Optical power detection-based optical fiber performance automatic test method, system and equipment

Abstract

The invention relates to the technical field of communication, and discloses an optical fiber performance automatic test method, system and device based on optical power detection. The method comprises the following steps: carrying out intelligent scanning detection on an optical fiber network, generating optical fiber resource mapping data, obtaining an OTDR measurement configuration table through dynamic analysis, executing optical pulse injection and signal acquisition to obtain a time domain response characteristic curve, obtaining an optical fiber performance parameter set through wavelet denoising and peak value identification processing, and obtaining the optical fiber performance parameter set. And performing fault analysis by using deep learning and expert rules to obtain diagnosis decision data, and finally generating an optical fiber state evaluation report through digital twin modeling. According to the invention, remote automatic testing, intelligent fault diagnosis and visual monitoring management of the optical fiber network can be realized, and the operation and maintenance efficiency and reliability of the optical fiber communication network are improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a method, system and device for automatically testing optical fiber performance based on optical power detection. Background Art

[0002] With the widespread adoption of fiber-optic communications in power systems, the requirements for the operation, maintenance, and monitoring of fiber-optic communication networks are increasing. Currently, fiber-optic network maintenance primarily relies on manual fiber patching operations performed on-site in substation equipment rooms to achieve cross-connections between different optical fibers. Manual fiber data entry is required to maintain statistical management of fiber core usage. Furthermore, on-site performance testing of spare fiber cores is required to identify potential safety hazards. Existing technology primarily uses OTDR testers for fiber performance testing. These instruments emit probe light pulses and receive reflected and scattered light from the fiber to assess fiber performance parameters. However, this method requires on-site operation and data analysis by specialized personnel.

[0003] However, existing fiber optic testing technology has the following shortcomings: First, due to the dispersion of computer room sites, personnel access security management requirements, and the number and professional level of maintenance personnel, even if a large amount of cost is invested in the operation and maintenance of fiber optic communication networks, there are still problems such as slow fiber patching service activation, inaccurate fiber resource statistics, and many potential fiber faults. Second, traditional OTDR testing requires manual setting of test parameters, and the accuracy and reliability of test results largely rely on the operator's experience level. In addition, existing technologies lack the ability to intelligently analyze test data and diagnose faults, and are unable to timely detect and warn of potential network failure risks. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present application provides a method, system and equipment for automatic testing of optical fiber performance based on optical power detection, which is used to realize remote automated testing, intelligent fault diagnosis and visual monitoring and management of optical fiber networks without relying on manual on-site operation, thereby improving the operation and maintenance efficiency and reliability of optical fiber communication networks.

[0005] In the first aspect, the present application provides an automatic testing method for optical fiber performance based on optical power detection, the method comprising: performing a comprehensive detection of the optical fiber network structure through intelligent optical fiber scanning to generate optical fiber resource mapping data including optical fiber numbering data, optical fiber spatial position data, and optical path connection relationship data; performing test parameter calculation on the optical fiber resource mapping data through a dynamic analysis algorithm to obtain an OTDR measurement configuration table including wavelength selection parameters, pulse width, and sampling frequency; performing optical pulse injection and signal acquisition on the optical fiber according to the parameters specified in the OTDR measurement configuration table to obtain a time domain response characteristic curve of the optical fiber link; performing data processing on the time domain response characteristic curve through wavelet noise reduction and peak recognition algorithm to obtain an optical fiber performance parameter set including fault point coordinates, attenuation coefficient, and connection loss; performing a comprehensive analysis on the optical fiber performance parameter set through a deep learning network and an expert rule base to obtain diagnostic decision data including fault level, abnormality type, and optimization suggestions; visually converting the diagnostic decision data through digital twin modeling to obtain an optical fiber status assessment report including a topology map, loss distribution, and health score.

[0006] In a second aspect, the present application provides an automatic optical fiber performance testing system based on optical power detection, the automatic optical fiber performance testing system based on optical power detection comprising: The detection module is used to conduct a comprehensive detection of the optical fiber network structure through optical fiber intelligent scanning, and generate optical fiber resource mapping data including optical fiber number data, optical fiber spatial position data, and optical path connection relationship data; A calculation module is used to calculate test parameters of the optical fiber resource mapping data through a dynamic analysis algorithm to obtain an OTDR measurement configuration table including wavelength selection parameters, pulse width, and sampling frequency; An acquisition module, configured to perform optical pulse injection and signal acquisition on the optical fiber according to the parameters specified in the OTDR measurement configuration table, and obtain a time domain response characteristic curve of the optical fiber link; A processing module, configured to perform data processing on the time domain response characteristic curve using a wavelet noise reduction and peak recognition algorithm to obtain a set of optical fiber performance parameters including fault point coordinates, attenuation coefficient, and connection loss; An analysis module is used to comprehensively analyze the optical fiber performance parameter set through a deep learning network and an expert rule base to obtain diagnostic decision data including fault level, abnormality type, and optimization suggestions; The conversion module is used to perform visual conversion on the diagnostic decision data through digital twin modeling to obtain an optical fiber status assessment report including a topology map, loss distribution, and health score.

[0007] In a third aspect, the present application provides an electronic device comprising: a memory and a processor, wherein the memory stores a computer program, and when the processor executes the program, the method described in the first aspect is implemented.

[0008] The technical solution provided in this application uses intelligent optical fiber scanning for comprehensive detection, which can accurately obtain the fiber network number, optical fiber spatial location data, and connection relationship information, effectively avoiding manual recording errors and significantly improving the accuracy and efficiency of resource information collection. In addition, the dynamic analysis algorithm is used to calculate the test parameters of the optical fiber resource mapping data, realizing the intelligent configuration of the OTDR measurement parameters, ensuring the optimal match between the test parameters and the optical fiber characteristics, and improving the accuracy and reliability of the test results. At the same time, during the process of optical pulse injection and signal acquisition into the optical fiber, precise timing control and real-time sampling technology are used to ensure the time resolution and integrity of the measurement data, providing high-quality raw data for subsequent analysis. The time domain response characteristic curve is processed by wavelet noise reduction and peak recognition algorithm, effectively eliminating the influence of measurement noise, accurately identifying the fault point location and performance parameters, and improving the accuracy of fault diagnosis. The deep learning network and expert rule base are used for comprehensive analysis, combining the adaptive capabilities of machine learning with expert experience to achieve more accurate and intelligent fault diagnosis and provide reliable optimization suggestions. Finally, the digital twin modeling is used for visualization conversion, intuitively displaying the status and performance distribution of the optical fiber network, facilitating maintenance personnel to quickly understand and locate problems, and improving operation and maintenance efficiency. It should be understood that the contents described in the Summary of the Invention are not intended to limit the key or important features of the embodiments of the present application, nor are they intended to limit the scope of the present application. Other features of the present application will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are provided for a better understanding of the present invention and do not constitute a limitation of the present application. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which: Figure 1 This is a schematic diagram of an embodiment of a method for automatically testing optical fiber performance based on optical power detection in an embodiment of the present application; Figure 2 Schematic diagram of an embodiment of an automatic optical fiber performance testing system based on optical power detection in an embodiment of the present application; Figure 3 This is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0010] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0011] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0012] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In one embodiment of the present application, an automatic optical fiber performance testing method based on optical power detection includes: Step S101: Perform a comprehensive inspection of the optical fiber network structure through optical fiber intelligent scanning to generate optical fiber resource mapping data including optical fiber numbering data, optical fiber spatial position data, and optical path connection relationship data; Step S102: Calculate test parameters for the optical fiber resource mapping data using a dynamic analysis algorithm to obtain an OTDR measurement configuration table including wavelength selection parameters, pulse width, and sampling frequency; Step S103: injecting light pulses and collecting signals into the optical fiber according to the parameters specified in the OTDR measurement configuration table to obtain a time domain response characteristic curve of the optical fiber link; Step S104: Processing the time domain response characteristic curve using wavelet noise reduction and peak recognition algorithms to obtain a set of optical fiber performance parameters including fault point coordinates, attenuation coefficient, and connection loss; Step S105: Comprehensively analyze the optical fiber performance parameter set using a deep learning network and an expert rule base to obtain diagnostic decision data including fault level, abnormality type, and optimization suggestions; Step S106: Visually convert the diagnostic decision data through digital twin modeling to obtain a fiber status assessment report including topology map, loss distribution, and health score.

[0013] It is understandable that the execution subject of the present application can be an automatic optical fiber performance test system based on optical power detection, or a terminal or a server, which is not limited here. The embodiment of the present application is described by taking a server as the execution subject as an example.

[0014] Specifically, the fiber network structure is comprehensively inspected using intelligent fiber scanning technology. This technology uses a port recognition algorithm to scan the fiber network structure. The port recognition algorithm combines image processing with deep learning. A high-definition camera captures fiber port images. After image preprocessing and feature extraction, the images are input into a pre-trained deep convolutional neural network for recognition, obtaining fiber number data. Subsequently, a distributed fiber Bragg grating (FBG) positioning algorithm uses distributed fiber Bragg grating (FBG) sensors to spatially locate the fiber. By measuring changes in the grating reflection wavelength, the fiber's three-dimensional coordinates are determined, generating fiber spatial position data. The fiber number data is then connected using an optical path tracing algorithm. The optical path tracing algorithm uses an improved depth-first search strategy to gradually trace the optical signal transmission path from the fiber's starting port, recording each node and connection relationship passed through, and generating an initial optical path connection relationship. A topology analysis algorithm then performs correlation verification based on the optical fiber spatial position data and the initial optical path connection relationship. The topology analysis algorithm, based on graph theory principles, constructs a network connection graph. By verifying the physical feasibility of node connections, it eliminates incorrect connections and ensures the accuracy of the optical path connection relationship data.

[0015] An attribute association algorithm is used to extract features from fiber number data, fiber spatial location data, and optical path connection relationship data. Data features from different dimensions are then correlated and analyzed to obtain a complete resource feature set. Finally, the resource feature set is standardized using a data aggregation algorithm to generate standardized fiber resource mapping data. The second phase dynamically analyzes the fiber resource mapping data and calculates OTDR test parameters. First, the path analysis algorithm segments the fiber resource mapping data into different test segments based on features such as fiber length and type, and assigns an appropriate wavelength range to each segment. The dynamic threshold algorithm calculates the optimal wavelength based on fiber feature groupings, selecting the appropriate test wavelength based on the attenuation characteristics of different fiber types. The adaptive allocation algorithm calculates the pulse width based on the fiber length, selecting narrow pulses for shorter distances to improve spatial resolution and wide pulses for longer distances to ensure the measurement range. The frequency optimization algorithm determines the minimum sampling frequency based on the sampling theorem, while also considering system bandwidth limitations to determine the actual sampling frequency.

[0016] The parameter combination algorithm then optimizes the combination of parameters such as wavelength, pulse width, and sampling frequency to generate a test parameter matrix. This matrix is then converted into a standard OTDR measurement configuration table using a format conversion algorithm. The third stage involves fiber testing. The parameter parsing algorithm first parses the OTDR configuration table, extracts the test parameters, and generates a control sequence. The pulse modulation algorithm generates an optical signal of specified wavelength and pulse width based on the control sequence. The photoelectric converter adjusts the power of the optical signal to produce a standardized detection signal. The real-time sampling algorithm collects the optical fiber return signal at the configured sampling frequency. The timing processing algorithm synchronizes the time and amplitude of the sampled data, and the feature extraction algorithm is used to fit the complete time-domain response characteristic curve. The fourth stage processes the characteristic curve. The wavelet decomposition algorithm decomposes the curve into coefficients of different scales. Noise components are filtered out using threshold processing. The peak recognition algorithm locates characteristic points in the denoised data to obtain a fault point coordinate sequence.

[0017] The slope analysis algorithm calculates the attenuation coefficient of each curve segment, the splice loss calculation algorithm analyzes sudden loss at characteristic points, and the data fusion algorithm integrates these parameters to form a complete set of fiber performance parameters. The fifth stage involves fault diagnosis. The feature conversion algorithm performs standardized preprocessing on the performance parameters. A deep learning network is trained based on historical fault samples to develop a classification model and classify the current fault. The expert rule base matches fault types based on a pre-set rule base, and the correlation analysis algorithm retrieves a historical case library to provide optimization recommendations. A multi-dimensional evaluation algorithm comprehensively evaluates and prioritizes fault levels and optimization recommendations, and a decision fusion algorithm integrates the analysis results into complete diagnostic decision data. The final stage involves visualization. A topology mapping algorithm combines spatial information to construct a network structure model, a graphics rendering algorithm generates three-dimensional visualizations, an interpolation calculation algorithm processes the spatial distribution of loss data, a scoring calculation algorithm quantifies health status indicators, and a spatial registration algorithm correlates and maps the visualization model with the status data, ultimately generating a complete evaluation report.

[0018] For example, intelligent scanning discovered a 2000-meter single-mode fiber. Grating positioning determined its spatial coordinates as (x=10.5m, y=15.2m, z=2.1m), and the port identification number was SMF-2024-01. Dynamic analysis selected a 1310nm wavelength, 30ns pulse width, and a 100MHz sampling frequency for OTDR testing. Wavelet denoising was performed on the measured time-domain response curve, identifying three characteristic points at 500m, 1200m, and 1800m. The corresponding attenuation coefficients were 0.35dB / km, 0.38dB / km, and 0.42dB / km, respectively, and the connection losses were 0.21dB, 0.25dB, and 0.28dB. Deep learning diagnosis determined the risk to be medium, and the expert system recommended optimizing the connector at 1800m. The digital twin model visually displayed the loss distribution, giving the fiber a health score of 85, suggesting that the connector be addressed during routine maintenance.

[0019] In the embodiment of the present application, comprehensive detection is performed through intelligent optical fiber scanning, which can accurately obtain the fiber network number, optical fiber spatial location data and connection relationship information, effectively avoid manual recording errors, and significantly improve the accuracy and efficiency of resource information collection. In addition, the test parameters of the optical fiber resource mapping data are calculated through a dynamic analysis algorithm, which realizes the intelligent configuration of the OTDR measurement parameters, ensures the optimal match between the test parameters and the optical fiber characteristics, and improves the accuracy and reliability of the test results. At the same time, during the process of optical pulse injection and signal acquisition into the optical fiber, precise timing control and real-time sampling technology are adopted to ensure the time resolution and integrity of the measurement data, providing high-quality raw data for subsequent analysis. The time domain response characteristic curve is processed by wavelet noise reduction and peak recognition algorithm, effectively eliminating the influence of measurement noise, accurately identifying the fault point location and performance parameters, and improving the accuracy of fault diagnosis. A deep learning network and an expert rule base are used for comprehensive analysis, combining the adaptive ability of machine learning with expert experience, achieving more accurate and intelligent fault diagnosis, and providing reliable optimization suggestions. Finally, through digital twin modeling for visualization conversion, the status and performance distribution of the optical fiber network are intuitively displayed, which facilitates maintenance personnel to quickly understand and locate problems and improves operation and maintenance efficiency.

[0020] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Scan and detect the optical fiber network structure through the port identification algorithm to obtain the optical fiber number data; (2) According to the fiber number data, three-dimensional coordinates are collected through the grating positioning algorithm to obtain the fiber spatial position data; (3) Using the optical path tracking algorithm to determine the connection of the optical fiber number data, the initial connection relationship of the optical path is obtained; (4) Based on the optical fiber spatial position data and the initial connection relationship of the optical path, the topology analysis algorithm is used to perform correlation verification to obtain the optical path connection relationship data; (5) Extract features from the fiber number data, fiber spatial location data, and optical path connection relationship data using an attribute association algorithm to obtain a resource feature set; (6) The resource feature set is standardized and organized through a data aggregation algorithm to generate optical fiber resource mapping data.

[0021] Specifically, a high-definition camera was used to capture images of the fiber distribution frame at a resolution of 4096×3072 pixels. The captured raw images underwent image preprocessing, including grayscale conversion, histogram equalization, and edge enhancement, to improve image quality and highlight the fiber port features. The preprocessed images were then analyzed using a deep learning-based object detection network. The network employed a modified YOLOv5 architecture and was trained on 5000 labeled samples to develop a detection model that accurately identified the fiber port locations and corresponding identification information. The port identification algorithm outputs structured data containing port type, location coordinates, and number information, forming the fiber number data. The grating positioning algorithm, based on fiber Bragg grating sensing technology, pre-placed multiple Bragg grating sensing points on the optical fiber. The reflected wavelength of each sensing point changes with strain and temperature. A wavelength scanner scans the reflected signal within the 1510-1590 nm wavelength range to obtain the center wavelength of each sensing point. The strain value at that sensing point is then calculated using the grating calibration curve. The grating positioning algorithm uses triangulation principles and strain data from at least three known sensing points to construct a set of spatial coordinate equations to solve the fiber's three-dimensional coordinates. Positioning accuracy is better than 1cm, and the fiber's entire spatial trajectory is recorded in real time to generate fiber spatial position data.

[0022] The lightpath tracing algorithm uses a modified depth-first search strategy, starting from the fiber's starting port and progressively exploring connections along the signal transmission direction. The algorithm maintains a set of visited ports and a queue of ports to be visited. Each time a port is removed from the queue, the next port connected to it is determined through optical power detection, and the newly discovered port is added to the queue. This process continues until the queue is empty, during which all confirmed port connection pairs are recorded, ultimately generating a complete initial lightpath connection relationship. The topology analysis algorithm verifies the lightpath connection relationship data based on graph theory, representing each fiber port as a node in a graph and fiber connections as edges to construct an initial network topology. The algorithm first checks the connectivity of the topology graph to ensure that all nodes can reach each other via valid paths. It then verifies the physical feasibility of each connection, combining fiber spatial location data, to check whether the connection distance exceeds the fiber length limit and whether there are significant bends in the connection path. Connections that do not meet physical constraints are marked, and the connection relationships are adjusted through iterative optimization, ultimately outputting verified lightpath connection relationship data.

[0023] The attribute association algorithm extracts features and performs correlation analysis on multi-source heterogeneous data. It first converts fiber numbering data, fiber spatial location data, and optical path connection relationship data into standard feature vectors. These feature vectors contain key attributes such as port type, location coordinates, connected port, and transmission direction. It then calculates a correlation matrix between different features and identifies significant correlations, such as the spatial clustering of adjacent ports and the numbering patterns of consecutive ports. Based on the correlation analysis results, a core feature set that characterizes fiber resources is extracted to form a unified resource feature set. The data aggregation algorithm standardizes and structures the resource feature set. First, it normalizes feature data of different dimensions, mapping all feature values to a unified numerical range. It then organizes the feature information according to a predefined data model, aggregating the scattered feature data into a structured resource description encompassing multiple dimensions, including basic fiber attributes, spatial topology, and connection configuration information. Ultimately, fiber resource mapping data in a standardized format is generated, providing a data foundation for subsequent performance testing and analysis.

[0024] For example, a high-definition camera scans the ODF rack, capturing port images with a resolution of 2048×1536. After image enhancement, this image is fed into a trained deep learning model to identify the port locations and port numbers (SMF-2024-001 to SMF-2024-120). A grating positioning system deploys Bragg grating sensors every 50 meters. A wavelength scanner measures the reflection wavelength of the first sensor point as 1550.241 nm, with a strain value of 103 microstrain. Combined with data from other sensor points, the spatial coordinate trajectory of the optical fiber is calculated. Optical path tracing begins at the starting port SMF-2024-001 and, using power detection, gradually tracks the fiber to the ending port SMF-2024-085, recording the complete connection path. Topological analysis identifies a suspected over-the-distance connection, which is later verified as a false positive and then corrected. Attribute association analysis extracts 15 key features from the data of 120 ports. After data aggregation, a standardized resource mapping dataset is generated, containing port attributes, spatial information, and connection relationships.

[0025] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) The fiber resource mapping data is segmented using a path analysis algorithm to obtain fiber feature groups containing different wavelength selection parameter ranges; (2) Based on the fiber feature grouping, wavelength preselection calculation is performed using a dynamic threshold algorithm to obtain wavelength selection parameters; (3) Calculate the fiber length based on the fiber resource mapping data through the adaptive allocation algorithm to obtain the corresponding pulse width; (4) Perform sampling calculation based on wavelength selection parameters and pulse width through frequency optimization algorithm to obtain the sampling frequency; (5) Integrate the wavelength selection parameters, pulse width and sampling frequency through the parameter combination algorithm to obtain the test parameter matrix; (6) The test parameter matrix is standardized through a format conversion algorithm to obtain an OTDR measurement configuration table.

[0026] Specifically, the algorithm analyzes the fiber links in the fiber resource mapping data and segments them based on characteristics such as fiber type, length, and application scenario. Different wavelength parameter ranges are used for different fiber types. Single-mode fiber is primarily segmented between 1310nm and 1550nm, while multimode fiber is segmented between 850nm and 1300nm. The algorithm divides the fiber links into several test segments based on the fiber attenuation characteristics and application requirements, assigning an appropriate wavelength selection range to each segment to generate fiber characteristic grouping data. The dynamic threshold algorithm performs wavelength preselection calculations based on the fiber characteristic grouping data and sets dynamic wavelength selection thresholds based on the characteristic parameters of each fiber segment. The algorithm determines the optimal wavelength parameters by considering factors such as the fiber's dispersion, attenuation, and nonlinear effects. For example, for standard single-mode fiber G.652, a wavelength of 1310nm is preferred for short-distance testing to achieve better spatial resolution, while a wavelength of 1550nm is preferred for long-distance testing to reduce transmission loss. The algorithm uses an iterative optimization process to gradually adjust the wavelength parameters until the test requirements are met, ultimately outputting the wavelength selection parameters.

[0027] The adaptive allocation algorithm calculates pulse width based on the fiber link length information recorded in the fiber resource mapping data. Shorter fiber segments use narrower pulse widths to improve spatial resolution, while longer segments use wider pulse widths to ensure adequate measurement range. The algorithm establishes a mapping between fiber length and pulse width, using a 3ns pulse width for segments 0-500 meters, a 10ns pulse width for segments 500-2000 meters, and a 30ns pulse width for segments above 2000 meters. Dynamic adjustments are made based on actual signal quality. The frequency optimization algorithm calculates the sampling frequency based on the selected wavelength parameters and pulse width. It determines the minimum sampling frequency requirement based on the Nyquist sampling theorem, while also considering the bandwidth limitations and signal processing capabilities of the OTDR measurement system to optimize the sampling frequency configuration. For tests with a 3ns pulse width, the sampling frequency is set to 500MHz to ensure sufficient sampling points. For wider pulses, the sampling frequency can be reduced, such as using a 100MHz sampling frequency for a 30ns pulse, to ensure signal integrity while improving data processing efficiency.

[0028] The parameter combination algorithm systematically integrates key parameters such as wavelength selection, pulse width, and sampling frequency, constructing a multidimensional parameter space for optimization analysis. The algorithm evaluates the impact of different parameter combinations on test performance and selects the optimal parameter combination. For example, when testing with a 1310nm wavelength and 10ns pulse, a 200MHz sampling frequency is used to achieve ideal measurement results. All parameter combination information is organized into a structured test parameter matrix, containing specific configuration information for each test segment. The format conversion algorithm converts the test parameter matrix into a standardized OTDR measurement configuration table, which uses a unified data format and units to ensure that the test parameters can be correctly identified and executed by the OTDR device. The conversion process includes steps such as data format standardization, unit unification, and parameter validity verification, ultimately generating an OTDR measurement configuration table containing complete test configuration information.

[0029] For example, the path analysis algorithm analyzes a 3000-meter G.652 single-mode fiber and divides it into three segments: 0-500 meters, 500-2000 meters, and 2000-3000 meters. The dynamic threshold algorithm selects 1310 nm as the primary test wavelength and 1550 nm as the secondary test wavelength based on the characteristics of each segment. The adaptive allocation algorithm configures pulse widths based on length segments: 3 ns pulses for the first segment, 10 ns pulses for the second segment, and 30 ns pulses for the third segment. The frequency optimization algorithm configures sampling frequencies of 500 MHz, 200 MHz, and 100 MHz accordingly. The parameter combination algorithm generates a 3×3 parameter matrix containing combinations of wavelength, pulse width, and sampling frequency. The format conversion algorithm converts the parameter matrix into a standard OTDR configuration table, containing standardized information such as the test segment number, start and end positions, wavelength selection, pulse width, and sampling frequency, to guide subsequent OTDR measurements.

[0030] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Extract parameters from the OTDR measurement configuration table through a parameter parsing algorithm to obtain the optical pulse injection control sequence; (2) Generate an optical signal using a pulse modulation algorithm according to the optical pulse injection control sequence to obtain an optical pulse signal to be injected; (3) The optical pulse signal is power-adjusted through a photoelectric converter to obtain a detection signal injected into the optical fiber; (4) According to the sampling frequency in the OTDR measurement configuration table, the optical fiber return signal is collected through the real-time sampling algorithm to obtain the original response data; (5) The original response data is time synchronized and amplitude calibrated through the time series processing algorithm to obtain standardized response data; (6) The standardized response data is curve fitted using a feature extraction algorithm to obtain the time domain response characteristic curve.

[0031] Specifically, the parameter information in the OTDR measurement configuration table is read, and parameters such as wavelength, pulse width, and sampling frequency of each test section are parsed and verified. The algorithm converts these parameters into timing control commands that can be recognized by the light source control system, including light source switching timing, wavelength switching sequence, pulse triggering timing, etc., to form a complete optical pulse injection control sequence. Each control sequence contains a precise timestamp and corresponding control instructions to ensure that the parameter switching during the test is accurate and controllable. The pulse modulation algorithm generates the required optical pulse signal according to the control sequence, generates continuous light of the specified wavelength through a semiconductor laser, and performs pulse modulation through an electro-optical modulator. During the modulation process, the algorithm accurately controls the driving voltage waveform of the modulator to ensure that the generated optical pulse has an ideal time waveform and spectral characteristics. According to different pulse width requirements, the algorithm automatically adjusts the rise time and fall time of the modulation signal to achieve fast switching of narrow pulses and stable output of wide pulses.

[0032] The optical-to-electrical converter uses a highly linear PIN photodiode for photoelectric conversion, and a precision transimpedance amplifier amplifies and regulates the power of the optical signal. During the conversion process, the gain setting is automatically adjusted based on the fiber length and expected attenuation to ensure that the detection signal power level meets the measurement requirements. Furthermore, the converter's output is equipped with an automatic gain control circuit that monitors and adjusts the output signal amplitude in real time to maintain detection signal stability. A real-time sampling algorithm uses a high-speed analog-to-digital converter to acquire data from the returned optical signal. The sampling clock is synchronized with the optical pulse injection to ensure that the sampling points precisely correspond to the signal timing. The algorithm sets the sampling clock according to the sampling frequency specified in the configuration table. For a sampling frequency of 500 MHz, a data point is collected every 2 nanoseconds, recording the intensity changes of the backscattered light signal in real time. Timestamp information is also recorded during the sampling process for subsequent data processing and analysis.

[0033] The timing processing algorithm first performs time base calibration on the collected raw response data, eliminating sampling timing jitter using a reference clock signal to ensure the time accuracy of the data points. It then performs amplitude calibration, using a built-in calibration light source for absolute power calibration, converting the sampled values into standardized optical power units. The algorithm also performs zero-drift compensation and linearity correction to eliminate systematic errors in the measurement system and obtain accurate standardized response data. The feature extraction algorithm performs mathematical processing and curve fitting on the standardized response data, first using a median filter to remove discrete noise points and then applying a polynomial fitting method to smooth the data points. The algorithm then uses the least squares method to determine the coefficients of the fitting curve, generating a continuous time-domain response characteristic curve that accurately reflects the transmission characteristics of the optical fiber link.

[0034] For example, the parameter parsing algorithm extracts test parameters from the OTDR configuration table: 1310nm wavelength, 10ns pulse width, and 200MHz sampling frequency, generating an injection sequence consisting of 100 control instructions. The pulse modulation algorithm then generates an optical pulse signal with a peak power of 100mW and a repetition frequency of 10kHz. The photoelectric converter adjusts the input signal to a detection power level of 20mW. The real-time sampling algorithm collects the return signal at a frequency of 200MHz, acquiring 1 million sample points within 5 seconds. The time series processing algorithm processes the raw data, converting the sampled values to a standard optical power range of -20dBm to -70dBm. The feature extraction algorithm generates a smooth response curve using a 9th-order polynomial fit, which shows a 0.5dB dropout at 2km.

[0035] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Perform multi-scale analysis on the time domain response characteristic curve using the wavelet decomposition algorithm to obtain the noise characteristic coefficient; (2) Filter the signal using a threshold processing algorithm based on the noise characteristic coefficient to obtain the denoised response data; (3) The peak recognition algorithm is used to locate the characteristic points of the denoised response data to obtain a characteristic point sequence containing the coordinates of the fault point; (4) Calculate the loss using the slope analysis algorithm based on the characteristic point sequence to obtain the optical fiber attenuation coefficient data; (5) Analyze the connection points of the characteristic point sequence using the joint loss calculation algorithm to obtain the optical fiber connection loss data; (6) The fault point coordinates, attenuation coefficient and connection loss are integrated through the data fusion algorithm to obtain the optical fiber performance parameter set.

[0036] Specifically, a wavelet basis function suitable for the characteristics of the optical fiber signal is selected, and the db4 wavelet is used to perform a multi-layer decomposition of the time-domain response characteristic curve, decomposing the original signal into different frequency components. During the decomposition process, the signal is iteratively decomposed using high-pass and low-pass filters to obtain a series of wavelet coefficients, including low-frequency approximation coefficients and high-frequency detail coefficients. The algorithm uses a five-layer wavelet decomposition to separate the trend and noise components in the signal, generating a noise characteristic coefficient matrix containing characteristics at different scales. The threshold processing algorithm uses statistical principles to adaptively set thresholds for the noise characteristic coefficients and employs a soft thresholding method to filter high-frequency noise coefficients. The algorithm estimates the optimal threshold based on the noise level at each decomposition layer, shrinks coefficients exceeding the threshold, and sets coefficients below the threshold to zero. After threshold processing, the processed coefficients are converted back to the time domain through wavelet reconstruction to obtain denoised response data.

[0037] The peak identification algorithm uses a gradient-based search method to detect feature points in the denoised response data. The algorithm first calculates the first-order difference of the signal, marking the points where the differential value changes sign as potential feature points. It then screens the data based on features such as peak amplitude and width to eliminate false peaks. For each confirmed feature point, its time position and amplitude information are recorded, and its spatial coordinates are converted using the speed of light to form a feature point sequence containing the fault point location information. The slope analysis algorithm performs linear regression analysis on the signal segments between the feature points and calculates the slope value of each signal segment. Because the optical power attenuation in optical fibers follows an exponential law, the algorithm takes the logarithm of the signal and performs a linear fit. The slope value directly corresponds to the attenuation coefficient of the optical fiber. Through segmented calculation and statistical analysis, the attenuation characteristics of each optical fiber segment are obtained, generating complete optical fiber attenuation coefficient data.

[0038] The splice loss calculation algorithm specifically analyzes the breakpoints within a sequence of characteristic points, determining the connection loss by calculating the difference in signal levels before and after the breakpoint. The algorithm uses a local averaging method to mitigate the effects of random fluctuations. A window of appropriately wide width is selected on either side of the characteristic point to calculate the average power level. The difference between the two represents the insertion loss introduced by the splice. Statistical analysis is performed on all connection points to generate fiber connection loss data containing location and loss values. The data fusion algorithm utilizes multi-source information fusion to comprehensively process multidimensional data such as fault point coordinates, attenuation coefficients, and connection loss. The algorithm establishes a unified data structure, correlating and integrating different types of performance parameters based on their spatial locations and assigning corresponding confidence weights. Correlation analysis verifies the consistency of each parameter, eliminates anomalous data, and ultimately generates a standardized set of fiber performance parameters.

[0039] For example, after five layers of wavelet decomposition, the RMS value of the noise coefficient at the highest layer was 0.02 dB. Thresholding processing resets high-frequency coefficients below 0.05 dB to zero, improving the signal-to-noise ratio of the reconstructed signal by 12 dB. Peak identification detected five characteristic points in the processed data: at 500, 1000, 1500, 2000, and 2500 meters. Slope analysis calculated the attenuation coefficients of each fiber segment to be 0.32 dB / km, 0.35 dB / km, 0.33 dB / km, and 0.34 dB / km, respectively. Splice loss calculations revealed that the connection losses at the five characteristic points were 0.15 dB, 0.18 dB, 0.21 dB, 0.23 dB, and 0.20 dB, respectively. Data fusion generated a complete performance parameter set, identifying a potential problem point at 2000 meters with a high attenuation coefficient and significant connection loss.

[0040] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Standardize the fiber performance parameter set through the feature conversion algorithm to obtain the normalized feature vector; (2) Perform fault classification calculation on the normalized feature vector through a deep learning network to obtain the fault level; (3) Match the fault level with the expert rule base to obtain the abnormal type; (4) Retrieve historical cases based on the anomaly type using a correlation analysis algorithm to obtain optimization suggestions; (5) Prioritize the fault levels and optimization suggestions through a multi-dimensional evaluation algorithm to obtain an optimization strategy sequence; (6) The fault level, abnormality type and optimization strategy sequence are integrated through the decision fusion algorithm to obtain diagnostic decision data.

[0041] Specifically, each indicator in the fiber performance parameter set is normalized, converting parameters of different dimensions to the range [0, 1]. This includes normalizing the attenuation coefficient to its maximum and minimum values, normalizing the connection loss data to z-scores, and converting the fault point coordinates to relative distances. The algorithm also constructs a feature descriptor, encoding the fiber's multidimensional performance indicators into a fixed-dimensional feature vector containing information on multiple dimensions, including attenuation, loss, and location. The deep learning network employs a multi-layer convolutional neural network architecture for fault classification, consisting of three convolutional layers and two fully connected layers. The input layer receives the normalized feature vector and extracts local correlations among the fiber performance characteristics through convolution operations. The pooling layer performs feature dimensionality reduction and extracts key features. The fully connected layer maps the features to the fault classification space, and the output layer calculates the probability distribution of each fault level using a softmax function. The network is trained on 10,000 historical fault samples to establish a mapping between feature vectors and fault levels.

[0042] The expert rule base performs reasoning analysis based on a pre-set fiber optic fault diagnosis rule base, which contains 200 fiber optic fault diagnosis rules covering common fault types and anomaly patterns. The algorithm employs a forward reasoning strategy, using the fault level derived from deep learning as factual input. Through rule matching and reasoning chains, it generates a detailed description of the anomaly type, including information such as the nature of the fault, scope of impact, and severity. A correlation analysis algorithm intelligently searches a historical fault case library containing 5,000 verified fiber optic fault handling cases. The algorithm calculates the similarity between the current anomaly type and historical cases, using cosine similarity to measure the correlation between feature vectors. The case with the highest similarity is selected as a reference. Based on the selected similar cases, the algorithm extracts treatment methods and evaluates their effectiveness, forming targeted optimization recommendations.

[0043] The multi-dimensional evaluation algorithm comprehensively evaluates fault handling solutions from multiple dimensions, including technical feasibility, implementation difficulty, and resource requirements. The algorithm establishes a scoring model to score each optimization suggestion, taking into account factors such as repair effectiveness, implementation cost, and time requirements. The weight of each evaluation dimension is determined through the hierarchical analysis method, and the comprehensive score of the solution is calculated. The optimization strategies are prioritized according to the score. The decision fusion algorithm uses the Dempster-Shafer evidence theory framework to integrate the credibility of multi-source information such as fault level, anomaly type, and optimization strategy. The algorithm calculates the basic probability distribution of each data source, integrates information from multiple evidence sources through evidence synthesis rules, and generates the final decision recommendation. The fused diagnostic decision data contains complete fault diagnosis information, processing suggestions, and credibility assessment.

[0044] For example, the feature conversion algorithm converts the original parameters into a 100-dimensional feature vector, containing normalized attenuation coefficients, loss values, and location information. Deep learning network analysis results indicate a fault level of 3 (moderate fault) with a confidence level of 0.92. Expert rule reasoning determined it to be a joint degradation fault, with the impact limited to the connection point at 2,000 meters. Correlation analysis retrieved three similar cases from the case library and recommended a re-welding solution. A multi-dimensional evaluation and comparison of multiple treatment options resulted in the re-welding solution being rated 85 out of 100, ranking first among the optimization strategies. After decision fusion, a diagnostic report was generated, confirming the fault as a joint degradation fault and recommending re-welding within two weeks, which is expected to reduce loss by 0.15dB.

[0045] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Perform spatial structure analysis on diagnostic decision data through topological mapping algorithm to obtain a topological map; (2) Based on the topological map, a view is constructed using a graphics rendering algorithm to obtain a three-dimensional visual topological map; (3) The loss information in the diagnosis decision data is processed spatially through interpolation calculation algorithm to obtain the loss distribution; (4) Quantify the health status of the diagnostic decision data through a scoring calculation algorithm to obtain a health score; (5) The three-dimensional visualization topology map and loss distribution are correlated and mapped through a spatial registration algorithm to obtain state visualization data; (6) The status visualization data and health score are integrated through the report generation algorithm to obtain the fiber status assessment report.

[0046] Specifically, the fiber optic network structure information in the diagnostic decision data is parsed to extract the fiber connection relationship and spatial location information. The algorithm uses graph theory to construct a network topology model, representing fiber ports as nodes and fiber connections as edges, while recording the three-dimensional spatial coordinate information of each node. The network structure layout is optimized using the minimum spanning tree algorithm to ensure the clarity and readability of the topology map, and a network topology structure diagram containing complete connection relationships is generated. The graphics rendering algorithm visualizes the topology map based on the WebGL three-dimensional graphics engine and maps the fiber optic network structure into a virtual three-dimensional space through three-dimensional modeling. The algorithm uses a geometry shader to generate a three-dimensional curve of the fiber optic path and enhances the visual effect through texture mapping and lighting calculation. The rendering process implements perspective roaming, zooming operations, and node interaction functions, supporting the observation of the network structure from different angles and the rapid location of areas of interest.

[0047] The interpolation algorithm uses Kriging spatial interpolation to estimate the continuous distribution of discrete loss measurement data. The algorithm considers spatial correlation and calculates loss estimates at any point based on the numerical and location information of known loss points. Spatial correlation is described using a variogram model, interpolation parameters are optimized, and a high-precision loss distribution map is generated, visually displaying the spatial variation characteristics of loss. The scoring algorithm establishes a multi-level health evaluation indicator system, encompassing dimensions such as transmission performance, structural integrity, and reliability. The algorithm uses a fuzzy comprehensive evaluation method to calculate a health score based on various performance indicators. The score is standardized to a range of 0-100, taking into account indicator weights and grading. Both sub-item scores and an overall score are generated to quantitatively reflect the overall health of the fiber link.

[0048] The spatial registration algorithm uses a feature point matching method to precisely align the 3D topology map with the loss distribution map. The algorithm extracts feature points from the two datasets, calculates the spatial transformation matrix, and achieves unified registration of the coordinate systems. The registration process optimizes the transformation parameters using an iterative closest point algorithm to ensure the accuracy of the visualization and generate status visualization data that integrates topology and loss information. The report generation algorithm uses a templated approach to construct assessment reports, formatting and displaying status visualization data and health score information in a standard format. The algorithm automatically generates fault diagnosis conclusions, health status assessments, and optimization recommendations, and intuitively presents the assessment results through charts, data tables, and other formats. The report also includes historical trend analysis and predictive maintenance recommendations, providing a basis for operation and maintenance decisions.

[0049] Taking a data center fiber optic network as an example, topology mapping analysis revealed that the network contains 120 fiber ports, forming a star-shaped connection structure. Graphical rendering generates a three-dimensional view, clearly showing the orientation and cross-connection relationships of each fiber layer. Interpolation calculations revealed a high-loss area at 2000 meters, with a peak loss of 0.45dB. The health scoring system rated this fiber segment 85 for transmission performance, 92 for structural integrity, and 88 for overall health. Spatial registration visually displays the correspondence between loss distribution and physical location, highlighting areas requiring special attention. The resulting assessment report details all diagnostic and assessment information and recommends optimizing high-loss areas during the next maintenance.

[0050] The above describes the automatic optical fiber performance test method based on optical power detection in the embodiment of the present application. The following describes the automatic optical fiber performance test system based on optical power detection in the embodiment of the present application. Figure 2 In one embodiment of the present application, an automatic optical fiber performance test system based on optical power detection includes: The detection module 201 is used to perform a comprehensive detection of the optical fiber network structure through optical fiber intelligent scanning, and generate optical fiber resource mapping data including optical fiber number data, optical fiber spatial position data, and optical path connection relationship data; A calculation module 202 is configured to calculate test parameters of the optical fiber resource mapping data using a dynamic analysis algorithm to obtain an OTDR measurement configuration table including wavelength selection parameters, pulse width, and sampling frequency; An acquisition module 203 is configured to perform optical pulse injection and signal acquisition on the optical fiber according to the parameters specified in the OTDR measurement configuration table, and obtain a time domain response characteristic curve of the optical fiber link; The processing module 204 is configured to perform data processing on the time domain response characteristic curve using a wavelet noise reduction and peak recognition algorithm to obtain a set of optical fiber performance parameters including the coordinates of the fault point, the attenuation coefficient, and the connection loss; An analysis module 205 is configured to perform a comprehensive analysis on the optical fiber performance parameter set using a deep learning network and an expert rule base to obtain diagnostic decision data including fault level, abnormality type, and optimization suggestions; The conversion module 206 is used to perform visual conversion on the diagnostic decision data through digital twin modeling to obtain an optical fiber status assessment report including a topology map, loss distribution, and health score.

[0051] Through the collaborative efforts of these components, comprehensive inspections performed through intelligent fiber scanning accurately capture fiber network numbers, fiber spatial location data, and connection relationship information, effectively avoiding manual recording errors and significantly improving the accuracy and efficiency of resource information collection. Furthermore, a dynamic analysis algorithm calculates test parameters based on fiber resource mapping data, enabling intelligent configuration of OTDR measurement parameters. This ensures optimal matching of test parameters with fiber characteristics, improving the accuracy and reliability of test results. Precise timing control and real-time sampling techniques are employed during optical pulse injection and signal acquisition to ensure the temporal resolution and integrity of the measurement data, providing high-quality raw data for subsequent analysis. Wavelet noise reduction and peak recognition algorithms are used to process the time-domain response characteristic curves, effectively eliminating the effects of measurement noise and accurately identifying fault locations and performance parameters, improving fault diagnosis accuracy. Comprehensive analysis using a deep learning network and expert rule base combines the adaptive capabilities of machine learning with expert experience, enabling more accurate and intelligent fault diagnosis and providing reliable optimization recommendations. Finally, digital twin modeling is used for visualization, visually displaying the status and performance distribution of the fiber network, enabling maintenance personnel to quickly understand and locate problems, thereby improving operation and maintenance efficiency.

[0052] Figure 3 A schematic block diagram of an electronic device 300 that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0053] The electronic device 300 includes a computing unit 301, which can perform various appropriate actions and processes according to a computer program stored in a ROM 302 or a computer program loaded from a storage unit 308 into a RAM 303. The RAM 303 may also store various programs and data required for the operation of the electronic device 300. The computing unit 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An I / O interface 305 is also connected to the bus 304.

[0054] Multiple components in the electronic device 300 are connected to the I / O interface 305, including an input unit 306, such as a keyboard, a mouse, etc.; an output unit 307, such as various types of displays, speakers, etc.; a storage unit 308, such as a magnetic disk, an optical disk, etc.; and a communication unit 309, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 309 allows the electronic device 300 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0055] The computing unit 301 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 301 performs the various methods and processes described above, such as the automatic optical fiber performance testing method based on optical power detection. For example, in some embodiments, the automatic optical fiber performance testing method based on optical power detection can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 308. In some embodiments, part or all of the computer program can be loaded and / or installed into the electronic device 300 via the ROM 302 and / or the communication unit 309. When the computer program is loaded into the RAM 303 and executed by the computing unit 301, one or more steps of the automatic optical fiber performance testing method based on optical power detection described above can be performed. Alternatively, in other embodiments, the computing unit 301 may be configured to execute the optical fiber performance automatic testing method based on optical power detection in any other appropriate manner (for example, by means of firmware).

[0056] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0057] The program code for implementing the methods of the present application can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow charts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0058] In the context of this application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0059] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0060] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0061] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0062] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved. This is not a limitation herein.

[0063] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A method for automatically testing optical fiber performance based on optical power detection, characterized in that: The method comprises: Comprehensively detect the optical fiber network structure through optical fiber intelligent scanning, and generate optical fiber resource mapping data including optical fiber number data, optical fiber spatial location data, and optical path connection relationship data; Calculating test parameters of the optical fiber resource mapping data using a dynamic analysis algorithm to obtain an OTDR measurement configuration table including wavelength selection parameters, pulse width, and sampling frequency; Performing optical pulse injection and signal acquisition on the optical fiber according to the parameters specified in the OTDR measurement configuration table to obtain a time domain response characteristic curve of the optical fiber link; Processing the time domain response characteristic curve using wavelet noise reduction and peak recognition algorithms to obtain a set of optical fiber performance parameters including fault point coordinates, attenuation coefficient, and connection loss; Comprehensively analyzing the optical fiber performance parameter set through a deep learning network and an expert rule base to obtain diagnostic decision data including fault level, abnormality type, and optimization suggestions; The diagnostic decision data is visualized and converted through digital twin modeling to obtain an optical fiber status assessment report including topology map, loss distribution, and health score.

2. The optical fiber performance automatic testing method based on optical power detection according to claim 1 is characterized in that: The optical fiber network structure is comprehensively detected by optical fiber intelligent scanning to generate optical fiber resource mapping data containing optical fiber numbering data, optical fiber spatial position data, and optical path connection relationship data, including: Scan and detect the optical fiber network structure through the port identification algorithm to obtain optical fiber number data; Performing three-dimensional coordinate acquisition using a grating positioning algorithm based on the optical fiber numbering data to obtain optical fiber spatial position data; Performing connection determination on the optical fiber number data by using an optical path tracking algorithm to obtain an initial optical path connection relationship; Performing correlation verification based on the optical fiber spatial position data and the initial connection relationship of the optical path through a topology analysis algorithm to obtain optical path connection relationship data; Extracting features from the optical fiber numbering data, the optical fiber spatial position data, and the optical path connection relationship data using an attribute association algorithm to obtain a resource feature set; The resource feature set is standardized and sorted through a data aggregation algorithm to generate the optical fiber resource mapping data.

3. The optical fiber performance automatic testing method based on optical power detection according to claim 1, characterized in that: The optical fiber resource mapping data is subjected to test parameter calculation by a dynamic analysis algorithm to obtain an OTDR measurement configuration table including wavelength selection parameters, pulse width, and sampling frequency, including: Dividing the optical fiber resource mapping data into sections using a path analysis algorithm to obtain optical fiber feature groups containing different wavelength selection parameter ranges; Performing wavelength preselection calculations based on the optical fiber feature groups using a dynamic threshold algorithm to obtain wavelength selection parameters; Calculating the fiber length of the fiber resource mapping data using an adaptive allocation algorithm to obtain a corresponding pulse width; Perform sampling calculation according to the wavelength selection parameter and the pulse width through a frequency optimization algorithm to obtain a sampling frequency; Integrating the wavelength selection parameter, the pulse width, and the sampling frequency through a parameter combination algorithm to obtain a test parameter matrix; The test parameter matrix is standardized through a format conversion algorithm to obtain the OTDR measurement configuration table.

4. The optical fiber performance automatic testing method based on optical power detection according to claim 1 is characterized in that: The step of injecting light pulses and collecting signals on the optical fiber according to the parameters specified in the OTDR measurement configuration table to obtain a time domain response characteristic curve of the optical fiber link includes: Extracting parameters from the OTDR measurement configuration table using a parameter parsing algorithm to obtain an optical pulse injection control sequence; generating an optical signal by a pulse modulation algorithm according to the optical pulse injection control sequence to obtain an optical pulse signal to be injected; The optical pulse signal is power-adjusted by a photoelectric converter to obtain a detection signal injected into the optical fiber; According to the sampling frequency in the OTDR measurement configuration table, data acquisition is performed on the optical fiber return signal through a real-time sampling algorithm to obtain original response data; Performing time synchronization and amplitude calibration on the original response data through a time series processing algorithm to obtain standardized response data; The standardized response data is subjected to curve fitting through a feature extraction algorithm to obtain the time domain response characteristic curve.

5. The optical fiber performance automatic testing method based on optical power detection according to claim 1, characterized in that: The time domain response characteristic curve is processed by wavelet noise reduction and peak recognition algorithm to obtain a set of optical fiber performance parameters including fault point coordinates, attenuation coefficient, and connection loss, including: Performing multi-scale analysis on the time domain response characteristic curve by wavelet decomposition algorithm to obtain noise characteristic coefficient; Performing signal filtering using a threshold processing algorithm according to the noise characteristic coefficient to obtain denoised response data; Performing feature point positioning on the denoised response data using a peak recognition algorithm to obtain a feature point sequence containing the coordinates of the fault point; Calculating the loss using a slope analysis algorithm based on the characteristic point sequence to obtain optical fiber attenuation coefficient data; Performing connection point analysis on the characteristic point sequence using a joint loss calculation algorithm to obtain optical fiber connection loss data; The fault point coordinates, the attenuation coefficient, and the connection loss are integrated through a data fusion algorithm to obtain the optical fiber performance parameter set.

6. The optical fiber performance automatic testing method based on optical power detection according to claim 1, characterized in that: The optical fiber performance parameter set is comprehensively analyzed through a deep learning network and an expert rule base to obtain diagnostic decision data including fault level, abnormality type, and optimization suggestions, including: Normalizing the optical fiber performance parameter set by a feature conversion algorithm to obtain a normalized feature vector; Performing fault classification calculation on the normalized feature vector through a deep learning network to obtain a fault level; Perform rule matching on the fault level through an expert rule base to obtain an abnormality type; According to the anomaly type, historical case retrieval is performed using a correlation analysis algorithm to obtain optimization suggestions; Prioritizing the fault levels and the optimization suggestions using a multi-dimensional evaluation algorithm to obtain an optimization strategy sequence; The fault level, the abnormality type and the optimization strategy sequence are integrated through a decision fusion algorithm to obtain the diagnosis decision data.

7. The optical fiber performance automatic testing method based on optical power detection according to claim 1, characterized in that: The diagnostic decision data is visualized and converted through digital twin modeling to obtain a fiber status assessment report containing a topology map, loss distribution, and health score, including: Performing spatial structure analysis on the diagnostic decision data using a topological mapping algorithm to obtain a topological map; According to the topological map, a view is constructed by a graphics rendering algorithm to obtain a three-dimensional visual topological map; Performing spatial distribution processing on the loss information in the diagnosis decision data by an interpolation calculation algorithm to obtain a loss distribution; quantifying the health status of the diagnostic decision data using a scoring calculation algorithm to obtain a health score; Correlation mapping is performed between the three-dimensional visualization topology map and the loss distribution through a spatial registration algorithm to obtain state visualization data; The status visualization data and the health score are integrated through a report generation algorithm to obtain the optical fiber status assessment report.

8. An automatic optical fiber performance test system based on optical power detection, used to implement the automatic optical fiber performance test method based on optical power detection as described in any one of claims 1 to 7, characterized in that: The optical fiber performance automatic testing system based on optical power detection includes: The detection module is used to conduct a comprehensive detection of the optical fiber network structure through optical fiber intelligent scanning, and generate optical fiber resource mapping data including optical fiber number data, optical fiber spatial position data, and optical path connection relationship data; A calculation module is used to calculate test parameters of the optical fiber resource mapping data through a dynamic analysis algorithm to obtain an OTDR measurement configuration table including wavelength selection parameters, pulse width, and sampling frequency; An acquisition module, configured to perform optical pulse injection and signal acquisition on the optical fiber according to the parameters specified in the OTDR measurement configuration table, and obtain a time domain response characteristic curve of the optical fiber link; A processing module, configured to perform data processing on the time domain response characteristic curve using a wavelet noise reduction and peak recognition algorithm to obtain a set of optical fiber performance parameters including fault point coordinates, attenuation coefficient, and connection loss; An analysis module is used to comprehensively analyze the optical fiber performance parameter set through a deep learning network and an expert rule base to obtain diagnostic decision data including fault level, abnormality type, and optimization suggestions; The conversion module is used to perform visual conversion on the diagnostic decision data through digital twin modeling to obtain an optical fiber status assessment report including a topology map, loss distribution, and health score.

9. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, so as to enable the at least one processor to perform the method according to any one of claims 1 to 7.

Images