Photoelectronic device function degradation evaluation method, device and equipment and storage medium

Abstract

The invention discloses an optoelectronic device function degradation evaluation method, device and equipment and a storage medium. According to the method, the function degradation degree of the optoelectronic device is evaluated based on analysis of the operation state data of the optoelectronic device, so that prevention processing can be performed in advance based on the evaluation result, equipment faults caused by function degradation are eliminated in the bud, and normal operation of an optical fiber communication network is guaranteed.

Description

Technical Field

[0001] This application relates to the field of communication technology, specifically to a method, apparatus, device, and computer-readable storage medium for evaluating the functional degradation of optoelectronic devices. Background Technology

[0002] Optoelectronic devices are a crucial component of fiber optic communication systems, undertaking key functions such as optical signal transmission and reception, modulation and demodulation, multiplexing and amplification. During operation, various physical and chemical processes can lead to functional degradation in optoelectronic devices, such as electrostatic discharge damage, dielectric breakdown, lead corrosion, scratches, and breakage. If this functional degradation is not addressed promptly, it can cause significant communication failures, including service interruptions. With the widespread application of optoelectronic devices in communication systems, experts and scholars in academia and industry have conducted research on the reliability assessment, failure mechanisms and modes, and degradation processes of optoelectronic devices, proposing various detection and evaluation methods.

[0003] In related technologies, semiconductor device failure analysis methods acquire relevant parameters of the surface and cross-section of the failure point area through physical testing, including the area's shape, structure, color, and chemical element composition. These parameters are then input into a preset simulation algorithm for analysis to determine the cause of semiconductor device failure. Thermal conduction-based lossless failure analysis methods set multiple curve segments based on the inflection points of the device's integral and differential function structural curves. A segment-by-segment comparison method is used to identify the abnormal thermal resistance layer of the device. Tomographic scan data is used to reconstruct the structural image of the device under test, which is then compared with images of qualified devices to determine the cause of failure in optoelectronic devices. Component failure zero-reset analysis methods convert physical parameters such as electrical performance, thermal performance, mechanical performance, surface characteristics, and hermeticity into observable events. A fault dictionary is constructed that corresponds to single-mechanism causes and node failure characteristics. Failure feature vector analysis is used to determine the failure mechanism corresponding to the component failure mode, identify mechanistic factors and influencing factors, and propose targeted failure control measures. The near-field detection failure determination method utilizes the principles of electromagnetic injection and detection. It determines the electrical distribution on the surface of the device by using the electromagnetic field information of the target height plane of the device under test, and fits the standing wave current distribution curve to determine the failure location of the device.

[0004] While the probability of optoelectronic devices failing during operation is relatively low, the consequences of such failures are extremely serious. Existing technologies, such as those targeting electrical, thermal, and mechanical loads, cannot accurately uncover the patterns of functional degradation in optoelectronic devices and struggle to identify deterioration damage caused by the interplay of multiple factors and the superposition of various physicochemical processes. Currently, the intelligent operation and maintenance processes for fiber optic communication equipment urgently require new technologies and methods for detecting the functional degradation of optoelectronic devices, eliminating significant potential risks such as failures at their inception. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a method, apparatus, device, and computer-readable storage medium for evaluating the functional degradation of optoelectronic devices.

[0006] In a first aspect, embodiments of this application provide a method for evaluating the functional degradation of optoelectronic devices, the method comprising: Multiple operational status data related to the causes of functional degradation of optoelectronic devices are collected at various times, and the sample matrix constructed based on the operational status data is subjected to deviation standardization processing to obtain the sample deviation standardization matrix. The sample deviation normalization matrix is ​​subjected to singular value decomposition and truncation to obtain the truncated left singular matrix and feature data matrix; Based on the sample deviation normalization matrix, the truncated left singular matrix, and the feature data matrix, the interval distance vector corresponding to the labeled sample and the interval distance vector corresponding to the matrix to be evaluated are obtained. Obtain the measured degradation category corresponding to the label sample, and combine it with the interval distance vector corresponding to the label sample to obtain the weight parameter vector. Based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated, obtain the degradation evaluation category of the optoelectronic device.

[0007] In conjunction with the first aspect, in one implementation, the step of performing deviation standardization on the sample matrix constructed based on the operational status data to obtain a sample deviation standardization matrix includes: The average column vector is obtained by summing the elements of each row of the column vector in the sample matrix constructed based on the running status data and then dividing by the number of columns. Subtract the average column vector from each column of the sample matrix to obtain the sample deviation matrix; Standardize each column vector of the sample deviation matrix to obtain the standardized sample deviation matrix.

[0008] In conjunction with the first aspect, in one implementation, the step of performing singular value decomposition and truncation on the sample deviation normalization matrix to obtain a truncated left singular matrix and a feature data matrix includes: Singular value decomposition is performed on the sample deviation standardization matrix to obtain the left singular matrix, the singular value matrix, and the right singular matrix. Calculate the cutoff value based on the diagonal elements of the singular value matrix and the set error threshold. Based on the truncated values, the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix are obtained. The feature data matrix is ​​obtained based on the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix.

[0009] In conjunction with the first aspect, in one implementation, obtaining the interval distance vector corresponding to the label sample and the interval distance vector corresponding to the matrix to be evaluated based on the sample deviation normalization matrix, the truncated left singular matrix, and the feature data matrix includes: A reference matrix is ​​constructed by selecting r column vectors from the feature data matrix. The reference matrix consists of m rows and r columns, where m is the number of rows in the sample matrix. In the sample deviation standardization matrix, select a column vector to be detected, and multiply the column vector with a row vector whose elements are all 1 and the number of elements is r to obtain the matrix to be evaluated, which includes m rows and r columns. Based on the reference matrix, the matrix to be evaluated, and the truncated left singular matrix, the interval distance vector corresponding to the matrix to be evaluated is obtained; Select several label samples from the feature data matrix, where one of the label samples is a column vector in the feature data matrix; For each label sample, multiply the label sample by a row vector with all elements being 1 and the number of elements being r to obtain the label sample matrix. Based on the reference matrix, the label sample matrix, and the truncated left singular matrix, obtain the interval distance vector corresponding to the label sample.

[0010] In conjunction with the first aspect, in one implementation, obtaining the interval distance vector corresponding to the matrix to be evaluated based on the reference matrix, the matrix to be evaluated, and the truncated left singular matrix includes: The truncated left singular matrix is ​​transposed and multiplied with the reference matrix. Then, the result of multiplying the truncated left singular matrix transposed with the matrix to be evaluated is subtracted to obtain the difference matrix, which consists of h rows and r columns, where h is the truncated value. Multiply the difference matrix by its transpose to obtain the distance matrix, which consists of h rows and h columns; Extract the diagonal elements of the distance matrix and calculate the square root of each diagonal element to obtain the interval distance vector corresponding to the matrix to be evaluated. The interval distance vector includes h elements.

[0011] In conjunction with the first aspect, in one implementation, obtaining the measured degradation category corresponding to the label sample, and combining it with the interval distance vector corresponding to the label sample to obtain the weight parameter vector includes: Obtain the measured degradation category corresponding to each labeled sample; Construct a distance coefficient matrix based on the interval distance vectors corresponding to all labeled samples. , The j-th element in the interval distance vector corresponding to the i-th label sample point in the i-th row; based on The weight parameter calculation formula is constructed and solved for the measured degradation degree of all labeled samples, resulting in a weight parameter vector. The weight parameter calculation formula is as follows:

[0012] Among them, the weight parameter vector The i-th element This indicates the importance of the i-th element in the interval distance vector; express The transpose of the matrix, express Its transpose Multiply them, then find the inverse matrix; This represents the measured category of degradation level corresponding to the i-th label sample point.

[0013] In conjunction with the first aspect, in one implementation, the degradation assessment category of the optoelectronic device is obtained based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated, including: Calculate the product of the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated; The degradation assessment category of the optoelectronic device is determined based on the product.

[0014] Secondly, embodiments of this application provide a device for evaluating the functional degradation of optoelectronic devices, the device comprising: The feature data selection module is used to acquire operating status data related to the causes of functional degradation of optoelectronic devices collected at multiple times, and to perform deviation standardization processing on the sample matrix constructed based on the operating status data to obtain a sample deviation standardization matrix; and to perform singular value decomposition and truncation processing on the sample deviation standardization matrix to obtain a truncated left singular matrix and a feature data matrix. The interval distance calculation module is used to obtain the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated based on the sample deviation standardization matrix, the truncated left singular matrix and the feature data matrix. The degradation classification module is used to obtain the degradation assessment category of optoelectronic devices based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated.

[0015] Thirdly, embodiments of this application provide an optoelectronic device functional degradation assessment device, which includes a processor, a memory, and an optoelectronic device functional degradation assessment program stored in the memory and executable by the processor. When the optoelectronic device functional degradation assessment program is executed by the processor, it implements the steps of the optoelectronic device functional degradation assessment method as described in the first aspect.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing an optoelectronic device functional degradation assessment program, wherein when the optoelectronic device functional degradation assessment program is executed by a processor, it implements the steps of the optoelectronic device functional degradation assessment method as described in the first aspect.

[0017] The beneficial effects of the technical solutions provided in this application include: In this embodiment, operational status data related to the causes of functional degradation of optoelectronic devices are acquired at multiple times. A sample matrix constructed based on the operational status data is subjected to deviation standardization to obtain a sample deviation standardization matrix. Singular value decomposition and truncation are then performed on the sample deviation standardization matrix to obtain a truncated left singular matrix and a feature data matrix. Based on the sample deviation standardization matrix, the truncated left singular matrix, and the feature data matrix, the interval distance vector corresponding to the label sample and the interval distance vector corresponding to the matrix to be evaluated are obtained. The measured degradation degree category corresponding to the label sample is obtained, and a weight parameter vector is obtained by combining the interval distance vector corresponding to the label sample. Based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated, the degradation degree evaluation category of the optoelectronic device is obtained. Through this embodiment, based on the analysis of the operational status data of optoelectronic devices, the degree of functional degradation of optoelectronic devices is evaluated, allowing for preventative measures to be taken in advance based on the evaluation results. This eliminates equipment failures caused by functional degradation in their early stages, ensuring the normal operation of the optical fiber communication network. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating an embodiment of the optoelectronic device functional degradation assessment method of this application; Figure 2 A flowchart illustrating an embodiment of a method for selecting feature data; Figure 3 This is a flowchart illustrating an embodiment of the interval distance calculation method; Figure 4 A flowchart illustrating an embodiment of a degradation degree classification method; Figure 5 A schematic diagram illustrating a method for classifying the degree of degradation; Figure 6 A schematic diagram of the architecture of an embodiment of a system for evaluating the functional degradation of optoelectronic devices; Figure 7 This is a schematic diagram of the functional modules of an embodiment of the optoelectronic device functional degradation assessment device of this application; Figure 8 This is a schematic diagram of the hardware structure of the optoelectronic device functional degradation assessment device involved in the embodiments of this application. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0020] First, the main concepts in this application are explained to enable those skilled in the art to understand this application.

[0021] Fiber optic communication equipment contains numerous optoelectronic devices. During operation, these devices can experience functional degradation due to various physical or chemical changes such as radiation damage, ion contamination, pin corrosion, and oxide layer deformation, affecting equipment performance and even causing network service interruptions. To build high-quality, highly reliable communication networks, fiber optic communication equipment requires timely identification of optoelectronic device functional degradation, locating degraded devices, and proactively eliminating potential risks through path switching and device replacement to ensure the normal operation of upper-layer network services. To address these requirements, this application proposes a method for detecting optoelectronic device functional degradation, including three specific implementation methods: feature data selection, interval distance calculation, and degradation degree classification. Specifically: I. Feature Data Selection Methods This application proposes a multi-dimensional sample data principal feature selection method. This method comprehensively evaluates all data features that lead to the functional degradation of optoelectronic devices, selecting the most important features as principal features to participate in subsequent interval distance calculations and degradation degree classification. The causes of optoelectronic device functional degradation mainly include five categories: bulk degradation such as dielectric breakdown or neutron radiation; surface degradation such as ion contamination or material composite degradation; encapsulation degradation such as lead cracks or damage / leakage; degradation caused by oxidation / sulfidation or moisture-sensitive deformation; and degradation caused by mechanical vibration or bonding applications. This application uses... , , , , These five types of causes represent the reasons why the relevant state characteristic values ​​of optoelectronic devices change during operation. After the fiber optic communication equipment management system collects the state characteristic values ​​reported by the optoelectronic devices, it constructs five sample data matrices. , , , , The elements in the matrix correspond to the operational status data caused by the five reasons mentioned above. In the five sample data matrices, each row represents a data feature, and each column represents the operational status data at a collection time. The matrix formed by combining the five sample data matrices is the sample matrix of the optoelectronic device, as shown in formula (1):

[0022] In the above formula (1), the sample matrix has m rows and n columns. The j-th row represents the j-th feature, and the k-th column represents the running status data at acquisition time k. In the sample matrix, the units of measurement for each element are different, and their values ​​are also different. For example, the unit of temperature is degrees Celsius, and the unit of bias current is milliamperes. Different units of measurement result in different ranges of data values, leading to problems with accuracy and convergence during matrix processing. To solve this problem, the sample matrix is ​​standardized. Specifically: First, all columns of the sample matrix are summed, and the result is divided by the number of columns n to obtain the average column vector. Then, the average column vector is subtracted from each column to obtain the sample deviation matrix. Each column vector of the sample deviation matrix is ​​standardized to obtain the sample deviation standardized matrix, denoted as […]. The above calculation process corresponds to formula (2) shown below.

[0023]

[0024] In formula (2) above, the symbol Represents the average column vector; symbol This represents the deviation column vector obtained by subtracting the mean column vector from the k-th column vector in the sample matrix. express The model, for example, The value is (1, 2, 3). The model is ;symbol This represents the final calculated sample deviation normalization matrix, containing all deviation column vectors. It is an m x n matrix.

[0025] The standardized sample deviation matrix calculated above is subjected to singular value decomposition to obtain the left singular matrix, singular value matrix, and right singular matrix. The diagonal elements of the decomposed singular value matrix are arranged from largest to smallest. The minimum h value is calculated such that the ratio of the sum of the first h elements on the diagonal to the sum of all elements is greater than or equal to the set threshold value d. The calculation process is as follows: formula (3).

[0026]

[0027] In formula (3) above, the symbol Represents the diagonal elements of the singular value matrix, with the first element denoted as... The second is represented as And so on. Symbols Represents a left singular matrix, with the symbol Denotes a right singular matrix, with the symbol Represents a singular value matrix, symbol Denotes a truncated left singular matrix, symbol Denotes a truncated right singular matrix, symbol Denotes the truncated singular value matrix, symbol This represents the truncated sample deviation standardization matrix (i.e., the feature data matrix). Characteristic data that characterizes the functional degradation of electronic devices.

[0028] II. Interval Distance Calculation Method This application proposes a method for calculating the distance between optoelectronic devices. The distance between the devices is used to measure the degree of degradation; a larger distance indicates a greater degree of degradation, and a smaller distance indicates a lesser degree of degradation. (In the feature data matrix...) Select r column vectors to construct a reference matrix , Includes m rows and r columns.

[0029] In the sample deviation standardization matrix Select a column vector to be detected, and multiply it by a row vector with r elements, all of which are 1, to obtain the matrix to be evaluated. , It includes m rows and r columns. For example, selecting the sample deviation standardization matrix. The j-th column vector Matrix to be evaluated The construction process is shown in formula (4).

[0030]

[0031] The truncated left singular matrix After transpose and reference matrix Multiply, then subtract the truncated left singular matrix. Transposed and the matrix to be evaluated The result of multiplication is the difference matrix. , Includes h rows and r columns. Difference matrix Rather than transpose Multiply to obtain the distance matrix , Includes h rows and h columns. Extract the distance matrix. Calculate the square root of each diagonal element to construct the interval distance vector. , It includes h elements. The calculation process is as follows: formula (5).

[0032]

[0033] In formula (5) above, the symbol The transpose of a truncated left singular matrix is ​​represented by the symbol. Represents the reference matrix, symbol Represents the matrix to be evaluated, with the symbol... Represents the difference matrix, with the symbol... This represents the distance matrix. The symbol `diag()` selects the diagonal elements of the matrix, constructs a new vector, and then takes the square root of each element to obtain the interval distance vector. .

[0034] III. Classification Methods for Degradation This application proposes a degradation degree classification method based on interval distance vectors. The degradation category to which the optoelectronic device belongs is calculated. The optoelectronic devices are divided into t categories based on their degradation degree, represented as a row vector. In the feature data matrix In this process, a subset of sample points are selected as label samples. Using instruments or expert assessment, the degradation level category to which these label samples belong is determined. The interval distance calculation method described earlier is used to obtain the interval distance vector of the label samples. If the label sample points contain... There are h column vectors, and since the distance vector corresponding to each sample point column vector contains h elements, the distance vector elements of the label sample points can form a distance coefficient matrix. , The j-th element in the interval distance vector corresponding to the i-th label sample point in the i-th row is used. The weight parameter values ​​of each element in the interval distance vector are calculated using the following formula (6).

[0035]

[0036] In formula (6) above, the symbol This represents the j-th element of the interval distance vector corresponding to the i-th labeled sample point. For example, the symbol... This indicates that the value is the third element of the distance vector corresponding to the second labeled sample point. (Symbol) This represents the i-th element of the weight parameter values. (Symbol) This represents the degradation category to which the i-th labeled sample point belongs. Using symbols... This represents the distance coefficient matrix, which is the leftmost matrix in formula (6) above. Because a large amount of data is needed to improve the detection accuracy in the process of detecting the functional degradation of optoelectronic devices, the number of column vectors of the label sample points is... The number of elements in the distance vector is greater than the number of rows in the distance coefficient matrix. This matrix is ​​an overdetermined matrix with no exact solution, but it has a least squares approximate solution. The weight parameters are solved using formula (7).

[0037]

[0038] In formula (7) above, the symbol Represents the distance coefficient matrix The transpose of the matrix, sign Represents the distance coefficient matrix Its transpose Multiply the matrices and then find the inverse matrix. The weight parameter vector is obtained through the rightmost operation of formula (7). The i-th element in this vector Represents the interval distance vector The importance of the i-th element value. In the process of detecting functional degradation of optoelectronic devices, this is combined with the calculated interval distance vector of the samples to be evaluated. Then use the formula Obtain optoelectronic devices in The corresponding degradation assessment category at the time of data acquisition.

[0039] This application embodiment achieves functional degradation detection of optoelectronic devices based on the feature data selection method, interval distance calculation method, and degradation degree classification method described above. Fiber optic communication network equipment collects operating status data of optoelectronic devices, reports it to the equipment management system, and uses the above methods to assess the degradation degree category, allowing for targeted processing of severely degraded optoelectronic devices.

[0040] In summary, the technical solutions conceived through the embodiments of this application have the following beneficial effects compared with the prior art: (1) The present application proposes a feature data selection method, which solves the problem of large iterative calculation error caused by different measurement units of functional degradation factors, extracts key feature data, supports numerical calculation, and improves detection accuracy.

[0041] (2) The embodiments of this application propose an interval distance calculation method, which uses multidimensional spatial deviation distance to measure the degree of functional degradation of optoelectronic devices. The projection components of degradation factors are calculated in the truncated left singular matrix, and an interval distance vector is constructed to realize the scientific measurement of the functional degradation results of optoelectronic devices caused by physical or chemical reaction processes.

[0042] (3) This application proposes a degradation degree classification method. By constructing a coefficient matrix and solving the linear equation of the overdetermined matrix, a weight parameter vector is obtained, and the degradation degree of the sample points to be evaluated is accurately classified. For optoelectronic devices whose degradation degree category exceeds the set threshold value, proactive prevention and control are carried out to eliminate potential failure risks and fault hazards.

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0044] In a first aspect, embodiments of this application provide a method for evaluating the functional degradation of optoelectronic devices.

[0045] In one embodiment, reference is made to Figure 1 , Figure 1 This is a schematic flowchart of an embodiment of the optoelectronic device functional degradation assessment method of this application. Figure 1 As shown, the methods for evaluating the functional degradation of optoelectronic devices include: Step S10: Obtain operating status data related to the causes of functional degradation of optoelectronic devices collected at multiple times, and perform deviation standardization processing on the sample matrix constructed based on the operating status data to obtain the sample deviation standardization matrix. Step S20: Perform singular value decomposition and truncation on the sample deviation normalization matrix to obtain the truncated left singular matrix and feature data matrix. Furthermore, in one embodiment, the sample matrix constructed based on the operational status data undergoes deviation standardization processing to obtain a sample deviation standardization matrix, including: The average column vector is obtained by summing the elements of each row of the column vectors in the sample matrix constructed based on the running status data and then dividing by the number of columns. The average column vector is then subtracted from each column of the sample matrix to obtain the sample deviation matrix. Each column vector of the sample deviation matrix is ​​then standardized to obtain the standardized sample deviation matrix.

[0046] Further, in one embodiment, step S20 includes: The sample deviation normalization matrix is ​​subjected to singular value decomposition to obtain a left singular matrix, a singular value matrix, and a right singular matrix. Based on the diagonal elements of the singular value matrix and a set error threshold, a truncated value is calculated. Based on the truncated value, the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix are obtained. Based on the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix, the feature data matrix is ​​obtained.

[0047] In this embodiment, refer to Figure 2 , Figure 2 A flowchart illustrating an embodiment of a method for selecting feature data. For example... Figure 2 As shown, feature data selection includes: S11, Select an optoelectronic device to be tested, extract operational status data related to the causes of functional degradation of the optoelectronic device from the operational status monitoring data collected at multiple times for that optoelectronic device, and construct a sample matrix. Each row of the matrix corresponds to a feature, and each column of the matrix corresponds to a sample point.

[0048] S12, all columns of the sample matrix Summing the results and dividing by the number of columns yields the average column vector. Subtracting the average column vector from each column gives the sample deviation matrix. Standardizing the column vectors gives the standardized sample deviation matrix. .

[0049] S13, Standardization of the sample deviation matrix Perform singular value decomposition According to the singular value matrix By truncating the diagonal elements and setting error thresholds, a feature data matrix containing the functional degradation of electronic devices is obtained. . As shown in step S11, feature data is extracted from the optoelectronic device operating status data reported by the fiber optic communication equipment. This feature data includes real-time performance data of optoelectronic devices after functional degradation caused by five physical or chemical reactions: internal degradation such as dielectric breakdown or neutron radiation, surface degradation such as ion contamination or material composite degradation, encapsulation degradation such as lead cracks or damage leakage, degradation caused by oxidation, sulfidation or moisture-sensitive deformation, and degradation caused by mechanical vibration or bonding applications. A sample matrix is ​​constructed using this feature data. Table 1 shows examples of operating status data collected for optoelectronic devices at multiple times related to the causes of optoelectronic device functional degradation.

[0050] Table 1

[0051] In Table 1, the first column lists the names and units of measurement for four characteristics of optoelectronic devices: temperature, bias current, luminous power, and luminous power. The corresponding units of measurement for these four characteristics are degree, milliampere, decibel, and decibel, respectively.

[0052] The first row of Table 1 lists the sample point numbers; for example, with 10 sample points. In the example above, the number 61.8 in the second row and second column of Table 1 indicates that the temperature of the optoelectronic device at time 1 was 61.8 degrees Celsius. In Table 1, the four characteristics correspond to two types of functional degradation causes: temperature corresponds to degradation caused by oxidation and sulfidation or moisture-sensitive deformation, while bias current, luminous power, and luminous power correspond to degradation in vivo such as dielectric breakdown or neutron radiation.

[0053] In step S11, a sample matrix can be constructed using the data in Table 1, with 4 rows and 10 columns, denoted as follows: The four rows of the matrix correspond to four features, and the ten column vectors of the matrix correspond to ten sample points.

[0054] In step S12, the 10 column vectors of the sample matrix are summed to obtain a sum vector, which is then divided by the number of columns in the sample matrix to obtain the average column vector. Subtracting the mean column vector from each column of the sample matrix yields the sample deviation matrix. Standardizing the column vectors yields the standardized sample deviation matrix. ,as follows:

[0055] As shown in step S13, singular value decomposition is performed on the sample deviation standardization matrix. The singular value matrix is ​​obtained. The diagonal elements are {2.8, 1.4, 0.39, 0.05}. Assuming the error threshold is 0.98, according to the previous formula (3), the pre-truncation value h is calculated to be 3, that is:

[0056] In practical engineering, error thresholds can be set based on the experience of business experts. In this embodiment, the threshold is 0.98, which means that the feature data truncated based on this threshold already contains 98% of the state information of the optoelectronic device's operating status. Based on this threshold, the truncated value is 3, and the truncated left singular matrix elements are as follows:

[0057] The feature data matrix is ​​calculated using the truncated left singular matrix, the truncated singular value vector matrix, and the truncated right singular matrix. The calculation formula is as follows: The calculation results are as follows:

[0058] At this point, the feature data matrix can be obtained. . Key information from the sample deviation standardization matrix was preserved.

[0059] Step S30: Based on the sample deviation normalization matrix, the truncated left singular matrix, and the feature data matrix, obtain the interval distance vector corresponding to the label sample and the interval distance vector corresponding to the matrix to be evaluated; Further, in one embodiment, step S30 includes: Step S301: Select r column vectors from the feature data matrix to construct a reference matrix. The reference matrix consists of m rows and r columns, where m is the number of rows in the sample matrix. Step S302: Select a column vector to be detected in the sample deviation standardization matrix, and multiply the column vector with a row vector whose elements are all 1 and whose number of elements is r to obtain the matrix to be evaluated. The matrix to be evaluated includes m rows and r rows. Step S303: Based on the reference matrix, the matrix to be evaluated, and the truncated left singular matrix, obtain the interval distance vector corresponding to the matrix to be evaluated; Further, in one embodiment, step S303 includes: The truncated left singular matrix is ​​transposed and multiplied by the reference matrix. Then, the result of multiplying the truncated left singular matrix transposed by the matrix to be evaluated is subtracted to obtain the difference matrix, which consists of h rows and r columns, where h is the truncated value. The difference matrix is ​​multiplied by its transpose to obtain the distance matrix, which consists of h rows and h columns. The diagonal elements of the distance matrix are extracted, and the square root of each diagonal element is calculated to obtain the interval distance vector corresponding to the matrix to be evaluated, which consists of h elements.

[0060] In this embodiment, refer to Figure 3 , Figure 3 This is a flowchart illustrating an embodiment of the interval distance calculation method. Figure 3 As shown, the interval distance calculation method includes: S21, in the feature data matrix Select r column vectors to construct a reference matrix The matrix consists of m rows and r columns. From the sample deviation standardization matrix, a column vector to be tested is selected, and this column vector is multiplied by a row vector containing r elements, all of which are 1, to obtain the matrix to be evaluated. The matrix consists of m rows and r columns.

[0061] S22, the truncated left singular matrix After transpose and reference matrix Multiply, then subtract the truncated left singular matrix. Transposed and the matrix to be evaluated The result of multiplication is a difference matrix P, which has h rows and r columns. Multiplying the difference matrix by its transpose yields a distance matrix Q, which has h rows and h columns.

[0062] S23, extract the diagonal elements of the distance matrix Q, calculate the square root of each element, and construct the interval distance vector q, which includes h elements. As shown in step S21, construct the reference matrix and the matrix to be evaluated. The feature data matrix constructed earlier... In this process, four column vectors are selected, i.e., the number of column vectors is r=4, to construct the reference matrix. Select the sample deviation standardization matrix. The fifth column vector is used as the sample point vector to be evaluated. This column vector is multiplied by a row vector with all elements equal to 1 and a total of 4 elements to obtain the matrix to be evaluated. Reference matrix With the matrix to be evaluated as follows:

[0063] As shown in step S22, following the described process, the difference matrix is ​​obtained. With distance matrix as follows:

[0064] As shown in step S23, the distance matrix is ​​extracted. Given the diagonal elements {0.52, 1.73, 0.07}, calculate the square root of each diagonal element to obtain the interval distance vector. ={0.72,1.32,0.26} T .

[0065] Step S304: Select several label samples from the feature data matrix, wherein one of the label samples is a column vector in the feature data matrix; Step S305: For each label sample, multiply the label sample with a row vector whose element values ​​are all 1 and whose number of elements is r to obtain the label sample matrix. Based on the reference matrix, the label sample matrix and the truncated left singular matrix, obtain the interval distance vector corresponding to the label sample.

[0066] Referring to the example of obtaining the interval distance vector corresponding to the matrix to be evaluated, in the feature data matrix In the above, column vectors 6 to 10 are selected as label samples, and the distance vector between these five label samples is calculated to be {0.66, 1.61, 0.26}. T {0.71, 1.41, 0.26} T {3.46, 1.32, 0.24} T {0.65, 1.49, 0.74} T {0.61, 1.24, 0.5} T .

[0067] Step S40: Obtain the measured degradation category corresponding to the label sample, and obtain the weight parameter vector by combining the interval distance vector corresponding to the label sample. Obtain the degradation evaluation category of the optoelectronic device based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated.

[0068] Further, in one embodiment, obtaining the measured degradation category corresponding to the label sample and combining it with the interval distance vector corresponding to the label sample to obtain the weight parameter vector includes: Obtain the measured degradation category corresponding to each labeled sample; Construct a distance coefficient matrix based on the interval distance vectors corresponding to all labeled samples. , The j-th element in the interval distance vector corresponding to the i-th label sample point in the i-th row; based on The weight parameter calculation formula is constructed and solved for the measured degradation degree of all labeled samples, resulting in a weight parameter vector. The weight parameter calculation formula is as follows:

[0069] Among them, the weight parameter vector The i-th element This indicates the importance of the i-th element in the interval distance vector; express The transpose of the matrix, express Its transpose Multiply them, then find the inverse matrix; This represents the measured category of degradation level corresponding to the i-th label sample point.

[0070] Furthermore, in one embodiment, the degradation assessment category of the optoelectronic device is obtained based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated, including: Calculate the product of the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated; determine the degradation assessment category of the optoelectronic device based on the product.

[0071] In this embodiment, refer to Figure 4 , Figure 4 This is a flowchart illustrating an embodiment of a method for classifying the degree of degradation. Figure 4 As shown, the methods for classifying the degree of degradation include: S31, in the feature data matrix In this process, a subset of column vectors are selected as label samples. Using instruments or expert assessment, the measured degradation category to which each label sample belongs is determined. The interval distance vector of the label samples is then obtained using an interval distance calculation method. S32. Using the distance vector of the label sample as the distance coefficient matrix, the measured degradation category to which the label sample belongs as the dependent variable, and the weight parameter vector as the independent variable, a system of linear equations is constructed. The approximate optimal solution of the system of equations is calculated to obtain the weight parameter vector.

[0072] S33, calculate the distance vector between the sample points to be evaluated using the formula. Calculate the degradation assessment category to which the sample point belongs.

[0073] As shown in step S31, label samples are selected from the feature data matrix, with each label sample corresponding to a measured degradation level category. In practical engineering, based on the operation and maintenance needs of fiber optic communication equipment, numerical values ​​can be used to represent the degree of functional degradation of optoelectronic devices, with each value representing a category. For example, using row vectors... This represents four categories: normal, slight degradation, moderate degradation, and severe degradation. (In the feature data matrix) In the dataset, column vectors 6 through 10 are selected as label samples, and the distance vector between these five samples is calculated to be {0.66, 1.61, 0.26}. T {0.71, 1.41, 0.26} T {3.46, 1.32, 0.24} T {0.65, 1.49, 0.74} T {0.61, 1.24, 0.5} T .

[0074] As shown in step S32, the distance vectors between the five labeled samples are used as the distance coefficient matrix, and the degradation category to which the sample belongs is used as the dependent variable to construct a system of linear equations. The distance coefficient matrix is ​​3 rows and 4 columns, with each row corresponding to the distance vector of a labeled sample and each column corresponding to a labeled sample. The independent variable is the weight parameter vector, which includes three elements, and the weight parameter vector is represented as follows: The dependent variable vector represents the degradation level categories to which the five labeled samples belong, expressed as... The system of linear equations constructed based on the distance coefficient matrix, independent variables, and dependent variables above is as follows:

[0075] Using formula Solving this system of linear equations yields an approximate optimal solution for the independent variables. The three elements of this solution set constitute the weight parameter vector. In the formula, and The values ​​are shown below:

[0076] As described in step S33, the weight parameter vector mentioned above is used. ={0.85,0.81,0.01} and the interval distance vector between the sample points to be evaluated. ={0.72,1.32,0.26} TThe classification result of the degradation degree of the sample point to be evaluated is calculated, and the result is rounded to the nearest integer. The final value is 2, which means that the classification result number of the sample point to be evaluated is 2, corresponding to moderate degradation.

[0077] Furthermore, referring to Figure 5 , Figure 5 A schematic diagram illustrating a method for classifying the degree of degradation. Figure 5 The bottom left corner is based on the truncated left singular matrix. The constructed vector space has a feature data matrix containing 10 sample points in the lower right corner. , The first four column vectors serve as reference samples, the fifth column vector as the sample to be evaluated (it should be noted that, for ease of explanation, this example selects the sample to be evaluated from the feature data matrix to construct the evaluation matrix; alternatively, a column vector to be detected can be selected from the sample deviation normalization matrix as the reference sample), and the sixth to tenth column vectors serve as label samples. The reference samples and the samples to be evaluated in the matrix are projected into the vector space, and the interval distance vector is calculated. Using the labeled samples, the weight parameter vector is calculated. . Figure 5 The above represents the interval distance vector. and weight parameter vector Enter formula The degree of degradation is then calculated and categorized.

[0078] Furthermore, referring to Figure 6 , Figure 6 This is a schematic diagram of the architecture of an embodiment of an optoelectronic device functional degradation assessment system. The system includes eight modules and six types of interactive interfaces. The status data acquisition module and status data reporting module are located within each network element, while the status data receiving module, feature data extraction module, interval distance calculation module, weight parameter calculation module, degradation detection and report generation module, and effect evaluation and feedback module are located on the network management platform. Each network element includes interactive interface I1, and the network management platform includes interactive interfaces I3, I4, I5, and I6. Network elements report status data to the network management platform through interactive interface I2.

[0079] Figure 6 At the bottom are the network elements, which house the status data acquisition module and the status data reporting module. In the fiber optic communication network, the status data acquisition module of each network element collects the operating status data of optoelectronic devices according to the set period, transmits it to the status data reporting module through the interaction interface I1, assembles the messages according to the protocol format between the device and the network management platform, and reports it to the network management platform through the interaction interface I2.

[0080] Figure 6The above describes the network management platform, which reads the status data reported by each network element through the status data receiving module and saves it to the database. The detection system calls the feature data extraction module, obtains the status data through the interaction interface I3, executes the feature data selection method, and calculates a feature data matrix containing the main information on the functional degradation of optoelectronic devices. The detection system uses the interval distance calculation module to obtain the distance between the sample to be evaluated and the reference sample, and calls the weight parameter calculation module to obtain a weight parameter vector that can correlate the interval distance with the degradation level category. The degradation detection and report generation module obtains the interval distance and weight parameter vector of the sample to be evaluated through the interaction interface I5, calculates the degradation level category to which the sample to be evaluated belongs, and creates an optoelectronic device functional degradation detection report. The optoelectronic device functional degradation assessment system provided in this application embodiment includes a back-feedback module for effect evaluation. This module can present the detection report to optical fiber communication network operation and maintenance experts, evaluate the detection results, and transmit the evaluation results to the weight parameter calculation module through the interaction interface I6. After receiving the evaluation result feedback data, the weight parameter calculation module of the optoelectronic device functional degradation assessment system adjusts the degradation level category to which the label sample belongs. This adjustment enables the weight parameter vector to more accurately correlate the interval distance with the degradation level category.

[0081] The optoelectronic device functional degradation assessment system provided in this application can create an optoelectronic device functional degradation detection report each time the degradation detection and report generation module is called. Based on the detection results provided in the report, optoelectronic devices whose functional degradation exceeds a set threshold are identified, and preventive measures are taken in advance to eliminate potential risks and ensure the stable operation of fiber optic communication equipment.

[0082] It should be noted that, for the sake of brevity and ease of understanding, the calculation results in this application's embodiments have been truncated to two decimal places. The rounding error caused by truncation may result in calculation errors in the values ​​of related matrix and vector elements; however, this error will not affect the description of the relevant steps in the embodiments. In practical engineering applications, the method of this application's embodiments can be improved by retaining more decimal places, using double-precision floating-point operations, or other methods to reduce calculation errors.

[0083] Secondly, embodiments of this application also provide a device for evaluating the functional degradation of optoelectronic devices.

[0084] In one embodiment, reference is made to Figure 7 , Figure 7 This is a schematic diagram of the functional modules of an embodiment of the optoelectronic device functional degradation assessment device of this application. Figure 7 As shown, the optoelectronic device functional degradation assessment device includes: The feature data selection module 10 is used to acquire multiple operational status data collected at different times related to the causes of functional degradation of the optoelectronic device, and to perform deviation standardization processing on the sample matrix constructed based on the operational status data to obtain a sample deviation standardization matrix; and to perform singular value decomposition and truncation processing on the sample deviation standardization matrix to obtain a truncated left singular matrix and a feature data matrix. The interval distance calculation module 20 is used to obtain the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated based on the sample deviation standardization matrix, the truncated left singular matrix and the feature data matrix. The degradation degree classification module 30 is used to obtain the degradation degree evaluation category of optoelectronic devices based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated.

[0085] Furthermore, in one embodiment, the feature data selection module 10 is used for: The average column vector is obtained by summing the elements of each row of the column vector in the sample matrix constructed based on the running status data and then dividing by the number of columns. Subtract the average column vector from each column of the sample matrix to obtain the sample deviation matrix; Standardize each column vector of the sample deviation matrix to obtain the standardized sample deviation matrix.

[0086] Furthermore, in one embodiment, the feature data selection module 10 is used for: Singular value decomposition is performed on the sample deviation standardization matrix to obtain the left singular matrix, the singular value matrix, and the right singular matrix. Calculate the cutoff value based on the diagonal elements of the singular value matrix and the set error threshold. Based on the truncated values, the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix are obtained. The feature data matrix is ​​obtained based on the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix.

[0087] Furthermore, in one embodiment, the interval distance calculation module 20 is used for: A reference matrix is ​​constructed by selecting r column vectors from the feature data matrix. The reference matrix consists of m rows and r columns, where m is the number of rows in the sample matrix. In the sample deviation standardization matrix, select a column vector to be detected, and multiply the column vector with a row vector whose elements are all 1 and the number of elements is r to obtain the matrix to be evaluated, which includes m rows and r columns. Based on the reference matrix, the matrix to be evaluated, and the truncated left singular matrix, the interval distance vector corresponding to the matrix to be evaluated is obtained; Select several label samples from the feature data matrix, where one of the label samples is a column vector in the feature data matrix; For each label sample, multiply the label sample by a row vector with all elements being 1 and the number of elements being r to obtain the label sample matrix. Based on the reference matrix, the label sample matrix, and the truncated left singular matrix, obtain the interval distance vector corresponding to the label sample.

[0088] Furthermore, in one embodiment, the interval distance calculation module 20 is used for: The truncated left singular matrix is ​​transposed and multiplied with the reference matrix. Then, the result of multiplying the truncated left singular matrix transposed with the matrix to be evaluated is subtracted to obtain the difference matrix, which consists of h rows and r columns, where h is the truncated value. Multiply the difference matrix by its transpose to obtain the distance matrix, which consists of h rows and h columns; Extract the diagonal elements of the distance matrix and calculate the square root of each diagonal element to obtain the interval distance vector corresponding to the matrix to be evaluated. The interval distance vector includes h elements.

[0089] Furthermore, in one embodiment, the degradation degree classification module 30 is used for: Obtain the measured degradation category corresponding to each labeled sample; Construct a distance coefficient matrix based on the interval distance vectors corresponding to all labeled samples. , The j-th element in the interval distance vector corresponding to the i-th label sample point in the i-th row; based on The weight parameter calculation formula is constructed and solved for the measured degradation degree of all labeled samples, resulting in a weight parameter vector. The weight parameter calculation formula is as follows:

[0090] Among them, the weight parameter vector The i-th element This indicates the importance of the i-th element in the interval distance vector; express The transpose of the matrix, express Its transpose Multiply them, then find the inverse matrix; This represents the measured category of degradation level corresponding to the i-th label sample point.

[0091] Furthermore, in one embodiment, the degradation degree classification module 30 is used for: Calculate the product of the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated; The degradation assessment category of the optoelectronic device is determined based on the product.

[0092] The functions of each module in the aforementioned optoelectronic device function degradation assessment device correspond to the steps in the aforementioned optoelectronic device function degradation assessment method embodiment, and their functions and implementation processes will not be described in detail here.

[0093] Thirdly, embodiments of this application provide an optoelectronic device functional degradation assessment device, which can be a personal computer (PC), laptop computer, server, or other device with data processing capabilities.

[0094] Reference Figure 8 , Figure 8 This is a schematic diagram of the hardware structure of the optoelectronic device functional degradation assessment device involved in the embodiments of this application. In the embodiments of this application, the optoelectronic device functional degradation assessment device may include a processor, a memory, a communication interface, and a communication bus.

[0095] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0096] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting devices within the optoelectronic device functional degradation assessment device, as well as interfaces used for interconnecting the optoelectronic device functional degradation assessment device with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0097] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0098] The processor can be a general-purpose processor, which can call the optoelectronic device functional degradation assessment program stored in the memory and execute the optoelectronic device functional degradation assessment method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the optoelectronic device functional degradation assessment program is called can refer to the various embodiments of the optoelectronic device functional degradation assessment method of this application, and will not be repeated here.

[0099] Those skilled in the art will understand that Figure 8 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0100] Fourthly, embodiments of this application also provide a computer-readable storage medium.

[0101] The present application stores a computer-readable storage medium containing a program for evaluating the functional degradation of optoelectronic devices, wherein when the program is executed by a processor, it implements the steps of the optoelectronic device functional degradation evaluation method described above.

[0102] The method implemented when the optoelectronic device functional degradation assessment procedure is executed can refer to the various embodiments of the optoelectronic device functional degradation assessment method of this application, and will not be repeated here.

[0103] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0104] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0105] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0106] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0107] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0108] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0109] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for evaluating the functional degradation of optoelectronic devices, characterized in that, The method for evaluating the functional degradation of optoelectronic devices includes: Multiple operational status data related to the causes of functional degradation of optoelectronic devices are collected at various times, and the sample matrix constructed based on the operational status data is subjected to deviation standardization processing to obtain the sample deviation standardization matrix. The sample deviation normalization matrix is ​​subjected to singular value decomposition and truncation to obtain the truncated left singular matrix and feature data matrix; Based on the sample deviation normalization matrix, the truncated left singular matrix, and the feature data matrix, the interval distance vector corresponding to the labeled sample and the interval distance vector corresponding to the matrix to be evaluated are obtained. Obtain the measured degradation category corresponding to the label sample, and combine it with the interval distance vector corresponding to the label sample to obtain the weight parameter vector. Based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated, obtain the degradation evaluation category of the optoelectronic device.

2. The method for evaluating the functional degradation of optoelectronic devices as described in claim 1, characterized in that, The step of performing deviation standardization on the sample matrix constructed based on the operating status data to obtain the sample deviation standardization matrix includes: The average column vector is obtained by summing the elements of each row of the column vector in the sample matrix constructed based on the running status data and then dividing by the number of columns. Subtract the average column vector from each column of the sample matrix to obtain the sample deviation matrix; Standardize each column vector of the sample deviation matrix to obtain the standardized sample deviation matrix.

3. The method for evaluating the functional degradation of optoelectronic devices as described in claim 1, characterized in that, The singular value decomposition and truncation of the sample deviation normalization matrix to obtain the truncated left singular matrix and feature data matrix include: Singular value decomposition is performed on the sample deviation standardization matrix to obtain the left singular matrix, the singular value matrix, and the right singular matrix. Calculate the cutoff value based on the diagonal elements of the singular value matrix and the set error threshold. Based on the truncated values, the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix are obtained. The feature data matrix is ​​obtained based on the truncated left singular matrix, the truncated singular value matrix, and the truncated right singular matrix.

4. The method for evaluating the functional degradation of optoelectronic devices as described in claim 3, characterized in that, Based on the sample deviation normalization matrix, the truncated left singular matrix, and the feature data matrix, the interval distance vectors corresponding to the labeled samples and the interval distance vectors corresponding to the matrices to be evaluated are obtained, including: A reference matrix is ​​constructed by selecting r column vectors from the feature data matrix. The reference matrix consists of m rows and r columns, where m is the number of rows in the sample matrix. In the sample deviation standardization matrix, select a column vector to be detected, and multiply the column vector with a row vector whose elements are all 1 and the number of elements is r to obtain the matrix to be evaluated, which includes m rows and r columns. Based on the reference matrix, the matrix to be evaluated, and the truncated left singular matrix, the interval distance vector corresponding to the matrix to be evaluated is obtained; Select several label samples from the feature data matrix, where one of the label samples is a column vector in the feature data matrix; For each label sample, multiply the label sample by a row vector with all elements being 1 and the number of elements being r to obtain the label sample matrix. Based on the reference matrix, the label sample matrix, and the truncated left singular matrix, obtain the interval distance vector corresponding to the label sample.

5. The method for evaluating the functional degradation of optoelectronic devices as described in claim 4, characterized in that, Based on the reference matrix, the matrix to be evaluated, and the truncated left singular matrix, the interval distance vector corresponding to the matrix to be evaluated is obtained as follows: The truncated left singular matrix is ​​transposed and multiplied with the reference matrix. Then, the result of multiplying the truncated left singular matrix transposed with the matrix to be evaluated is subtracted to obtain the difference matrix, which consists of h rows and r columns, where h is the truncated value. Multiply the difference matrix by its transpose to obtain the distance matrix, which consists of h rows and h columns; Extract the diagonal elements of the distance matrix and calculate the square root of each diagonal element to obtain the interval distance vector corresponding to the matrix to be evaluated. The interval distance vector includes h elements.

6. The method for evaluating the functional degradation of optoelectronic devices as described in claim 4, characterized in that, Obtain the measured degradation level category corresponding to the labeled sample, and combine it with the interval distance vector corresponding to the labeled sample to obtain the weight parameter vector, including: Obtain the measured degradation category corresponding to each labeled sample; Construct a distance coefficient matrix based on the interval distance vectors corresponding to all labeled samples. , The j-th element in the interval distance vector corresponding to the i-th label sample point in the i-th row; based on The weight parameter calculation formula is constructed and solved for the measured degradation degree of all labeled samples, resulting in a weight parameter vector. The weight parameter calculation formula is as follows: Among them, the weight parameter vector The i-th element This indicates the importance of the i-th element in the interval distance vector; express The transpose of the matrix, express Its transpose Multiply them, then find the inverse matrix; This represents the measured category of degradation level corresponding to the i-th label sample point.

7. The method for evaluating the functional degradation of optoelectronic devices as described in claim 6, characterized in that, The degradation assessment categories of optoelectronic devices are obtained based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated, including: Calculate the product of the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated; The degradation assessment category of the optoelectronic device is determined based on the product.

8. A device for evaluating the functional degradation of optoelectronic devices, characterized in that, The optoelectronic device functional degradation assessment device includes: The feature data selection module is used to acquire operating status data related to the causes of functional degradation of optoelectronic devices collected at multiple times, and to perform deviation standardization processing on the sample matrix constructed based on the operating status data to obtain a sample deviation standardization matrix; and to perform singular value decomposition and truncation processing on the sample deviation standardization matrix to obtain a truncated left singular matrix and a feature data matrix. The interval distance calculation module is used to obtain the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated based on the sample deviation standardization matrix, the truncated left singular matrix and the feature data matrix. The degradation classification module is used to obtain the degradation assessment category of optoelectronic devices based on the weight parameter vector and the interval distance vector corresponding to the matrix to be evaluated.

9. A device for evaluating the functional degradation of optoelectronic devices, characterized in that, The optoelectronic device functional degradation assessment device includes a processor, a memory, and an optoelectronic device functional degradation assessment program stored in the memory and executable by the processor, wherein when the optoelectronic device functional degradation assessment program is executed by the processor, it implements the steps of the optoelectronic device functional degradation assessment method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an optoelectronic device functional degradation assessment program, wherein when the optoelectronic device functional degradation assessment program is executed by a processor, it implements the steps of the optoelectronic device functional degradation assessment method as described in any one of claims 1 to 7.

Images