MT ferrule multichannel parallel optical transmission system and automatic alignment method

Abstract

The invention discloses an MT ferrule multichannel parallel optical transmission system and an automatic alignment method, and the method comprises the steps: collecting the end face array structure information of an MT ferrule, carrying out the optical fiber arrangement topology analysis and channel space coordinate mapping based on the end face array structure information, and forming a channel space reference map; analyzing the signal feature code of each parallel channel, performing channel state attribute deconstruction according to the signal feature code, generating a channel identity label library, and setting a multi-dimensional alignment anchor point matrix according to the channel space reference map and the channel identity label library; performing offset detection and analysis on the real-time channel transmission state of each parallel channel, generating a coupling offset field, performing adjustment action serialization arrangement processing according to the coupling offset field, and generating a precision calibration instruction chain; displacement trend prediction and mechanical stability evaluation are carried out according to the end face array structure information and the working condition data, a pre-judgment precision calibration instruction chain is formed, and the alignment efficiency, the dynamic stability and the signal transmission consistency of multi-channel transmission are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of optical fiber communication technology, specifically to an MT ferrule multi-channel parallel optical transmission system and an automatic alignment method. Background Technology

[0002] With the rapid development of 5G communication, cloud computing, and data centers, unprecedented demands have been placed on the bandwidth, density, and reliability of optical transmission systems. Multi-channel parallel optical transmission technology, by integrating multiple independent optical channels (such as 12-core, 24-core, or 48-core) into a single path, can achieve transmission rates in the tens of terabit range (Tbps), becoming a core solution for high-density data exchange. Among these, the MT (Mechanical Transfer Ferrule), as a key connecting component of multi-channel fiber optic arrays, achieves simultaneous splicing of multiple optical fibers through a precise mechanical structure and is widely used in core equipment such as optical modules, fiber optic connectors, and optical backplanes.

[0003] The core performance of MT ferrules lies in multi-channel alignment accuracy—the fiber arrays at the transmitter and receiver must achieve micron-level or even nanometer-level spatial matching to ensure efficient coupling of optical signals (coupling efficiency typically needs to be ≥90%). If the alignment deviation exceeds the threshold (e.g., the core diameter of a single-mode fiber is only 9μm, and the allowable deviation is typically ≤0.5μm), it will directly lead to signal attenuation, increased crosstalk, and even system failure. In existing technologies, factors such as changes in environmental temperature and humidity and vibration can easily cause channel offset, and there is a lack of real-time compensation mechanisms; moreover, the materials of MT ferrules and fixing structures (such as ceramics, metals, and plastics) have long-term creep characteristics, and will produce slow displacement (0.1~0.5μm drift per month) under continuous stress; existing technologies can only passively respond to the offset that has occurred and cannot predict long-term drift trends, resulting in frequent system maintenance and high operation and maintenance costs. Summary of the Invention

[0004] To solve the above-mentioned technical problems, the present invention aims to provide an MT ferrule multi-channel parallel optical transmission system, including a cloud, wherein the cloud communication connection includes a data acquisition module, an anchoring module, an offset monitoring and processing module, and a dynamic compensation module; The data acquisition module is used to collect the end-face array structure information of the MT ferrule, and to perform fiber arrangement topology analysis and channel space coordinate mapping based on the end-face array structure information to form a channel space reference map. The anchoring module is used to parse the signal feature codes of each parallel channel, deconstruct the channel state attributes based on the signal feature codes, generate a channel identity database, and set a multi-dimensional alignment anchor point matrix based on the channel spatial reference map and the channel identity database. The offset monitoring and processing module is used to detect and analyze the real-time channel transmission status of each parallel channel, generate a coupled offset field, and perform adjustment action sequence arrangement processing based on the coupled offset field to generate a precision calibration instruction chain. The dynamic compensation module is used to predict displacement trends and assess mechanical stability based on end-face array structure information and operating data, forming a predictive precision calibration instruction chain.

[0005] Furthermore, the process of acquiring end-face array structure information of the MT ferrule includes: Before optical transmission in the MT ferrule, an end-face image of the MT ferrule is acquired. The end-face image is then denoised and enhanced. Feature extraction is performed on the denoised and enhanced end-face image to obtain the core feature points of the optical fiber. The distribution parameters of the core feature points of the optical fiber are acquired, and a preset arrangement rule is established. Based on the arrangement rule, the distribution parameters of the core feature points of the optical fiber are clustered and matched to obtain the arrangement rule of the MT ferrule end face. A standard topology template of the MT ferrule end face is obtained. The distribution parameters and arrangement rule of the MT ferrule end face are compared with the standard topology template for consistency. If the distribution parameters and arrangement rule of the MT ferrule end face are consistent with the standard topology template, channel space coordinate mapping is performed. If the distribution parameters or arrangement rule of the MT ferrule end face are inconsistent with the standard topology template, a production defect warning for the MT ferrule is generated.

[0006] Furthermore, the process of forming a spatial reference map of the channel includes: The arrangement rules of the MT ferrule end face are abstracted into a logical association graph. Based on the logical association graph, the relative position index of each fiber channel in the MT ferrule end face is obtained, and the physical reference point of the MT ferrule end face is obtained. With the physical reference point as the origin, the coordinate axis direction is defined, and a spatial coordinate system is constructed. Based on the distribution parameters of the MT ferrule and the relative position index of each fiber channel in the MT ferrule end face, the design coordinates and logical index ID of each fiber channel in the spatial coordinate system are obtained. Map each fiber core feature point in the end face image to a spatial coordinate system to obtain the actual coordinates of each fiber channel. Compare the actual coordinates of each fiber channel with the design coordinates and calibrate the spatial coordinate system based on the comparison results. Obtain the design coordinates of each fiber channel in the calibrated spatial coordinate system and associate the design coordinates of each fiber channel with the logical index ID to construct a channel spatial reference map.

[0007] Furthermore, the process of generating the channel identity database includes: Preset test signals are synchronously injected into all fiber optic channels of the MT ferrule, and the output signals of all fiber optic channels are synchronously acquired. Core feature parameters are extracted from the output signals of each fiber optic channel, and signal feature codes are constructed based on the core feature parameters. The output signals corresponding to the signal feature codes are classified and analyzed for state attributes to obtain the standard state attributes corresponding to the signal feature codes. The design coordinates and logical index ID of each optical fiber channel in the channel spatial reference map are bound to the signal feature code and standard status attribute corresponding to each optical fiber channel to generate a channel identity library.

[0008] Furthermore, the process of setting up the multidimensional alignment anchor point matrix includes: Based on the channel spatial reference map, obtain the physical reference points in the spatial coordinate system, the design coordinates and arrangement rules of all fiber optic channels, set inherent structural anchor points in the spatial coordinate system according to the physical reference points and arrangement rules, obtain the standard state attributes of all fiber optic channels in the spatial coordinate system based on the channel identity database, divide the spatial coordinate system into several sub-regions, set regional performance anchor points in each sub-region of the spatial coordinate system according to the standard state attributes, and construct a multi-dimensional alignment anchor point matrix based on the inherent structural anchor points, regional performance anchor points and non-anchor points in the spatial coordinate system.

[0009] Furthermore, the process of generating the coupled migration field includes: A quantitative comparison table was constructed based on previous calibration experiments. The quantitative comparison table includes the quantitative relationship between the offset characteristics and performance parameter changes of all fiber channels, as well as the quantitative relationship between the offset characteristics and performance parameter changes of a single fiber channel. When the MT ferrule transmits optical light, the real-time channel transmission status of each fiber channel of the MT ferrule is collected and mapped to a multi-dimensional alignment anchor point matrix. The real-time channel transmission status of all non-anchor points is compared with the real-time channel transmission status of the inherent structural anchor points to obtain the performance parameter changes of all non-anchor points. If all non-anchor points have the same performance parameter change, the performance parameter change of that type is marked as the global performance parameter change. Based on the global performance parameter change and the quantization lookup table, the global offset characteristics of all non-anchor points relative to the inherent structural anchor points are obtained. The real-time channel transmission status of each non-anchor point is compared with the real-time channel transmission status of the regional performance anchor point of the sub-region to which each non-anchor point belongs. The change amount of local performance parameters of each non-anchor point is obtained. The local performance parameter change threshold of each non-anchor point is preset. If the change amount of local performance parameters of a non-anchor point is greater than the corresponding local performance parameter change threshold, it is determined that the non-anchor point has shifted. Based on the change amount of local performance parameters of the non-anchor point that has shifted, the local shift characteristics of the non-anchor point relative to the regional performance anchor point are obtained by combining the quantization comparison table. An offset distribution map is constructed in a spatial coordinate system based on the local and global offset characteristics of the fiber optic channel.

[0010] Furthermore, the process of adjusting the sequence of actions based on the coupled offset field to generate a precision calibration command chain includes: Based on the local offset characteristics of each non-anchor point relative to the regional performance anchor point in the offset distribution map, local offset correction values ​​for each anchor point are generated. The magnitudes of the offsets of each non-anchor point relative to the regional performance anchor point in the offset distribution map are compared. The local offset correction values ​​of each non-anchor point are sorted in ascending order according to the offsets, and an action correction sequence for each non-anchor point is generated. It is determined whether there are global offset characteristics in the offset distribution map. If there are global offset characteristics, a global offset correction value is generated based on the global offset characteristics and inserted at the beginning of the action correction sequence. A precision calibration command chain is generated based on the action correction sequence, and the fiber optic channel of the MT ferrule is scheduled according to the precision calibration command chain.

[0011] Furthermore, the process of forming a predictive precision calibration command chain includes: A displacement trend prediction model is constructed, and historical operating data and arrangement rules of various MT ferrules are extracted as training data. The historical operating data includes the offset distribution map of MT ferrules under different working conditions within the collection period. The displacement trend prediction model is trained using the training data to obtain the completed displacement trend prediction model. The system collects real-time operating data of the MT ferrule and marks the collection period. It then performs a mechanical stability assessment of the MT ferrule to obtain its compensation period. Based on the compensation period, the system divides the next collection period of the MT ferrule into several compensation sub-cycles. The end timestamps of each compensation sub-cycle and the end timestamp of the collection period are used as compensation time points. The operating data and arrangement rules of the MT ferrule in the current collection period are input into the shift trend prediction model. Based on the shift trend prediction model, the system outputs the offset distribution prediction map of each compensation time point in the next collection period. Based on the offset distribution prediction map, the system performs sequential arrangement processing of the entire action to generate a predictive precision calibration command chain for each compensation time point. Finally, the system schedules the fiber optic channel of the MT ferrule according to the predictive precision calibration command chain.

[0012] Furthermore, the process of conducting a mechanical stability assessment of the MT ferrule and obtaining its compensation period includes: The core evaluation indicators of the mechanical structure of the MT ferrule, the corresponding telecom-grade qualification thresholds for each core evaluation indicator, and the compensation periods for stable and unstable states are set. The core evaluation indicators of the mechanical structure of the MT ferrule are measured to obtain the specific values ​​of each core evaluation indicator. The specific values ​​of each core evaluation indicator are compared with the corresponding telecom-grade qualification thresholds. If the specific values ​​of each core evaluation indicator are all less than the corresponding telecom-grade qualification thresholds, the mechanical structure is marked as stable and the compensation period corresponding to the stable state is obtained. If the specific value of any core evaluation indicator is greater than the corresponding telecom-grade qualification threshold, the mechanical structure is marked as unstable and the compensation period corresponding to the unstable state is obtained.

[0013] An automatic alignment method includes the following steps: Step 1: Collect the end-face array structure information of the MT ferrule, and perform fiber arrangement topology analysis and channel space coordinate mapping based on the end-face array structure information to form a channel space reference map; Step 2: Analyze the signal feature codes of each parallel channel, deconstruct the channel state attributes based on the signal feature codes, generate a channel identity database, and set a multi-dimensional alignment anchor point matrix based on the channel spatial reference map and the channel identity database. Step 3: Perform offset detection and analysis on the real-time channel transmission status of each parallel channel to generate a coupled offset field. Based on the coupled offset field, perform sequential arrangement of adjustment actions to generate a precision calibration instruction chain. Step 4: Based on the end-face array structure information and operating data, perform displacement trend prediction and mechanical stability assessment to form a predictive precision calibration instruction chain.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. By systematically collecting the structural information of the MT ferrule end face array, and combining topology analysis and channel spatial coordinate mapping, a precise channel spatial reference map is constructed from the source. This provides reliable data support for all subsequent alignment and compensation actions, effectively avoiding alignment deviations caused by fuzzy structural information, ensuring the core accuracy of multi-channel transmission. Furthermore, by setting up a multi-dimensional alignment anchor matrix, the identity, spatial position, and status attributes of the channels are deeply bound together, forming a unified and traceable alignment reference framework. This avoids problems such as misconnection, missing connection, or inefficient coupling in multi-channel parallel transmission, making the spatial correlation and performance matching of each channel clearer and more reliable.

[0015] 2. By monitoring the channel transmission status in real time and generating a coupling offset field, various spatial offsets can be quickly captured. Then, precise correction is achieved through an ordered calibration command chain, avoiding transmission performance degradation caused by offset accumulation. This ensures that the system maintains a stable transmission state during operation, reducing signal interruptions or quality degradation. Based on shift trend prediction and mechanical stability assessment, it breaks through the limitations of traditional "passive calibration" by proactively addressing potential shift risks through a predictive calibration command chain. This effectively reduces the occurrence of sudden failures, extends the system's stable operation cycle, and lowers the probability of transmission interruptions caused by shifts. Attached Figure Description

[0016] Figure 1 This is a flowchart of an MT ferrule multi-channel parallel optical transmission system according to an embodiment of this application.

[0017] Figure 2 This is a flowchart of an automatic alignment method according to an embodiment of this application. 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, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] like Figure 1 As shown, an MT ferrule multi-channel parallel optical transmission system includes a cloud platform, which is connected to a data acquisition module, an anchoring module, an offset monitoring and processing module, and a dynamic compensation module. The data acquisition module is used to collect the end-face array structure information of the MT ferrule, and to perform fiber arrangement topology analysis and channel space coordinate mapping based on the end-face array structure information to form a channel space reference map. The anchoring module is used to parse the signal feature codes of each parallel channel, deconstruct the channel state attributes based on the signal feature codes, generate a channel identity database, and set a multi-dimensional alignment anchor point matrix based on the channel spatial reference map and the channel identity database. The offset monitoring and processing module is used to detect and analyze the real-time channel transmission status of each parallel channel, generate a coupled offset field, and perform adjustment action sequence arrangement processing based on the coupled offset field to generate a precision calibration instruction chain. The dynamic compensation module is used to predict displacement trends and assess mechanical stability based on end-face array structure information and operating data, forming a predictive precision calibration instruction chain.

[0020] It should be further explained that, in the specific implementation process, the process of collecting the end face array structure information of the MT ferrule includes: Before optical transmission in the MT ferrule, an end-face image of the MT ferrule is acquired. The end-face image undergoes denoising and enhancement processing (including eliminating ambient light interference through Gaussian filtering and using edge detection algorithms (such as the Canny operator) to enhance the boundary contrast of the fiber core / cladding, avoiding feature misidentification caused by blurred areas). Feature extraction is performed on the denoised and enhanced end-face image to obtain the core feature points of the fiber (the center coordinates of each fiber). Distribution parameters of the core feature points are collected (number of cores, the spacing between the X-axis coordinates of the center coordinates of each fiber, and the spacing between the Y-axis coordinates of the center coordinates of each fiber). Pre-defined arrangement rules are used, and clustering matching is performed on the distribution parameters of the core feature points based on these rules. For example, if the X-axis coordinates are consistent and the Y-axis coordinates increase at equal intervals, it is determined to be a linear arrangement; if both the X and Y axis coordinates are equally spaced, it is determined to be a rectangular matrix arrangement. The arrangement rules of the MT ferrule end face are obtained, and a standard topology template (such as a 12-core rectangular matrix) is obtained from the MT ferrule design specifications. The standard spacing of the MT ferrules in the array arrangement is 250μm. The distribution parameters and arrangement rules of the MT ferrule end face are compared with the standard topology template for consistency. Specifically, the fiber arrangement rules, core count, and spacing obtained from the analysis are compared item by item with the standard topology template of the MT ferrule, and the deviation of key parameters (the difference between the actual spacing and the design spacing) is calculated. If the deviation exceeds the industry standard (±10μm), a production defect warning for the MT ferrule is generated. If the distribution parameters and arrangement rules of the MT ferrule end face are consistent with the standard topology template (the arrangement rules and core count are consistent, and the deviation of key parameters does not exceed the industry standard), channel space coordinate mapping is performed. If the distribution parameters or arrangement rules of the MT ferrule end face are inconsistent with the standard topology template (the arrangement rules or core count are inconsistent, or the deviation of key parameters exceeds the industry standard), a production defect warning for the MT ferrule is generated. For example, if the spacing between adjacent fibers in the actual arrangement deviates from the design value, it is judged as a fiber misalignment defect; if the actual core count is less than the design value (e.g., 23 cores are analyzed when the design is 24 cores), it is judged as a fiber missing defect.

[0021] It should be further explained that, in the specific implementation process, the process of performing fiber optic topology analysis and channel spatial coordinate mapping based on the end-face array structure information to form a channel spatial reference map includes: The arrangement rules of the MT ferrule endface are abstracted into a logical association diagram. Based on this diagram, the relative position index of each fiber channel within the MT ferrule endface is obtained. For example, the physical structure of a 24-core MT ferrule endface is a "2x12 rectangular array, with a lateral spacing (column spacing) of 250μm and a longitudinal spacing (row spacing) of 250μm between adjacent fibers." This is abstracted into a logical association diagram of "(row index, column index) → fiber number." For instance, the 1st row, 3rd column corresponds to fiber number 5, and the 2nd row, 5th column corresponds to fiber number 17 (associated through the rule "1st row 1-12, 2nd row 13-24"). This provides the physical basis of the MT ferrule endface. The reference point (guide hole for ferrule mating) is used as the origin to define the coordinate axis directions: X-axis: along the "horizontal" direction of the MT ferrule end face (such as the column direction of a rectangular array), positive to the right; Y-axis: along the "vertical" direction of the MT ferrule end face (such as the row direction of a rectangular array), positive downwards; Z-axis: along the fiber axis (perpendicular to the end face), with the end face plane as Z=0, negative towards the inside of the ferrule (used to reflect whether the fiber end face is flush). A spatial coordinate system is constructed. Based on the distribution parameters of the MT ferrule and the relative position index of each fiber channel in the MT ferrule end face, the design coordinates and logical index ID of each fiber channel in the spatial coordinate system are obtained. The specific process includes: Key physical quantities are obtained from the distribution parameters, including: Fiber spacing: Horizontal spacing Px = 250μm (distance between columns), vertical spacing Py = 250μm (distance between rows); Offset between the reference point and the first fiber channel: For example, the distance from the origin to the center of the first fiber channel in the first row and first column is (X0, Y0) = (500μm, 500μm); The design coordinates and logical index ID of each fiber channel are obtained based on key physical quantities and relative position indices. For example, for a 2×12 rectangular array, the fiber channel in the i-th row and j-th column (i=1,2; j=1,12) has a logical index ID of (i-1)×12+j, and the corresponding physical position rules are as follows: X=X0+(j-1)×Px, Y=Y0+(i-1)×Py, Z=0 (ideally, all fiber end faces are flush). Substitute the parameters (X0=500μm, Y0=500μm, Px=250μm, Py=250μm): Row 1, Column 1 (ID=1): X=500+0=500μm, Y=500+0=500μm→(500,500,0); Row 1, Column 2 (ID=2): X=500+250=750μm, Y=500→(750,500,0); Row 2, Column 1 (ID=13): X=500, Y=500+250=750μm→(500,750,0).

[0022] The core feature points of each fiber in the end-face image are mapped to a spatial coordinate system to obtain the actual coordinates of each fiber channel. The actual coordinates of each fiber channel are then compared with the design coordinates for consistency. Based on the comparison results, the spatial coordinate system is calibrated. The specific process includes: comparing the actual coordinates with the design coordinates and calculating the deviation (e.g., ΔX = X_actual - X_theoretical, ΔY = Y_actual - Y_theoretical). ΔX ≤ ±10μm is considered consistent. If the coordinate system deviation is caused by ferrule end-face tilt or lens fisheye effect (e.g., the X-axis is not parallel to the array's transverse direction), the spatial coordinate system is adjusted using the least squares fitting method to minimize the deviation of most fiber channels.

[0023] Obtain the design coordinates of each fiber channel in the calibrated spatial coordinate system, associate the design coordinates of each fiber channel with the logical index ID, and construct a channel spatial reference map (index → ​​coordinate lookup table, such as ID:1→(502,499,0.2), ID:2→(751,500,-0.1), etc.).

[0024] It should be further explained that, in the specific implementation process, the process of parsing the signal feature codes of each parallel channel, deconstructing the channel state attributes based on the signal feature codes, and generating a channel identity database includes: Preset test signals are synchronously injected into all fiber channels of the MT ferrule (the signal type must be unique and distinguishable, and the signal amplitude, modulation frequency, pulse width, etc. of the test signals must be consistent). Output signals of all fiber channels are synchronously acquired, and core feature parameters (including frequency characteristics, phase characteristics, and pulse characteristics) are extracted from the output signals of each fiber channel. Based on the core feature parameters, a signal feature code is constructed. The output signals corresponding to the signal feature codes are classified and analyzed for state attributes to obtain the standard state attributes (including insertion loss, optical power fluctuation, eye diagram parameters, and bit error rate) corresponding to the signal feature codes. The design coordinates and logical index ID of each optical fiber channel in the channel spatial reference map are bound to the signal feature code and standard status attribute corresponding to each optical fiber channel to generate a channel identity library.

[0025] It should be further explained that, in the specific implementation process, the process of setting up a multi-dimensional alignment anchor point matrix based on the channel spatial reference map and the channel identity database includes: Based on the channel spatial reference map, physical reference points, design coordinates, and arrangement rules of all fiber optic channels are obtained in the spatial coordinate system. Inherent structural anchor points are then set in the spatial coordinate system according to these physical reference points and arrangement rules. For example, the inherent physical structure of the MT ferrule end face, such as positioning holes, is fixed in position and unaffected by transmission status, serving as the global reference core and thus set as inherent structural anchor points. Simultaneously, channels whose inherent structural anchor points conform to the fiber array topology rules and reflect the overall arrangement logic are also selected. These channels can cover critical areas of the array, avoiding local alignment deviations. The selection logic for setting inherent structural anchor points based on arrangement rules is as follows: Linear array (e.g., 1×12 cores): Select the two end channels + the middle channel (e.g., cores 1, 6, and 12) to cover the entire X-axis range; Rectangular array (e.g., 2×12 cores): Select four corner channels + center channel (e.g., (1,1), (1,12), (2,1), (2,12), (1,6) cores) to cover the entire X / Y axis; High-density arrays (e.g., 4×12 cores): Filter by "sub-region division" (each 12 cores is a sub-region, and 2 channels are selected for each sub-region) to ensure that each sub-region has a reference anchor point.

[0026] Based on the channel identity database, the standard status attributes of all fiber optic channels in the spatial coordinate system are obtained. The spatial coordinate system is then divided into several sub-regions, and regional performance anchor points are set in each sub-region according to the standard status attributes. These regional performance anchor points are the channels with the "optimal transmission performance and most stable status" in the channel identity database. The alignment status of these channels directly determines the overall transmission quality and is the core performance reference. The selection criteria are as follows: Transmission parameters meet the following standards: insertion loss ≤ 0.2dB (better than the industry standard of 0.3dB), bit error rate ≤ (below the acceptable value) Optical power fluctuation ≤ 0.05dB; Stability verification: Three consecutive repeated tests (5-minute intervals) showed channel coordinate deviation ≤ ±0.2μm and performance parameter fluctuation ≤ 3%; Selection quantity: determined according to array size, ≥3 for 12-core, ≥4 for 24-core, and ≥6 for 48-core and above, to ensure redundancy backup.

[0027] Construct a multidimensional alignment anchor matrix based on the inherent structural anchor points, regional performance anchor points, and non-anchor points in the spatial coordinate system.

[0028] It should be further explained that, in the specific implementation process, the process of offset detection and analysis of the real-time channel transmission status to generate a coupled offset field includes: Based on previous calibration experiments (such as manually offsetting the ferrule by 0.5μm along the X-axis and collecting changes in all performance parameters, such as power attenuation of 0.2dB and bit error rate increase to...), A quantitative comparison table is constructed, which includes the quantitative relationship between the offset characteristics (including offset dimensions (X-axis, Y-axis and Z-axis) and offset amount) of all fiber channels and the changes in performance parameters (including optical power attenuation, bit error rate increase, return loss increase, etc.), as well as the quantitative relationship between the offset characteristics and the changes in performance parameters of a single fiber channel. When the MT ferrule transmits optical light, the real-time channel transmission status of each fiber channel of the MT ferrule is collected and mapped to a multi-dimensional alignment anchor point matrix. The real-time channel transmission status of all non-anchor points is compared with the real-time channel transmission status of the inherent structural anchor points to obtain the performance parameter changes of all non-anchor points. If all non-anchor points have a consistent performance parameter change of a certain type (such as average power attenuation of 0.15dB), then the performance parameter change of that type is marked as the global performance parameter change. Based on the global performance parameter change and the quantization lookup table, the global offset characteristics of all non-anchor points relative to the inherent structural anchor points are obtained. The real-time channel transmission status of each non-anchor point is compared with the real-time channel transmission status of the regional performance anchor point in the sub-region to which each non-anchor point belongs. The change in local performance parameters of each non-anchor point is obtained. A threshold for the change in local performance parameters of each non-anchor point is preset. If the change in local performance parameters of a non-anchor point exceeds the corresponding threshold, it is determined that the non-anchor point has shifted. Based on the change in local performance parameters of the shifted non-anchor point and a quantization lookup table, the local shift characteristics of the shifted non-anchor point relative to the regional performance anchor point are obtained. For example: If the optical power of a certain channel decreases by 0.2dB while other parameters are normal: according to the quantization comparison table, the local offset characteristic of this channel is "axial (Z-axis) offset of 0.5μm" (single channel offset, judged as local offset); If the optical power of all channels is attenuated by 0.15dB and other parameters are normal, then the global offset characteristic is "global axial offset 0.375μm" (multi-channel synchronous offset, determined to be global offset). An offset distribution map is constructed in a spatial coordinate system based on the local and global offset characteristics of the fiber optic channels. The offset distribution map includes offsets common to all channels (e.g., ΔX global = 0.3 μm, ΔY global = 0.2 μm) and offsets unique to individual / partial channels (e.g., ΔZ local = 0.4 μm for channel 5, ΔZ local = 0 for other channels).

[0029] It should be further explained that, in the specific implementation process, the process of adjusting the sequence of actions based on the coupled offset field to generate a precision calibration command chain includes: Based on the local offset characteristics of each non-anchor point relative to the regional performance anchor point in the offset distribution map, local offset correction values ​​are generated for each anchor point. The offset magnitudes of the local offset characteristics of each non-anchor point relative to the regional performance anchor point are compared. The local offset correction values ​​of each non-anchor point are sorted in ascending order according to the offset magnitude, and an action correction sequence is generated for each non-anchor point (the larger the offset, the higher the priority in the action sequence). It is determined whether there are global offset characteristics in the offset distribution map. If there are global offset characteristics, a global offset correction value (specifically, the X-axis correction value is -0.2μm) is generated based on the global offset characteristics (e.g., X-axis offset is ΔX=+0.2μm). The global offset correction value is inserted at the beginning of the action correction sequence. A precision calibration command chain is generated based on the action correction sequence, and the fiber channel of the MT ferrule is scheduled according to the precision calibration command chain.

[0030] For precision calibration command chains, targeted calibration commands (such as offset repair detected by coupled offset field) are provided to solve existing displacement problems. These commands include: logical index IDs of non-anchor points; mechanical mechanisms: specified drive components (such as global X-axis drivers, local channel 5 drivers), support structures; and adjustment parameters: direction (+X / -X / +Y / -Y / +Z / -Z / ±θ), speed (such as 1μm / s to avoid overshoot due to excessive speed) and offset correction values ​​(such as ΔX target = 0μm).

[0031] It should be further explained that, in the specific implementation process, the process of predicting displacement trends and assessing mechanical stability based on end-face array structure information and operating condition data, and forming a predictive precision calibration command chain, includes: A displacement trend prediction model is constructed based on the LSTM time series model. Historical operating data and arrangement rules of various MT ferrules are extracted as training data. The historical operating data includes the displacement distribution map of MT ferrules under different operating conditions (temperature, vibration) within the collection period. The displacement trend prediction model is trained using the training data to obtain the completed displacement trend prediction model. The system collects real-time operating data of the MT ferrule and marks the acquisition period. It then performs a mechanical stability assessment of the MT ferrule to obtain its compensation period. Based on this compensation period, the system divides the next acquisition period of the MT ferrule into several sub-compensation periods. The end timestamps of each sub-compensation period and the end timestamp of the acquisition period are used as compensation time points. The operating data and arrangement rules of the MT ferrule in the current acquisition period are input into the shift trend prediction model. Based on this model, the system outputs a predicted offset distribution map for each compensation time point in the next acquisition period. The system then performs sequential arrangement processing based on the offset distribution prediction map to generate a predictive precision calibration instruction chain for each compensation time point (providing predictive compensation instructions such as periodic fine-tuning due to gradual temperature changes and real-time vibration-resistant adjustment for sudden vibrations, addressing potential shift problems). Finally, the system schedules the fiber optic channel of the MT ferrule based on this predictive precision calibration instruction chain.

[0032] It should be further explained that, in the specific implementation process, the process of conducting a mechanical stability assessment of the MT ferrule and obtaining the compensation period of the MT ferrule includes: The core evaluation indicators of the mechanical structure of the MT ferrule, the corresponding telecom-grade qualification thresholds for each core evaluation indicator, and the compensation periods for stable and unstable states are set. The core evaluation indicators of the mechanical structure of the MT ferrule are measured to obtain the specific values ​​of each core evaluation indicator. The specific values ​​of each core evaluation indicator are compared with the corresponding telecom-grade qualification thresholds. If the specific values ​​of each core evaluation indicator are all less than the corresponding telecom-grade qualification thresholds, the mechanical structure is marked as stable and the compensation period corresponding to the stable state is obtained (consistent with the acquisition period). If the specific value of any core evaluation indicator is greater than the corresponding telecom-grade qualification threshold, the mechanical structure is marked as unstable and the compensation period corresponding to the unstable state (100ms) is obtained.

[0033] The core evaluation indicators of the mechanical structure of the MT ferrule and the corresponding telecom-grade qualification thresholds for each core evaluation indicator are shown in Table 1 below:

[0034] An automatic alignment method includes the following steps: Step 1: Collect the end-face array structure information of the MT ferrule, and perform fiber arrangement topology analysis and channel space coordinate mapping based on the end-face array structure information to form a channel space reference map; Step 2: Analyze the signal feature codes of each parallel channel, deconstruct the channel state attributes based on the signal feature codes, generate a channel identity database, and set a multi-dimensional alignment anchor point matrix based on the channel spatial reference map and the channel identity database. Step 3: Perform offset detection and analysis on the real-time channel transmission status of each parallel channel to generate a coupled offset field. Based on the coupled offset field, perform sequential arrangement of adjustment actions to generate a precision calibration instruction chain. Step 4: Based on the end-face array structure information and operating data, perform displacement trend prediction and mechanical stability assessment to form a predictive precision calibration instruction chain.

[0035] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A multi-channel parallel optical transmission system with an MT ferrule, characterized in that, Including the cloud, the cloud communication connection includes a data acquisition module, an anchoring module, an offset monitoring and processing module, and a dynamic compensation module; The data acquisition module is used to collect the end-face array structure information of the MT ferrule, and to perform fiber arrangement topology analysis and channel space coordinate mapping based on the end-face array structure information to form a channel space reference map. The anchoring module is used to parse the signal feature codes of each parallel channel, deconstruct the channel state attributes based on the signal feature codes, generate a channel identity database, and set a multi-dimensional alignment anchor point matrix based on the channel spatial reference map and the channel identity database. The offset monitoring and processing module is used to detect and analyze the real-time channel transmission status of each parallel channel, generate a coupled offset field, and perform adjustment action sequence arrangement processing based on the coupled offset field to generate a precision calibration instruction chain. The dynamic compensation module is used to predict displacement trends and assess mechanical stability based on end-face array structure information and operating data, forming a predictive precision calibration instruction chain.

2. The MT ferrule multi-channel parallel optical transmission system according to claim 1, characterized in that, The process of acquiring end-face array structure information of the MT ferrule includes: Before optical transmission in the MT ferrule, an end-face image of the MT ferrule is acquired. The end-face image is then denoised and enhanced. Feature extraction is performed on the denoised and enhanced end-face image to obtain the core feature points of the optical fiber. The distribution parameters of the core feature points of the optical fiber are acquired, and a preset arrangement rule is established. Based on the arrangement rule, the distribution parameters of the core feature points of the optical fiber are clustered and matched to obtain the arrangement rule of the MT ferrule end face. A standard topology template of the MT ferrule end face is obtained. The distribution parameters and arrangement rule of the MT ferrule end face are compared with the standard topology template for consistency. If the distribution parameters and arrangement rule of the MT ferrule end face are consistent with the standard topology template, channel space coordinate mapping is performed. If the distribution parameters or arrangement rule of the MT ferrule end face are inconsistent with the standard topology template, a production defect warning for the MT ferrule is generated.

3. The MT ferrule multi-channel parallel optical transmission system according to claim 2, characterized in that, The process of forming a channel spatial reference map includes: The arrangement rules of the MT ferrule end face are abstracted into a logical association graph. Based on the logical association graph, the relative position index of each fiber channel in the MT ferrule end face is obtained, and the physical reference point of the MT ferrule end face is obtained. With the physical reference point as the origin, the coordinate axis direction is defined, and a spatial coordinate system is constructed. Based on the distribution parameters of the MT ferrule and the relative position index of each fiber channel in the MT ferrule end face, the design coordinates and logical index ID of each fiber channel in the spatial coordinate system are obtained. Map each fiber core feature point in the end face image to a spatial coordinate system to obtain the actual coordinates of each fiber channel. Compare the actual coordinates of each fiber channel with the design coordinates and calibrate the spatial coordinate system based on the comparison results. Obtain the design coordinates of each fiber channel in the calibrated spatial coordinate system and associate the design coordinates of each fiber channel with the logical index ID to construct a channel spatial reference map.

4. The MT ferrule multi-channel parallel optical transmission system according to claim 3, characterized in that, The process of generating a channel identity database includes: Preset test signals are synchronously injected into all fiber optic channels of the MT ferrule, and the output signals of all fiber optic channels are synchronously acquired. Core feature parameters are extracted from the output signals of each fiber optic channel, and signal feature codes are constructed based on the core feature parameters. The output signals corresponding to the signal feature codes are classified and analyzed for state attributes to obtain the standard state attributes corresponding to the signal feature codes. The design coordinates and logical index ID of each optical fiber channel in the channel spatial reference map are bound to the signal feature code and standard status attribute corresponding to each optical fiber channel to generate a channel identity library.

5. The MT ferrule multi-channel parallel optical transmission system according to claim 4, characterized in that, The process of setting up a multidimensional alignment anchor point matrix includes: Based on the channel spatial reference map, obtain the physical reference points in the spatial coordinate system, the design coordinates and arrangement rules of all fiber optic channels, set inherent structural anchor points in the spatial coordinate system according to the physical reference points and arrangement rules, obtain the standard state attributes of all fiber optic channels in the spatial coordinate system based on the channel identity database, divide the spatial coordinate system into several sub-regions, set regional performance anchor points in each sub-region of the spatial coordinate system according to the standard state attributes, and construct a multi-dimensional alignment anchor point matrix based on the inherent structural anchor points, regional performance anchor points and non-anchor points in the spatial coordinate system.

6. The MT ferrule multi-channel parallel optical transmission system according to claim 5, characterized in that, The process of generating the coupled migration field includes: A quantitative comparison table was constructed based on previous calibration experiments. The quantitative comparison table includes the quantitative relationship between the offset characteristics and performance parameter changes of all fiber channels, as well as the quantitative relationship between the offset characteristics and performance parameter changes of a single fiber channel. When the MT ferrule transmits optical light, the real-time channel transmission status of each fiber channel of the MT ferrule is collected and mapped to a multi-dimensional alignment anchor point matrix. The real-time channel transmission status of all non-anchor points is compared with the real-time channel transmission status of the inherent structural anchor points to obtain the performance parameter changes of all non-anchor points. If all non-anchor points have the same performance parameter change, the performance parameter change of that type is marked as the global performance parameter change. Based on the global performance parameter change and the quantization lookup table, the global offset characteristics of all non-anchor points relative to the inherent structural anchor points are obtained. The real-time channel transmission status of each non-anchor point is compared with the real-time channel transmission status of the regional performance anchor point of the sub-region to which each non-anchor point belongs. The change amount of local performance parameters of each non-anchor point is obtained. The local performance parameter change threshold of each non-anchor point is preset. If the change amount of local performance parameters of a non-anchor point is greater than the corresponding local performance parameter change threshold, it is determined that the non-anchor point has shifted. Based on the change amount of local performance parameters of the non-anchor point that has shifted, the local shift characteristics of the non-anchor point relative to the regional performance anchor point are obtained by combining the quantization comparison table. An offset distribution map is constructed in a spatial coordinate system based on the local and global offset characteristics of the fiber optic channel.

7. The MT ferrule multi-channel parallel optical transmission system according to claim 6, characterized in that, The process of adjusting the sequence of actions based on the coupled offset field to generate a precision calibration command chain includes: Based on the local offset characteristics of each non-anchor point relative to the regional performance anchor point in the offset distribution map, local offset correction values ​​for each anchor point are generated. The magnitudes of the offsets of each non-anchor point relative to the regional performance anchor point in the offset distribution map are compared. The local offset correction values ​​of each non-anchor point are sorted in ascending order according to the offsets, and an action correction sequence for each non-anchor point is generated. It is determined whether there are global offset characteristics in the offset distribution map. If there are global offset characteristics, a global offset correction value is generated based on the global offset characteristics and inserted at the beginning of the action correction sequence. A precision calibration command chain is generated based on the action correction sequence, and the fiber optic channel of the MT ferrule is scheduled according to the precision calibration command chain.

8. The MT ferrule multi-channel parallel optical transmission system according to claim 7, characterized in that, The process of forming a predictive precision calibration instruction chain includes: A displacement trend prediction model is constructed, and historical operating data and arrangement rules of various MT ferrules are extracted as training data. The historical operating data includes the offset distribution map of MT ferrules under different working conditions within the collection period. The displacement trend prediction model is trained using the training data to obtain the completed displacement trend prediction model. The system collects real-time operating data of the MT ferrule and marks the collection period. It then performs a mechanical stability assessment of the MT ferrule to obtain its compensation period. Based on the compensation period, the system divides the next collection period of the MT ferrule into several compensation sub-cycles. The end timestamps of each compensation sub-cycle and the end timestamp of the collection period are used as compensation time points. The operating data and arrangement rules of the MT ferrule in the current collection period are input into the shift trend prediction model. Based on the shift trend prediction model, the system outputs the offset distribution prediction map of each compensation time point in the next collection period. Based on the offset distribution prediction map, the system performs sequential arrangement processing of the entire action to generate a predictive precision calibration command chain for each compensation time point. Finally, the system schedules the fiber optic channel of the MT ferrule according to the predictive precision calibration command chain.

9. A multi-channel parallel optical transmission system with an MT ferrule according to claim 8, characterized in that, The process of conducting a mechanical stability assessment of the MT ferrule and obtaining its compensation period includes: The core evaluation indicators of the mechanical structure of the MT ferrule, the corresponding telecom-grade qualification thresholds for each core evaluation indicator, and the compensation periods for stable and unstable states are set. The core evaluation indicators of the mechanical structure of the MT ferrule are measured to obtain the specific values ​​of each core evaluation indicator. The specific values ​​of each core evaluation indicator are compared with the corresponding telecom-grade qualification thresholds. If the specific values ​​of each core evaluation indicator are all less than the corresponding telecom-grade qualification thresholds, the mechanical structure is marked as stable and the compensation period corresponding to the stable state is obtained. If the specific value of any core evaluation indicator is greater than the corresponding telecom-grade qualification threshold, the mechanical structure is marked as unstable and the compensation period corresponding to the unstable state is obtained.

10. An automatic alignment method, specifically applied to the MT ferrule multi-channel parallel optical transmission system according to any one of claims 1 to 9, characterized in that, Includes the following steps: Step 1: Collect the end-face array structure information of the MT ferrule, and perform fiber arrangement topology analysis and channel space coordinate mapping based on the end-face array structure information to form a channel space reference map; Step 2: Analyze the signal feature codes of each parallel channel, deconstruct the channel state attributes based on the signal feature codes, generate a channel identity database, and set a multi-dimensional alignment anchor point matrix based on the channel spatial reference map and the channel identity database. Step 3: Perform offset detection and analysis on the real-time channel transmission status of each parallel channel to generate a coupled offset field. Based on the coupled offset field, perform sequential arrangement of adjustment actions to generate a precision calibration instruction chain. Step 4: Based on the end-face array structure information and operating data, predict the displacement trend and assess the mechanical stability to form a predictive precision calibration instruction chain.

Images