Optical fiber array coupling system and method based on multi-source data fusion

Abstract

The invention discloses an optical fiber array coupling system and method based on multi-source data fusion, and relates to the technical field of optical fiber array coupling system.When an automatic coupling machine receives a working instruction, the system firstly starts a pre-coupling process of an optical fiber array, fuses and analyzes multi-source vibration state data, corrects operating parameters in real time, effectively suppresses initial mechanical artifacts, and finally, the optical fiber array is subjected to multi-source data fusion analysis; the adaptability of equipment to external disturbance and signal ground noise is improved, after initialization of operation parameters is completed, the system enters an actual coupling process, key working condition parameters are continuously monitored, and whether further optimization and adjustment are needed or not is dynamically judged, so that the mechanical vibration amplitude and response dispersion in the coupling process are reduced, and the coupling efficiency is improved. According to the optical fiber array coupling detection system, signal bottom noise and mechanical artifact superposition amplification are prevented, the signal quality and the structural stability are improved, when it is detected that an optical fiber array completes coupling, the system executes locking and steady-state processing according to a preset sequence, and the connection anti-interference performance and the long-term stability are improved through technological solidification and closed-loop parameter adjustment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fiber array coupling systems, in particular to a fiber array coupling system and method based on multi-source data fusion. BACKGROUND

[0002] The automatic coupling machine is oriented to fiber array alignment and coupling, and is usually composed of a mechanical base and a clamping / locking mechanism, a nanometer displacement table and a piezoelectric actuator, a vibration and pose sensor, an optical power detection link (detector-front end amplifier-A / D-locking phase / filtering), an industrial controller and a glue application and curing module. The principle is to implement micro-sweeping / gradient optimization and closed-loop control under multi-source measurement feedback, so that the array end face is locked and cured after reaching maximum coupling. In engineering fields (such as high-density packaging production lines, field maintenance, and multi-station parallel operation), the coupling gap and the optical path phase / amplitude are modulated due to environmental vibration, thermal drift, mechanism gap, and assembly tolerance, etc., which causes short-term fluctuations and slow drifts in optical power readings. In order to realize process observability and controllability, the automatic coupling machine is provided with vibration / pose sensors and optical power monitoring units on the coupling structure, which are input into the fusion algorithm after front-end amplification and data acquisition, so as to maintain coupling efficiency and system stability in a dynamic environment.

[0003] For example, the adaptive control method of the fiber coupler array disclosed in the Chinese patent with the publication number CN112485870B includes building a fiber coupler array control system, using the built fiber coupler array control system for data transmission, obtaining real data sets and dividing them into training data sets and test data sets, building an adaptive preliminary control model for the fiber coupler array and training to obtain an adaptive control model for the fiber coupler array, and using the obtained adaptive control model to perform adaptive control on the fiber coupler array.

[0004] For example, the focusing method of a zero-gap fiber bundle coupler disclosed in Chinese patent application No. CN118884617A belongs to the technical field of fiber bundle couplers. It comprises a fiber bundle coupler and a confocal probe, the fiber bundle coupler comprises a base, a connecting sleeve, an objective sleeve and a coupling objective, the confocal probe comprises a plug seat, a plug sleeve and a fiber bundle, the focusing method comprises the following steps: S1, holding the plug sleeve, inserting the plug seat together with the fiber bundle into the connecting sleeve; installing the coupling objective in the objective sleeve, and then installing the objective sleeve in the connecting sleeve; S2, moving the sliding table to the farthest position from the plug seat; S3, driving the sliding table to move gradually in the direction close to the fiber bundle with a step size less than 5 microns, the flexible connection end pushes the coupling objective to move in the direction of the fiber bundle through the flexible contact with the sliding table, and the interference between the two linear motion pairs of the sliding table and the coupling objective is corrected; wherein, the fiber bundle end face image obtained by the coupling objective at each step is analyzed to find the position with the highest sharpness, which is the best focusing position of the coupling objective, so that the focusing mechanism can still maintain accurate focusing under the condition of vibration interference.

[0005] The above-mentioned technology has the following technical problems: In the existing fiber array coupling control, the automatic coupling machine serves as the execution and control body, and the control instructions are mainly distributed to each execution mechanism (pose / force channel). The signal link (detector-front-end amplification-collection-filtering-evaluation) basically operates according to the established noise model and fixed parameters, and is not cooperatively designed with the mechanical disturbance. In actual working conditions, platform vibration and impact will introduce modulation of the coupling gap of the automatic coupling machine and disturbance of the optical path phase / amplitude, causing the automatic coupling machine to present non-stationary noise and artifacts in the power reading. At the same time, the execution side closed-loop bandwidth and the measurement side effective bandwidth / delay of the automatic coupling machine do not match, which easily leads to coexistence of strong anti-disturbance on the execution side and high noise floor on the measurement side, forming an anti-disturbance-measurement contradiction, prolonging the recovery time (hysteresis), reducing the overall coupling stability and energy utilization efficiency of the automatic coupling machine. Therefore, the existing technology lacks a cross-layer cooperation and self-adaptive trimming mechanism between the mechanical disturbance channel and the signal link processing, and fails to realize joint optimization of bandwidth matching, delay compensation and noise floor suppression in the vibration abnormal scene, thereby causing control misjudgment, closed-loop oscillation and coupling efficiency fluctuation. SUMMARY

[0006] In order to solve the technical problems that the existing technology lacks a cross-layer cooperation and self-adaptive trimming mechanism between the mechanical disturbance channel and the signal link processing, and fails to realize joint optimization of bandwidth matching, delay compensation and noise floor suppression in the vibration abnormal scene, thereby causing control misjudgment, closed-loop oscillation and coupling efficiency fluctuation, the embodiments of the present application provide a fiber array coupling system and method based on multi-source data fusion. The technical solution is as follows: In one aspect, a fiber array coupling system based on multi-source data fusion is provided, which comprises: a running parameter initialization module configured to start a fiber array pre-coupling process when an automatic coupling machine receives a work instruction, collect and fuse multi-source vibration state data of the automatic coupling machine, and correct running parameters of the automatic coupling machine based on an analysis result of the vibration state data; a running process optimization adjustment module configured to start a fiber array actual coupling process after the running parameter initialization of the automatic coupling machine is completed, continuously collect and monitor running condition parameters of the automatic coupling machine, and determine whether to carry out optimization adjustment on a running process of the automatic coupling machine; and a coupling steady module configured to execute locking and steady processing in a predetermined sequence after monitoring that the fiber array is coupled, so as to improve the anti-interference and long-term stability of the connection.

[0007] In another aspect, a fiber array coupling method based on multi-source data fusion is provided, which comprises: step one, starting a fiber array pre-coupling process when an automatic coupling machine receives a work instruction, collecting and fusing multi-source vibration state data of the automatic coupling machine, and correcting running parameters of the automatic coupling machine based on an analysis result of the vibration state data; step two, starting a fiber array actual coupling process after the running parameter initialization of the automatic coupling machine is completed, continuously collecting and monitoring running condition parameters of the automatic coupling machine, and determining whether to carry out optimization adjustment on a running process of the automatic coupling machine; and step three, executing locking and steady processing in a predetermined sequence after monitoring that the fiber array is coupled, so as to improve the anti-interference and long-term stability of the connection.

[0008] The technical scheme provided by the embodiments of the present application has at least the following beneficial effects: (1) The application realizes systematic improvement on the key problems of misjudgment, remodulation, resonance amplification, bandwidth mismatch and long-term drift in the coupling process through the closed-loop mechanism of differential criterion-adaptive amplitude-resonance peak avoidance-bandwidth matching-locking solidification: firstly, the vibration abnormality coefficient difference is calculated in the running / static monitoring window and is divided according to the hierarchical threshold to avoid lifting the noise floor as coupling degradation; when falling into the second interval, the actuator jitter amplitude / micro-scan step is adaptively adjusted to suppress the remodulation of power readings; when falling into the third interval, spectrum diagnosis is carried out, the process injection frequency and the detection jitter frequency are reconstructed, and the control bandwidth is reconfigured to improve the closed-loop phase margin; the vibration / pose / power are fused to realize adaptive initialization of running parameters during the pre-coupling stage, and the first locking time is shortened; the measurement-control bandwidth matching and gain / amplitude limiting scheduling are continuously carried out during the coupling process to reduce the response dispersion; after the coupling is completed, the mechanical locking and steady state are sequentially executed, and the process solidified parameters are improved, which improves the batch consistency and long-term anti-disturbance; when the vibration abnormality coefficient reaches the threshold without resonance evidence, the early warning is triggered to promote the front-end treatment. The comprehensive effect is that the power sequence fluctuation and false alarm rate are reduced, the signal-to-noise ratio is improved, the phase margin and stability are enhanced, the setting and locking time are shortened, and the long-term drift and reassembly dispersion are reduced, thereby significantly improving the coupling efficiency and reliability of the fiber array.

[0009] (2) The application differentiates and thresholds the vibration abnormality coefficient in different working condition windows to eliminate the state misjudgment and excessive parameter adjustment caused by inconsistent running / static noise floor; when falling into the second comparison interval, the actuator jitter amplitude is adaptively adjusted to suppress the remodulation of the micro-scan to the measurement link, and the power artifact and energy loss caused by slight vibration drift are reduced; when falling into the third comparison interval, spectrum diagnosis is implemented, and the process injection frequency and the detection jitter frequency are peak-avoiding reconfigured to weaken the risk of structural resonance amplification and closed-loop phase margin reduction from the source; if there is no resonance evidence, an abnormal early warning is issued in advance to compress the treatment lag and reduce the instability probability, thereby realizing the systematic suppression of misjudgment, remodulation, resonance amplification and maintenance lag without increasing additional hardware, improving the adaptive and anti-interference ability of the automatic coupling machine under multiple disturbance conditions, significantly improving the coupling power fluctuation and response dispersion, and ensuring the consistency, reliability and long-term stability of the fiber array coupling.

[0010] (3) The application is aimed at the running characteristics of the automatic coupling machine under different vibration levels and response dispersion conditions, adopts a working condition perception and self-adaptive adjustment mechanism, effectively solves the technical problems of state misjudgment and excessive parameter adjustment caused by inconsistent running / static noise floors, measurement remodulation and power artifacts caused by light to moderate vibration, closed-loop phase margin reduction caused by structural resonance, gradient estimation deviation caused by response dispersion amplification, and signal-to-noise ratio decline caused by acquisition link gain / time delay drift, when vibration abnormalities are detected, the system can automatically adjust the running parameters, quickly identify and suppress vibration artifacts caused by mechanical vibration, and ensure the stability of the mechanical structure during coupling. For response dispersion fluctuations, the system effectively avoids the superimposed distortion of noise and signal by adaptively improving the observation accuracy, and improves the signal consistency and detection accuracy in the key monitoring window. At the same time, in the presence of complex interference or measurement chain state fluctuations, the system can maintain the high stability of the acquisition link, realize self-recovery under abnormal conditions, thereby significantly reducing the coupling power fluctuation and false alarm rate, suppressing resonance amplification, shortening the locking and setting time, improving the coupling repeatability and long-term stability, and ensuring the high consistency and high reliability of the precision coupling task of the optical fiber array.

[0011] (4) The application performs locking and steady-state processing according to a predetermined sequence, and adopts an average vibration abnormality coefficient as a curing progress criterion to real-time evaluate the dynamic stability and energy dissipation level of the assembly, determine the glue curing time and locking holding time in stages, and cooperate with temperature / stress release strategies when necessary, in order to effectively avoid curing deficiency or overcuring, reduce the risk of residual stress concentration and interface creep, suppress mechanical loosening and coupling drift, improve the connection anti-interference ability, repeatability consistency and long-term operation stability, and realize high reliability and high consistency of the optical fiber array coupling. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0013] Figure 1 is a structural schematic diagram of an optical fiber array coupling system based on multi-source data fusion provided by the embodiments of the present application; Figure 2 is a flow chart of an optical fiber array coupling method based on multi-source data fusion provided by the embodiments of the present application; Figure 3 is a detailed flow chart of fiber array coupling provided by the embodiment of the present application; Figure 4 is an automatic coupling machine running parameter display interface diagram provided by the embodiment of the present application; Figure 5 is an automatic coupling machine test display interface diagram provided by the embodiment of the present application; Figure 6 is an automatic coupling machine instruction display interface diagram provided by the embodiment of the present application. DETAILED DESCRIPTION

[0014] The technical solutions in the present application will be described below with reference to the drawings.

[0015] In the embodiments of the present application, the words such as "example", "for example" and the like are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0016] In the embodiments of the present application, sometimes the subscript such as W1 may be written in the form of non-subscript such as W1, and the meanings expressed thereby are consistent when the difference is not emphasized.

[0017] In order to make the technical problems, technical solutions and advantages of the present application more clear, the following will be described in detail with reference to the drawings and specific embodiments.

[0018] The embodiments of the present application provide a fiber array coupling system based on multi-source data fusion. As shown in the fiber array coupling system structure schematic diagram based on multi-source data fusion, the system comprises a running parameter initialization module, a running process optimization adjustment module, a coupling steady-state module and a database. Figure 1

[0019] The database is used to store the parameters involved in the fiber array coupling system based on multi-source data fusion, and the data sources include but are not limited to original / preprocessed signals of various sensing channels, controller parameters and strategy versions, device and process configurations, alarm and audit logs. The data structure is organized in a mapping relationship (including mapping table and mapping function and the like), and is stored and managed according to a hierarchical scheme; the system provides retrieval capability based on multi-dimensional index, and can perform fast lookup and correlation query.

[0020] ​The running parameter initialization module is connected with a running process optimization adjustment module, the running process optimization adjustment module is connected with a coupling steady-state module, and the running parameter initialization module, the running process optimization adjustment module and the coupling steady-state module are all connected with a database.

[0021] Figure 4 is an automatic coupling machine running parameter display interface diagram provided by an embodiment of the application, which focuses on detailed presentation of core running parameters, Figure 5 is an automatic coupling machine test display interface diagram provided by an embodiment of the application, which is used for displaying dynamic data and results in a coupling test process, Figure 6 is an automatic coupling machine instruction display interface diagram provided by an embodiment of the application, which feeds back operation instructions and execution states in a system running process in real time, comprehensively covers parameter monitoring, test tracking and instruction tracing requirements in a coupling process, and all exhibits real-time collection and visual monitoring of key running parameters of the system in a fiber array coupling process, which helps to improve the intuitiveness of operation and process transparency, and provides data support for intelligent decision-making such as abnormal early warning, parameter optimization and fault tracing.

[0022] Figure 5 The four signal channels in the automatic coupling machine test are presented in the figure, wherein the abscissa represents a test position (Loc), the ordinate represents a phase value (unit: Opm), the red curve is a phase maximum value trajectory of each position sampled in a current statistical window of the channel, and the blue curve is a phase average value trajectory. The relative position and fitting degree of the red and blue curves can be used to judge the channel stability and abnormal peak value: when the two curves are basically coincident, it indicates that the phase fluctuation in the position section is small, the distribution is concentrated, and the coupling process is stable; when the red line is significantly higher than the blue line and local protrusions appear, it indicates that there are instantaneous sharp peaks or outlying fluctuations, and whether the safety margin is touched needs to be evaluated in combination with a threshold and an alarm strategy, thereby providing a basis for channel adjustment and quality determination.

[0023] The running parameter initialization module is used for starting a fiber array pre-coupling process after the automatic coupling machine receives a work instruction, collecting and fusing multi-source vibration state data of the automatic coupling machine, and obtaining and correcting running parameters of the automatic coupling machine based on vibration state data analysis results.

[0024] The running parameters include a jitter amplitude of an actuator, a process injection frequency and a platform detection jitter frequency.

[0025] The work instruction is a control command issued by the operation terminal to the automatic coupling machine. The instruction usually includes the type of target fiber array (such as a single-row 12-core array), the process parameters of the coupling task (such as an insertion force control between 0.5N and 1.0N, an end face inclination control within 0.1°, and a coupling loss of no more than 0.3dB), and a start / stop signal of the execution process, etc. Its role is to trigger the automatic coupling machine to enter the corresponding process preparation stage, and to ensure that the subsequent coupling operation has a unified task orientation and control basis.

[0026] The fiber array pre-coupling process is a preliminary adjustment and matching process of the position, posture and end face state of the fiber array before the formal coupling operation. The process usually includes preliminary positioning of the fiber array, coarse adjustment of the alignment reference, detection of end face cleanliness and insertion force, etc. Through this process, deviations and losses in the formal coupling process can be effectively reduced, providing a basic guarantee for efficient and stable optical coupling.

[0027] The working principle of the automatic coupling machine is as follows: after the automatic coupling machine is started, the fiber array and the chip to be coupled are fixed on the six-degree-of-freedom precision platform (including piezoelectric ceramic or screw drive) by the feeding mechanism, and then the visual positioning module (such as a microscopic camera and image algorithm) completes coarse alignment, reducing the pose deviation to the order of hundreds of microns. Then, the laser source injects light signals into the fiber array at a set frequency, and the signals are emitted from the fiber to the chip coupling window. The light signal received by the chip is converted into an electrical signal by a photodetector, and a power meter or a lock-in demodulation module collects the coupling power in real time as a feedback basis for alignment accuracy. Subsequently, the control unit (programmable logic controller) drives the platform to perform nanoscale micro-adjustment, and optimizes the translation and rotation parameters through adaptive algorithms (such as SPGD). When the coupling power reaches the peak threshold (such as insertion loss ≤0.5dB) and is stable (coupling power root mean square value ≤0.02dB), the solidification mechanism (dispensing or laser welding) is started to complete the fixation. The coupling accuracy, convergence time and other indicators are recorded throughout the process to ensure the consistency of batch coupling.

[0028] Specifically, the vibration state data analysis result specifically includes a first comparison result, a second comparison result, and a third comparison result.

[0029] The vibration abnormal coefficient of the automatic coupling machine in the running state monitoring window is differentially processed with the vibration abnormal coefficient of the automatic coupling machine in the static monitoring window, that is, difference processing, and the processing result is marked as vibration abnormal coefficient increment, and is compared with the vibration abnormal coefficient increment and the vibration abnormal coefficient increment warning threshold, wherein the vibration abnormal coefficient increment is less than the vibration abnormal coefficient increment warning threshold, and the vibration abnormal coefficient increment is a demarcation value for quantifying the vibration state of the automatic coupling machine, which is used to determine whether the vibration fluctuation exceeds the normal operation range. The setting of the demarcation value is based on the fact that its value is less than the vibration abnormal coefficient increment warning threshold, which reflects that it is only used to identify slight vibration abnormalities, but not serious instability. The vibration abnormal coefficient increment warning threshold is an upper limit value set according to the process requirements and the equipment safety margin, which is used to determine that the vibration enters a serious stage that needs to be intervened. The formulation process of the two includes: first, collecting vibration data under different working conditions through experiments to obtain the statistical distribution of the vibration abnormal coefficient; second, according to the stability requirements of the coupling mass, the vibration abnormal coefficient is divided into non-abnormal, slight abnormal and serious abnormal; finally, the value distinguishing between non-abnormal and slight abnormal is determined as the demarcation value of the vibration abnormal coefficient increment, and the value distinguishing between slight abnormal and serious abnormal is determined as the vibration abnormal coefficient increment warning threshold, which are stored in the database to form a hierarchical judgment benchmark.

[0030] The first comparison result refers to the vibration abnormal coefficient increment being less than the demarcation vibration abnormal coefficient increment; the second comparison result refers to the vibration abnormal coefficient increment being greater than or equal to the demarcation vibration abnormal coefficient increment and less than the vibration abnormal coefficient increment warning threshold; and the third comparison result refers to the vibration abnormal coefficient increment being greater than or equal to the vibration abnormal coefficient increment warning threshold.

[0031] Optionally, the multi-source vibration state data of the automatic coupling machine is collected and fused for analysis. The specific analysis process is as follows: the multi-source vibration state data of the automatic coupling machine includes the real-time vibration signal intensity of the automatic coupling machine in the static monitoring window, the real-time vibration kurtosis of the automatic coupling machine in the static monitoring window, and the real-time sideband asymmetry of the automatic coupling machine in the static monitoring window; the vibration signal intensity refers to the amplitude characteristic of the vibration acceleration of the automatic coupling machine, which is collected by an acceleration sensor or a laser vibration meter and processed by filtering; the vibration kurtosis is a statistical description of the sharpness of the vibration signal waveform, which is obtained by high-order statistical analysis of the time domain signal, and its change often reflects the existence of local impact or minor defects; the sideband asymmetry is a frequency domain characteristic parameter, which is used to measure whether the energy distribution of the modulation sidebands on both sides of the main frequency of the vibration signal is balanced, and is calculated by fast Fourier transform (FFT), and its abnormality often means that the system is under uneven stress or nonlinear excitation.

[0032] The static monitoring window refers to a time interval before the automatic coupling machine receives a work instruction. During this time interval, the equipment is in standby mode, and its vibration characteristics can objectively reflect the influence of the environment on the automatic coupling machine. By collecting data in the static window, a baseline reference before operation can be effectively established, providing a stable basis for subsequent abnormal comparison in dynamic conditions.

[0033] The multi-source vibration state data generated by the automatic coupling machine in different working condition stages are collected and archived by time period. By integrating the data of the static window and the running window into the same database for multi-source fusion storage, a stable pre-operation baseline can be formed, and dynamic tracking of the running state can be realized, thereby constructing a complete, continuous, and traceable vibration state archive. By comparing each real-time parameter of the automatic coupling machine in the static monitoring window with the corresponding defined value, the different dimensions are converted into dimensionless indicators on a unified scale, thereby eliminating dimensional differences and enhancing the horizontal comparability and fusion feasibility of heterogeneous data. The dimensionless indicators are then aggregated by a weighting algorithm to obtain a window-by-window comprehensive abnormality quantity. Finally, the window-by-window comprehensive abnormality quantity is time-integrated within the static monitoring window to obtain the vibration abnormality coefficient of the automatic coupling machine within the static monitoring window.

[0035] The vibration abnormality coefficient of the automatic coupling machine within the static monitoring window represents the degree of vibration abnormality of the automatic coupling machine within the static monitoring window, and the specific expression is: ; In the formula, VAC is the vibration abnormality coefficient of the automatic coupling machine within the static monitoring window, RT_RMS t is the vibration signal intensity of the automatic coupling machine at time t within the static monitoring window, JRMS is the defined vibration signal intensity, representing the upper limit value of the allowed range of vibration signal intensity, RT_EK t is the kurtosis of the automatic coupling machine at time t within the static monitoring window, JEK is the defined kurtosis, representing the upper limit value of the allowed range of kurtosis, RT_ASB t is the sideband asymmetry of the automatic coupling machine at time t within the static monitoring window, JASB is the defined sideband asymmetry, representing the upper limit value of the allowed range of sideband asymmetry, S1 is the vibration signal intensity measurement weight, representing the weight of vibration signal intensity in the vibration abnormality coefficient, with a value range of 0-1, S2 is the kurtosis measurement weight, representing the weight of kurtosis in the vibration abnormality coefficient, with a value range of 0-1, S3 is the sideband asymmetry measurement weight, representing the weight of sideband asymmetry in the vibration abnormality coefficient, with a value range of 0-1, t0 is the start time point of the static monitoring window, and t1 is the end time point of the static monitoring window.

[0036] The sideband asymmetry can present a positive value or a negative value in the frequency domain analysis, and to avoid the interference of the difference in directionality, generally the absolute value thereof is taken as a characterization index for quantifying the deviation degree of the sideband energy distribution from the symmetric state.

[0037] The vibration anomaly coefficient is calculated in an integral form, which is to accumulate and smooth the instantaneous anomaly in the time dimension, so as to reflect the comprehensive degree of the overall vibration anomaly in a specific time period, rather than relying on the transient fluctuation at a single moment.

[0038] The increase in the vibration signal intensity is usually accompanied by the increase in the amplitude of the vibration waveform, which will directly lead to the increase in the vibration kurtosis, because the vibration kurtosis measures the sharpness or transient impact characteristics of the signal, and the higher amplitude fluctuation will enhance the peak characteristics of the waveform, at the same time, the amplitude variation and uneven distribution of the vibration signal will also cause the change of the sideband asymmetry, because the sideband asymmetry describes the deviation of the modulation energy on both sides of the main frequency, when the local impact or uneven vibration occurs, the sideband energy distribution appears asymmetric, thereby affecting the index. Therefore, the three of them jointly determine the size of the vibration anomaly coefficient through the interaction of amplitude fluctuation and energy distribution, so as to quantify the abnormal degree in the whole time period.

[0039] The principle of setting these limit values is usually based on the historical operation data of the equipment and the process tolerance analysis, that is, by statistically obtaining the typical fluctuation range of each parameter from long-term static monitoring data, and combining the coupling quality requirement and the vibration stability requirement to determine the upper limit value, so as to be able to distinguish between slight abnormality and serious abnormality. The measurement weights S1, S2 and S3 are used to reflect the importance or contribution of each parameter in the calculation of the vibration anomaly coefficient, and the design rules generally follow the principles of process sensitivity and parameter influence: the parameters with greater influence on the vibration anomaly coefficient are given higher weights, reflecting the dominant role of the parameter on the overall abnormal state, while the parameters with smaller influence are given relatively lower weights, to ensure that the comprehensive anomaly coefficient can truly and balancedly represent the overall vibration deviation, and these parameters are stored in the database.

[0040] The multi-source vibration state data of the automatic coupling machine also includes the real-time vibration signal intensity of the automatic coupling machine in the running state monitoring window, the real-time vibration kurtosis of the automatic coupling machine in the running state monitoring window, and the real-time sideband asymmetry of the automatic coupling machine in the running state monitoring window.

[0041] In other embodiments, the parameters contained in the vibration anomaly coefficient can be equivalently replaced and expanded according to the actual execution of the automatic coupling machine to form a unified measure driven by multi-source data. In addition to the vibration signal intensity, the vibration kurtosis and the sideband asymmetry, the observation quantities representing the running state and stability can also be introduced, such as the actuator jitter amplitude / root mean square of driving current, the control error variance, the short-term fluctuation of optical power / signal-to-noise ratio, the stripe contrast, the encoder jitter, the temperature and thermal drift indicators, the base motion acceleration, the clock jitter and the cross-channel cross-correlation time delay, etc. The above replaced channel data are first normalized and mapped according to the corresponding defined values, and then the weights are set according to the channel reliability and the sensitivity to coupling degradation, and the weighted integration is completed to obtain the vibration anomaly coefficient. Thus, the coefficient can adaptively reflect the comprehensive abnormality degree of heterogeneous sensing and execution side information. In order to avoid arbitrary replacement, in other embodiments, the adjustment of the parameters involved in the vibration anomaly coefficient must be based on the actual working conditions and verifiable basis, and must follow the principles of equivalence, comparability and repeatability, and must be confirmed by calibration test, sensitivity analysis and correlation test before it can be included. The replacement process includes: candidate index screening to defined value and normalization strategy determination to weight configuration and consistency check to offline playback / online A / B verification to versioned solidification and audit retention. If the verification is not up to standard, it cannot be used and must be rolled back to the confirmed parameter configuration.

[0042] By comparing each real-time parameter of the automatic coupling machine in the running state monitoring window with the corresponding defined value, the dimensionless index of the same scale is obtained, so as to eliminate the dimensional difference and enhance the horizontal comparability and the fusion feasibility of heterogeneous data. The weighted algorithm is used to integrate each dimensionless index to obtain the window-by-window comprehensive abnormality quantity. Finally, the time integral of the window-by-window comprehensive abnormality quantity in the running state monitoring window is performed to obtain the vibration anomaly coefficient of the automatic coupling machine in the running state monitoring window.

[0043] The running state monitoring window is a time interval for continuously collecting vibration state parameters during the execution of the fiber array pre-coupling process of the automatic coupling machine. This window covers the entire pre-coupling operation process and is used to record the vibration signal intensity, vibration kurtosis and sideband asymmetry of the device in dynamic working conditions in real time, so as to reflect the transient fluctuations and abnormal trends that may occur during the execution process. In order to facilitate comparative analysis with the static reference, the time length of this running state monitoring window is consistent with that of the static monitoring window, so as to ensure that the static and running data can be directly corresponding in time scale, and the vibration anomaly coefficient can be effectively compared and comprehensively evaluated.

[0044] Further, the running parameters of the automatic coupling machine are corrected based on the analysis results of the vibration state data. The specific initialization process is as follows: If the first comparison result exists, it means that the vibration generated by the automatic coupling machine during the start and execution of the pre-coupling process is mainly derived from the inherent vibration of the automatic coupling machine itself and the influence of environmental disturbance, and the superposition of the two is still within the allowable range. Therefore, there is no need to adjust the operation parameters, the controller keeps the existing configuration and coupling strategy, and the original parameters are used to continue to execute the subsequent process to ensure the process continuity and operation stability.

[0045] If the second comparison result exists, it means that the vibration of the automatic coupling machine itself and the environmental vibration coexist in the running state, causing slight abnormal fluctuations, which may be caused by the micro-shaking of the actuator, the local structure response or slight external disturbance. In order to reduce the influence of the superimposed vibration on the coupling accuracy, the operation parameters of the automatic coupling machine are corrected, that is, the vibration abnormality coefficient of the automatic coupling machine in the running state monitoring window is extracted, and the reference shaking amplitude (such as the initial setting of ±30 nm) of the current actuator (such as piezoelectric ceramic PZT) is recorded; then, according to the preset vibration abnormality coefficient-shaking amplitude correction table in the database, the downshift amplitude of the shaking amplitude is determined, for example, when the vibration abnormality coefficient is 5%-10%, the actuator shaking amplitude is downshifted by 10%-15% from the reference value, so as to weaken the superposition effect of the actuator self-shaking and environmental vibration.

[0046] The amplitude downshift strategy under the above-mentioned second comparison result can directly weaken the remodulation effect of shaking on the measurement link and the short-term fluctuation of power reading in view of the technical problem of slight abnormal fluctuations caused by the superposition of micro-shaking of the actuator and environmental vibration in the running state. On the one hand, through the amplitude self-adaptive downshift based on the vibration abnormality coefficient, the energy of the execution side micro-scan is controlled and distributed, the peak-valley shaking caused by excessive micro-scan and threshold false triggering are reduced, and the unnecessary adjustment and integral saturation risk of the control loop are reduced. On the other hand, the variance and coefficient of variation of the power sequence are reduced, the instantaneous signal-to-noise ratio is improved, and the stability of the extreme value search / gradient estimation is enhanced, thereby shortening the locking and setting time, improving the coupling retention ability and anti-disturbance margin; at the process level, the reproducible parameter correction path is realized by combining the vibration abnormality coefficient-shaking amplitude table, the batch consistency is improved, and the long-term drift and reassembly dispersion are reduced, and finally the coupling accuracy, robustness and operation efficiency of the automatic coupling machine under light and moderate disturbance working conditions are improved.

[0047] If there is a third comparison result, it means that the superposition effect of environmental vibration and equipment vibration is significant, the equipment is in a state of abnormal vibration, and resonance phenomenon may occur between environmental disturbance and mechanical vibration. Whether resonance phenomenon occurs is determined through frequency spectrum analysis. If resonance phenomenon occurs, the operation parameters of the automatic coupling machine are modified, that is, the process injection frequency and the platform detection jitter frequency are adjusted based on the vibration abnormality coefficient of the automatic coupling machine in the running state monitoring window, so that the system avoids the resonance interval and reduces the overall vibration level. If it does not occur, an abnormal warning is given, and visual instructions such as information are used to prompt manual intervention to prevent vibration from affecting coupling accuracy and equipment safety, while providing reference data for subsequent processes.

[0048] The frequency spectrum diagnosis and peak avoidance and reconfiguration strategy under the third comparison result are used to solve the technical problems of superposition of environmental vibration and self vibration and possible triggering of structure resonance amplification, closed-loop phase margin reduction and self-excitation oscillation. First, the resonance main peak and bandwidth are identified based on the running state vibration abnormality coefficient to trigger frequency spectrum analysis. Then, the process injection frequency and the platform detection jitter frequency are adjusted to avoid peaks and reconfigure the control bandwidth according to the resonance center frequency, so as to realize bandwidth matching and phase margin recovery of the measurement-control link. This can significantly reduce the root mean square / peak-to-peak vibration in the resonance band, suppress the peak-to-valley jitter of power readings and false alarms, improve the instantaneous signal-to-noise ratio and extreme value search stability, reduce integral saturation and amplitude clipping bounce, shorten the locking and setting time, and reduce coupling drift and long-term instability risk. If there is no evidence of resonance in the frequency spectrum, an abnormal warning is given and characteristic quantities are recorded to support pre-intervention and post-tracing, reduce the average repair time, and improve batch consistency and operation safety, thereby improving the coupling accuracy, robustness and long-term stability of the automatic coupling machine under strong disturbance conditions.

[0049] Whether resonance phenomenon occurs is determined through frequency spectrum analysis, which specifically refers to extracting vibration data (such as six-degree-of-freedom platform, fixture acceleration time domain signal) in the running state monitoring window of the automatic coupling machine and synchronously collected process injection frequency and platform detection jitter frequency signals, and using fast Fourier transform (FFT) to convert the two types of time domain signals into frequency domain power spectral density (PSD) graphs. Then, the main peak frequency is identified in the PSD graph. If the vibration power spectrum amplitude of a certain frequency point is significantly higher than that of other frequencies (such as more than 3 times the amplitude of adjacent frequencies), and the deviation of the frequency from the structural modal frequency (previously obtained through modal testing, such as platform support natural frequency and fixture resonance frequency) or external device conversion frequency (such as the running frequency of workshop water pump and machine tool, which is recorded through pre-survey) is ≤5Hz, it is determined that resonance phenomenon occurs.

[0050] Based on the vibration abnormality coefficient of the automatic coupling machine in the running state monitoring window, the process injection frequency and the platform detection jitter frequency are adjusted, the resonance risk level corresponding to the current vibration abnormality coefficient is indicated, the basic offset of frequency adjustment is determined according to the “vibration abnormality coefficient-resonance risk” classification standard established by the technical personnel and stored in the database (for example, the vibration abnormality coefficient is 15%-25%, which is moderate resonance risk, and more than 25% is severe resonance risk), for example, the basic offset of frequency is set to 10-15 Hz under moderate resonance risk, and 15-20 Hz under severe resonance risk, and then the offset direction is determined combined with the spectrum analysis result: if the resonance frequency is close to the process injection frequency, the process injection frequency is offset away from the resonance frequency (for example, the resonance frequency is 100 Hz, and the current process injection frequency is 98 Hz, then it is adjusted to 108-113 Hz or 83-88 Hz); if the resonance frequency is close to the platform detection jitter frequency, the same logic offset is performed on the detection jitter frequency, to ensure that the frequency difference between the two types of frequencies and the resonance frequency is ≥10 Hz (to avoid entering the near resonance interval again), the adjustment relationship between the process injection frequency and the platform detection jitter frequency needs to be formulated based on the corresponding relationship between the vibration abnormality coefficient and the platform detection jitter frequency increment established by the technical personnel and stored in the database, the core logic is as follows: the vibration abnormality coefficient quantifies the resonance strength, the larger the vibration abnormality coefficient, the more significant the vibration deviation caused by resonance, and a larger frequency offset is needed to cut off the energy coupling (for example, for every 5% increase in the vibration abnormality coefficient, the platform detection jitter frequency increases by 2-3 Hz), to avoid being unable to escape from the near resonance interval, the corresponding relationship between the vibration abnormality coefficient and the platform detection jitter frequency increment is fixed through batch experiments, under different vibration abnormality coefficient conditions, a plurality of vibration abnormality coefficient-platform detection jitter frequency increment data are recorded, a piecewise fitting method is used to determine the platform detection jitter frequency increment corresponding to each coefficient interval, to form a corresponding relationship that can be directly called, to ensure that parameter correction does not need to repeat experiments, and directly matches the adjustment scheme based on the current vibration abnormality coefficient, to balance efficiency and accuracy.

[0051] The running process optimization adjustment module is used to start the actual coupling process of the fiber array after the initialization of the running parameters of the automatic coupling machine, and continuously collect and monitor the running condition parameters of the automatic coupling machine, so as to determine whether to carry out optimization adjustment on the running process of the automatic coupling machine.

[0052] Specifically, the operation condition parameters of the automatic coupling machine are continuously collected and monitored. The specific monitoring process is as follows: in the actual coupling process of the optical fiber array, a plurality of monitoring windows are divided in time sequence; in the actual coupling process of the optical fiber array, a plurality of monitoring windows are divided in time sequence. The essence is to divide the complete coupling operation cycle of the automatic coupling machine from the start of the equipment, alignment, glue application to curing into a plurality of continuous time segments, and each time segment is a monitoring window. By monitoring the key parameters in different windows, fine management and control of the whole coupling process are realized.

[0053] Each monitoring window has a corresponding vibration abnormality coefficient of the automatic coupling machine and a receiving response dispersion of the photodetector; the receiving response of the photodetector is A / W, i.e., ampere / watt. Before the start of the monitoring window, a known and stable reference light power (denoted as P, unit W, calibrated by a high-precision standard light power meter) is output by a laser source, the light signal is accurately incident to the light-sensitive area of the photodetector through the optical fiber link, a temperature control module is used to stabilize the working environment temperature of the detector at 25±0.5℃, an electromagnetic shield is added to eliminate external interference, and the influence of temperature drift and electromagnetic noise on current output accuracy is avoided; then, in the monitoring window, a high-precision current collection module with a resolution ≥1nA and a sampling rate ≥1kHz is used to continuously sample the photocurrent output by the detector, and the average value (denoted as I_avg, unit A) of the photocurrent in each sampling period of the monitoring window is calculated, abnormal current values caused by transient pulse interference are excluded, finally, according to the definition formula (R=I_avg / P) of the receiving response, the receiving response of the photodetector in the monitoring window is calculated, and the receiving response dispersion of the photodetector is the standard deviation of all the receiving responses monitored in the monitoring window.

[0054] The vibration abnormality coefficient of the automatic coupling machine is compared with the vibration abnormality threshold value, and the receiving response dispersion of the photodetector is compared with the defined receiving response dispersion, so as to determine whether to carry out optimization adjustment on the operation process of the automatic coupling machine; the receiving response dispersion represents the signal response dispersion degree in the actual coupling process of the fiber array, the vibration abnormality threshold value is the maximum value of the reasonable range of the vibration abnormality coefficient preset in the database, and the defined receiving response dispersion is the maximum value of the reasonable range of the receiving response dispersion preset in the database. The vibration abnormality threshold value and the defined receiving response dispersion are both based on the historical data accumulated in the database and the coupling quality requirements as the core basis: the vibration abnormality threshold value needs to collect the running state vibration data of the automatic coupling machine under the qualified coupling working condition (insertion loss ≤ 0.5 dB), calculate the vibration abnormality coefficients of multiple batches of operations, and take the 95% quantile value as the maximum value of the reasonable range (i.e. the vibration abnormality threshold value), so as to ensure that the vibration abnormality coefficients of more than 95% of the qualified working conditions do not exceed the value. The defined receiving response dispersion needs to record the receiving response dispersion of the photodetector under the same qualified working condition, and also takes the 95% quantile value of multiple batches of data as the maximum value of the reasonable range (i.e. the defined receiving response dispersion), so as to ensure that the signal response dispersion degree of more than 95% of the qualified couplings meets the requirements, and finally forms the two types of determination threshold values preset in the database, providing quantitative standards for optimization adjustment of the operation process.

[0055] Specifically, whether to carry out optimization adjustment on the operation process of the automatic coupling machine is determined, and the specific determination process is: if there is a normal comparison condition in a certain monitoring window, it is determined that optimization adjustment is not carried out on the operation process of the automatic coupling machine; if there is no normal comparison condition in a certain monitoring window, the monitoring window is marked as an abnormal monitoring window, and it is judged that optimization adjustment is carried out on the operation process of the automatic coupling machine; the normal comparison condition refers to that the vibration abnormality coefficient of the automatic coupling machine is less than or equal to the vibration abnormality threshold value, and the receiving response dispersion of the photodetector is less than or equal to the defined receiving response dispersion.

[0056] Further, the operation process of the automatic coupling machine is optimized and adjusted, and the specific adjustment process is that in the abnormal monitoring window, if the vibration abnormality coefficient of the automatic coupling machine is greater than the vibration abnormality threshold value, and the receiving response dispersion of the photodetector is less than or equal to the defined receiving response dispersion, an abnormal optimization adjustment strategy is executed, that is, the dithering amplitude of the actuator and the integral gain of the closed-loop controller are reduced based on the vibration abnormality coefficient of the automatic coupling machine.

[0057] The judgment and execution mechanism of normal comparison condition-abnormal monitoring window-hierarchical optimization adjustment is adopted. For the technical problems of inconsistent running / static noise floor, re-modulation of measurement link by micro-scan of execution side, and sensitivity of integral ring to noise, the verifiable suppression path is given: when the normal comparison condition is met, the adjustment is not triggered to avoid the closed-loop disturbance and efficiency loss caused by unnecessary parameter change; when the abnormal monitoring window is entered and the vibration abnormality coefficient exceeds the threshold while the received response dispersion is still controlled, the amplitude-integral dual-channel optimization is performed, that is, the vibration abnormality coefficient is adaptively adjusted to reduce the amplitude of the actuator jitter, weaken the re-modulation of power reading by micro-scan and peak-to-valley jitter, and simultaneously reduce the integral gain of the closed-loop controller to suppress integral saturation and low-frequency oscillation driven by noise. Thus, the power sequence variance and false alarm rate are reduced, the instantaneous signal-to-noise ratio and extreme value search stability are improved, the overshoot and setting time are shortened, the phase margin and robustness are enhanced, and the batch consistency and long-term stability are improved without increasing hardware, ensuring the coupling accuracy and operation reliability of the automatic coupling machine under multiple disturbance conditions.

[0058] First, the quantitative correlation between the vibration abnormality coefficient and the adjustment amount is established, the matching data of the vibration abnormality coefficient, the jitter amplitude, and the integral gain in the historical qualified working conditions are extracted, and the adjustment rules are set according to the vibration abnormality coefficient interval. For example, if the vibration abnormality coefficient exceeds the vibration abnormality threshold by 10% or less, the jitter amplitude is reduced by 10%-15% from the current value, and the integral gain is reduced by 8%-12% from the current value. During adjustment, the corresponding relationship stored in the database is first determined according to the current vibration abnormality coefficient, and then the two parameters are simultaneously reduced according to the rules. The corresponding relationship is developed based on the core of predictable narrow-band disturbance characteristics.

[0059] Because the observation chain is stable (the received response dispersion is less than or equal to the defined received response dispersion), the disturbance has no random interference, so the jitter amplitude and the integral gain are adjusted according to the vibration abnormality coefficient. By reducing the jitter amplitude, the superposition of actuator active vibration and narrow-band disturbance is reduced, and by reducing the integral gain, the controller over-response to small power fluctuations caused by disturbance is avoided. The coupling accuracy is guaranteed while the vibration is suppressed.

[0060] If the vibration abnormality coefficient of the automatic coupling machine is less than or equal to the vibration abnormality threshold value, and the receiving response dispersion of the photoelectric detector is greater than the defined receiving response dispersion, a response abnormality optimization adjustment strategy is executed, that is, the sampling rate of the data acquisition module to which the measurement chain belongs and the low-pass filter time constant of the phase-locked demodulation chain are increased based on the receiving response dispersion of the photoelectric detector; when the receiving response dispersion of the photoelectric detector is adjusted, the quantized correlation between the receiving response dispersion and the sampling rate and the low-pass time constant is extracted from the database, so that corresponding adjustment is performed, for example, when the receiving response dispersion exceeds the defined receiving response dispersion within 10%, the sampling rate of the data acquisition module is increased to 1.2-1.5 times of the current value, and the low-pass time constant of the phase-locked demodulation is increased to 1.2-1.3 times of the current value.

[0061] By using the response abnormality optimization adjustment strategy, the technical problem of the receiving response dispersion being too large in the photoelectric detection link when the vibration level is controlled is limited to the measurement side: on the one hand, the sampling rate of the data acquisition module is increased, the proportion of aliasing and quantization noise is reduced, the statistical description of transient fluctuations is refined, and the confidence of dispersion estimation and threshold determination is improved; on the other hand, the time constant of the low-pass filter of the phase-locked demodulation chain is increased, the high-frequency noise and amplitude remodulation sidebands are suppressed without changing the parameters of the mechanical side, the reading fluctuation is smoothed, and the power evaluation sequence is stabilized. The effective bandwidth and noise floor of the measurement-estimation link are re-leveled by the two, which directly reduces the receiving response dispersion and the false alarm rate, improves the signal-to-noise ratio and the robustness of the extreme value search / gradient estimation, avoids excessive tuning and phase margin reduction caused by the fluctuation of the measurement side to the control loop; at the same time, the actuator and the structure state remain unchanged, the disturbance to the closed-loop dynamics is reduced, and the coupling accuracy, consistency and long-term stability are improved without increasing the hardware cost.

[0062] The vibration abnormality coefficient of the automatic coupling machine is less than or equal to the vibration abnormality threshold value, and the receiving response dispersion of the photoelectric detector is greater than the defined receiving response dispersion, which indicates that the measurement chain has noise or signal distortion (such as insufficient sampling and mismatched filtering), increasing the sampling rate can capture high-frequency signal details and avoid information loss; increasing the low-pass time constant can enhance the noise filtering capability, both of which improve the signal consistency, the corresponding relationship is based on historical data fitting, the dispersion out-of-tolerance amplitude is positively correlated with the parameter adjustment amount, and the effective signal is ensured to be retained while the noise is suppressed.

[0063] If the vibration abnormality coefficient of the automatic coupling machine is greater than the vibration abnormality threshold value, and the receiving response dispersion of the photoelectric detector is greater than the defined receiving response dispersion, an abnormal optimization adjustment strategy is executed, and in response to the abnormal optimization adjustment strategy, the grid step and the sampling time length in the parameter search optimization algorithm are increased, and the phase calibration process is re-executed. For example, in each adjustment process, the grid step of the parameter search optimization algorithm is increased to 1.2-1.5 times the current value, the sampling time length is increased to 1.5-2 times the current value, and relevant personnel need to establish a working condition parameter matching table based on the database accumulated in the early stage, and ensure the adjustment accuracy and efficiency according to the preset multiple in the real-time monitored working condition parameter matching table each time.

[0064] The vibration abnormality coefficient of the automatic coupling machine is greater than the vibration abnormality threshold value, and the receiving response dispersion of the photoelectric detector is greater than the defined receiving response dispersion, and the synchronous optimization of the vibration and the response parameters can respectively suppress mechanical disturbance and improve signal quality; increasing the grid step and the sampling time length can improve the parameter search anti-interference ability and avoid noise misjudgment; re-labeling the phase can correct the phase shift caused by vibration and noise, and ensure the accuracy of the alignment direction. The corresponding relationship is based on historical data fitting, and the disturbance suppression and alignment accuracy are cooperatively guaranteed.

[0065] The linkage strategy of abnormal optimization adjustment plus response abnormal optimization adjustment plus parameter search reconfiguration plus phase re-labeling is adopted to face the strong disturbance working condition with high vibration and high dispersion, and three types of core problems of the measurement-control link are cooperatively solved: first, by reducing the actuator jitter amplitude and the controller integral gain, micro-scan remodulation and low-frequency oscillation driven by noise are suppressed, and the closed-loop phase margin and anti-saturation ability are restored; second, the acquisition sampling rate is increased and the phase-locked demodulation low-pass time constant is increased, the aliasing and high-frequency sideband are reduced, the power reading is smoothed, and the reliability of the dispersion and threshold determination is improved; third, the grid step and the sampling time length are increased in the search layer to improve the distinguishability and extreme value identification confidence of parameter identification in a strong noise scene, and the phase calibration is re-executed to correct the demodulation phase drift and quadrature error, and avoid the search misdirection caused by bias accumulation. The comprehensive effect is that the vibration energy and the reading variance are reduced, the signal-to-noise ratio and the extreme value search stability are improved, the locking and setting time converges, the resonance / self-excitation risk is reduced, and the coupling accuracy, robustness and batch consistency in abnormal scenes are significantly improved, and the long-term stable operation of the automatic coupling machine under strong disturbance conditions is ensured.

[0066] After the optimization adjustment is completed, in the review window, the vibration abnormality coefficient of the automatic coupling machine and the receiving response dispersion of the photoelectric detector are obtained, and it is determined whether to optimize the warning of the operation process of the automatic coupling machine.

[0067] Further, it is determined whether to optimize the running process of the automatic coupling machine. The specific determination process is as follows: in the review window, if there is a normal comparison condition, it is determined that the running process of the automatic coupling machine is not optimized; the review window refers to a time interval set for verification and confirmation after the automatic coupling machine completes the first round of vibration and receives the response parameter comparison, and the core function is to ensure the accuracy of the optimization warning determination. The length of the window needs to match the characteristics of the key stage of the coupling process (such as setting the review window to 3-5s in the fine alignment stage and 5-8s in the curing stage), and the vibration abnormality coefficient and the photoelectric detector receiving response dispersion data need to be continuously collected in the window.

[0068] If there is no normal comparison condition, the length of the monitoring window is increased, and the running process of the automatic coupling machine is continuously optimized and adjusted. If there is a normal comparison condition within the maximum number of cycles stored in the database, it is determined that the running process of the automatic coupling machine is not optimized. If there is still no normal comparison condition, it is determined that the running process of the automatic coupling machine is optimized through a visual method (such as a short message) to prompt relevant personnel to intervene in time and start targeted parameter correction and process adjustment to avoid abnormal expansion affecting the coupling quality.

[0069] The coupling steady-state module is used to perform locking and steady-state processing according to a predetermined sequence after the fiber array is coupled, so as to improve the noise immunity and long-term stability of the connection.

[0070] In one example embodiment, the fiber array coupling completion flag is that the coupling power collected by the photoelectric detector reaches the peak threshold (such as insertion loss ≤0.5dB), and the power fluctuation RMS is ≤0.02dB within a set time (usually 3-5s), indicating that the optical signal transmission efficiency and stability of the fiber and the chip meet the standard.

[0071] Specifically, locking and steady-state processing are performed according to a predetermined sequence. The specific processing process is as follows: according to the predetermined process steps, the structure is first positioned and mechanically locked, and then glue is applied to the key parts. Specifically, according to the predetermined process sequence, the fiber array is first positioned, the positions of the array are adjusted to ensure accurate alignment of each fiber core, and mechanical locking is performed at key parts such as the fiber end face contact area, the array fixed support and the interface connection point to prevent displacement or loosening in subsequent operations. Subsequently, glue is applied to these key parts to enhance the structural strength and vibration suppression, thereby entering the steady-state processing stage.

[0072] The average vibration anomaly coefficient of the automatic coupling machine is obtained, the single solidification duration of the glue is determined, the solidification process is accurately controlled, the glue is solidified, and when the solidification completion result appears, the solidification is stopped, thereby completing the fiber array coupling. The solidification completion result refers to that the temperature of the glue reaches the set solidification temperature and the duration meets the preset duration. Once it is confirmed that the solidification completion result is met, the controller can stop the solidification operation, thereby completing the steady-state coupling of the fiber array and ensuring that it has reliable anti-disturbance performance and coupling stability in long-term use.

[0073] The average vibration anomaly coefficient of the automatic coupling machine refers to the average value of the vibration anomaly coefficients collected at each time node in the complete work cycle from receiving the work instruction to applying the glue to the key parts. In an ideal environment without vibration, the solidification time-bonding strength curve of the glue is obtained from the production log by differential scanning calorimetry (DSC) or real-time bonding strength test, the basic single solidification duration T0 that meets the target bonding strength (such as the minimum strength that meets the subsequent process anti-vibration requirement) is determined, and the basic solidification duration T0 is corrected according to the size of the average vibration anomaly coefficient. The specific correction relationship is formulated by relevant technical personnel and stored in the database. For example, if the average vibration anomaly coefficient is ≤5% (slight vibration, no obvious interference), the single solidification duration is 1.0×T0-1.2×T0, which is slightly extended to offset the influence of slight vibration on the initial bonding of the glue. If the average vibration anomaly coefficient is between 5% and 15% (moderate vibration, with some interference), the solidification duration is 1.2T0-1.5×T0, which is extended to ensure that the glue is fully crosslinked in the vibration environment and avoid incomplete solidification leading to subsequent falling off. Because vibration interference will directly affect the solidification process of the glue in this case, the glue in the viscous state of this solidification is prone to interference, so the solidification duration needs to be extended to ensure sufficient crosslinking. If the vibration is slight, the duration of this time does not need to be excessively extended to avoid wasting single batch operation time.

[0074] In the fiber array and chip coupling operation of the automatic coupling machine, the glue is not solidified in a single global solidification mode, but follows a step-by-step point-by-point process logic. For multi-channel fiber array coupling scenarios (such as 16-channel, 32-channel fiber and chip interface docking), precise alignment and glue application of a single channel (or a local number of channels) need to be completed first. After the glue in this channel is solidified according to the preset single solidification duration (to support and fix the strength, and ensure that the alignment pose of this channel does not deviate), the alignment, glue application, and solidification operation of the next channel are performed.

[0075] Figure 3The optical fiber array coupling detailed flow chart provided by the embodiment of the present application is as follows: when the automatic coupling machine receives a work instruction, the optical fiber array pre-coupling flow is started, the multi-source vibration state data of the automatic coupling machine is collected and fused for analysis, so that the vibration abnormality coefficient increment is compared with the defined vibration abnormality coefficient increment and the vibration abnormality coefficient increment alarm threshold, if there is a first comparison result, the operation parameters of the automatic coupling machine do not need to be corrected, the current coupling control strategy is continued to execute the subsequent flow, if there is a second comparison result, the operation parameters of the automatic coupling machine are corrected, if there is a third comparison result, whether resonance phenomenon occurs is determined through frequency spectrum analysis, if it occurs, the operation parameters of the automatic coupling machine are corrected, if it does not occur, an abnormality early warning is performed, after the operation parameters of the automatic coupling machine are initialized, the optical fiber array actual coupling flow is started, and the operation condition parameters of the automatic coupling machine are continuously collected and monitored, the vibration abnormality coefficient of the automatic coupling machine is compared with the vibration abnormality threshold, the receiving response dispersion of the photoelectric detector is compared with the defined receiving response dispersion, so that whether there is a normal comparison condition is determined, if there is, the operation process of the automatic coupling machine is not optimized and adjusted, and locking and steady processing are performed according to a predetermined sequence, so as to improve the anti-interference and long-term stability of the connection, if there is not, the operation process of the automatic coupling machine is optimized and adjusted, after the adjustment is completed, whether optimization early warning is needed is determined, if yes, the optimization early warning is performed, if not, whether the optical fiber array is coupled is determined, if yes, locking and steady processing are performed according to a predetermined sequence, if not, the operation condition parameters of the automatic coupling machine are continuously collected and monitored.

[0076] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software mode depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0077] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0078] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and the division of units is merely a logical function division. In actual implementation, another division manner can be used, for example, a plurality of units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0079] The above merely provides a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A fiber array coupling system based on multi-source data fusion, characterized in that, The system comprises: an operation parameter initialization module, configured to start a fiber array pre-coupling process after the automatic coupling machine receives a work instruction, collect and fuse analysis of multi-source vibration state data of the automatic coupling machine, obtain and correct operation parameters of the automatic coupling machine based on an analysis result of the vibration state data; an operation process optimization adjustment module, configured to start a fiber array actual coupling process after the operation parameter initialization of the automatic coupling machine is completed, and continuously collect and monitor operation condition parameters of the automatic coupling machine, so as to determine whether to carry out optimization adjustment on the operation process of the automatic coupling machine; a coupling steady module, configured to perform locking and steady processing in a predetermined sequence after monitoring that the fiber array is completed coupling, so as to improve the anti-interference and long-term stability of the connection.

2. The multi-source data fusion based fiber array coupling system of claim 1, wherein, The collection and fusion analysis of the multi-source vibration state data of the automatic coupling machine comprises the following specific analysis process: The multi-source vibration state data of the automatic coupling machine comprises real-time vibration signal intensity of the automatic coupling machine in a static monitoring window, real-time vibration kurtosis of the automatic coupling machine in the static monitoring window, and real-time sideband asymmetry of the automatic coupling machine in the static monitoring window. By comparing each real-time parameter of the automatic coupling machine in the static monitoring window with a corresponding defined value, non-dimensional indexes of different dimensions are converted into a unified scale, so as to eliminate dimensional differences and enhance the fusion feasibility of horizontal comparison and heterogeneous data, and each non-dimensional index is summarized by a weighting algorithm to obtain a window-by-window comprehensive abnormal quantity, and finally the window-by-window comprehensive abnormal quantity is time-integrated in the static monitoring window to obtain a vibration abnormality coefficient of the automatic coupling machine in the static monitoring window. The vibration abnormality coefficient of the automatic coupling machine in the static monitoring window represents the vibration abnormality degree of the automatic coupling machine in the static monitoring window. The multi-source vibration state data of the automatic coupling machine further comprises real-time vibration signal intensity of the automatic coupling machine in a running state monitoring window, real-time vibration kurtosis of the automatic coupling machine in the running state monitoring window, and real-time sideband asymmetry of the automatic coupling machine in the running state monitoring window. By comparing each real-time parameter of the automatic coupling machine in the running state monitoring window with a corresponding defined value, non-dimensional indexes of different dimensions are converted into a unified scale, so as to eliminate dimensional differences and enhance the fusion feasibility of horizontal comparison and heterogeneous data, and each non-dimensional index is summarized by a weighting algorithm to obtain a window-by-window comprehensive abnormal quantity, and finally the window-by-window comprehensive abnormal quantity is time-integrated in the running state monitoring window to obtain a vibration abnormality coefficient of the automatic coupling machine in the running state monitoring window.

3. The multi-source data fusion based fiber array coupling system of claim 1, wherein, The vibration state data analysis result specifically comprises a first comparison result, a second comparison result and a third comparison result. The vibration abnormality coefficient of the automatic coupling machine in the running state monitoring window is differentially processed with the vibration abnormality coefficient of the automatic coupling machine in the static monitoring window, and the processing result is marked as a vibration abnormality coefficient increment and compared with a defined vibration abnormality coefficient increment and a vibration abnormality coefficient increment alarm threshold. The first comparison result is that the vibration abnormality coefficient increment is less than the defined vibration abnormality coefficient increment. The second comparison result refers to that the vibration abnormality coefficient increment is greater than or equal to the defined vibration abnormality coefficient increment and less than the vibration abnormality coefficient increment warning threshold; The third comparison result refers to that the vibration abnormality coefficient increment is greater than or equal to the vibration abnormality coefficient increment warning threshold.

4. The multi-source data fusion based fiber array coupling system of claim 1, wherein, The operation parameters of the automatic coupling machine are corrected based on the vibration state data analysis result, and the specific initialization process is as follows: If the first comparison result exists, the operation parameters of the automatic coupling machine do not need to be corrected, the controller maintains the existing parameter configuration and running state, and the current coupling control strategy is continued to be executed to execute the subsequent process; If the second comparison result exists, the operation parameters of the automatic coupling machine are corrected, that is, the vibration abnormality coefficient of the automatic coupling machine in the running state monitoring window is used to reduce the jitter amplitude of the actuator; If the third comparison result exists, it is determined whether resonance occurs through frequency spectrum analysis, if yes, the operation parameters of the automatic coupling machine are corrected, that is, the vibration abnormality coefficient of the automatic coupling machine in the running state monitoring window is used to adjust the process injection frequency and the platform detection jitter frequency, and if not, an abnormality warning is given.

5. The multi-source data fusion based fiber array coupling system of claim 1, wherein, The running condition parameters of the automatic coupling machine are continuously collected and monitored, and the specific monitoring process is as follows: In the actual coupling process of the optical fiber array, a plurality of monitoring windows are divided in time sequence; Each monitoring window has a corresponding vibration abnormality coefficient of the automatic coupling machine and a receiving response dispersion of the photodetector; The vibration abnormality coefficient of the automatic coupling machine is compared with the vibration abnormality threshold, and the receiving response dispersion of the photodetector is compared with the defined receiving response dispersion, so as to determine whether the running process of the automatic coupling machine needs to be optimized and adjusted; The receiving response dispersion represents the signal response dispersion degree in the actual coupling process of the optical fiber array.

6. The multi-source data fusion based fiber array coupling system of claim 5, wherein, The determination of whether the running process of the automatic coupling machine needs to be optimized and adjusted is as follows: If there is a normal comparison condition in a certain monitoring window, it is determined that the running process of the automatic coupling machine does not need to be optimized and adjusted; If there is no normal comparison condition in a certain monitoring window, the monitoring window is marked as an abnormal monitoring window, and it is determined that the running process of the automatic coupling machine needs to be optimized and adjusted; The normal comparison condition refers to that the vibration abnormality coefficient of the automatic coupling machine is less than or equal to the vibration abnormality threshold, and the receiving response dispersion of the photodetector is less than or equal to the defined receiving response dispersion.

7. The multi-source data fusion based fiber array coupling system of claim 6, wherein, The optimization and adjustment of the running process of the automatic coupling machine are as follows: In the abnormal monitoring window, if the vibration abnormality coefficient of the automatic coupling machine is greater than the vibration abnormality threshold and the receiving response dispersion of the photodetector is less than or equal to the defined receiving response dispersion, an abnormal optimization and adjustment strategy is executed, that is, the vibration abnormality coefficient of the automatic coupling machine is used to reduce the jitter amplitude of the actuator and the integral gain of the closed-loop controller; If the vibration abnormality coefficient of the automatic coupling machine is less than or equal to the vibration abnormality threshold and the receiving response dispersion of the photodetector is greater than the defined receiving response dispersion, a response abnormality optimization and adjustment strategy is executed, that is, the receiving response dispersion of the photodetector is used to increase the sampling rate of the data acquisition module belonging to the measurement chain and the time constant of the low-pass filter of the phase-locked demodulation chain. If the vibration abnormality coefficient of the automatic coupling machine is greater than the vibration abnormality threshold value, and the receiving response dispersion of the photoelectric detector is greater than the defined receiving response dispersion, an abnormal optimization adjustment strategy is executed, and in response to the abnormal optimization adjustment strategy, the grid step and the sampling time length in the parameter search optimization algorithm are increased, and the phase calibration process is re-executed; After the optimization adjustment is completed, in the review window, the vibration abnormality coefficient of the automatic coupling machine and the receiving response dispersion of the photoelectric detector are obtained, and it is determined whether to optimize the warning of the running process of the automatic coupling machine.

8. The multi-source data fusion based fiber array coupling system of claim 7, wherein, The determination of whether to optimize the warning of the running process of the automatic coupling machine is specifically determined as follows: In the review window, if there is a normal comparison condition, it is determined that the running process of the automatic coupling machine is not optimized for warning; If there is no normal comparison condition, the time length corresponding to the monitoring window is increased, and the optimization adjustment of the running process of the automatic coupling machine is continuously performed. Within the defined number of cycles, if there is a normal comparison condition, it is determined that the running process of the automatic coupling machine is not optimized for warning, and if there is still no normal comparison condition, it is determined that the running process of the automatic coupling machine is optimized for warning.

9. The multi-source data fusion based fiber array coupling system of claim 1, wherein, The locking and stabilization process is performed in a predetermined sequence, and the specific processing process is as follows: According to the predetermined process steps, after the initial positioning and mechanical locking of the structure are completed, glue is applied to the key parts, and the stabilization processing stage is entered; The average vibration abnormality coefficient of the automatic coupling machine is obtained, the single solidification time length of the glue is determined, and the glue is solidified. When the solidification completion result appears, the solidification is stopped, and the fiber array coupling is completed.

10. The method for coupling fiber array based on multi-source data fusion, applied to the system for coupling fiber array based on multi-source data fusion according to any one of claims 1-9, characterized in that, It includes: Step one, after the automatic coupling machine receives the work instruction, the fiber array pre-coupling process is started, the multi-source vibration state data of the automatic coupling machine is collected and fused, the running parameters of the automatic coupling machine are corrected based on the vibration state data analysis result; Step two, after the initialization of the running parameters of the automatic coupling machine is completed, the fiber array actual coupling process is started, and the running condition parameters of the automatic coupling machine are continuously collected and monitored, so as to determine whether to optimize the running process of the automatic coupling machine; Step three, after the fiber array coupling is completed, locking and stabilization processing is performed in a predetermined sequence to improve the anti-interference and long-term stability of the connection.

Images