A method for detecting dynamic performance of a hollow-core anti-resonant optical fiber

Abstract

The present application relates to the technical field of material damage detection, and particularly relates to a kind of dynamic performance detection method of hollow core anti-resonant optical fiber, in the process of robot driving optical fiber movement, the relative distance and instantaneous acceleration of pedestal are collected in real time to construct the cable curvature vector and impact conduction vector representing the stress state of optical fiber along different positions of body, and the relative thermal eigenvalue and relative structural eigenvalue decoupled from periodic pulse data are combined, the local thermal damage vector and structural damage vector of optical fiber along body are iteratively inverted using recursive parameter estimation algorithm;The method creatively uses the robot motion trajectory as an active scanning excitation source, converts the traditional endpoint total station measurement into distributed sensing of the local state of optical fiber along body, not only can accurately locate the position of hidden local damage, but also can effectively distinguish the two different damage mechanisms of material aging and structural crack, so that the accuracy of dynamic performance detection of hollow core anti-resonant optical fiber by the present application is higher.

Description

Technical Field

[0001] This invention relates to the field of material damage detection technology, specifically to a method for detecting the dynamic performance of hollow anti-resonant optical fibers. Background Technology

[0002] Hollow-core anti-resonant fiber (HC-ARF), with its extremely low light-material overlap factor, exhibits a superior high damage threshold and low nonlinear effects, and is gradually replacing solid fiber as the preferred transmission medium in high-end ultrafast laser precision machining. In typical industrial applications, optical fibers are usually installed at the end effector of a six-axis industrial robot, moving at high speed and in multiple dimensions with the robot flange to cut or weld complex curved parts. However, long-term dynamic motion can easily lead to aging or breakage of the optical fiber, resulting in processing accidents. Therefore, real-time status monitoring of the optical fiber is crucial.

[0003] Existing technologies typically employ online monitoring by integrating photodiodes at the fiber optic output end. This involves real-time acquisition of the total optical power transmitted through the fiber, triggering a shutdown alarm upon detecting a significant power attenuation. However, this endpoint monitoring method only obtains the cumulative total transmission loss of the entire fiber, failing to capture information on the local damage distribution along the fiber, thus unable to pinpoint the specific location of damage. Furthermore, it cannot distinguish the physical mechanism of the damage, making it difficult to differentiate between material thermal aging and structural microcracks. In other words, existing methods for dynamic performance testing of optical fibers using endpoint photodiodes cannot achieve precise local damage location and mechanism differentiation, resulting in low accuracy in dynamic performance testing of hollow-core antiresonant fibers. Summary of the Invention

[0004] To address the low accuracy of existing technologies using endpoint photodiodes for dynamic performance testing of optical fibers, this invention aims to provide a method for testing the dynamic performance of hollow-core anti-resonant optical fibers. The specific technical solution adopted is as follows:

[0005] The first aspect of this invention provides a method for detecting the dynamic performance of a hollow-core anti-resonant optical fiber, comprising:

[0006] During the process of the robot driving the optical fiber to move in space, all periodic pulse data corresponding to each sampling moment, as well as the relative distance and instantaneous acceleration of the robot to the base, are acquired; wherein, the optical cable is located between the robot's end flange and the base;

[0007] Based on the relative distance between the base and the total length of the optical cable, the curvature vector of the optical cable at each sampling moment is determined; based on the instantaneous acceleration changes in time sequence, the impact transmission vector at each sampling moment is determined.

[0008] At each sampling time, the corresponding relative thermal characteristic value is determined based on the reference relative magnitude of the trailing edge energy of the pulse main peak of all periodic pulse data; the corresponding relative structural characteristic value is determined based on the relative stability of the peak voltage reference of all periodic pulse data.

[0009] Based on the recursive parameter estimation algorithm, the optical cable thermal damage vector at each sampling time is determined by iteratively updating the relative thermal characteristic value and the optical cable curvature vector according to the regression relationship. Based on the principle of obtaining the optical cable thermal damage vector, the optical cable structural damage vector at each sampling time is determined by iteratively updating the relative structural characteristic value and the impact transmission vector according to the regression relationship.

[0010] Dynamic performance testing of hollow anti-resonant optical fiber is performed based on the optical cable thermal damage vector and the optical cable structural damage vector.

[0011] Furthermore, the process of obtaining the optical cable curvature vector includes:

[0012] At each sampling time, the optical cable center sag is determined based on the relative deviation between the total length of the optical cable and the relative distance to the base.

[0013] The optical cable is divided into equal intervals by a preset interval length to obtain at least two local optical cable segments; parabolic fitting is performed based on the position of each local optical cable segment in the optical cable to determine the corresponding morphological distribution characteristic value.

[0014] Based on the product of the morphological distribution characteristic value and the optical cable center sag, the equivalent bending curvature of each local optical cable segment at each sampling time is determined; the equivalent bending curvatures of all local optical cable segments are arranged sequentially to determine the optical cable curvature vector.

[0015] Furthermore, the process of obtaining the sag of the optical cable center includes:

[0016] The difference between the square of the total length of the optical cable and the square of the relative distance to the base is positively correlated to determine the center sag of the optical cable at each sampling time.

[0017] Furthermore, the process of obtaining the morphological distribution feature values ​​includes:

[0018] All local optical cable segments are arranged along the direction from the robot end flange to the base to determine the local optical cable segment sequence; in the local optical cable segment sequence, the index value corresponding to each local optical cable segment is normalized to determine the corresponding index coefficient; the negative correlation mapping value of the index coefficient of each local optical cable segment is positively correlated with the product of the corresponding index coefficient to determine the morphological distribution characteristic value of each local optical cable segment.

[0019] Furthermore, the process of obtaining the impact transmission vector includes:

[0020] The acceleration at each sampling moment is differentially calculated with the acceleration at the previous sampling moment to determine the instantaneous impact scalar at each sampling moment; the product of the index coefficient of each local optical cable segment and the total length of the optical cable is exponentially negatively correlated and mapped to determine the corresponding exponential attenuation coefficient.

[0021] The effective impact intensity of each local optical cable segment at each sampling time is determined by the product of the exponential decay coefficient and the instantaneous impact scalar.

[0022] The effective impact intensity of all local optical cable segments is arranged along the direction from the robot end flange to the base, and the impact transmission vector at each sampling time is determined.

[0023] Furthermore, the process of obtaining the relative thermal characteristic value includes:

[0024] Acquire the start time, end time, and trailing edge integration cutoff time of the main pulse for each cycle of pulse data; integrate the waveform amplitude of each cycle of pulse data within the interval corresponding to the start time and end time of the main pulse to obtain the main peak energy; integrate the waveform amplitude of each cycle of pulse data within the interval corresponding to the end time and end time of the main pulse to obtain the trailing edge energy; determine the trailing edge energy ratio of each cycle of pulse data based on the ratio between the main peak energy and the trailing edge energy; determine the corresponding absolute thermal characteristic value based on the average of the trailing edge energy ratios of all sampling times corresponding to each sampling time; determine the relative thermal characteristic value of each sampling time based on the difference between the absolute thermal characteristic value and the preset baseline thermal characteristic value.

[0025] Furthermore, the process of obtaining the relative structural feature values ​​includes:

[0026] Calculate the coefficient of variation of all periodic pulse peaks corresponding to each sampling time to determine the corresponding absolute structural feature value; determine the relative structural feature value of each sampling time based on the difference between the absolute structural feature value and the preset baseline structural feature value.

[0027] Furthermore, the process of obtaining the optical cable thermal damage vector includes:

[0028] After concatenating a constant 1 to the end of the optical cable curvature vector, the vector is transposed to determine the augmented excitation vector at each sampling time.

[0029] The predicted thermal feature value is determined by multiplying the transpose of the augmented excitation vector at each sampling time with the optical cable thermal damage vector at the previous sampling time; the corresponding prediction error is determined by the difference between the relative thermal feature value at each sampling time and the predicted thermal feature value.

[0030] The Kalman gain vector for each sampling time is calculated based on the iterative covariance matrix of the previous sampling time and the augmented excitation vector.

[0031] The damage parameter vector correction amount is determined based on the product between the Kalman gain vector and the prediction error; the sum vector between the damage parameter vector correction amount at each sampling time and the optical cable thermal damage vector at the corresponding previous sampling time is calculated to determine the optical cable thermal damage vector at each sampling time.

[0032] Furthermore, after determining the optical cable thermal damage vector at each sampling time, the process also includes:

[0033] At each sampling time, the covariance correction matrix is ​​determined by performing matrix multiplication sequentially between the Kalman gain vector, the transpose of the augmented excitation vector, and the covariance matrix of the previous sampling time. The matrix obtained by subtracting the iterative covariance matrix of the previous sampling time from the covariance correction matrix is ​​then weighted by a preset forgetting factor to determine the iterative covariance matrix for each sampling time.

[0034] Furthermore, the process of performing dynamic performance testing of hollow-core anti-resonant optical fiber based on the optical cable thermal damage vector and the optical cable structural damage vector includes:

[0035] Remove the last element from the optical cable thermal damage vector to determine the effective thermal damage vector; in the effective thermal damage vector, the index value corresponding to the element that is greater than the preset thermal damage threshold is used as the thermal damage index value; in the local optical cable segment sequence, the local optical cable segment with the index value equal to the thermal damage index value is marked as the thermally damaged optical cable segment.

[0036] Remove the last element from the optical cable structure damage vector to determine the effective structural damage vector; in the effective structural damage vector, the index value corresponding to the element that is greater than the preset structural damage threshold is used as the structural damage index value; in the sequence of local optical cable segments, the local optical cable segment with the index value equal to the structural damage index value is marked as the structural damage optical cable segment.

[0037] Dynamic performance testing of hollow anti-resonant optical fibers was conducted on optical cable segments with thermal damage and structural damage.

[0038] Secondly, the present invention provides a dynamic performance testing system for hollow-core anti-resonant optical fiber, the system comprising:

[0039] The data acquisition module is used to acquire all periodic pulse data corresponding to each sampling moment, as well as the relative distance and instantaneous acceleration of the robot to the base, during the spatial motion of the robot driving the optical fiber; wherein, the optical cable is located between the robot's end flange and the base;

[0040] The vector determination module is used to determine the optical cable curvature vector at each sampling moment based on the relative distance between the base and the total length of the optical cable; and to determine the impact transmission vector at each sampling moment based on the instantaneous acceleration changes in time sequence.

[0041] The eigenvalue determination module is used to determine the corresponding relative thermal eigenvalue at each sampling time based on the reference relative magnitude of the trailing edge energy of the pulse main peak of all periodic pulse data; and to determine the corresponding relative structural eigenvalue based on the relative stability of the peak voltage reference of all periodic pulse data.

[0042] The damage determination module is used to determine the optical cable thermal damage vector at each sampling time by iteratively updating the relative thermal feature value and the optical cable curvature vector based on the recursive parameter estimation algorithm; and to determine the optical cable structural damage vector at each sampling time by iteratively updating the relative structural feature value and the impact conduction vector based on the principle of obtaining the optical cable thermal damage vector.

[0043] The dynamic performance testing module is used to perform dynamic performance testing of the hollow anti-resonant optical fiber based on the optical cable thermal damage vector and the optical cable structural damage vector.

[0044] Thirdly, the present invention provides a computer device including a memory and a processor. The memory is used to store computer program code, and the processor is used to call and run the computer program code from the memory to perform the method as described in the first aspect or any embodiment of the first aspect of the present invention.

[0045] Fourthly, the present invention provides a computer program product comprising computer program code, which, when executed, performs the method as described in the first aspect or any embodiment of the first aspect of the present invention.

[0046] Fifthly, the present invention provides a computer-readable storage medium storing computer program code that, when executed, performs the method as described in the first aspect or any embodiment of the first aspect of the present invention.

[0047] The present invention has the following beneficial effects:

[0048] This invention constructs cable curvature vectors and impact conduction vectors characterizing the stress state at different locations along the fiber by real-time acquisition of the relative distance and instantaneous acceleration of the base during the robot-driven fiber movement. Combined with relative thermal and structural characteristic values ​​extracted from periodic pulse data, a recursive parameter estimation algorithm iteratively inverts the local thermal damage vectors and structural damage vectors along the fiber. This method creatively utilizes the robot's motion trajectory as an active scanning excitation source, transforming traditional endpoint measurement into distributed sensing of the local state along the fiber. This not only accurately locates hidden local damage locations but also effectively distinguishes between two different damage mechanisms: material aging and structural cracks. Therefore, the accuracy of dynamic performance testing of hollow anti-resonant optical fibers using this invention is significantly improved. Attached Figure Description

[0049] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A flowchart illustrating a dynamic performance testing method for hollow anti-resonant optical fiber according to an embodiment of the present invention;

[0051] Figure 2 This is a structural diagram of a dynamic performance testing system for hollow anti-resonant optical fiber provided in one embodiment of the present invention;

[0052] Figure 3 This is a schematic diagram of a computer device structure provided in one embodiment of the present invention. Detailed Implementation

[0053] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a dynamic performance testing method for hollow-core anti-resonant optical fiber proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0055] The following description, in conjunction with the accompanying drawings, details the specific scheme of the dynamic performance testing method for hollow anti-resonant optical fiber provided by this invention.

[0056] This invention provides a method for testing the dynamic performance of hollow-core anti-resonant optical fiber. Please refer to [link to relevant documentation]. Figure 1 The diagram illustrates a flowchart of a dynamic performance testing method for hollow anti-resonant optical fiber according to an embodiment of the present invention. The method includes:

[0057] Step S101: During the process of the robot driving the optical fiber to move in space, acquire all periodic pulse data corresponding to each sampling moment, as well as the relative distance and instantaneous acceleration of the robot to the base; wherein, the optical cable is located between the robot's end flange and the base.

[0058] The method of this invention operates in a hardware environment consisting of a six-axis industrial robot, a hollow-core anti-resonant fiber laser transmission system, and a high-speed data acquisition unit. The two ends of the optical cable are rigidly fixed between the robot's stationary base and its moving end flange, respectively. It should be noted that optical fibers face complex damage risks during long-term dynamic service. On the one hand, the glass material inside the fiber may develop photo-darkening or color center defects (material aging) under high-energy laser irradiation. This type of damage is highly sensitive to static bending and easily induces localized heat accumulation at the bending point. On the other hand, the cladding microstructure may develop microcracks or structural loosening (structural damage) under repeated mechanical acceleration and deceleration oscillations. This type of damage is sensitive to dynamic impacts and easily leads to transmission instability or even fiber breakage.

[0059] To achieve online decoupled detection of the two types of hidden damage mentioned above, the system synchronously performs data acquisition tasks at a fixed sampling period during the process of the robot driving the optical fiber to perform multi-degree-of-freedom spatial motion (such as cutting and welding trajectories). Specifically, the system reads the spatial coordinates of the end flange relative to the base and the three-axis instantaneous acceleration vector in real time through the communication interface of the robot controller, and determines the relative distance to the base and the instantaneous acceleration at each sampling moment accordingly; in this embodiment of the invention, the relative distance to the base is based on the Euclidean distance between the robot's end flange and the base; simultaneously, the system continuously acquires all periodic pulse waveform data emitted by the laser through a high-speed photodetector located at the output end of the optical fiber, ensuring that the optical signal and mechanical motion data are strictly aligned on the time axis, providing the original data foundation for subsequent feature extraction and spatial inversion. In a specific implementation of this embodiment of the invention, the sampling period is set to 1ms, which can be adjusted according to the specific implementation environment.

[0060] It should be noted that the instantaneous acceleration in the embodiments of the present invention is a directional vector, and all periodic pulse data corresponding to each sampling moment are specifically: all periodic pulse waveform data collected within the time period between each sampling moment and the previous sampling moment.

[0061] The relative distance between the bases can indirectly characterize the overall expansion and contraction state of the optical cable, and the instantaneous acceleration can indirectly characterize the dynamic impact state of the optical cable. By synchronously collecting these data that characterize the mechanical motion state and the periodic pulse data that characterize the optical transmission quality in real time, a spatiotemporal correlation dataset of mechanical motion-photoelectric response was established. This provides the necessary physical input for decoupling the hidden damage along the optical fiber from the overall signal and solves the problem of the lack of spatial location information in the existing technology.

[0062] Step S102: Determine the optical cable curvature vector at each sampling moment based on the relative distance between the base and the total length of the optical cable; determine the impact transmission vector at each sampling moment based on the instantaneous acceleration changes in time sequence.

[0063] After acquiring macroscopic motion data such as the relative distance between the base and the robot during its movement, this single scalar value alone cannot directly characterize the specific physical deformation distribution along different parts of the optical fiber. Furthermore, the thermal damage mechanism of hollow anti-resonant optical fibers is highly sensitive to local static bending. Without a detailed description of the optical fiber's spatial morphology, it is impossible to map subsequent observed optical signal anomalies to specific spatial locations. Therefore, to construct a spatial index for inverting damage locations, it is necessary to further derive and determine the optical cable curvature vector at each sampling moment based on the geometric constraints of the relative distance between the base and the total length of the optical cable. This step aims to map the robot's end-effector position state to the static bending stress field distribution along the optical fiber, thereby reconstructing a time-varying "virtual spatial probe" at the software level. This provides the necessary spatial dimension for accurately decoupling and locating local thermal damage from the overall optical signal, significantly improving the spatial resolution and logical closed-loop performance of the detection scheme.

[0064] While the fiber optic cable curvature vector constructs a spatial excitation field characterizing static bending, it cannot provide an effective physical mapping for structural microcracks that are more sensitive to dynamic mechanical oscillations. Therefore, to capture and locate structural damage caused by dynamic mechanical impacts, it is necessary to determine the impact propagation vector at each sampling moment based on the instantaneous acceleration changes in time sequence. This process transforms the instantaneous acceleration changes at the robot's end effector into a dynamic oscillation intensity distribution propagating along the fiber optic cable, thereby constructing a dynamic excitation field specifically for detecting structural microcracks. This provides the necessary spatial independent variable input for subsequent accurate inversion and location of impact-sensitive local structural defects by combining structural feature values, thus completing the physical model of the detection scheme in the dynamic dimension.

[0065] Step S103: At each sampling time, determine the corresponding relative thermal characteristic value based on the reference relative magnitude of the trailing edge energy of the pulse main peak of all periodic pulse data; determine the corresponding relative structural characteristic value based on the relative stability of the peak voltage reference of all periodic pulse data.

[0066] While the optical cable curvature vector constructs an excitation field characterizing the static bending degree of the optical fiber in a spatial dimension, providing a necessary spatial index for locating thermal damage, the lack of observational indicators that can specifically characterize the thermal response of materials from photoelectric signals makes it impossible to establish a quantitative relationship between physical excitation and the thermal aging state inside the optical fiber. Furthermore, existing total power monitoring methods obfuscate various losses, failing to isolate the thermal effect signal caused by bending. Therefore, to obtain thermal damage observation data matching the bending excitation field, this embodiment of the invention determines the corresponding relative thermal characteristic value at each sampling time based on the baseline relative magnitude of the trailing edge energy of the pulse main peak of all periodic pulse data. This step utilizes the pulse tailing characteristics exhibited in the time domain by photon relaxation or fluorescence effects caused by thermal aging to successfully extract the thermal response increment sensitive to bending height from the complex original waveform. This provides a high-purity characteristic response input for subsequent inversion of the local thermal damage vector using a recursive algorithm, ensuring a closed-loop physical logic for thermal damage detection.

[0067] While the impact propagation vector effectively reconstructs the intensity of dynamic mechanical oscillations along each segment of the optical fiber in space and establishes the physical excitation model required for inverting structural damage, the external excitation field alone is insufficient to diagnose internal structural defects. It is also necessary to extract structural features from the photoelectric signal that specifically respond to mechanical impacts, distinguishing them from changes in the optical signal caused by thermal effects. Therefore, to construct structural health status observation indicators corresponding to dynamic impact excitation, this embodiment of the invention determines the corresponding relative structural characteristic values ​​based on the relative stability of the peak voltage reference of all periodic pulse data. This process, through quantitative analysis of the statistical dispersion of pulse peaks, can keenly capture transient jitter in the transmission mode caused by microcracks or structural loosening, thereby establishing a structural response channel independent of thermal features at the signal level. This provides crucial observational evidence for subsequent precise decoupling and localization of local structural damage using the impact propagation vector, greatly enhancing the detection system's ability to identify structural faults.

[0068] Step S104: Based on the recursive parameter estimation algorithm, the optical cable thermal damage vector at each sampling time is determined by iteratively updating the relationship between the relative thermal characteristic value and the optical cable curvature vector; based on the principle of obtaining the optical cable thermal damage vector, the optical cable structural damage vector at each sampling time is determined by iteratively updating the relationship between the relative structural characteristic value and the impact transmission vector.

[0069] Relative thermal characteristic values ​​and optical cable curvature vectors characterize the thermal aging state of optical fibers from two dimensions: time-domain observation response and spatial physical excitation, respectively. However, the former is a scalar representation of the cumulative effect of the entire fiber, while the latter is a vector field distributed along the fiber. A complex spatial integral relationship exists between the two, making it impossible to accurately map the observed thermal anomalies back to specific damaged segments through simple algebraic operations. Therefore, to solve this inverse problem of resolving multi-point distribution from single-point observation, a recursive parameter estimation algorithm is needed. This algorithm iteratively updates the relative thermal characteristic value and the optical cable curvature vector based on their regression relationship to determine the optical cable thermal damage vector at each sampling moment. This step utilizes the multi-pose excitation changes generated by the robot during continuous movement, continuously refining the estimation of thermal damage coefficients for each segment through recursive iteration of the algorithm. This mathematically achieves decoupling and inversion from the overall cumulative signal to the local spatial distribution, ultimately enabling precise localization of the thermal damage location along the fiber without introducing distributed sensors.

[0070] After successfully constructing the inversion logic for thermal damage, for structural damage with completely different physical mechanisms, although the corresponding excitation source (impact transmission vector) and observed index (relative structural characteristic value) are different, the mathematical essence of solving the local distribution from the overall response is consistent with thermal damage inversion; both belong to parameter estimation problems of underdetermined systems. Therefore, in order to simultaneously achieve spatial localization of structural microcracks, it is necessary to adopt the same recursive parameter estimation strategy based on the principle of obtaining the optical cable thermal damage vector, and iteratively update the optical cable structural damage vector according to the regression relationship between the relative structural characteristic value and the impact transmission vector to determine the optical cable structural damage vector at each sampling time. This process reuses the validated mathematical inversion framework, spatiotemporally correlates the dynamic impact excitation and the structural jitter response, thereby achieving independent decoupling and quantification of the fiber's structural damage distribution in parallel computing channels, and completing a full-dimensional analysis of the optical fiber's health status.

[0071] Step S105: Perform dynamic performance testing of the hollow anti-resonant optical fiber based on the optical cable thermal damage vector and the optical cable structural damage vector.

[0072] After calculating the optical cable thermal damage vector characterizing material aging distribution and the optical cable structural damage vector characterizing structural crack distribution, the system has obtained the spatial distribution information of the microscopic health status of the optical fiber. However, these high-dimensional numerical matrices still need to be further transformed into detection conclusions with engineering guidance significance to support on-site operation and maintenance decisions. Therefore, in order to complete the final closed loop from data inversion to condition diagnosis, it is necessary to perform dynamic performance testing of hollow-core anti-resonant optical fibers based on the optical cable thermal damage vector and the optical cable structural damage vector. This step, through quantitative evaluation and threshold discrimination of the two types of damage vectors, achieves accurate identification of the local damage location and damage type of the optical fiber, thereby providing a reliable decision basis for targeted implementation of power reduction thermal protection or speed reduction structural protection, and significantly improving the accuracy of dynamic performance testing of hollow-core anti-resonant optical fibers.

[0073] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the optical cable curvature vector includes:

[0074] At each sampling time, the optical cable center sag is determined based on the relative deviation between the total length of the optical cable and the relative distance to the base. In a specific implementation of this invention, the process of obtaining the optical cable center sag includes: performing a positive correlation mapping between the difference between the square of the total length of the optical cable and the square of the relative distance to the base to determine the optical cable center sag at each sampling time.

[0075] This process is based on the quadratic parabola approximation method, approximating the optical cable between the robot's end flange and the base as an isosceles triangle for fitting analysis. The relative distance between the base and the base is the base of this isosceles triangle, and the optical cable as a whole forms the legs of the isosceles triangle. Based on the principle that half the base length, the base height, and the leg length of an isosceles triangle conform to the Pythagorean theorem, the formula for calculating the maximum sag at the center of the optical cable, i.e., the base height, is derived as follows: ;in, Sampling time The sag of the optical cable center; This is the total length of the optical cable; this parameter is a fixed value obtained in advance. Sampling time The relative distance between the bases; The preset correction coefficient based on the stiffness of the optical cable is set to 0.4 in this embodiment of the invention for common industrial-grade flexible optical cables, after experimental calibration. This value is a dimensionless factor reflecting the bending stiffness characteristics of the optical cable. In the ultimate case of considering the optical cable as a completely flexible catenary with no bending stiffness, the preset correction coefficient should be 0.5, which corresponds to the geometric height of an isosceles triangle. However, the actual optical cable is affected by the stiffness of the sheath material and the internal filling structure, and it has a certain bending stiffness, which can resist some sag under the action of gravity, making the actual sag less than the theoretical value of the ideal geometric model. It can be adjusted according to the specific implementation environment. The larger the overall stiffness of the optical cable, the smaller its natural sag under the action of gravity, and the lower the ratio of the actual sag to the theoretical geometric sag, so the smaller the preset correction coefficient. Conversely, the more flexible the optical cable, the closer its shape is to the ideal catenary or zigzag line, the larger the preset correction coefficient.

[0076] Further, the optical cable is divided into equal intervals by a preset interval length to obtain at least two local optical cable segments; parabolic fitting is performed based on the position of each local optical cable segment in the optical cable to determine the corresponding morphological distribution characteristic value; specifically: all local optical cable segments are arranged along the direction from the robot end flange to the base to determine the local optical cable segment sequence; in the local optical cable segment sequence, the index value corresponding to each local optical cable segment is normalized to determine the corresponding index coefficient; the negative correlation mapping value of the index coefficient of each local optical cable segment is positively correlated with the product between the corresponding index coefficient to determine the morphological distribution characteristic value of each local optical cable segment.

[0077] In one specific implementation of this invention, the process of obtaining the morphological distribution feature values ​​is expressed by the following formula: ;in, For local optical cable sections morphological distribution characteristic values; For local optical cable sections The corresponding index value in the local optical cable segment sequence; This represents the number of local optical cable segments in the local optical cable segment sequence. For local optical cable sections The index coefficient.

[0078] This formula, based on the parabolic approximation principle of catenaries, constructs a normalized morphological distribution function, making the morphological distribution characteristic value at both ends of the optical cable 0, while the characteristic value at the center point of the optical cable, i.e., the point of maximum sag, is 1, thus simulating the geometric morphological distribution of the optical cable's natural sag. Furthermore, based on the product of the morphological distribution characteristic value and the center sag of the optical cable, the equivalent bending curvature of each local optical cable segment at each sampling time is determined. This product operation effectively maps the single maximum sag scalar of the optical cable center to each segment along the fiber, calculating the degree to which each segment deviates from the straight path. This degree is proportional to the local static bending radius; a larger value indicates more severe local bending. Finally, the equivalent bending curvatures of all local optical cable segments are arranged sequentially to determine the optical cable curvature vector, thereby constructing a time-varying spatial excitation field characterizing the static bending stress distribution along the fiber, providing the necessary spatial independent variables for subsequent inversion of the thermal damage location.

[0079] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the impact transmission vector includes:

[0080] The acceleration at each sampling moment is differentially calculated with the acceleration at the previous sampling moment to determine the instantaneous impact scalar at each sampling moment. Specifically, in this embodiment of the invention, the vector corresponding to the instantaneous acceleration at each sampling moment is subtracted from the vector corresponding to the instantaneous acceleration at the previous sampling moment to determine the corresponding difference vector. The ratio between the magnitude of the difference vector and the sampling period is used as the instantaneous impact scalar at each sampling moment to quantify the intensity of transient mechanical oscillations caused by the sudden change in the robot's end effector action, i.e., the jerk.

[0081] The end flange of the robot is the source of mechanical impact, and the optical cable, as a flexible medium, connects the source and the base. When the mechanical wave propagates along the optical cable, its energy is attenuated due to the material damping effect. The farther away from the source, the smaller the impact. Therefore, this embodiment of the invention takes into account the attenuation characteristics of mechanical impact along the length of the optical cable and performs an exponential negative correlation mapping between the index coefficient of each local optical cable segment and the total length of the optical cable to determine the exponential attenuation coefficient that characterizes the lower the proportion of impact energy retained as the center point of each local optical cable segment is farther away from the source.

[0082] Then, combining the source intensity and transmission attenuation, the effective impact intensity of each local optical cable segment at each sampling moment is determined based on the product between the exponential attenuation coefficient and the instantaneous impact scalar. The effective impact intensities of all local optical cable segments are arranged along the direction from the robot end flange to the base to determine the impact transmission vector at each sampling moment. This process reconstructs the physical distribution field of dynamic mechanical oscillation along the optical fiber, providing the necessary spatial variables for subsequent accurate inversion of the location of impact-sensitive local microcracks by combining structural characteristic values.

[0083] In one specific implementation of this invention, the process of obtaining the effective impact strength is expressed by the formula: ;in, Sampling time Lower local optical cable section Effective impact strength; Sampling time The instantaneous impact scalar; Sampling time Lower local optical cable section The exponential decay coefficient; The preset mechanical damping coefficient for the optical cable sheath is 0.5 in this embodiment of the invention. On the one hand, it is used to control the rate at which impact energy decays with distance; on the other hand, its unit is the reciprocal of length, used to balance the influence of dimensions so that the input of the exponential function is dimensionless. The preset mechanical damping coefficient can be adjusted automatically according to the specific implementation environment. Its size depends on the viscoelasticity of the optical cable sheath material and the shock absorption capacity of the braided layer. The larger the preset mechanical damping coefficient, the stronger the absorption and dissipation capacity of the optical cable material for mechanical waves, the faster the impact energy decays with distance, and the shorter the propagation distance; conversely, if the damping coefficient is smaller, the shock wave can propagate further.

[0084] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the relative thermal characteristic value includes:

[0085] The main pulse start time, main pulse end time, and trailing edge integration cutoff time are obtained for each cycle of pulse data. Specifically, firstly, the main peak time of each cycle of pulse data is identified using a threshold detection or peak search algorithm. The moment when the leading edge of the main peak rises to a preset specific rise ratio of the peak value is defined as the main pulse start time, and the moment when the trailing edge of the main peak falls to a preset specific fall ratio of the peak value is defined as the main pulse end time. Further, the sampling time corresponding to a preset multiple of the main pulse half-width time length after the main pulse end time is defined as the trailing edge integration cutoff time, thereby clearly defining the time window used to capture the thermal response tail signal. In a specific implementation of this invention, the preset specific rise ratio is set to 10%, and the preset specific fall ratio is set to 37%. The preset multiple is set to 5. In this embodiment of the invention, the preset rise ratio, the preset fall ratio and the preset multiple are all determined a priori based on the characteristic thermal relaxation time constant of the optical fiber material and the detector response bandwidth, and can be adjusted according to the specific implementation environment.

[0086] The waveform amplitude of each periodic pulse data in the interval from the start time to the end time of the main pulse is integrated to obtain the main peak energy. The main peak energy mainly reflects the intrinsic emission intensity of the laser pulse and is the denominator reference for normalization. The waveform amplitude of each periodic pulse data in the interval from the end time of the main pulse to the end time of the trailing edge integration is integrated to obtain the trailing edge energy. The energy range of the trailing edge energy avoids the strong signal of the main pulse and is used to characterize the weak trailing signal derived from the fiber thermal effect.

[0087] Based on the ratio between the peak energy and the trailing edge energy, the proportion of trailing edge energy in each period of pulse data is determined. By calculating the ratio of peak energy to trailing edge energy, the overall light intensity jitter caused by laser source power fluctuations or changes in optical path coupling efficiency is effectively eliminated, so that the subsequently calculated relative thermal characteristic values ​​are only related to pulse waveform distortion, thus improving the robustness of thermal characteristics.

[0088] The absolute thermal characteristic value is determined based on the average of the trailing edge energy percentages of all sampling times corresponding to each sampling time. The relative thermal characteristic value is determined based on the difference between the absolute thermal characteristic value and the preset baseline thermal characteristic value. This process eliminates the inherent background thermal response of the optical fiber when it is not under stress, ensuring that the extracted relative thermal characteristic value purely reflects the dynamic thermal damage increment induced by the current bending excitation, providing high signal-to-noise ratio input data for subsequent recursive inversion.

[0089] In one specific implementation of this invention, the preset baseline thermal characteristic value is obtained through the system initialization calibration process. Specifically, with the robot driving the optical fiber in a straight and stress-free posture, the laser is controlled to emit a preset number of calibration pulse sequences. The trailing edge energy ratio of each calibration pulse is calculated using the same method as described above, and its arithmetic mean is taken as the preset baseline thermal characteristic value under healthy system conditions. The process of obtaining the preset baseline structural characteristic value in the subsequent analysis process is similar, that is, the coefficient of variation of the peak values ​​in the calibration pulse sequence is calculated as the baseline, which will not be further limited or elaborated here.

[0090] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the relative structural feature values ​​includes:

[0091] Calculate the coefficient of variation of all periodic pulse peaks corresponding to each sampling time to determine the corresponding absolute structural feature value; determine the relative structural feature value of each sampling time based on the difference between the absolute structural feature value and the preset baseline structural feature value.

[0092] The coefficient of variation, as a normalized statistic, can eliminate the influence of the absolute value of light intensity and focus on characterizing the degree of signal fluctuation. Therefore, it is very suitable for characterizing transmission mode instability caused by structural microcracks. By subtracting the baseline, the inherent background noise of the system is further filtered out, making the relative structural characteristic value a high-purity structural health indicator that is only sensitive to dynamic mechanical shock. It should be noted that, in order to avoid the problem of the denominator being 0, when calculating the absolute structural characteristic value, if the mean of the peak values ​​of all periodic pulses is 0, that is, the detector does not detect a light signal, it means that the laser is in a turned-off state or the optical path is completely interrupted. At this time, in order to prevent the program from crashing due to division by zero error, the absolute structural characteristic value at this moment is directly forced to be 0 and an abnormal light-off state alarm is issued to ensure that the system can maintain the continuous operation of logic even in non-processing intervals or abnormal light-off states.

[0093] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the optical cable thermal damage vector includes:

[0094] A constant 1 is appended to the end of the optical cable curvature vector, and then transposed to determine the augmented excitation vector at each sampling time; that is, the augmented excitation vector is a column vector. Although the optical cable curvature vector captures the excitation components related to bending, actual optical path systems often have fixed losses (such as coupling efficiency substrate loss) or slow drifts (such as thermal background caused by changes in ambient temperature) that are unrelated to robot motion. If these constant components are ignored, the algorithm will forcibly attribute all observed thermal anomalies to bending excitation, leading to biased inversion results. Therefore, a constant term is introduced to construct a bias channel that can absorb global system drift or background thermal loss, thereby enhancing the model's fitting ability and robustness.

[0095] The predicted thermal characteristic value is determined by multiplying the transpose of the augmented excitation vector at each sampling time with the optical cable thermal damage vector at the previous sampling time; the corresponding prediction error is determined by the difference between the relative thermal characteristic value and the predicted thermal characteristic value at each sampling time.

[0096] The process mathematically constructs a closed-loop feedback mechanism based on the current state estimation: First, the damage distribution state calculated in the previous time step is convolved with the spatial excitation field at the current time step to forward deduce the theoretically expected thermal response observation value, establishing a mapping from microscopic local parameters to macroscopic overall signal; then, the theoretical prediction value is compared with the actual acquired relative thermal characteristic value, and the residual signal between the two is extracted; this residual objectively reflects the degree of deviation between the current damage model and the real physical system, serving as the core driving force for the recursive algorithm to continuously correct the parameter estimation value, ensuring that the inversion process can gradually approach the true health state of the optical fiber over time.

[0097] In one specific implementation of this invention, the process of obtaining the prediction error is expressed by the following formula: ;in, Sampling time The corresponding prediction error; Sampling time The corresponding relative thermal characteristic value; Sampling time The transpose of the corresponding augmented activation vector; Sampling time The optical cable thermal damage vector corresponding to the previous sampling time; Sampling time The predicted thermal characteristic values; where, The number of columns and The number of rows is the same, which is the number of local optical cable segments plus one. Therefore, the predicted thermal characteristic value is a numerical value, which represents the theoretical relative thermal characteristic value generated based on the current robot thermal excitation, assuming that the current optical cable thermal damage state is the estimated value at the previous moment.

[0098] Based on the iterative covariance matrix and augmented excitation vector of the previous sampling time, the Kalman gain vector at each sampling time is calculated. Specifically, the process of obtaining the Kalman gain vector is expressed by the formula: ;in, Sampling time The corresponding Kalman gain vector; Sampling time The iterative covariance matrix corresponding to the previous sampling time; Sampling time The corresponding augmented activation vector; The forgetting factor is set to 0.99 in this embodiment of the invention. It can be adjusted according to the specific implementation environment. The larger the forgetting factor, the higher the weight of the algorithm in remembering historical data, and the better the smoothness and noise resistance of parameter estimation. However, the tracking response speed to rapid changes in fiber damage parameters will be slower. Conversely, the smaller the forgetting factor, the faster the algorithm forgets old data, and the stronger the system's ability to dynamically capture parameter mutations. However, the estimation results are easily affected by instantaneous measurement noise and fluctuate.

[0099] It should be noted that the formula for obtaining the Kalman gain vector belongs to the conventional technique for calculating the gain matrix in the Recursive Least Squares (RLS) algorithm. In the physical model constructed in this embodiment, the augmented excitation vector corresponds to the regression vector in the standard algorithm. This is because, in the physical mechanism of fiber damage, the robot's motion excitation (bending and impact) is the direct cause of incremental changes in the optical signal. This vector value characterizes the distribution of physical excitation experienced by the system at the current moment. The iterative covariance matrix corresponds to the estimation error covariance matrix in the standard algorithm. This is because this matrix mathematically records the confidence level of the estimated damage coefficients of each fiber segment and the correlation between the parameters of each segment. Its magnitude reflects the uncertainty of the algorithm's assessment of the damage state of each segment at the current moment and is used to adjust the correction weights of different segment parameters during the update process.

[0100] The Kalman gain vector plays a crucial role in the recursive estimation algorithm by dynamically adjusting the error allocation weights. Its physical significance lies in quantifying the contribution of each local optical cable segment to the overall observation error at the current sampling moment. Specifically, the algorithm adaptively calculates the gain coefficient based on the historical estimation uncertainty information recorded by the iterative covariance matrix and the spatial force distribution represented by the augmented excitation vector at the current moment. If a local optical cable segment is subjected to strong physical excitation at the current moment and its corresponding parameter estimation uncertainty is high, the algorithm will assign a larger gain value to that optical cable segment, thereby guiding subsequent correction steps to allocate the observation error mainly to that region. This ensures that the parameter update process can accurately respond to changes in spatial excitation, achieving efficient and robust local damage localization.

[0101] The damage parameter vector correction is determined based on the product of the Kalman gain vector and the prediction error. The sum vector between the damage parameter vector correction at each sampling time and the corresponding optical cable thermal damage vector at the previous sampling time is calculated to determine the optical cable thermal damage vector at each sampling time. The process of obtaining the optical cable thermal damage vector is expressed by the following formula: ;in, Sampling time The thermal damage vector of the optical cable; Sampling time The thermal damage vector of the optical cable at the previous sampling time.

[0102] It should be noted that the formula used to obtain the optical cable thermal damage vector belongs to the conventional technique for parameter updating in the recursive least squares algorithm. The optical cable thermal damage vector... The role of the system parameter vector to be estimated in the standard algorithm is because this embodiment of the invention treats the local thermal damage degree of each segment along the fiber as the unknown internal state of the system. The numerical change of this vector directly determines the magnitude of the system output under a specific excitation, and therefore it is the target object that the algorithm needs to approximate. The prediction error is the prediction residual in the standard algorithm. This prediction error quantifies the degree of deviation between the currently observed thermal characteristics of the fiber and the theoretical characteristics predicted based on the old model. This degree of deviation reflects the system state change information contained in the latest acquired data that has not yet been captured by the parameter model at the previous moment. The product of the prediction error and the Kalman gain vector corresponds to the parameter correction term in the standard algorithm. The Kalman gain vector, according to the current physical excitation distribution, reasonably maps and distributes the prediction residual containing the new information to specific segments in space, thereby realizing the directional correction of the optical cable thermal damage vector at the previous moment using the latest observation data.

[0103] This step constitutes the state update stage of the recursive parameter estimation algorithm. It combines the Kalman gain carrying spatial allocation weights with the prediction error characterizing the overall model deviation to generate specific correction instructions for each segment along the fiber, and superimposes them into the state estimate of the previous moment, thereby realizing the dynamic evolution of the damage vector. This process utilizes the time-varying spatial excitation field generated during the robot's movement, and through continuous "prediction-feedback-correction" cycles, the thermal damage vector of the optical cable gradually converges to a steady-state value that reflects the true physical damage distribution, completing the reverse analysis from macroscopic endpoint observation signals to microscopic local health status.

[0104] Preferably, in some possible implementations of the embodiments of the present invention, after determining the optical cable thermal damage vector at each sampling time, the method further includes:

[0105] At each sampling time, the covariance correction matrix is ​​determined by performing matrix multiplication between the Kalman gain vector, the transpose of the augmented excitation vector, and the covariance matrix of the previous sampling time. The matrix obtained by subtracting the iterative covariance matrix of the previous sampling time from the covariance correction matrix is ​​then weighted by a preset forgetting factor to determine the iterative covariance matrix of each sampling time.

[0106] In one specific implementation of this invention, the process of obtaining the iterative covariance matrix at each sampling time includes: ;in, Sampling time The iterative covariance matrix; it should be noted that the formula for obtaining this iterative covariance matrix belongs to the conventional technique for updating the covariance matrix in the recursive least squares algorithm, where... Its function is to perform covariance dilation to prevent the covariance matrix from becoming too small over time, which would cause the algorithm to lose its sensitivity to new changes and ensure that it can continuously track the dynamic changes in the fiber damage state.

[0107] Furthermore, it should be noted that since there is no preceding sampling time for the first sampling time, to avoid boundary issues, the iterative covariance matrix of the sampling time preceding the first sampling time is set as a preset initial diagonal matrix. Specifically, the values ​​corresponding to the non-zero elements in this diagonal matrix are all set to a preset large positive real number. In this embodiment, the preset large positive real number is set to 10 to the power of 5. This setting serves as an initialization condition with large prior uncertainty in the algorithm. Its physical meaning is that it indicates that the algorithm is almost completely unaware of the true values ​​of the damage parameters of each fiber segment in the early stage, thus giving the initial stage observation data a very high correction weight, enabling the algorithm to quickly converge to the true parameter state when new data is input. The optical cable thermal damage vector of the sampling time preceding the first sampling time is set to a zero vector, representing that the initial optical cable is in a damage-free state; this serves as the starting condition for the recursive algorithm.

[0108] It should be noted that the principle for obtaining the optical cable structural damage vector is completely consistent with the principle for obtaining the optical cable thermal damage vector (i.e., the iterative update process based on the recursive parameter estimation algorithm) in terms of mathematical logic and algorithm structure. The only difference between the two is the input physical quantity and the output physical object of the algorithm. That is, the optical cable curvature vector in the process of obtaining the optical cable thermal damage vector is replaced with the impact conduction vector, the relative thermal characteristic value is replaced with the relative structural characteristic value, and the target object of the iterative update is replaced with the optical cable thermal damage vector. Thus, the optical cable structural damage vector at each sampling moment can be determined. In addition, the start conditions of the recursive algorithm are also the same, which will not be elaborated further here.

[0109] Preferably, in some possible implementations of the embodiments of the present invention, the process of detecting the dynamic performance of hollow anti-resonant optical fiber based on the optical cable thermal damage vector and the optical cable structural damage vector includes:

[0110] To determine the effective thermal damage vector, the last element is removed from the optical cable thermal damage vector. Specifically, after removing the last element, min-max normalization is performed on the remaining elements to determine the effective thermal damage vector where all element values ​​are normalized values. The min-max normalization is performed on all optical cable thermal damage vectors of the same model stored in the database (the normalized objects do not include the last element). Within the effective thermal damage vector, the index value corresponding to the element greater than a preset thermal damage threshold is used as the thermal damage index value. In the local optical cable segment sequence, the local optical cable segment with an index value equal to the thermal damage index value is marked as a thermally damaged optical cable segment.

[0111] Since the last element in the optical cable thermal damage vector is a constant term, and the other elements each correspond to a specific local optical cable segment; and according to the process of obtaining the optical cable thermal damage vector, the magnitude of each element corresponds to the gain intensity of the thermal response signal generated by the local optical cable segment under unit bending excitation, which characterizes the aging degree or thermal sensitivity of the optical fiber material at that location; therefore, for each local optical cable segment, the larger the element corresponding to it in the optical cable thermal damage vector, the more severe the aging of the optical fiber material at that location, and the more likely it is to generate local overheating or increased transmission loss under the same bending action; the embodiment of the present invention sets... A preset thermal damage threshold is used to screen out potential fault areas with aging levels significantly higher than normal as thermally damaged optical cable segments for analysis, thereby achieving precise location of hidden thermal fault points along the optical fiber. In one specific implementation of this invention, the preset thermal damage threshold is set to 0.5, which can be adjusted according to the specific implementation environment. The larger the preset thermal damage threshold, the stricter the judgment standard for thermal damage, and the lower the false alarm rate of the system, but it may miss minor aging hazards. Conversely, the more lenient the judgment standard for thermal damage, the more likely the system can detect early and weak signs of aging, but it may generate false alarms due to noise interference.

[0112] The last element in the optical cable structural damage vector is removed to determine the effective structural damage vector. Specifically, after removing the last element in the optical cable structural damage vector, the other elements are subjected to minimum-maximum normalization to determine the effective structural damage vector whose element values ​​are all normalized values. The normalization object range is the same as the normalization object range of the effective thermal damage vector mentioned above, and will not be elaborated further here.

[0113] In the effective structural damage vector, the index value corresponding to the element that is greater than the preset structural damage threshold is used as the structural damage index value; in the sequence of local optical cable segments, the local optical cable segment whose index value is equal to the structural damage index value is marked as the structural damage optical cable segment.

[0114] The structurally damaged optical cable segment obtained represents the physical location where there is a risk of microcracks, structural loosening, or breakage along the optical fiber. It exhibits an exceptionally high oscillation sensitivity to mechanical impact. In a specific implementation of this invention, the preset structural damage threshold is set to 0.4, which is adjusted according to the specific implementation environment. The larger the preset structural damage threshold, the more stringent the criteria for judging structural damage, and the system will only alarm for severe crack risks, which is beneficial to ensuring production efficiency. Conversely, the more sensitive the criteria for judging structural damage, the more the system can capture early micro-structural defects, which is beneficial to improving equipment safety, but may increase the frequency of downtime inspections.

[0115] Finally, the dynamic performance of the hollow-core anti-resonant fiber is tested based on the thermally damaged and structurally damaged optical cable segments. The detected thermally damaged and structurally damaged optical cable segments are marked and transmitted to a visual interactive interface for real-time dynamic performance testing of the hollow-core anti-resonant fiber.

[0116] In summary, a dynamic performance testing method for hollow anti-resonant optical fibers constructs cable curvature vectors and impact transmission vectors characterizing the stress state at different locations along the fiber by real-time acquisition of the relative distance to the base and instantaneous acceleration during the robot-driven fiber movement. Combined with relative thermal and structural characteristic values ​​decoupled from periodic pulse data, a recursive parameter estimation algorithm iteratively inverts the local thermal damage vectors and structural damage vectors along the fiber. This method creatively utilizes the robot's motion trajectory as an active scanning excitation source, transforming traditional endpoint measurement into distributed sensing of the local state along the fiber. This not only accurately locates hidden local damage locations but also effectively distinguishes between material aging and structural cracks, thus improving the accuracy of dynamic performance testing of hollow anti-resonant optical fibers using this invention.

[0117] This invention also provides a dynamic performance testing system for hollow-core antiresonant optical fibers. Please refer to [link to relevant documentation]. Figure 2 The diagram shows a structural diagram of a dynamic performance testing system for hollow anti-resonant optical fiber according to an embodiment of the present invention. The system includes: a data acquisition module 201, a vector determination module 202, an eigenvalue determination module 203, a damage determination module 204, and a dynamic performance testing module 205.

[0118] The data acquisition module 201 is used to acquire all periodic pulse data corresponding to each sampling moment, as well as the relative distance and instantaneous acceleration of the robot to the base, during the process of the robot driving the optical fiber to move in space; wherein, the optical cable is located between the robot's end flange and the base;

[0119] The vector determination module 202 is used to determine the optical cable curvature vector at each sampling moment based on the relative distance between the base and the total length of the optical cable; and to determine the impact transmission vector at each sampling moment based on the instantaneous acceleration changes in the time sequence.

[0120] The eigenvalue determination module 203 is used to determine the corresponding relative thermal eigenvalue at each sampling time based on the reference relative magnitude of the trailing edge energy of the pulse main peak of all periodic pulse data; and to determine the corresponding relative structural eigenvalue based on the relative stability of the peak voltage reference of all periodic pulse data.

[0121] The damage determination module 204 is used to determine the optical cable thermal damage vector at each sampling time by iteratively updating the relative thermal characteristic value and the optical cable curvature vector based on the recursive parameter estimation algorithm; and to determine the optical cable structural damage vector at each sampling time by iteratively updating the relative structural characteristic value and the impact conduction vector based on the principle of obtaining the optical cable thermal damage vector.

[0122] The dynamic performance testing module 205 is used to perform dynamic performance testing of hollow anti-resonant optical fiber based on the optical cable thermal damage vector and the optical cable structural damage vector.

[0123] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the dynamic performance testing system for hollow anti-resonant optical fiber and the dynamic performance testing method for hollow anti-resonant optical fiber provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0124] This invention also provides a computer device; please refer to [link / reference]. Figure 3 The illustration shows a schematic diagram of a computer device structure according to an embodiment of the present invention. The computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302. When the processor 302 executes the computer program 303, the computer device can execute any of the aforementioned methods for dynamic performance testing of hollow anti-resonant optical fibers.

[0125] This invention also provides a computer program product that, when run on a computer device, enables the computer device to execute any of the aforementioned methods for detecting the dynamic performance of hollow anti-resonant optical fibers.

[0126] This invention also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer device, the computer device can execute any of the aforementioned methods for detecting the dynamic performance of hollow anti-resonant optical fibers.

[0127] In the embodiments provided by the present invention, it should be understood that the computer device, computer program product and computer-readable storage medium provided are all used to execute the corresponding methods provided above, and therefore the beneficial effects they can achieve can be referred to the beneficial effects of the methods provided above, which will not be repeated here.

[0128] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0129] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for detecting the dynamic performance of a hollow-core anti-resonant optical fiber, characterized in that, The method includes: During the process of the robot driving the optical fiber to move in space, all periodic pulse data corresponding to each sampling moment, as well as the relative distance and instantaneous acceleration of the robot to the base, are acquired; the optical cable is located between the robot's end flange and the base; Based on the relative distance between the base and the total length of the optical cable, the curvature vector of the optical cable at each sampling moment is determined; based on the instantaneous acceleration changes in time sequence, the impact transmission vector at each sampling moment is determined. At each sampling time, the corresponding relative thermal characteristic value is determined based on the reference relative magnitude of the trailing edge energy of the pulse main peak of all periodic pulse data; the corresponding relative structural characteristic value is determined based on the relative stability of the peak voltage reference of all periodic pulse data. Based on the recursive parameter estimation algorithm, the optical cable thermal damage vector at each sampling time is determined by iteratively updating the relative thermal characteristic value and the optical cable curvature vector according to the regression relationship. Based on the principle of obtaining the optical cable thermal damage vector, the optical cable structural damage vector at each sampling time is determined by iteratively updating the relative structural characteristic value and the impact transmission vector according to the regression relationship. Dynamic performance testing of hollow anti-resonant optical fiber is performed based on the optical cable thermal damage vector and the optical cable structural damage vector.

2. The method for dynamic performance testing of hollow-core anti-resonant optical fiber according to claim 1, characterized in that, The process of obtaining the optical cable curvature vector includes: At each sampling time, the optical cable center sag is determined based on the relative deviation between the total length of the optical cable and the relative distance to the base. The optical cable is divided into equal intervals by a preset interval length to obtain at least two local optical cable segments; parabolic fitting is performed based on the position of each local optical cable segment in the optical cable to determine the corresponding morphological distribution characteristic value. Based on the product of the morphological distribution characteristic value and the optical cable center sag, the equivalent bending curvature of each local optical cable segment at each sampling time is determined; the equivalent bending curvatures of all local optical cable segments are arranged sequentially to determine the optical cable curvature vector.

3. The method for dynamic performance testing of hollow-core anti-resonant optical fiber according to claim 2, characterized in that, The process of obtaining the center sag of the optical cable includes: The difference between the square of the total length of the optical cable and the square of the relative distance to the base is positively correlated to determine the center sag of the optical cable at each sampling time.

4. The method for dynamic performance testing of hollow-core anti-resonant optical fiber according to claim 2, characterized in that, The process of obtaining the morphological distribution feature values ​​includes: All local optical cable segments are arranged along the direction from the robot end flange to the base to determine the local optical cable segment sequence; in the local optical cable segment sequence, the index value corresponding to each local optical cable segment is normalized to determine the corresponding index coefficient; the negative correlation mapping value of the index coefficient of each local optical cable segment is positively correlated with the product of the corresponding index coefficient to determine the morphological distribution characteristic value of each local optical cable segment.

5. The method for dynamic performance testing of hollow anti-resonant optical fiber according to claim 4, characterized in that, The process of obtaining the impact transmission vector includes: The acceleration at each sampling moment is differentially calculated with the acceleration at the previous sampling moment to determine the instantaneous impact scalar at each sampling moment; the product of the index coefficient of each local optical cable segment and the total length of the optical cable is exponentially negatively correlated and mapped to determine the corresponding exponential attenuation coefficient. The effective impact intensity of each local optical cable segment at each sampling time is determined by the product of the exponential decay coefficient and the instantaneous impact scalar. The effective impact intensity of all local optical cable segments is arranged along the direction from the robot end flange to the base, and the impact transmission vector at each sampling time is determined.

6. The method for dynamic performance testing of a hollow anti-resonant optical fiber according to claim 1, characterized in that, The process of obtaining the relative thermal characteristic value includes: Acquire the start time, end time, and trailing edge integration cutoff time of the main pulse for each cycle of pulse data; integrate the waveform amplitude of each cycle of pulse data within the interval corresponding to the start time and end time of the main pulse to obtain the main peak energy; integrate the waveform amplitude of each cycle of pulse data within the interval corresponding to the end time and end time of the main pulse to obtain the trailing edge energy; determine the trailing edge energy ratio of each cycle of pulse data based on the ratio between the main peak energy and the trailing edge energy; determine the corresponding absolute thermal characteristic value based on the average of the trailing edge energy ratios of all sampling times corresponding to each sampling time; determine the relative thermal characteristic value of each sampling time based on the difference between the absolute thermal characteristic value and the preset baseline thermal characteristic value.

7. The method for dynamic performance testing of hollow-core anti-resonant optical fiber according to claim 1, characterized in that, The process of obtaining the relative structural feature values ​​includes: Calculate the coefficient of variation of all periodic pulse peaks corresponding to each sampling time to determine the corresponding absolute structural feature value; determine the relative structural feature value of each sampling time based on the difference between the absolute structural feature value and the preset baseline structural feature value.

8. The method for dynamic performance testing of hollow-core anti-resonant optical fiber according to claim 1, characterized in that, The process of obtaining the optical cable thermal damage vector includes: After concatenating a constant 1 to the end of the optical cable curvature vector, the vector is transposed to determine the augmented excitation vector at each sampling time. The predicted thermal feature value is determined by multiplying the transpose of the augmented excitation vector at each sampling time with the optical cable thermal damage vector at the previous sampling time; the corresponding prediction error is determined by the difference between the relative thermal feature value at each sampling time and the predicted thermal feature value. The Kalman gain vector for each sampling time is calculated based on the iterative covariance matrix of the previous sampling time and the augmented excitation vector. The damage parameter vector correction amount is determined based on the product between the Kalman gain vector and the prediction error; the sum vector between the damage parameter vector correction amount at each sampling time and the optical cable thermal damage vector at the corresponding previous sampling time is calculated to determine the optical cable thermal damage vector at each sampling time.

9. The method for dynamic performance testing of hollow-core anti-resonant optical fiber according to claim 8, characterized in that, After determining the optical cable thermal damage vector at each sampling time, the process also includes: At each sampling time, the covariance correction matrix is ​​determined by performing matrix multiplication sequentially between the Kalman gain vector, the transpose of the augmented excitation vector, and the covariance matrix of the previous sampling time. The matrix obtained by subtracting the iterative covariance matrix of the previous sampling time from the covariance correction matrix is ​​then weighted by a preset forgetting factor to determine the iterative covariance matrix for each sampling time.

10. The method for dynamic performance testing of a hollow-core anti-resonant optical fiber according to claim 4, characterized in that, The process of dynamic performance testing of hollow anti-resonant optical fiber based on the optical cable thermal damage vector and the optical cable structural damage vector includes: Remove the last element from the optical cable thermal damage vector to determine the effective thermal damage vector; in the effective thermal damage vector, the index value corresponding to the element that is greater than the preset thermal damage threshold is used as the thermal damage index value; in the local optical cable segment sequence, the local optical cable segment with the index value equal to the thermal damage index value is marked as the thermally damaged optical cable segment. Remove the last element from the optical cable structure damage vector to determine the effective structural damage vector; in the effective structural damage vector, the index value corresponding to the element that is greater than the preset structural damage threshold is used as the structural damage index value; in the sequence of local optical cable segments, the local optical cable segment with the index value equal to the structural damage index value is marked as the structural damage optical cable segment. Dynamic performance testing of hollow anti-resonant optical fibers was conducted on optical cable segments with thermal damage and structural damage.

Images