopts 15 shape descriptors to obtain a spectrum feature vector; The spectral feature vector and the parameter importance score matrix are input into the first compensation calculation layer, the first compensation calculation layer includes 3 fully connected layers, the number of nodes in the first fully connected layer is 128, the number of nodes in the second fully connected layer is 64, the number of nodes in the third fully connected layer is 32, and the ReLU activation function and the Dropout layer are used between each layer to obtain the wavelength compensation parameters; Inputting the wavelength compensation parameter into the first output adjustment layer, the first output adjustment layer comprises a proportional integral loop and a feedforward compensation loop, the proportional coefficient of the proportional integral loop is 0.8, the integral coefficient is 0.2, and the feedforward compensation loop uses a 5th order FIR filter to generate a wavelength compensation value; Inputting the real-time power data and the wavelength compensation value into the second data preprocessing layer, wherein the second data preprocessing layer comprises a data synchronization unit and a standardization unit, wherein the data synchronization unit adopts a nearest neighbor interpolation algorithm, and the standardization unit adopts a Z-score standardization algorithm, to obtain preprocessed power data; Inputting the preprocessed power data into the second feature extraction layer, the second feature extraction layer comprises a 7-layer time domain feature extraction network and a 5-layer frequency domain feature extraction network, the time domain feature extraction network adopts statistical moment features and trend features, and the frequency domain feature extraction network adopts wavelet packet decomposition and Hilbert transform to obtain a power feature vector; Based on the power characteristic vector, the wavelength compensation value and the parameter importance scoring matrix, cascade compensation control is performed in the second compensation calculation layer and the second output adjustment layer. The second compensation calculation layer adopts a dynamic matrix control algorithm, and the second output adjustment layer adopts a composite control structure that combines forward channel compensation and feedback channel compensation to generate a power compensation value.

5. The wavelength power calibration method of a wavelength tunable light source according to claim 4, characterized in that: The step of inputting the wavelength compensation value into a wavelength tuning mechanism, inputting the power compensation value into a power adjustment mechanism, and performing joint compensation control on the laser to obtain a calibrated output optical signal comprises: The wavelength compensation value is converted into voltage by a first digital-to-analog conversion unit to obtain a wavelength tuning control voltage, and the wavelength tuning control voltage is input into the wavelength tuning mechanism by a first driving amplifier circuit, wherein the first driving amplifier circuit comprises a three-stage amplifier unit, wherein the first stage is a differential amplifier, the second stage is a voltage follower, and the third stage is a power operational amplifier; Performing current conversion on the power compensation value through a second digital-to-analog conversion unit to obtain a power regulation control current, and inputting the power regulation control current into the power regulation mechanism through a second drive amplifier circuit, wherein the second drive amplifier circuit includes a constant current source unit and a current detection unit; Acquire the compensated initial calibration optical signal, convert the initial calibration optical signal into an electrical signal through a photoelectric conversion unit to obtain a photoelectric conversion signal, and digitally sample the photoelectric conversion signal through a third digital-to-analog conversion unit to obtain digital feedback data; The digital feedback data is input into a dual closed-loop control system, wherein the dual closed-loop control system comprises a wavelength outer-loop controller and a power inner-loop controller, wherein the wavelength outer-loop controller adopts a feedforward-feedback composite control structure, and the power inner-loop controller adopts an incremental PID algorithm, so as to obtain the control quantity output by the dual closed-loop control system, and respectively connect to the wavelength tuning mechanism and the power adjustment mechanism to obtain a calibrated output optical signal.

6. The wavelength power calibration method of a wavelength tunable light source according to claim 5, characterized in that: The step of performing wavelength and power tracking measurement on the calibrated output optical signal and calculating calibration accuracy data includes: The calibrated output optical signal is subjected to wavelength power sampling by a high-precision wavelength measurement unit and a power measurement unit respectively to obtain original wavelength power sampling data, and the original wavelength power sampling data is subjected to time synchronization and data alignment processing to generate a synchronous measurement data stream; Dividing the synchronous measurement data stream into bands according to S band, C band and L band to generate a sub-band measurement matrix, and performing wavelength reference correction on the wavelength data in the sub-band measurement matrix to obtain corrected wavelength measurement data; Calculating wavelength deviation parameters based on the corrected wavelength measurement data, the wavelength deviation parameters including instantaneous deviation, root mean square deviation and maximum deviation, to obtain a wavelength accuracy evaluation result; and calculating power deviation parameters based on the power data in the synchronous measurement data stream, the power deviation parameters including instantaneous deviation, root mean square deviation and maximum deviation, to obtain a power accuracy evaluation result; The wavelength accuracy evaluation results and the power accuracy evaluation results are subjected to band correlation analysis, performance evaluation matrices of the S band, C band and L band are constructed, and the performance evaluation matrices are subjected to index normalization processing to generate calibration accuracy data.

7. The wavelength power calibration method of a wavelength tunable light source according to claim 6, characterized in that: The generating a wavelength power calibration report based on the calibration accuracy data, the parameter importance score matrix and the wavelength and power joint calibration parameter matrix comprises: Arrange and analyze the wavelength and power joint calibration parameter matrix according to the band to generate a parameter distribution feature table, and input the parameter distribution feature table and the parameter importance score matrix into a data association processing unit to sort the parameter importance and generate a parameter optimization weight table; Performing statistical analysis on the wavelength root mean square error, wavelength maximum deviation, power root mean square error and power maximum deviation in the calibration accuracy data to generate a performance index analysis curve; Inputting the performance index analysis curve into a trend analysis unit to generate stability evaluation data, and dividing the stability evaluation data according to the time dimension to generate multi-time scale stability data; Comparing and analyzing the multi-time scale stability data with preset control indicators to generate a calibration control effect evaluation table; A comprehensive evaluation matrix is ​​constructed based on the parameter optimization weight table, the performance index analysis curve and the calibration control effect evaluation table, and a hierarchical analysis method is used to calculate the normalized weight of each evaluation index to generate a comprehensive calibration quality score; The calibration quality comprehensive score, the parameter distribution characteristic table, the multi-time scale stability data and the calibration control effect evaluation table are converted into charts through a data visualization engine to generate a wavelength power calibration report.

8. A wavelength power calibration device for a wavelength tunable light source, characterized in that: Used to perform the wavelength power calibration method of the wavelength tunable light source according to any one of claims 1 to 7, the wavelength power calibration device of the wavelength tunable light source comprising: The acquisition module is used to collect the spectrum and power of the laser output optical signals in the S-band, C-band and L-band, and generate a wavelength and power joint calibration parameter matrix; A calculation module, used for scoring and calculating the wavelength compensation coefficient, the power compensation coefficient and the coupling influence coefficient in the wavelength and power joint calibration parameter matrix to obtain a parameter importance scoring matrix; A processing module, configured to construct a first adaptive control unit and a second adaptive control unit based on the parameter importance scoring matrix, generate a wavelength compensation value through the first adaptive control unit, and generate a power compensation value through the second adaptive control unit; A compensation control module, used to input the wavelength compensation value into a wavelength tuning mechanism, input the power compensation value into a power adjustment mechanism, perform joint compensation control on the laser, and obtain a calibrated output optical signal; A measurement module, used to perform wavelength and power tracking measurement on the calibrated output optical signal, and calculate calibration accuracy data; A generation module is used to generate a wavelength power calibration report based on the calibration accuracy data, the parameter importance scoring matrix and the wavelength and power joint calibration parameter matrix.

9. An electronic device, characterized in that: The electronic device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the electronic device to execute the wavelength power calibration method for a wavelength tunable light source according to any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the wavelength power calibration method of the wavelength tunable light source according to any one of claims 1 to 7 is implemented.

Images