Method, system and equipment for managing optical module

Abstract

The invention provides an optical module management method, system and device, relates to the technical field of optical modules, and can be applied to scenes such as dynamic prediction and intelligent scheduling of the optical modules, and the method comprises the following steps: data acquisition of operation state data and link state data of each optical module in a communication device, state modeling of the optical modules by using the operation state data, and state modeling of the optical modules by using the link state data; obtaining a state vector corresponding to each optical module; and inputting the state vector and the link state data into a preset parameter evolution analysis model to obtain a potential risk event, and uploading the potential risk event to a remote network control platform. Through obtaining and processing the operation state data and the link state data, a potential risk event is obtained, the potential risk event is identified and synchronized to a remote network control platform, fine state evaluation and intelligent scheduling of the optical module are realized, and the defect that the optical module is difficult to carry out in an actual complex communication environment is overcome. And the potential performance degradation and aging trend of the optical module cannot be judged to perform intelligent scheduling on the optical module.

Description

Technical Field

[0001] This application relates to the technical field of optical modules, and in particular to methods, systems and devices for managing optical modules. Background Technology

[0002] With the continuous improvement of network communication capabilities driven by data centers, 5G base stations, and high-speed interconnect devices, optical communication systems face higher demands in terms of transmission rate, bandwidth capacity, and system stability. Optical modules, as key components connecting optical and electrical signals, play a crucial role in communication links. Communication equipment typically integrates multiple optical modules to achieve large-scale parallel data transmission. To ensure the continuity and high reliability of system operation, effective management of the optical modules deployed in communication equipment is necessary, including operational status monitoring, performance evaluation, fault detection, and maintenance scheduling.

[0003] In existing technologies, basic operating parameters of the optical module, such as current, voltage, temperature, transmitted optical power, and received optical power, are typically collected through a communication management unit or the monitoring chip within the optical module itself. These parameters are then periodically recorded in conjunction with link-side indicators such as bit error rate and packet loss rate. Some systems also introduce a simple threshold judgment mechanism, comparing operating parameters with preset upper and lower limits to preliminarily determine whether the optical module is in an abnormal state. When a threshold is exceeded, the system triggers an alarm or executes a fixed-rule channel switching strategy to improve overall communication stability, thus achieving monitoring of the basic operating status of the optical module and coarse fault response management.

[0004] While the above technical solutions can achieve preliminary identification of abnormal states of optical modules through parameter monitoring and threshold comparison, and reduce risks by switching channels according to fixed rules, in actual complex communication environments, existing solutions lack dynamic prediction of the operating trend of optical modules and cannot determine the potential performance degradation and aging trend of optical modules to intelligently schedule optical modules. Summary of the Invention

[0005] To overcome the problem of not being able to intelligently schedule optical modules in real-world complex communication environments due to the inability to judge potential performance degradation and aging trends, this application provides a management method, system, and device for optical modules, enabling more intelligent, predictive, and interconnected end-to-end management of optical modules.

[0006] On one hand, the present invention provides a management method for optical modules, applied to communication equipment. The management method includes: acquiring the operating status data and link status data of each optical module in the communication equipment; using the operating status data to perform state modeling on the optical modules to obtain a state vector corresponding to each optical module; inputting the state vector and the link status data into a preset parameter evolution analysis model to obtain a state change curve and trend evaluation result; performing performance analysis on the optical modules using the trend evaluation result and the state change curve to obtain a health status level and availability score; prioritizing all optical modules based on the health status level and the availability score to obtain an optical module scheduling strategy; adjusting the optical module channels in the communication equipment according to the optical module scheduling strategy to obtain a channel mapping relationship; fusing and analyzing the channel mapping relationship, the state change curve, the trend evaluation result, and the health status level to obtain potential risk events; and uploading the potential risk events to a remote network control platform.

[0007] Optionally, the step of acquiring the operating status data and link status data of each optical module in the communication device, and using the operating status data to perform state modeling on the optical modules to obtain a state vector corresponding to each optical module includes: reading the status of each optical module through the data acquisition unit built into the communication device to obtain operating status data and link status data; normalizing the operating status data to obtain normalized data; combining the normalized data with a reference range in a preset equipment specification table to construct a standardized index set; fusing the standardized index set with historical operating status data to generate a dynamic feature window; wherein, the reference range in the equipment specification table refers to a parameter tolerance range preset according to the manufacturing parameters and factory performance standards of each type of optical module; calculating the correlation matrix and rate of change vector between parameters based on the dynamic feature window to obtain parameter association features; and using the parameter association features and the normalized data to perform state modeling on the optical modules to obtain a state vector corresponding to each optical module.

[0008] Optionally, the step of inputting the state vector and the link state data into a preset parameter evolution analysis model to obtain the state change curve and trend evaluation result includes: extracting features from the link state data to obtain the link quality indicators for each optical module, including bit error rate, packet loss rate, and retransmission ratio; fusing the state vector and the link quality indicators to construct a multidimensional input tensor; inputting the multidimensional input tensor into the preset parameter evolution analysis model to output a short-term prediction vector and a long-term trend vector; performing a difference analysis between the short-term prediction vector and the state vector to obtain a state offset; and performing a fitting analysis between the long-term trend vector and historical state change data to obtain a state change curve; wherein, the historical state change data refers to the time series set of optical module operating state data and link state data recorded by the communication device in multiple historical operating cycles; calculating the stability index and trend score of the optical module state using the state offset and the state change curve, and constructing a trend evaluation result based on the stability index and the trend score.

[0009] Optionally, the step of performing performance analysis on the optical module using the trend assessment results and the state change curve to obtain a health status level and availability score includes: performing stability screening on the trend assessment results, extracting a trend score and state volatility based on the screening results, calculating a performance degradation factor for the optical module using the state volatility, and combining the performance degradation factor with historical maintenance cycles to determine an aging coefficient; wherein, the historical maintenance cycle refers to the cumulative maintenance records recorded by the communication equipment for each optical module during its lifecycle; performing combined calculations on the trend score, the performance degradation factor, and the aging coefficient to obtain a comprehensive health score, mapping the comprehensive health score to a preset health status level range to obtain a corresponding health status level; analyzing the performance availability trend within a few future time slices based on the state change curve to obtain a trend availability index, and matching and comparing the health status level with the trend availability index to obtain an availability score.

[0010] Optionally, the step of combining the trend score, the performance degradation factor, and the aging coefficient to obtain a comprehensive health score, and mapping the comprehensive health score to a preset health status level range to obtain a corresponding health status level, includes: normalizing the trend score to an interval value to obtain a normalized trend score; adjusting the performance degradation factor with exponential weight to obtain a weighted degradation factor; multiplying and fusing the weighted degradation factor and the aging coefficient to obtain a long-term degradation intensity index; performing a weighted linear superposition of the normalized trend score and the long-term degradation intensity index to obtain a comprehensive health score; and mapping the comprehensive health score to a preset health status level range to obtain a corresponding health status level.

[0011] Optionally, the step of prioritizing all optical modules based on the health status level and the availability score to obtain an optical module scheduling strategy, and adjusting the optical module channels in the communication device according to the optical module scheduling strategy to obtain a channel mapping relationship, includes: standardizing the health status level and the availability score of all optical modules to generate a two-dimensional scheduling index matrix; clustering all optical modules based on the two-dimensional scheduling index matrix to obtain different scheduling priority groups; inputting all the scheduling priority groups into a preset scheduling weight model to output the optical module scheduling strategy; inputting the optical module scheduling strategy into a channel configuration module to perform optical module channel switching operations; after completing the channel switching, collecting link quality indicators and state vectors before and after the switching to perform stability evaluation, obtaining stability evaluation results, and generating a channel mapping relationship based on the stability evaluation results.

[0012] Optionally, the step of fusing and analyzing the channel mapping relationship, the state change curve, the trend assessment result, and the health status level to obtain potential risk events, and uploading the potential risk events to the remote network control platform, includes: aligning the channel mapping relationship of each optical module with its corresponding state change curve in time to construct a channel-state association table; performing statistical analysis on the channel-state association table to extract channel stability indicators; fusing and analyzing the channel stability indicators, the trend assessment result, and the health status level to obtain an abnormal pattern label set; filtering out potential risk events based on the abnormal pattern label set; and uploading the potential risk events to the remote network control platform.

[0013] On the other hand, this application also provides a management system for optical modules, including: a data acquisition module, used to acquire the operating status data and link status data of each optical module in the communication device, and to perform state modeling on the optical modules using the operating status data to obtain a state vector corresponding to each optical module; a performance analysis module, used to input the state vector and the link status data into a preset parameter evolution analysis model to obtain a state change curve and a trend evaluation result, and to perform performance analysis on the optical modules using the trend evaluation result and the state change curve to obtain a health status level and an availability score; a priority ranking module, used to rank the scheduling priorities of all optical modules based on the health status level and the availability score to obtain an optical module scheduling strategy, and to adjust the optical module channels in the communication device according to the optical module scheduling strategy to obtain a channel mapping relationship; and a fusion analysis module, used to perform fusion analysis on the channel mapping relationship, the state change curve, the trend evaluation result, and the health status level to obtain potential risk events, and to upload the potential risk events to a remote network control platform.

[0014] On the other hand, this application also provides an electronic device, the device including a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the optical module management method as described in any of the above.

[0015] The optical module management method, system, and equipment provided in this application have the following technical advantages: strong intelligent scheduling and high flexibility. By acquiring the operating status data and link status data of each optical module in the communication equipment, and combining this data with the equipment specification table for normalization and dynamic modeling, a state vector reflecting the current operating characteristics of the optical module is generated. Furthermore, the state vector is fused with the link status data, and the input parameter evolution analysis model is used to obtain state change curves and trend evaluation results from both short-term and long-term perspectives. A comprehensive evaluation is performed using trend scores, aging coefficients, and performance degradation factors to generate a health status level and availability score. Then, based on the scheduling priority strategy, the channels of the optical modules are dynamically adjusted to obtain channel mapping relationships reflecting the system resource adaptation relationship. Combining the channel mapping relationship, state change curves, trend evaluation results, and health status level for fusion analysis, potential risk events are identified and synchronized to a remote network control platform, achieving refined status evaluation and intelligent scheduling of optical modules. This overcomes the problem of being unable to intelligently schedule optical modules in complex actual communication environments by judging potential performance degradation and aging trends. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the optical module management method provided in an embodiment of the present invention; Figure 2 This is a schematic block diagram of the structure of the optical module management system provided in an embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of the electronic device provided in the embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or sub-modules is not necessarily limited to those steps or sub-modules explicitly listed, but may include other steps or sub-modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0020] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0021] Example 1: like Figure 1 As shown, this embodiment 1 provides a management method for an optical module, applied to a communication device. The management method specifically includes: Step S10: Obtain the operating status data and link status data of each optical module in the communication device, and use the operating status data to perform state modeling of the optical modules to obtain the state vector corresponding to each optical module.

[0022] In step S10, the data acquisition unit built into the communication equipment periodically reads the operating status data and link status data of each optical module. Specifically, the operating status data includes parameters from multiple dimensions such as the optical module's voltage, current, operating temperature, transmitted optical power, and received optical power. The link status data includes bit error rate, packet loss rate, and retransmission ratio. After acquisition, the operating status data is normalized, and a standardized index set is constructed by combining it with reference intervals in the preset equipment specification table. This standardized index set is then fused with historical operating status data to construct a dynamic feature window. Further, based on the dynamic feature window, the correlation matrix and rate of change vector between parameters are calculated to obtain parameter correlation features. These parameter correlation features, along with the normalized data, are input into the state modeling logic to generate the corresponding state vector.

[0023] For example, the 3.2mW value collected in real time is normalized to the medium-high range according to the range of "emitted optical power: 2mW-5mW" given in the equipment specification table. After integrating 30 days of historical temperature and power change data, the change rate vector is calculated to be higher than the threshold. The state modeling logic marks it as a high fluctuation state.

[0024] Step S20: Input the state vector and link state data into the preset parameter evolution analysis model to obtain the state change curve and trend evaluation results. Perform performance analysis on the optical module based on the trend evaluation results and state change curves to obtain the health status level and availability score.

[0025] In step S20, the state vector and link state data are dimensionally fused to construct a multidimensional input tensor, which is then input into the pre-trained parameter evolution analysis model. Specifically, this model outputs a short-term prediction vector and a long-term trend vector through joint analysis of short-term trends and long-term changes. The short-term prediction vector is then compared with the original state vector to obtain the state offset. The long-term trend vector is then fitted with historical state change data to obtain a state change curve. The state stability index and trend score are further calculated using the state offset and state change curve, ultimately constructing a trend assessment result. Combining the trend assessment result, the state volatility, performance degradation factor, and aging coefficient of the optical module are analyzed to generate a comprehensive health score, which is mapped to a health status level. Finally, an availability score is generated based on the availability trend within future time slices.

[0026] For example, if the status offset of a certain optical module is 15% and the trend score is 75, combined with the aging coefficient corresponding to its 3-month maintenance cycle of 1.2, the comprehensive health score is 62.5, which is mapped to the "medium health" level, and the availability score is 67.

[0027] Step S30: Based on the health status level and availability score, sort all optical modules by scheduling priority to obtain the optical module scheduling strategy. Adjust the optical module channels in the communication equipment according to the optical module scheduling strategy to obtain the channel mapping relationship.

[0028] In step S30, the scheduling priority of all optical modules is calculated by weighting their health status level and availability score. Specifically, the health status level is divided into three levels: high, medium, and low, with corresponding weights of 0.5, 0.3, and 0.2. This is then combined with the availability score to form a scheduling priority queue. The scheduling system uses this priority queue to sequentially select optical modules with unstable channel performance, performs channel mapping replacement, and records the corresponding adjustment relationships.

[0029] If module A is "high health level" and has an availability score of 88, it is assigned to the core data channel; while module C is "low health level" and is downgraded to the backup channel, forming a new channel mapping table.

[0030] Step S40: Perform integrated analysis on the channel mapping relationship, state change curve, trend assessment results and health status level to obtain potential risk events, and upload the potential risk events to the remote network control platform.

[0031] In step S40, a fusion analysis matrix is ​​constructed, using the number of channel mapping relationship changes, the volatility of the state change curve, negative indicators in the trend assessment results, and the downward trend of the health status level as input dimensions. In detail, a rule engine and threshold judgment mechanism are used to perform cross-analysis on the above indicators. When a certain optical module experiences frequent channel switching, drastic fluctuations in the state curve, and a continuous decline in the health level, it is marked as a potential risk event. Subsequently, the potential risk event is encapsulated into a log package and uploaded to the remote network control platform through the remote management interface of the communication device to achieve cloud-based early warning management of potential faults.

[0032] For example, if the optical module numbered M21 switches channels 3 times within 48 hours, the maximum fluctuation of the status curve reaches 28%, and the health level drops from "medium" to "low", it is identified as a potential risk event, and the corresponding event package is uploaded to the control platform.

[0033] In Example 1, by acquiring the operating status data and link status data of each optical module in the communication device, the operating status data is used to model the state of the optical modules, resulting in a state vector corresponding to each optical module. Subsequently, the state vector and link status data are input into a preset parameter evolution analysis model to obtain state change curves and trend evaluation results. Next, the performance of the optical modules is analyzed based on the trend evaluation results and state change curves to obtain health status levels and availability scores. Furthermore, all optical modules are prioritized based on health status levels and availability scores to obtain optical module scheduling strategies. The optical module channels in the communication device are adjusted according to the optical module scheduling strategies to obtain channel mapping relationships. Finally, the channel mapping relationships, state change curves, trend evaluation results, and health status levels are fused and analyzed to obtain potential risk events, which are then uploaded to a remote network control platform. This system enables collaborative perception of the operational status data and link status data of multiple optical modules in communication equipment. Based on state modeling and parameter evolution analysis models, it completes trend assessment and performance prediction of the operational status of optical modules, thereby obtaining more accurate health status levels and availability scores. Combining health status levels and availability scores for dynamic priority ranking and channel adjustment helps to build more real-time and adaptive optical module scheduling strategies. At the same time, by fusing analysis of channel mapping relationships, state change curves, trend assessment results, and health status levels, it can achieve early identification and remote uploading of potential risk events, enhance the intelligent maintenance capabilities of optical modules and the overall stability of the system, improve the operation and maintenance efficiency and reliability of optical communication systems, and overcome the problem of not being able to judge the potential performance degradation and aging trends of optical modules in real complex communication environments and perform intelligent scheduling of optical modules.

[0034] Example 2: In step S10, the operating status data and link status data of each optical module in the communication device are obtained. The operating status data is used to perform state modeling on the optical modules to obtain the state vector corresponding to each optical module. Specifically, this includes: The status of each optical module is read by the data acquisition unit built into the communication equipment to obtain operating status data and link status data.

[0035] The automated management system of the communication equipment is used to identify, parse, and clean the collected operating status data and link status data. In detail, the automated management system performs unified formatting processing on the sensor output data (such as operating current, bias voltage, internal temperature, and optical power) of different types of optical modules, and builds a data dictionary based on the unique identifier of each optical module to achieve accurate traceability and dynamic updates of the data source.

[0036] For example, when there are multiple optical modules of different models in a communication device, the system can distinguish and process different data structures based on the module ID in its internal EEPROM, ensuring that a data record table with a unified structure and standardized format is finally obtained.

[0037] The operational status data is normalized to obtain normalized data. This normalized data is then combined with reference intervals in the preset equipment specification table to construct a standardized indicator set. This standardized indicator set is then fused with historical operational status data to generate a dynamic feature window. The reference intervals in the equipment specification table refer to the parameter tolerance ranges preset based on the manufacturing parameters and factory performance standards for each type of optical module. These ranges include the upper and lower limits of the normal range for each operating parameter (including current, voltage, temperature, transmit power, and receive power) of the optical module under rated operating conditions. The reference intervals are used for boundary verification and difference assessment of the normalized data to ensure that deviations caused by obvious anomalies or drifting data are excluded during status modeling.

[0038] The normalized data is filtered by time sliding window using historical operating status data and reference intervals. Specifically, the system extracts continuous segments of operating data at fixed time steps (e.g., every 5 minutes) and eliminates outliers and drift errors by comparing the fluctuation range of the normalized data within the reference interval of the equipment specification table, thus forming a stable and time-series-based feature window.

[0039] For example, the temperature fluctuation of a certain optical module has been steadily increasing within a specified range over the past 20 cycles, with the normalized value slowly rising from 0.42 to 0.56. This indicates that the equipment is operating stably, but it will face future temperature risks. The abnormal trend can be identified in advance through the feature window.

[0040] Based on the dynamic feature window, the correlation matrix and rate of change vector between parameters are calculated to obtain parameter correlation features. The optical modules are then modeled using the parameter correlation features and normalized data to obtain the state vector corresponding to each optical module.

[0041] The system extracts the co-fluctuation relationship between parameters based on the Pearson correlation analysis method and the local moving average strategy. In detail, the system first calculates the correlation coefficient matrix between all parameter pairs in the dynamic feature window, identifies parameter combinations with significant correlation (such as absolute values ​​greater than 0.6), and then calculates the fluctuation trend vector based on the mean and rate of change within the window. The system then obtains the parameter association features that represent the overall coupling degree of the current state of the optical module.

[0042] For example, when the bias voltage and transmit power of an optical module are highly positively correlated (correlation coefficient of 0.82) over a certain period of time, and its temperature change rate continues to rise, the system generates a feature vector [0.82, 0.05] based on this, and combines it with normalized data [0.65, 0.72, 0.56] to perform state modeling, and finally outputs the three-dimensional state vector of the optical module [0.73, 0.62, 0.81], which is used for subsequent performance prediction and health assessment.

[0043] In step S20, the state vector and link state data are input into a preset parameter evolution analysis model to obtain the state change curve and trend evaluation results, specifically including: Feature extraction is performed on the link status data to obtain the link quality indicators for each optical module, including bit error rate, packet loss rate, and retransmission ratio.

[0044] Based on the link quality assessment logic set in the communication equipment, key quality indicators are extracted from the link status data. In detail, the system calculates the bit error rate, packet loss rate and retransmission ratio per unit time by statistically analyzing the transmission results of link data packets, and binds them with optical module identifiers through data structure alignment to form a link quality indicator sequence.

[0045] For example, if an optical module has a packet loss rate of 0.3%, a bit error rate of 1.2E-7, and a retransmission ratio of 4.5% in the last 10 minutes, the system records this three-dimensional index vector as the link quality assessment result of the optical module.

[0046] The state vector is fused with the link quality index to construct a multidimensional input tensor. The multidimensional input tensor is then input into a preset parameter evolution analysis model to output short-term prediction vectors and long-term trend vectors.

[0047] The state vector and link quality index are concatenated into a multi-dimensional tensor input model on a unified time dimension through a feature fusion mechanism. In detail, a tensor quantization tool is used to stack the state vector and link quality index of each optical module in units of time steps to construct a three-dimensional tensor (optical module × feature dimension × time step), which is then input into a parameter evolution analysis model based on the Transformer architecture for time series feature learning, and outputs short-term prediction vectors and long-term trend vectors for several future time steps.

[0048] For example, the optical module state vector is [0.73, 0.62, 0.81], and the link quality index is [0.003, 0.0012, 0.045]. After merging, a tensor input model is formed, which outputs a short-term prediction vector [0.75, 0.65, 0.83] and a trend vector [0.78, 0.68, 0.85] for subsequent analysis.

[0049] The difference analysis between the short-term prediction vector and the state vector is performed to obtain the state offset. The long-term trend vector is fitted with historical state change data to obtain the state change curve. The historical state change data refers to the time series set of optical module operation status data and link status data recorded by communication equipment in multiple historical operating cycles. The time series set, after being aligned with timestamps, reflects the state evolution trend of the optical module under different network load conditions, and is used to support the trend fitting and long-term prediction of the parameter evolution analysis model.

[0050] The system uses a distance calculation method to perform Euclidean distance analysis between the short-term prediction vector and the current state vector to measure the degree of state deviation. Specifically, it obtains the state deviation index by summing and taking the square root of the squared differences between the current state vector and the short-term prediction vector in each dimension. At the same time, the system uses historical state change data to construct a time series regression curve, fits the trend vector, and generates a state change curve that reflects the trajectory of state evolution.

[0051] For example, if the current state vector is [0.73, 0.62, 0.81] and the predicted vector is [0.75, 0.65, 0.83], the Euclidean distance between them is: This reflects a slight upward trend in the state in the short term; at the same time, the trend vector is fitted with the data of the past 7 days using a polynomial to obtain the temperature trend curve and the stable change trend of the transmission power of the optical module.

[0052] The stability index and trend score of the optical module state are calculated using the state offset and state change curve, and the trend evaluation results are constructed based on the stability index and trend score.

[0053] The stability analysis module jointly evaluates the slope and offset of the state change curve. In detail, the system performs linear fitting on the state change curve, extracts the curve slope as a trend score, and then performs normalization processing in combination with the state offset. Finally, it generates a stability index (representing the strength of state fluctuations) and a trend score (representing the direction and magnitude of change), and combines them to construct the trend evaluation result.

[0054] For example, the slope of the state change curve is 0.015 and the state offset is 0.041. After normalization and weighting, the stability index is calculated to be 0.92 and the trend score is 0.78. Based on this, the system judges that the current state of the optical module is relatively stable but has a slight growth trend, and it is suitable to be included in the medium and long-term performance prediction model.

[0055] In step S20, the performance of the optical module is analyzed based on the trend assessment results and state change curves to obtain the health status level and availability score, specifically including: The trend assessment results are subjected to stability screening, and trend scores and state volatility are extracted based on the screening results. The performance degradation factor of the optical module is calculated using the state volatility. The performance degradation factor is combined with the historical maintenance cycle to determine the aging coefficient. The historical maintenance cycle refers to the cumulative maintenance records of each optical module recorded by the communication equipment during its life cycle, including the time interval and frequency of events such as optical module fault alarms, performance degradation detection, firmware reconstruction, parameter calibration and hot-swappable replacement, which are used to infer the long-term stability and physical aging degree of the optical module.

[0056] Performance degradation is calculated using the steepness of the trend reflected by the trend score and the magnitude of parameter changes measured by volatility. In detail, the state volatility is estimated by the mean square error of adjacent time windows, the performance degradation factor is normalized to the [0,1] interval according to the degradation magnitude, and a mapping model is constructed with historical maintenance cycles (such as maintenance frequency and maintenance interval) to obtain the aging coefficient.

[0057] For example, if a certain optical module was maintained once a month in the past 90 days and its performance degradation factor is 0.6, then the system sets its aging factor to 0.72 based on the historical maintenance frequency.

[0058] The trend score, performance degradation factor and aging coefficient are combined to obtain a comprehensive health score. The comprehensive health score is then mapped to a preset health status level range to obtain the corresponding health status level.

[0059] The comprehensive health score is calculated based on a three-factor weighted scoring mechanism. Specifically, the trend score has a weight of 40%, the performance degradation factor has a weight of 35%, and the aging coefficient has a weight of 25%. The results are weighted and summed and mapped to a health level scoring system of 0-100, and divided into five level intervals (such as excellent, good, average, poor, and very poor).

[0060] For example, if the trend score is 0.85, the performance degradation factor is 0.6, and the aging coefficient is 0.72, then the weighted health score is 0.751, which is mapped to the "good" level, corresponding to a grade B.

[0061] Based on the state change curve analysis, the performance availability trend within a certain number of time slices in the future is obtained to obtain the trend availability index. The health status level is matched and compared with the trend availability index to obtain the availability score.

[0062] The slope of the trend line and historical fluctuation range are used to predict future availability trends. In detail, N future periods (such as 30 minutes) are set as evaluation windows to analyze the trend of each dimension in the status change curve and cross-compare it with the health status level standard to generate a two-dimensional scoring matrix. The rows of the matrix represent the status level and the columns represent the trend category.

[0063] For example, the status change curve of a certain optical module shows a slow downward trend (slope of -0.004), the current level is C, and the future availability prediction obtained by matching the scoring matrix is ​​"moderately low", which is used for subsequent scheduling evaluation.

[0064] Then, the two-dimensional scoring matrix is ​​weighted and attributed based on the AHP (Analytic Hierarchy Process). Specifically, the system sets the weight of the health level influence factor to 0.6 and the weight of the trend availability index to 0.4. The final availability score of each optical module is calculated through weighted scoring and used for sorting and scheduling.

[0065] For example, a certain optical module has a health score of 0.751, corresponding to level B, and its availability trend prediction is "good". After weighted calculation by the system, the final availability score is 0.78.

[0066] Specifically, a comprehensive health score is obtained by combining trend scores, performance degradation factors, and aging coefficients. This comprehensive health score is then mapped to a preset health status level range to obtain the corresponding health status level, which includes: The trend score is normalized to an interval value [0,1] to measure the operating trend performance of the optical module in the recent time window, resulting in a normalized trend score. The performance degradation factor is then adjusted with an exponential weight to improve the sensitivity of recent performance degradation to health assessment, resulting in a weighted degradation factor.

[0067] The performance degradation factor is processed using an exponential decay algorithm; specifically, the system employs e -λt The form assigns higher weight to recent observations to more sensitively capture short-term performance degradation trends of optical modules and outputs a weighted attenuation factor; where λ is a hyperparameter for adjusting sensitivity and t is the sample position within the time window.

[0068] For example, for a set of highly volatile transmit power data, if the rate of decrease is significant in the last 3 hours, the system calculates a weighted attenuation factor of 0.68 based on the set λ=0.5, which improves the detection sensitivity compared to the original value.

[0069] The weighted attenuation factor and the aging coefficient are multiplied and fused to obtain the long-term degradation intensity index, which is used to quantify the overall lifespan attenuation of the optical module. The normalized trend score and the long-term degradation intensity index are weighted and linearly superimposed to obtain the comprehensive health score, where the weighting factor is dynamically determined based on the current usage time and scheduling frequency of the optical module.

[0070] Based on the scheduling records, the system calculates the usage time and call frequency of each optical module. In detail, the system normalizes the continuous online time and channel allocation frequency of each optical module over the past month and sets it as a reference parameter for weight generation factors. The system then dynamically adjusts the influence ratio between trend score and degradation intensity in the comprehensive model to reflect the current actual load status of the optical module.

[0071] For example, if a certain optical module has an 80% utilization rate and a scheduling frequency of 18 times in the past 30 days, the system calculates its trend weight factor as 0.35 and degradation intensity weight as 0.65 based on these data, and finally outputs a comprehensive health score of 0.73.

[0072] The comprehensive health score is used as input and mapped to a preset health status level range to obtain the corresponding health status level.

[0073] The continuous scores are graded based on the set health status level classification table. In detail, the system predefines 5 health level ranges (e.g., very healthy, healthy, slightly abnormal, moderately abnormal, and severely abnormal) and sets the comprehensive health score range for each level. The comprehensive health score of the current optical module is matched with the level table to generate the final health status level identifier.

[0074] For example, when the overall health score of a certain optical module is 0.47, the system looks up the table and finds that the value falls within the range of [0.3, 0.5], the corresponding health status level is "mildly abnormal", and it is marked as a yellow alarm status in the management system to prompt subsequent strategy optimization.

[0075] In step S30, all optical modules are prioritized based on health status level and availability score to obtain an optical module scheduling strategy. The optical module channels in the communication equipment are then adjusted according to this strategy to obtain the channel mapping relationship, specifically including: The health status level and availability score of all optical modules are standardized to generate a two-dimensional scheduling index matrix. Based on the two-dimensional scheduling index matrix, all optical modules are clustered to obtain different scheduling priority groups.

[0076] The health status level and availability score are processed separately using the Z-score standardization algorithm. In detail, the system performs a standardization operation on the score value of each optical module by subtracting the mean and dividing by the standard deviation to ensure that the scores of different dimensions have uniform distribution characteristics. Then, the K-means clustering algorithm is used to classify the two-dimensional score matrix to obtain three groups of scheduling priorities: high, medium and low.

[0077] For example, an optical module with a score of [1.23, 0.88] after standardization is classified into a high-priority group after clustering, while an optical module with a score of [-0.57, -1.12] is classified into a low-priority group.

[0078] All scheduling priorities are grouped and input into a preset scheduling weight model, and the optical module scheduling strategy is output. The optical module scheduling strategy is then input into the channel configuration module to perform optical module channel switching operations.

[0079] The system sets scheduling priority weight values ​​based on a multi-factor weight model and generates a scheduling strategy list. In detail, the system maps priority levels to weight coefficients, for example, 0.8 for high priority, 0.5 for medium priority, and 0.2 for low priority. It also combines the current service load of the optical module with the historical scheduling frequency to output the optimal scheduling strategy. The strategy is then passed to the channel configuration module for dynamic channel remapping.

[0080] For example, if the current service load of a certain medium-priority optical module surges, its scheduling weight will be adjusted and it will be switched to the main channel to ensure performance.

[0081] After the channel switching is completed, the link quality indicators and state vectors before and after the switching are collected to perform stability assessment, and the channel mapping relationship is generated based on the stability assessment results.

[0082] The difference comparison module is used to calculate the trend of link quality indicators before and after the handover. In detail, the system calculates the bit error rate, packet loss rate and state offset before and after the handover. If the fluctuation is within the acceptable threshold range (e.g., bit error rate change <5%, offset <0.03), the handover is considered stable and the current mapping is recorded as an effective channel mapping relationship.

[0083] For example, after a certain optical module switches, the bit error rate increases from 1.1e-6 to 1.15e-6, and the state offset is 0.02. The system determines that there is no abnormality in its channel switching and generates a mapping record: port A - port C.

[0084] In step S40, the channel mapping relationship, state change curve, trend assessment results, and health status level are fused and analyzed to obtain potential risk events. Uploading these potential risk events to the remote network control platform specifically includes: The channel mapping relationship of each optical module is time-aligned with its corresponding state change curve to construct a channel-state association table. Statistical analysis is then performed on the channel-state association table to extract channel stability indicators.

[0085] Using key change points in the state change curve as a reference, a time-series comparison is performed with the channel mapping change record. In detail, the system aligns the channel switching time of the optical module with the abrupt change nodes in the state curve, statistically analyzes the stability fluctuation range of the state vector before and after the channel switching, and extracts the stability factor reflecting the channel adaptability by combining the continuous decline segment or periodic fluctuation segment in the trend evaluation results.

[0086] For example, if the received power status value of an optical module drops sharply within 5 minutes after a channel switch (e.g., from 0.78 to 0.52), and the status change curve has a clear inflection point within the corresponding time period, the system will determine that the channel has an adaptation risk and write the risk information into the channel-status association table.

[0087] By integrating and analyzing channel stability indicators, trend assessment results, and health status levels, an abnormal pattern label set is obtained. Potential risk events are then screened based on the abnormal pattern label set and uploaded to the remote network control platform.

[0088] Based on a multi-factor analysis mechanism, the channel stability index, trend evaluation results, and health status level of each optical module are vector-concatenated and fused. In detail, the system calculates normalized scores for the above three types of indicators, sets anomaly identification threshold, constructs anomaly identification rule set, and marks optical modules that do not meet the threshold rules with anomaly pattern, forming anomaly pattern label set.

[0089] For example, if a certain optical module has a trend score below 0.3, a health status level of "secondary", and a channel stability score below 0.55, the system will mark the optical module as having a risk of "degradation link fluctuation" and upload its corresponding risk event information (including timestamp, module ID, anomaly type, and suggested handling measures) to the remote network control platform for the operation and maintenance system to make real-time alarms and subsequent scheduling decisions.

[0090] In Example 2, the operating status data and link status data of each optical module are obtained from the communication device. These are then normalized and standardized in conjunction with the device specification table to construct a dynamic feature window. Based on the correlation and change trend between parameters, a state model is performed to obtain a state vector. Furthermore, the state vector and link quality index are fused into a multi-dimensional input tensor, which is then input into the parameter evolution analysis model to obtain a short-term prediction vector and a long-term change trend vector. A state change curve is generated by fitting historical state change data. Combined with the state offset, the stability index and trend score of the optical module are calculated to construct a trend evaluation result.

[0091] In the performance analysis phase, the aging coefficient is calculated using trend scores, state volatility, performance degradation factors, and historical maintenance cycles to obtain a comprehensive health score, which is then mapped to a health status level. Combined with trend availability indicators, a scoring matrix is ​​constructed, and an availability score is calculated. In the resource scheduling phase, a two-dimensional scheduling indicator matrix is ​​constructed using the health status level and availability score, and clustering is performed to generate optical module scheduling strategies. After channel switching, stability is assessed using link quality indicators and state vectors to obtain channel mapping relationships. Finally, time-aligned analysis is performed using the channel mapping relationships and state change curves to extract channel stability indicators. Combined with trend assessment results and health status level fusion analysis, an abnormal pattern marker set is generated, potential risk events are screened, and uploaded to a remote network control platform. This achieves closed-loop management of the optical module's operational status, including fine-grained modeling, trend prediction, performance assessment, and dynamic scheduling.

[0092] Example 3: like Figure 2 As shown, this embodiment provides a management system 10 for optical modules, which includes the following modules: The data acquisition module 11 is used to acquire the operating status data and link status data of each optical module in the communication device, and to perform state modeling of the optical modules using the operating status data to obtain the state vector corresponding to each optical module.

[0093] The performance analysis module 12 is used to input the state vector and link state data into the preset parameter evolution analysis model to obtain the state change curve and trend evaluation results. The performance of the optical module is analyzed by the trend evaluation results and state change curves to obtain the health status level and availability score.

[0094] The priority sorting module 13 is used to sort the scheduling priorities of all optical modules based on health status level and availability score, obtain the optical module scheduling strategy, and adjust the optical module channels in the communication equipment according to the optical module scheduling strategy to obtain the channel mapping relationship.

[0095] The fusion analysis module 14 is used to perform fusion analysis on channel mapping relationships, state change curves, trend assessment results, and health status levels to obtain potential risk events, and upload the potential risk events to the remote network control platform.

[0096] In Example 3, by integrating a data acquisition module 11, a performance analysis module 12, a priority ranking module 13, and a fusion analysis module 14 into the management system, full lifecycle management of each optical module in the communication equipment is achieved. The data acquisition module 11 interacts with the data acquisition unit of the communication equipment to accurately collect the operating status data and link status data of each optical module. Based on normalization processing, reference interval verification, and a dynamic feature window construction mechanism, it generates a state vector with temporal sequence and structure. The performance analysis module 12 constructs a multidimensional input tensor based on the state vector and link quality indicators. After inputting it into the parameter evolution analysis model, it outputs short-term prediction vectors and long-term trend vectors. It then combines the state offset and historical state change sequences to generate state change curves and trend evaluation results, and further evaluates the health status level and availability score. The priority ranking module 13 generates an optical module scheduling strategy based on a two-dimensional scheduling indicator matrix through clustering and a scheduling weight model, and dynamically configures and switches the channels of the optical modules in the communication equipment. The fusion analysis module 14 performs correlation analysis on information such as channel mapping relationship, state change curve and trend evaluation results, extracts channel stability indicators, and generates an abnormal mode label set. Then, it identifies potential risk events and reports them to the remote network control platform in real time, realizing the full-process closed-loop management of optical module operation status, including fine modeling, dynamic evaluation, intelligent scheduling and fault early warning.

[0097] Example 4: like Figure 3 As shown, this embodiment provides an electronic device 20, which includes a processor 22 and a memory 24. The memory 24 stores at least one instruction or at least one program. The at least one instruction or at least one program is loaded and executed by the processor 22 to implement the optical module management method as described in Embodiment 1 above.

[0098] In Embodiment 4, by integrating a processor 22 and a memory 24 into the electronic device 20, and pre-setting control instructions or programs for implementing the optical module management method in the memory 24, the device is able to automatically complete core functions such as acquiring optical module operating data, state vector modeling, parameter evolution analysis, health level assessment, scheduling strategy generation, and abnormal event identification during execution. Specifically, after the program is loaded by the processor 22, it calls various analysis modules in the system to monitor and schedule the optical module's operating status according to the steps of the optical module management method, ultimately achieving intelligent identification, dynamic control, and fault reporting of the optical module's status, thereby improving the operational reliability and intelligent maintenance level of the communication equipment. This electronic device 20 can be widely used in various communication racks, switches, or optical transmission platforms, supporting unified management and remote interaction of multiple types of optical modules, and possesses strong deployment adaptability and engineering practical value.

[0099] In one alternative embodiment, an electronic device 20 is provided, such as Figure 3 As shown, Figure 3 The illustrated electronic device includes a processor 22 and a memory 24. The processor 22 and the memory 24 are connected, for example, via a bus 21. Optionally, the electronic device 20 may further include a transceiver 23, which can be used for data interaction between the electronic device 20 and other electronic devices 20, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 23 is not limited to one unit, and the structure of the electronic device 20 does not constitute a limitation on the embodiments of this application.

[0100] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0101] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device, equipment, and storage medium embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0102] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware, or by a program instructing the relevant hardware to implement them. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0103] The above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for managing optical modules, applied to communication equipment, characterized in that, The management method includes: The operating status data and link status data of each optical module in the communication device are obtained, and the operating status data is used to perform state modeling of the optical module to obtain the state vector corresponding to each optical module. The state vector and the link state data are input into a preset parameter evolution analysis model to obtain state change curves and trend evaluation results. The performance of the optical module is analyzed based on the trend evaluation results and the state change curves to obtain the health status level and availability score. Based on the health status level and the availability score, all optical modules are prioritized for scheduling to obtain an optical module scheduling strategy. The optical module channels in the communication device are then adjusted according to the optical module scheduling strategy to obtain a channel mapping relationship. The channel mapping relationship, the state change curve, the trend assessment result, and the health status level are fused and analyzed to obtain potential risk events, which are then uploaded to the remote network control platform.

2. The optical module management method according to claim 1, characterized in that, The step of acquiring the operating status data and link status data of each optical module in the communication device, and using the operating status data to perform state modeling of the optical modules to obtain the state vector corresponding to each optical module includes: The communication device uses its built-in data acquisition unit to read the status of each optical module, thereby obtaining operational status data and link status data. The operating status data is normalized to obtain normalized data. The normalized data is combined with the reference range in the preset equipment specification table to construct a standardized indicator set. The standardized indicator set is then fused with historical operating status data to generate a dynamic feature window. The reference range in the equipment specification table refers to the parameter tolerance range preset according to the manufacturing parameters and factory performance standards of each type of optical module. Based on the dynamic feature window, the correlation matrix and rate of change vector between parameters are calculated to obtain parameter association features. The parameter association features and the normalized data are used to model the state of the optical module to obtain the state vector corresponding to each optical module.

3. The optical module management method according to claim 1, characterized in that, The step of inputting the state vector and the link state data into a preset parameter evolution analysis model to obtain the state change curve and trend evaluation results includes: Feature extraction is performed on the link status data to obtain the link quality indicators for each optical module, including bit error rate, packet loss rate, and retransmission ratio. The state vector is fused with the link quality index to construct a multidimensional input tensor. The multidimensional input tensor is then input into a preset parameter evolution analysis model to output a short-term prediction vector and a long-term trend vector. The short-term prediction vector and the state vector are compared to obtain the state offset. The long-term trend vector is fitted to the historical state change data to obtain the state change curve. The historical state change data refers to the time series set of optical module operation status data and link status data recorded by the communication device in multiple historical operating cycles. The stability index and trend score of the optical module state are calculated using the state offset and the state change curve, and a trend evaluation result is constructed based on the stability index and the trend score.

4. The optical module management method according to claim 1, characterized in that, The step of performing performance analysis on the optical module based on the trend assessment results and the state change curve to obtain a health status level and availability score includes: The trend assessment results are subjected to stability screening, and a trend score and state volatility are extracted based on the screening results. The state volatility is used to calculate the performance degradation factor of the optical module, and the performance degradation factor is combined with the historical maintenance cycle to determine the aging coefficient. The historical maintenance cycle refers to the cumulative maintenance records of each optical module recorded by the communication equipment during its life cycle. The trend score, the performance degradation factor and the aging coefficient are combined to obtain a comprehensive health score. The comprehensive health score is then mapped to a preset health status level range to obtain the corresponding health status level. Based on the state change curve, the performance availability trend within a few future time slices is analyzed to obtain a trend availability index. The health status level is then matched and compared with the trend availability index to obtain an availability score.

5. The optical module management method according to claim 4, characterized in that, The step of combining the trend score, the performance degradation factor, and the aging coefficient to obtain a comprehensive health score, and mapping the comprehensive health score to a preset health status level range to obtain the corresponding health status level, includes: The trend score is normalized to an interval value to obtain a normalized trend score. The performance decay factor is then adjusted by exponential weighting to obtain a weighted decay factor. The weighted decay factor and the aging coefficient are multiplied and fused to obtain the long-term degradation intensity index. The normalized trend score and the long-term degradation intensity index are weighted and linearly superimposed to obtain the comprehensive health score. The comprehensive health score is mapped to a preset health status level range to obtain the corresponding health status level.

6. The optical module management method according to claim 1, characterized in that, The steps of prioritizing all optical modules based on the health status level and the availability score to obtain an optical module scheduling policy, and adjusting the optical module channels in the communication device according to the optical module scheduling policy to obtain the channel mapping relationship, include: The health status level and availability score of all optical modules are standardized to generate a two-dimensional scheduling index matrix. Based on the two-dimensional scheduling index matrix, all optical modules are clustered to obtain different scheduling priority groups. All the scheduling priorities are grouped and input into a preset scheduling weight model, and the optical module scheduling strategy is output. The optical module scheduling strategy is then input into the channel configuration module to perform optical module channel switching operation. After the channel switching is completed, the link quality indicators and state vectors before and after the switching are collected to perform stability assessment, and the stability assessment results are obtained. Based on the stability assessment results, the channel mapping relationship is generated.

7. The optical module management method according to claim 1, characterized in that, The step of fusing and analyzing the channel mapping relationship, the state change curve, the trend assessment result, and the health status level to obtain potential risk events, and uploading the potential risk events to the remote network control platform, includes: Align the channel mapping relationship of each optical module with its corresponding state change curve in time to construct a channel-state association table, perform statistical analysis on the channel-state association table, and extract channel stability indicators. The channel stability index, the trend assessment result, and the health status level are fused and analyzed to obtain an abnormal pattern label set. Potential risk events are screened based on the abnormal pattern label set and uploaded to the remote network control platform.

8. A management system for an optical module, characterized in that, include: The data acquisition module is used to acquire the operating status data and link status data of each optical module in the communication device, and to perform state modeling on the optical module using the operating status data to obtain the state vector corresponding to each optical module. The performance analysis module is used to input the state vector and the link state data into a preset parameter evolution analysis model to obtain state change curves and trend evaluation results. The performance of the optical module is analyzed based on the trend evaluation results and the state change curves to obtain the health status level and availability score. The priority sorting module is used to sort the scheduling priorities of all optical modules based on the health status level and the availability score to obtain an optical module scheduling strategy, and adjust the optical module channels in the communication device according to the optical module scheduling strategy to obtain a channel mapping relationship. The fusion analysis module is used to fuse and analyze the channel mapping relationship, the state change curve, the trend assessment result, and the health status level to obtain potential risk events, and upload the potential risk events to the remote network control platform.

9. An electronic device, characterized in that, The device includes a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the optical module management method as described in any one of claims 1-8.

Images