Steel wire rope detection method based on quantum magnetic sensor and AI cooperative signal enhancement

Abstract

The application discloses a steel wire rope detection method based on quantum magnetic sensor and AI cooperative signal enhancement, relates to the technical field of steel wire rope detection, and retains real dynamic information of quantum spin decoherence through a constructed decoherence feature set Dcs, so that the detection process no longer completely depends on single change of magnetic flux density; a disturbance feature set Dpd can make subtle disturbance structures caused by noise explicit; an enhanced decoherence feature set Den generated can restore covered physical attenuation details in a high-noise on-site environment; a curve form difference amount Dff and a damage deviation set Dev formed can quantitatively express decoherence microshifts that are difficult to distinguish with naked eyes, so that early damage that is difficult to identify by a traditional method can be accurately captured; by comparing a damage deviation amplitude Fdv with a preset damage criterion threshold Thr and generating a damage result set Drs, accurate determination of the damage state of the steel wire rope is realized.

Description

Technical Field

[0001] This invention relates to the field of wire rope detection technology, specifically a wire rope detection method based on quantum magnetic sensors and AI-assisted signal enhancement. Background Technology

[0002] As the service environments for wire ropes continue to expand towards high loads, high-frequency vibrations, and strong magnetic noise, identifying minute internal damage within the wire rope amidst complex interference has become a crucial area requiring continuous breakthroughs in wire rope safety monitoring technology. With the gradual penetration of quantum magnetic measurement technology into engineering applications, its highly sensitive magnetic response characteristics have brought new possibilities to wire rope inspection. Meanwhile, AI-enhanced signal processing technology has demonstrated stronger potential advantages in processing difficult-to-analyze signals such as weak magnetic fields, noise overload, and subtle perturbations.

[0003] Current magnetic detection systems for steel wire ropes still largely rely on traditional magnetic features such as changes in magnetic flux density and gradients in the magnetic leakage field for damage identification. However, in noisy environments, during high-speed operation of the steel wire rope, and in the early stages of internal microcrack formation, traditional magnetic flux characteristics are often masked by background magnetic field fluctuations, resulting in a severe deficiency in the ability to identify early damage. While current quantum magnetic sensors possess high sensitivity, engineering applications still primarily utilize traditional magnetic measurement features and do not fully leverage the perturbation changes in the curve during quantum spin decoherence. Therefore, defect characteristics are not fully revealed at the physical level. Without introducing techniques capable of resolving perturbations in the decoherence curve morphology, it will not only be difficult to improve the detectability of weak defects but will also keep the monitoring of early damage to steel wire ropes within the limits of traditional methods, preventing the achievement of truly high-precision identification. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a wire rope detection method based on quantum magnetic sensors and AI-assisted signal enhancement, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a steel wire rope detection method based on quantum magnetic sensors and AI-assisted signal enhancement, comprising the following steps: S1. Collect the output signal of the quantum magnetic sensor to form the raw sequence set Raw, extract the short-time decoherence change T2s from the raw sequence set Raw, and construct the decoherence feature set Dcs. S2. Perform noise perturbation analysis on the decoherence feature set Dcs, extract the perturbation change Mdv, and generate the perturbation feature set Dpd based on the perturbation change Mdv. S3. Input the perturbation feature set Dpd into the AI ​​enhancement model to enhance and restore the perturbation curve, and obtain the enhanced decoherence feature set Den. S4. Compare the enhanced decoherence feature set Den with the preset normal decoherence template set Ntp, calculate the curve shape difference Dff, and form the damage deviation set Dev based on the curve shape difference Dff. S5. Generate the damage deviation amplitude Fdv based on the damage deviation set Dev, and compare the damage deviation amplitude Fdv with the preset damage criterion threshold Thr to generate the damage result set Drs as the judgment result of the wire rope damage detection.

[0006] Preferably, S1 includes S11; S11. Execute the quantum state excitation and readout process of the quantum magnetic sensor. The quantum spin is brought into a coherent state by the excitation pulse, and the magnetic response amplitude generated by the spin during the decoherence process is continuously read out within a fixed sampling period to form a time sequence of magnetic response amplitude arranged in the readout order. The acquired magnetic response amplitude time series is buffered, time synchronization calibrated, and amplitude reference corrected. Abnormal jump points are corrected by amplitude smoothing algorithm, and the readout interval error is made consistent with the time axis. The processed magnetic response amplitude time series is packaged according to the read batches to form the raw sequence set Raw.

[0007] Preferably, S1 further includes S12; S12. Based on the original sequence set Raw, each magnetic response amplitude time series is divided into short time windows. The magnetic response amplitude time series is divided into multiple continuous local time periods according to a fixed number of samples. Within each local time period, the magnetic response amplitude points contained therein are used as the basis for calculation. By fitting the local decay trend of these magnetic response amplitude points, the local magnetic response amplitude decay rate within the local time period is calculated. Perform differential comparison on the local decay rate of adjacent local time periods to form a short-time variation structure representing the change in short-time decoherence rate, and obtain the short-time decoherence change amount T2s; The short-term decoherence changes T2s corresponding to multiple local time periods are combined, normalized, and encapsulated in chronological order to form the decoherence feature set Dcs.

[0008] Preferably, S2 includes S21; S21. Based on multiple short-term decoherence changes T2s arranged in chronological order in the decoherence feature set Dcs, the statistical characteristics of each short-term decoherence change T2s in each time period are calculated: the dispersion of the short-term decoherence change T2s in each time period is calculated, the absolute amplitude of the difference between adjacent short-term decoherence changes T2s is obtained by taking the absolute amplitude of the difference between consecutive short-term decoherence changes T2s, and then averaging these differences to obtain the discrete value reflecting the amplitude of the change fluctuation; the amplitude of each short-term decoherence change T2s is directly compared with the short-term decoherence change T2s corresponding to the previous time period to obtain the instantaneous jump amplitude, which is used to identify the spike disturbance caused by sudden noise. The direction of change of multiple consecutive short-term decoherence variables T2s is statistically analyzed, and the trend of change is recorded by comparing the positive and negative values ​​of the difference before and after, forming a local fluctuation trend sequence; Based on the obtained discrete values, instantaneous jump amplitude, and local fluctuation trend, they are combined into a set of structured data describing the noise intensity and noise fluctuation characteristics, which is labeled as the noise density structure Nds; Based on the noise density structure Nds, the difference analysis of the short-term decoherence change T2s in the corresponding time period is performed. The residual offset of the noise influence is calculated by noise stripping method, and the perturbation change Mdv representing the noise disturbance amplitude is obtained.

[0009] Preferably, S2 further includes S22; S22. Arrange the perturbation changes Mdv corresponding to all time periods in chronological order, and perform perturbation structure analysis on each segment of the arranged perturbation changes Mdv. By comparing the perturbation changes Mdv of each time period with the perturbation changes Mdv of the previous time period, obtain the perturbation direction index that represents the direction of perturbation change. By calculating the difference in the change of perturbation Mdv between adjacent time periods, a perturbation intensity index representing the degree of perturbation enhancement or weakening is obtained. By detecting whether the change in perturbation Mdv increases or decreases over multiple consecutive time periods, the perturbation enhancement segment and the perturbation decay segment are identified, forming a perturbation trend structure that represents the evolution of perturbation over time. The disturbance direction index, disturbance intensity index, and disturbance trend structure are combined in chronological order and encapsulated into structured sequence feature data to construct the disturbance feature set Dpd.

[0010] Preferably, S3 includes S31; S31. Based on the disturbance feature set Dpd, preprocess the disturbance direction index sequence, disturbance intensity index sequence and disturbance trend structure sequence arranged in time order. This includes using the minimum-maximum scaling method to perform amplitude normalization processing on the disturbance direction index, disturbance intensity index and disturbance trend marker for each time period. By mapping each index data to the maximum and minimum values, the dimensional differences between different data with different dimensions are eliminated. Based on the fixed-dimensional vector recombination method, the perturbation direction index, perturbation intensity index, and perturbation trend label of each time period are rearranged into a fixed-dimensional feature vector structure, and all feature vectors are recombined and encapsulated in chronological order into an augmented input set Ain for model input.

[0011] Preferably, S3 further includes S32; S32. Input the enhanced input set Ain into the AI ​​enhancement model, and use the temporal correlation analysis algorithm inside the AI ​​enhancement model to perform reverse inference and enhanced restoration of the disturbance information; The AI ​​enhancement model embeds the feature vectors of each time period in the enhancement input set Ain using a sequence feature extraction method. It then extracts the internal correlations between perturbation direction indicators, perturbation intensity indicators, and perturbation trend markers through sequence embedding operations. Next, based on local dependency modeling using temporal convolution, it jointly analyzes the perturbation change structure between adjacent time periods, establishing local dynamic associations between features and identifying the enhancement or decay patterns of noise perturbations within a short timeframe. Finally, it applies a global association modeling mechanism based on bidirectional dependency inference, performing bidirectional information fusion on the entire sequence of the enhancement input set Ain to jointly infer the perturbation patterns of consecutive time periods, thereby restoring the global trend of perturbation behavior. After completing the reverse inference and enhancement restoration, the AI ​​enhancement model uses a perturbation suppression algorithm to reduce the weight of noise components based on the local change structure and global evolution trend of noise perturbation, and uses a morphological reconstruction algorithm to restore the natural evolution law of decoherence features. The enhanced decoherence features output by the AI ​​enhancement model are encapsulated in chronological order to form the enhanced decoherence feature set Den, which represents the true form of decoherence.

[0012] Preferably, S4 includes S41; S41. Taking the enhanced decoherence feature set Den as the analysis object and the preset normal decoherence template set Ntp as the benchmark data, the decoherence features of the two in the same time period are compared time period by time. For each corresponding time period, three types of difference parameters are calculated: by directly comparing the amplitude of the enhanced decoherence feature and the amplitude of the template decoherence feature in that time period, the amplitude difference reflecting the degree of amplitude shift is obtained. By calculating the local change rates of the enhanced decoherence feature and the template decoherence feature in the same time period, and performing a difference operation on the two local change rates, the difference in local change rates, representing the degree of deviation of the local change rate, is obtained. By comparing the changing directions of enhanced decoherence features and template decoherence features, we can obtain the curve trend differences that indicate whether the curve changing trends are consistent. The amplitude difference, local rate of change difference, and curve trend difference corresponding to each time period are combined in chronological order and encapsulated into structured difference data to form the curve shape difference quantity Dff.

[0013] Preferably, S4 further includes S42; S42. Based on the curve shape difference Dff, the maximum and minimum normalization methods are used to normalize the amplitude difference, the local rate of change difference and the curve trend difference. Then, the three types of difference parameters after normalization are added together to obtain the comprehensive offset value that reflects the overall offset degree of the time period. After obtaining the comprehensive offset value of the entire sequence in the curve morphology difference quantity Dff, a continuity analysis is performed on the comprehensive offset value of the entire sequence. By comparing the comprehensive offset values ​​of adjacent time periods, offset enhancement segments, offset surge segments, and offset abnormal trend segments are identified. The comprehensive offset value of the entire sequence, as well as the offset enhancement segments, offset surge segments, and offset abnormal trend segments, are encapsulated into structured deviation data in chronological order to form the damage deviation set Dev.

[0014] The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement includes S51 in step S5; S51. Taking the damage deviation set Dev as input, the damage deviation amplitude Fdv, which represents the overall offset intensity, is obtained by summing all the comprehensive offset values ​​in the set. The damage deviation magnitude Fdv is compared with the preset damage criterion threshold Thr: When the damage deviation magnitude Fdv > the damage criterion threshold Thr, the damage state along with the damage deviation magnitude Fdv is encapsulated into the damage result set Drs; When the damage deviation amplitude Fdv ≤ the damage criterion threshold Thr, the no-damage state is encapsulated into the damage result set Drs; The damage result set Drs is output as the result of the wire rope damage assessment.

[0015] This invention provides a method for detecting steel wire ropes based on quantum magnetic sensors and AI-assisted signal enhancement, which has the following advantages: (1) The constructed decoherence feature set Dcs retains the true dynamic information of quantum spin decoherence, making the detection process no longer entirely dependent on the single change of magnetic flux density; the perturbation feature set Dpd can make the subtle perturbation structure caused by noise explicit, creating a learnable differential basis for subsequent enhancement; the generated enhancement decoherence feature set Den can recover the masked physical attenuation details in a noisy field environment, significantly improving the readability of decoherence features; through the formed curve shape difference Dff and damage deviation set Dev, the micro-shifts of decoherence that are difficult to distinguish with the naked eye can be quantitatively expressed, enabling the accurate capture of early damage that is difficult to identify by traditional methods; finally, by comparing the damage deviation amplitude Fdv with the preset damage criterion threshold Thr and generating the damage result set Drs, the accurate determination of the wire rope damage state is realized. This enables early detection and early warning of micro-damage, greatly improving the reliability and practicality of wire rope safety monitoring.

[0016] (2) By transforming the complex dynamic behavior under noise perturbation conditions into a unified expression that the AI ​​model can effectively understand, high-precision recovery of decoherent features is achieved. Based on the minimum-maximum scaling normalization and fixed-dimensional vector reorganization constructed by the perturbation feature set Dpd, the perturbation direction index, perturbation intensity index, and perturbation trend label of different time periods are all expressed in a numerically consistent and structurally unified manner in the enhanced input set Ain, avoiding feature distortion or model misunderstanding caused by scale differences and dimensional inconsistencies in the original perturbation features. On this basis, the AI ​​enhancement model performs sequence embedding, temporal convolution, and bidirectional dependency inference on the enhanced input set Ain, enabling the model to simultaneously identify the local short-term jumps and global evolution trends of noise perturbation across time periods. Thus, when restoring decoherent features, it can suppress short-term burst noise and faithfully preserve the physically continuous decoherent trajectory, ultimately forming an enhanced decoherent feature set Den with significantly improved physical consistency. This processing method is particularly outstanding in practical application scenarios, significantly improving the detection reliability and early damage identifiability in complex scenarios.

[0017] (3) By calculating the amplitude difference, local rate of change difference, and curve trend difference to form the curve shape difference quantity Dff, the decoherence process can be compared comprehensively from the amplitude level, the rate of change level, and the physical trend level, avoiding the missed detection caused by traditional methods that rely solely on amplitude or magnetic flux changes. Furthermore, based on the normalization processing of the curve shape difference quantity Dff and the calculation of the comprehensive offset value, a unified offset metric can be established between different time periods. Then, through continuous analysis, a damage deviation set Dev is formed, which can automatically identify offset enhancement sections, offset surge sections, and offset abnormal trend sections, thereby providing a complete characterization of the damage development path, occurrence time, and abnormal nature. This significantly improves the positioning accuracy and temporal diagnostic capability of decoherence anomalies, giving this method an early identification advantage that traditional magnetic detection cannot replace in complex stress environments. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the steel wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to the present invention; Figure 2 This is an interactive trend diagram of the short-term decoherence change T2s. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. Example 1

[0020] This invention provides a method for detecting steel wire ropes based on quantum magnetic sensors and AI-assisted signal enhancement. Please refer to [link to relevant documentation]. Figure 1 This includes the following steps: S1. Collect the output signal of the quantum magnetic sensor to form the raw sequence set Raw, extract the short-time decoherence change T2s from the raw sequence set Raw, and construct the decoherence feature set Dcs. S2. Perform noise perturbation analysis on the decoherence feature set Dcs, extract the perturbation change Mdv, and generate the perturbation feature set Dpd based on the perturbation change Mdv. S3. Input the perturbation feature set Dpd into the AI ​​enhancement model to enhance and restore the perturbation curve, and obtain the enhanced decoherence feature set Den. S4. Compare the enhanced decoherence feature set Den with the preset normal decoherence template set Ntp, calculate the curve shape difference Dff, and form the damage deviation set Dev based on the curve shape difference Dff. S5. Generate the damage deviation amplitude Fdv based on the damage deviation set Dev, and compare the damage deviation amplitude Fdv with the preset damage criterion threshold Thr to generate the damage result set Drs as the judgment result of the wire rope damage detection.

[0021] In this embodiment, by adopting a wire rope detection method based on quantum magnetic sensors and AI-assisted signal enhancement, the problem of traditional magnetic detection being unable to identify weak defect responses in strong noise environments can be effectively solved. The decoherence feature set Dcs constructed in step S1 retains the true dynamic information of quantum spin decoherence, making the detection process no longer entirely dependent on the single change of magnetic flux density; the perturbation feature set Dpd formed in step S2 can make the subtle perturbation structure caused by noise explicit, creating a learnable differential basis for subsequent enhancement; the enhanced decoherence feature set Den generated in step S3 can recover the masked physical attenuation details in a highly noisy field environment, significantly improving the readability of decoherence features; through the curve shape difference Dff and damage deviation set Dev formed in step S4, the micro-decoherence shifts that are difficult to distinguish with the naked eye can be quantitatively expressed, enabling the accurate capture of early damage that is difficult to identify by traditional methods (such as the extremely slight decoherence trend shift generated in the initial stage of internal wire breakage); finally, in step S5, by comparing the damage deviation amplitude Fdv with the preset damage criterion threshold Thr and generating the damage result set Drs, the accurate determination of the wire rope damage state is achieved. Especially in practical engineering scenarios, such as the inspection of wind turbine tower hoisting ropes, deep well hoisting wire ropes, or cross-sea cables, vibration, electromagnetic interference, and temperature fluctuations often distort traditional magnetic flux leakage signals. However, this method utilizes quantum decoherence characteristics as the basis for physical detection and combines it with AI enhancement processing. It can maintain high sensitivity recognition capability even under conditions of extremely high noise and extremely weak defects, thereby achieving early detection and early warning of micro-damage and significantly improving the reliability and practicality of wire rope safety monitoring. Example 2

[0022] Please see Figure 1 and Figure 2 Specifically: S1 includes S11; S11. Execute the quantum state excitation and readout process of the quantum magnetic sensor. The quantum spin is brought into a coherent state by the excitation pulse, and the magnetic response amplitude generated by the spin during the decoherence process is continuously read out within a fixed sampling period to form a time sequence of magnetic response amplitude arranged in the readout order. The acquired magnetic response amplitude time series is buffered, time synchronization calibrated, and amplitude reference corrected. Abnormal jump points are corrected by amplitude smoothing algorithm, and the readout interval error is processed to ensure the physical continuity and readout stability of the time series. The processed magnetic response amplitude time series is packaged according to the read batches to form the raw sequence set Raw, which is used to represent the real decoherence physical process; It should be noted that: The magnetic response amplitude is acquired by using an excitation-readout loop method. Each magnetic response amplitude point corresponds to the decoherent state of the quantum spin at a certain time point, thus forming a complete magnetic response amplitude time series.

[0023] S1 further includes S12; S12. Based on the original sequence set Raw, each magnetic response amplitude time series is divided into short time windows. The magnetic response amplitude time series is divided into multiple continuous local time periods according to a fixed number of samples. Within each local time period, the magnetic response amplitude points contained therein are used as the basis for calculation. By fitting the local decay trend of these magnetic response amplitude points, the local magnetic response amplitude decay rate within the local time period is calculated. Perform differential comparison on the local decay rate of adjacent local time periods to form a short-time variation structure representing the short-time decoherence rate change, and obtain the short-time decoherence variation amount T2s used to describe the local change behavior of decoherence; The short-term decoherence changes T2s corresponding to multiple local time periods are combined, normalized, and encapsulated in chronological order to form a decoherence feature set Dcs; It should be noted that: The window is divided into time series of magnetic response amplitudes. Therefore, each local time period contains multiple magnetic response amplitude points, forming dynamic decay data within a small range. The decay rate is obtained by linearly fitting the magnetic response amplitude points within a local time period, and the fitting slope reflects the decoherent decay rate within that segment. The short-term decoherence change T2s is formed by the difference in local decay rates between adjacent time periods. It can enhance the decoherence disturbance caused by minor damage and make the weak change characteristics more prominent. The decoherence feature set Dcs consists of multiple short-time decoherence changes T2s arranged in time sequence, thus forming a feature vector targeting the local morphological changes of the decoherence curve.

[0024] In this embodiment, a data foundation with high physical content can be obtained during the signal acquisition stage. The magnetic response amplitude time series formed after the excitation pulse causes the quantum spin to enter the coherent state can completely reflect the real dynamic changes of the decoherence process. Moreover, the raw sequence set Raw, constructed after buffering, time synchronization calibration, amplitude reference correction, and smoothing correction, has the characteristics of strong continuity and high time axis consistency, so that subsequent analysis is no longer affected by waveform misalignment caused by field jitter, sensor micro-offset, or unstable readout period. In the subsequent local fitting processing based on short time windows, by performing decay trend fitting on the magnetic response amplitude points in each local time period and differentiating the local decay rate of adjacent local time periods, the short-time decoherence change T2s sensitive to subtle damage perturbations can be obtained, and a decoherence feature set Dcs that can characterize the changes in local decoherence behavior can be constructed. This data construction method can capture slight shifts in the decoherence decay rate within an extremely short sampling interval, thus detecting early signs of damage before the traditional magnetic signal shows significant changes. For example, during routine inspections of continuous suspension cables on viaducts, the slight structural fatigue caused by long-term temperature variations in the wire ropes does not immediately result in significant magnetic flux changes. However, this method, by enhancing the differential characteristics of the short-term decoherence change T2s, can expose these weak decoherence disturbances caused by fatigue in advance. This allows maintenance personnel to obtain early warnings of degradation before the damage develops to a macroscopically visible stage, thereby improving the safety redundancy of the structure for long-term use. Example 3

[0025] Specifically: S2 includes S21; S21. Based on multiple short-term decoherence changes T2s arranged in chronological order in the decoherence feature set Dcs, the statistical characteristics of each short-term decoherence change T2s in each time period are calculated: the dispersion of the short-term decoherence change T2s in each time period is calculated, the absolute amplitude of the difference between adjacent short-term decoherence changes T2s is obtained by taking the absolute amplitude of the difference between consecutive short-term decoherence changes T2s, and then averaging these differences to obtain the discrete value reflecting the amplitude of the change fluctuation; the amplitude of each short-term decoherence change T2s is directly compared with the short-term decoherence change T2s corresponding to the previous time period to obtain the instantaneous jump amplitude, which is used to identify the spike disturbance caused by sudden noise. The direction of change of multiple consecutive short-term decoherence variables T2s is statistically analyzed, and the trend of change is recorded by comparing the positive and negative values ​​of the difference before and after, forming a local fluctuation trend sequence; Based on the obtained discrete values, instantaneous jump amplitude, and local fluctuation trend, they are combined into a set of structured data describing the noise intensity and noise fluctuation characteristics, which is labeled as the noise density structure Nds; Based on the noise density structure Nds, the difference analysis of the short-term decoherence change T2s in the corresponding time period is performed. The remaining offset of the noise influence is calculated by noise stripping method, and the perturbation change Mdv representing the noise disturbance amplitude is obtained. It should be noted that: The degree of dispersion is used to quantify the stationarity of short-term decoherence change T2s in adjacent time periods. It is calculated as follows: take multiple consecutive short-term decoherence changes T2s, calculate the absolute value of their adjacent differences, and take the average of these absolute values. This average value represents the average degree of data fluctuation over time. The higher the degree of dispersion, the greater the noise influence. The instantaneous jump amplitude is used to identify sharp jumps caused by noise. It is calculated as follows: take two short-time decoherence changes T2s corresponding to time period n and time period n-1, calculate the absolute value of the difference between the two. The larger the value, the stronger the noise impact may be at that moment. Local fluctuation trends are used to identify the direction of noise disturbances. They are formed by comparing the magnitude of the decoherence change T2s in adjacent short time intervals. If it increases, it is recorded as a positive trend; if it decreases, it is recorded as a negative trend. The continuous trend sequence constitutes the local fluctuation trend structure, which helps to determine whether the noise is in a state of continuous enhancement or continuous decay. The noise density structure Nds is a structured set of three types of data: "discretion value" for each time period, "instantaneous jump amplitude" for each time period, and "local fluctuation trend mark" for each time period. Its structure is a sequence of triplets arranged in chronological order, namely: a continuous sequence of discrete values, jump amplitudes, and trend marks. The perturbation change Mdv originates from the original short-time decoherence change T2s - the decoherence change structure after noise stripping. Its value reflects the true magnitude of the perturbation caused by the noise during this time period and is the input for subsequent perturbation modeling.

[0026] S2 further includes S22; S22. Arrange the perturbation changes Mdv corresponding to all time periods in chronological order, and perform perturbation structure analysis on each segment of the arranged perturbation changes Mdv. By comparing the perturbation changes Mdv of each time period with the perturbation changes Mdv of the previous time period, obtain the perturbation direction index that represents the direction of perturbation change. By calculating the difference in the change of perturbation Mdv between adjacent time periods, a perturbation intensity index representing the degree of perturbation enhancement or weakening is obtained. By detecting whether the change in perturbation Mdv increases or decreases over multiple consecutive time periods, the perturbation enhancement segment and the perturbation decay segment are identified, forming a perturbation trend structure that represents the evolution of perturbation over time. The disturbance direction index, disturbance intensity index, and disturbance trend structure are combined in chronological order and encapsulated into structured sequence feature data to construct the disturbance feature set Dpd; It should be noted that: The disturbance direction index is used to record whether the noise disturbance is stronger or weaker in the current time period compared to the previous time period. It is defined as follows: if the current perturbation change Mdv is greater than the perturbation change Mdv in the previous time period, it is recorded as the direction of enhancement; if it is less than, it is recorded as the direction of weakening. The disturbance intensity index is used to measure the absolute magnitude of the change in disturbance amplitude. It is calculated by taking the absolute value of the difference between two adjacent micro-disturbance changes Mdv as the disturbance intensity. This index is used to characterize the degree of abrupt change in noise disturbance. The disturbance trend structure consists of disturbance direction indicators for multiple consecutive time periods. If multiple consecutive time periods are in an increasing direction, it constitutes a disturbance enhancement segment; if multiple consecutive time periods are in a decreasing direction, it constitutes a disturbance attenuation segment. This structure is used to express the overall development trend of noise disturbance. The perturbation feature set Dpd is composed of three types of structured data encapsulated in chronological order: perturbation direction index sequence, perturbation intensity index sequence, and perturbation trend structure sequence; its structure is a perturbation structured triple combination sequence arranged in time period, used to fully express the dynamic behavior of noise perturbation.

[0027] In this embodiment, the noise density structure Nds, composed of discrete values, instantaneous jump amplitudes, and local fluctuation trends based on the short-term decoherence change T2s, can accurately characterize the intensity distribution and fluctuation pattern of noise in different time periods. It can not only identify the energy level of the noise but also whether it exhibits spike-type impacts, stable fluctuations, or trend-driven drift. Furthermore, by stripping the noise influence from the noise density structure Nds to obtain the perturbation change Mdv, the actual disturbances caused by noise can be made explicit independently from the decoherence features. This allows signal segments often misjudged as "random fluctuations" in traditional methods to be restored to perturbation information with clear physical direction. Subsequently, the perturbation feature set Dpd, composed of the perturbation direction index, perturbation intensity index, and perturbation trend structure extracted from the perturbation change Mdv for all time periods, provides a complete dynamic link that can continuously track the evolution of noise perturbation. The above processing enables the identification of "noise pseudo-defects" and "real decoherence disturbances" in complex environments. For example, in the scenario of wire rope detection for port cranes, the collision vibrations generated by the operation of the equipment will form a large number of spike noises in the quantum magnetic sensor. Traditional magnetic detection usually mistakenly identifies this as a wire breakage signal. However, this method quantitatively describes the noise state through the noise density structure Nds and separates the disturbance properties caused by the noise through the perturbation change Mdv and the perturbation feature set Dpd. It can accurately distinguish between "instantaneous impact noise caused by light hook impact" and "out-of-phase interference caused by internal damage to the wire rope", significantly reducing the false judgment rate and greatly improving the detection reliability under complex working conditions.

[0028] Example 4 Specifically: S3 includes S31; S31. Based on the disturbance feature set Dpd, preprocess the disturbance direction index sequence, disturbance intensity index sequence and disturbance trend structure sequence arranged in time order. This includes using the minimum-maximum scaling method to perform amplitude normalization processing on the disturbance direction index, disturbance intensity index and disturbance trend marker for each time period. By mapping each index data to the maximum and minimum values, the dimensional differences between different data with different dimensions are eliminated. Based on the fixed-dimensional vector recombination method, the perturbation direction index, perturbation intensity index, and perturbation trend label of each time period are rearranged into a fixed-dimensional feature vector structure, and all feature vectors are recombined and encapsulated in chronological order into an augmented input set Ain for model input; It should be noted that: The min-max scaling method maps each perturbation index to a uniform interval by calculating the minimum and maximum values ​​of each perturbation index in the entire sequence, so that all perturbation direction indices, perturbation intensity indices, and perturbation trend markers are expressed on the same scale, effectively eliminating dimensional differences; the normalization results are used to ensure that the augmented input set Ain can maintain numerical stability in the AI ​​model; Fixed-dimensional vector reorganization method description: Each time period consists of three different types of data. By constructing a fixed-dimensional vector, each time period is represented by the same structure: Dimension 1: Disturbance direction index; Dimension 2: Disturbance intensity index; Dimension 3: Disturbance trend marker.

[0029] S3 further includes S32; S32. Input the enhanced input set Ain into the AI ​​enhancement model, and use the temporal correlation analysis algorithm inside the AI ​​enhancement model to perform reverse inference and enhanced restoration of the disturbance information; The AI ​​enhancement model embeds the feature vectors of each time period in the enhancement input set Ain using a sequence feature extraction method. It then extracts the internal correlations between perturbation direction indicators, perturbation intensity indicators, and perturbation trend markers through sequence embedding operations. Next, based on local dependency modeling using temporal convolution, it jointly analyzes the perturbation change structure between adjacent time periods, establishing local dynamic associations between features and identifying the enhancement or decay patterns of noise perturbations within a short timeframe. Finally, it applies a global association modeling mechanism based on bidirectional dependency inference, performing bidirectional information fusion on the entire sequence of the enhancement input set Ain to jointly infer the perturbation patterns of consecutive time periods, thereby restoring the global trend of perturbation behavior. After completing the reverse inference and enhancement restoration, the AI ​​enhancement model uses a perturbation suppression algorithm to reduce the weight of noise components based on the local change structure and global evolution trend of noise perturbation, and uses a morphological reconstruction algorithm to restore the natural evolution law of decoherence features. The enhanced decoherence features output by the AI ​​enhancement model are encapsulated in chronological order to form the enhanced decoherence feature set Den, which represents the true form of decoherence. It should be noted that: Sequence embedding is used to map the feature vectors of each time period in the augmented input set Ain to a unified high-dimensional representation space, enabling the model to recognize the semantic relationships between perturbation direction indicators, perturbation intensity indicators, and perturbation trend labels; sequence embedding ensures that the model can fully understand the internal parameter relationships of the feature vectors of each time period; Temporal convolution methods slide convolution kernels along the time dimension to jointly process features of adjacent time periods, enabling the model to capture: local changes in perturbation intensity indicators, short-term turning points in perturbation direction indicators, and local continuity of perturbation trend structures, thereby building the ability to understand the local dependencies of short-term noise perturbations. The bidirectional dependency inference method consists of time series information in two directions: forward sequence inference (analyzing the trend from the beginning to the end) and reverse sequence inference (analyzing the trend from the end to the beginning). By fusing the feature representations of the forward and reverse sequences, the model can understand the global trend of the disturbance based on the complete time axis, which can be used to determine whether the noise disturbance is continuously increasing, decreasing, or changing in stages. The perturbation suppression algorithm is based on the following factors: perturbation direction index, perturbation intensity index, and perturbation trend structure. The global perturbation trend obtained by bidirectional dependency inference reduces the weight or eliminates the parts that the model considers to be noise perturbations, so that the influence of noise on the decoherence change characteristics is gradually reduced.

[0030] In this embodiment, by transforming the complex dynamic behavior under noise perturbation conditions into a unified expression that the AI ​​model can effectively understand, high-precision recovery of decoherent true features is achieved. Based on the minimum-maximum scaling normalization and fixed-dimensional vector reorganization constructed from the perturbation feature set Dpd, the perturbation direction index, perturbation intensity index, and perturbation trend markers at different time periods are all expressed in a numerically consistent and structurally unified manner in the enhanced input set Ain, avoiding feature distortion or model misunderstanding caused by scale differences and dimensional inconsistencies in the original perturbation features. On this basis, the AI ​​enhancement model performs sequence embedding, temporal convolution, and bidirectional dependency inference on the enhanced input set Ain, enabling the model to simultaneously identify local short-term jumps in noise perturbation and global evolution trends across time periods. This allows the model to suppress short-term burst noise while faithfully preserving the physically continuous decoherence trajectory when restoring decoherent features, ultimately forming an enhanced decoherence feature set Den with significantly improved physical consistency. This processing method is particularly effective in practical applications. For example, in the monitoring of steel wire ropes in mine hoisting, repeated collisions and resonant vibrations of the inner wall of the shaft can generate high-frequency disturbances with random directions and unstable amplitudes in a short period of time. Traditional filtering methods often weaken the real decoherence changes along with the noise, causing delays in damage identification. However, this method, through the joint processing of the enhanced input set Ain and the AI ​​enhancement model, can automatically attenuate the noise components while maintaining the real decoherence attenuation pattern. Even under high vibration and strong interference environments, it can still accurately recover the decoherence pattern, significantly improving the detection reliability and early damage identification in complex scenarios.

[0031] Example 5 Specifically: S4 includes S41; S41. Taking the enhanced decoherence feature set Den as the analysis object and the preset normal decoherence template set Ntp as the benchmark data, the decoherence features of the two in the same time period are compared time period by time. For each corresponding time period, three types of difference parameters are calculated: by directly comparing the amplitude of the enhanced decoherence feature and the amplitude of the template decoherence feature in that time period, the amplitude difference reflecting the degree of amplitude shift is obtained. By calculating the local change rates of the enhanced decoherence feature and the template decoherence feature in the same time period, and performing a difference operation on the two local change rates, the difference in local change rates, representing the degree of deviation of the local change rate, is obtained. By comparing the changing directions (enhancement, weakening, or maintenance) of enhanced decoherence features and template decoherence features, we can obtain the curve trend differences that indicate whether the curve changing trends are consistent. The amplitude difference, local rate of change difference, and curve trend difference corresponding to each time period are combined in chronological order and encapsulated into structured difference data to form a curve shape difference quantity Dff used to characterize the degree of decoherence feature shift. It should be noted that: The enhanced decoherence feature (Den) is a decoherence feature output by the AI ​​enhancement model. It represents the true decoherence behavior feature recovered after noise is suppressed. The amplitude of the enhanced decoherence feature refers to the magnetic response amplitude of the feature at a certain time point, which is a quantitative indicator reflecting the internal decoherence of the wire rope. The normal decoherence template feature Ntp is a set of templates constructed based on the decoherence characteristics of wire rope under normal conditions. It is used as a reference benchmark. The template is formed by statistical analysis of measurement data under long-term stable conditions and represents the ideal decoherence curve without damage or degradation. The amplitude difference is calculated by comparing the amplitudes of the enhanced decoherence feature and the normal decoherence template feature within the same time period. The formula is: Amplitude difference = |Amplitude of enhanced decoherence feature - Amplitude of template decoherence feature|. This difference reflects the offset of the amplitudes of the two features and indicates the degree of amplitude deviation during the decoherence process within that time period. The difference in local rate of change is calculated by comparing the rate of change of enhanced decoherence features and normal decoherence template features within the same time period. The rate difference is calculated by the following steps: calculating the local decay rate of enhanced decoherence features and normal decoherence template features within each time period; comparing the rate difference between the two and taking their absolute value to reflect the degree of deviation between the two in terms of rate of change. The curve trend difference is used to describe whether the changing trends of the enhanced decoherence feature and the normal decoherence template feature are consistent in the same time period; if the two change in the same direction (e.g., both increase or both decrease), then the trend difference = 0; if the change in opposite directions (one increases and the other decreases), then the trend difference = 1; this difference reflects whether the decoherence process exhibits the expected physical trend.

[0032] S4 also includes S42; S42. Based on the curve shape difference Dff, the maximum and minimum normalization methods are used to normalize the amplitude difference, the local rate of change difference, and the curve trend difference. Then, the three types of difference parameters after normalization are added together to obtain a comprehensive offset value that reflects the overall offset degree of the time period. The comprehensive offset value is used to reflect the degree of deviation of the decoherence behavior of the time period from the normal template. After obtaining the comprehensive offset value of the entire sequence in the curve morphology difference quantity Dff, the continuity analysis of the comprehensive offset value of the entire sequence is performed. By comparing the comprehensive offset values ​​of adjacent time periods, the offset enhancement segment, offset surge segment, and offset abnormal trend segment are identified. The comprehensive offset value of the entire sequence, as well as the offset enhancement segment, offset surge segment, and offset abnormal trend segment, are encapsulated into structured deviation data in chronological order to form the damage deviation set Dev. It should be noted that: Anomaly zones are identified based on changes in the overall offset value, and include three categories: Increased offset segment: If the composite offset value shows a continuous upward trend over multiple consecutive time periods, it is defined as an increased offset segment, indicating that there may be signs of cumulative decoherence deviation. A sudden increase in offset segment: If the overall offset value in a certain time period shows a significant jump compared to the previous time period, it is defined as a sudden increase in offset segment, indicating that there may be sudden or localized damage behavior. Anomaly trend deviation segment: If the direction of change of the overall deviation value does not conform to the expected decoherence trend (such as a reversal change that should not occur physically), it is defined as an anomaly trend deviation segment, used to identify anomalies in the decoherence behavior.

[0033] S5 includes S51; S51. Taking the damage deviation set Dev as input, the damage deviation amplitude Fdv, which represents the overall offset intensity, is obtained by summing all the comprehensive offset values ​​in the set. The damage deviation magnitude Fdv is obtained by summing the comprehensive offset value sequence and is used to describe the overall degree of decoherence offset. The damage deviation magnitude Fdv is compared with the preset damage criterion threshold Thr: When the damage deviation magnitude Fdv > the damage criterion threshold Thr, the damage state along with the damage deviation magnitude Fdv is encapsulated into the damage result set Drs; When the damage deviation amplitude Fdv ≤ the damage criterion threshold Thr, the no-damage state is encapsulated into the damage result set Drs; The damage result set Drs is output as the result of the wire rope damage assessment.

[0034] In this embodiment, by calculating the amplitude difference, local rate of change difference, and curve trend difference to form a curve shape difference quantity Dff, a comprehensive comparison of the decoherence process can be performed from the amplitude level, rate of change level, and physical trend level, avoiding the missed detections caused by traditional methods that rely solely on amplitude or magnetic flux changes. Furthermore, based on the normalization processing of the curve shape difference quantity Dff and the calculation of the comprehensive offset value, a unified offset metric can be established across different time periods. Then, through continuous analysis, a damage deviation set Dev is formed, which can automatically identify offset enhancement sections, offset abrupt increase sections, and offset abnormal trend sections, thereby providing a complete characterization of the damage development path, occurrence time, and abnormal nature. For example, in the operation of aerial cableways, high humidity and periodic loads can cause early fatigue in some strands of the wire rope, resulting in a slight drift in the local attenuation rate on the decoherence curve. However, traditional magnetic detection, due to interference from cable swaying and wind loads, often only reveals significant signal changes after the damage has progressed. This method, however, uses the curve morphology difference (Dff) to promptly capture abnormal reversals in the direction of decoherence change within a specific time period, forming continuous abrupt shifts in the damage deviation set (Dev). This allows maintenance personnel to receive early warnings before fatigue develops into wire breakage. This mechanism significantly improves the location accuracy and temporal diagnostic capabilities of decoherence anomalies, giving this method an early identification advantage that traditional magnetic detection cannot replace in complex stress environments.

[0035] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for detecting steel wire ropes based on quantum magnetic sensors and AI-assisted signal enhancement, characterized in that: Includes the following steps: S1. Collect the output signal of the quantum magnetic sensor to form the raw sequence set Raw, extract the short-time decoherence change T2s from the raw sequence set Raw, and construct the decoherence feature set Dcs. S2. Perform noise perturbation analysis on the decoherence feature set Dcs, extract the perturbation change Mdv, and generate the perturbation feature set Dpd based on the perturbation change Mdv. S3. Input the perturbation feature set Dpd into the AI ​​enhancement model to enhance and restore the perturbation curve, and obtain the enhanced decoherence feature set Den. S4. Compare the enhanced decoherence feature set Den with the preset normal decoherence template set Ntp, calculate the curve shape difference Dff, and form the damage deviation set Dev based on the curve shape difference Dff. S5. Generate the damage deviation amplitude Fdv based on the damage deviation set Dev, and compare the damage deviation amplitude Fdv with the preset damage criterion threshold Thr to generate the damage result set Drs as the judgment result of the wire rope damage detection.

2. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 1, characterized in that: S1 includes S11; S11. Execute the quantum state excitation and readout process of the quantum magnetic sensor. The quantum spin is brought into a coherent state by the excitation pulse, and the magnetic response amplitude generated by the spin during the decoherence process is continuously read out within a fixed sampling period to form a time sequence of magnetic response amplitude arranged in the readout order. The acquired magnetic response amplitude time series is buffered, time synchronization calibrated, and amplitude reference corrected. Abnormal jump points are corrected by amplitude smoothing algorithm, and the readout interval error is made consistent with the time axis. The processed magnetic response amplitude time series is packaged according to the read batches to form the raw sequence set Raw.

3. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 2, characterized in that: S1 further includes S12; S12. Based on the original sequence set Raw, each magnetic response amplitude time series is divided into short time windows. The magnetic response amplitude time series is divided into multiple continuous local time periods according to a fixed number of samples. Within each local time period, the magnetic response amplitude points contained therein are used as the basis for calculation. By fitting the local decay trend of these magnetic response amplitude points, the local magnetic response amplitude decay rate within the local time period is calculated. Perform differential comparison on the local decay rate of adjacent local time periods to form a short-time variation structure representing the change in short-time decoherence rate, and obtain the short-time decoherence change amount T2s; The short-term decoherence changes T2s corresponding to multiple local time periods are combined, normalized, and encapsulated in chronological order to form the decoherence feature set Dcs.

4. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 3, characterized in that: S2 includes S21; S21. Based on multiple short-term decoherence changes T2s arranged in chronological order in the decoherence feature set Dcs, the statistical characteristics of each short-term decoherence change T2s in each time period are calculated: the dispersion of the short-term decoherence change T2s in each time period is calculated, the absolute amplitude of the difference between adjacent short-term decoherence changes T2s is obtained by taking the absolute amplitude of the difference between consecutive short-term decoherence changes T2s, and then averaging these differences to obtain the discrete value reflecting the amplitude of the change fluctuation; the amplitude of each short-term decoherence change T2s is directly compared with the short-term decoherence change T2s corresponding to the previous time period to obtain the instantaneous jump amplitude, which is used to identify the spike disturbance caused by sudden noise. The direction of change of multiple consecutive short-term decoherence variables T2s is statistically analyzed, and the trend of change is recorded by comparing the positive and negative values ​​of the difference before and after, forming a local fluctuation trend sequence; Based on the obtained discrete values, instantaneous jump amplitude, and local fluctuation trend, they are combined into a set of structured data describing the noise intensity and noise fluctuation characteristics, which is labeled as the noise density structure Nds; Based on the noise density structure Nds, the difference analysis of the short-term decoherence change T2s in the corresponding time period is performed. The residual offset of the noise influence is calculated by noise stripping method, and the perturbation change Mdv representing the noise disturbance amplitude is obtained.

5. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 4, characterized in that: S2 further includes S22; S22. Arrange the perturbation changes Mdv corresponding to all time periods in chronological order, and perform perturbation structure analysis on each segment of the arranged perturbation changes Mdv. By comparing the perturbation changes Mdv of each time period with the perturbation changes Mdv of the previous time period, obtain the perturbation direction index that represents the direction of perturbation change. By calculating the difference in the change of perturbation Mdv between adjacent time periods, a perturbation intensity index representing the degree of perturbation enhancement or weakening is obtained. By detecting whether the change in perturbation Mdv increases or decreases over multiple consecutive time periods, the perturbation enhancement segment and the perturbation decay segment are identified, forming a perturbation trend structure that represents the evolution of perturbation over time. The disturbance direction index, disturbance intensity index, and disturbance trend structure are combined in chronological order and encapsulated into structured sequence feature data to construct the disturbance feature set Dpd.

6. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 5, characterized in that: S3 includes S31; S31. Based on the disturbance feature set Dpd, preprocess the disturbance direction index sequence, disturbance intensity index sequence and disturbance trend structure sequence arranged in time order. This includes using the minimum-maximum scaling method to perform amplitude normalization processing on the disturbance direction index, disturbance intensity index and disturbance trend marker for each time period. By mapping each index data to the maximum and minimum values, the dimensional differences between different data with different dimensions are eliminated. Based on the fixed-dimensional vector recombination method, the perturbation direction index, perturbation intensity index, and perturbation trend label of each time period are rearranged into a fixed-dimensional feature vector structure, and all feature vectors are recombined and encapsulated in chronological order into an augmented input set Ain for model input.

7. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 6, characterized in that: S3 further includes S32; S32. Input the enhanced input set Ain into the AI ​​enhancement model, and use the temporal correlation analysis algorithm inside the AI ​​enhancement model to perform reverse inference and enhanced restoration of the disturbance information; The AI ​​enhancement model embeds the feature vectors of each time period in the enhancement input set Ain using a sequence feature extraction method. It then extracts the internal correlations between perturbation direction indicators, perturbation intensity indicators, and perturbation trend markers through sequence embedding operations. Next, based on local dependency modeling using temporal convolution, it jointly analyzes the perturbation change structure between adjacent time periods, establishing local dynamic associations between features and identifying the enhancement or decay patterns of noise perturbations within a short timeframe. Finally, it applies a global association modeling mechanism based on bidirectional dependency inference, performing bidirectional information fusion on the entire sequence of the enhancement input set Ain to jointly infer the perturbation patterns of consecutive time periods, thereby restoring the global trend of perturbation behavior. After completing the reverse inference and enhancement restoration, the AI ​​enhancement model uses a perturbation suppression algorithm to reduce the weight of noise components based on the local change structure and global evolution trend of noise perturbation, and uses a morphological reconstruction algorithm to restore the natural evolution law of decoherence features. The enhanced decoherence features output by the AI ​​enhancement model are encapsulated in chronological order to form the enhanced decoherence feature set Den, which represents the true form of decoherence.

8. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 7, characterized in that: S4 includes S41; S41. Taking the enhanced decoherence feature set Den as the analysis object and the preset normal decoherence template set Ntp as the benchmark data, the decoherence features of the two in the same time period are compared time period by time. For each corresponding time period, three types of difference parameters are calculated: by directly comparing the amplitude of the enhanced decoherence feature and the amplitude of the template decoherence feature in that time period, the amplitude difference reflecting the degree of amplitude shift is obtained. By calculating the local change rates of the enhanced decoherence feature and the template decoherence feature in the same time period, and performing a difference operation on the two local change rates, the difference in local change rates, representing the degree of deviation of the local change rate, is obtained. By comparing the changing directions of enhanced decoherence features and template decoherence features, we can obtain the curve trend differences that indicate whether the curve changing trends are consistent. The amplitude difference, local rate of change difference, and curve trend difference corresponding to each time period are combined in chronological order and encapsulated into structured difference data to form the curve shape difference quantity Dff.

9. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 8, characterized in that: S4 also includes S42; S42. Based on the curve shape difference Dff, the maximum and minimum normalization methods are used to normalize the amplitude difference, the local rate of change difference and the curve trend difference. Then, the three types of difference parameters after normalization are added together to obtain the comprehensive offset value that reflects the overall offset degree of the time period. After obtaining the comprehensive offset value of the entire sequence in the curve morphology difference quantity Dff, the continuity analysis of the comprehensive offset value of the entire sequence is performed. By comparing the comprehensive offset values ​​of adjacent time periods, the offset enhancement segment, the offset sudden increase segment, and the offset abnormal trend segment are identified. The overall offset value of the entire sequence, as well as the offset enhancement segment, offset surge segment, and offset abnormal trend segment, are encapsulated in chronological order into structured offset data to form the damage offset set Dev.

10. The wire rope detection method based on quantum magnetic sensor and AI collaborative signal enhancement according to claim 9, characterized in that: S5 includes S51; S51. Taking the damage deviation set Dev as input, the damage deviation amplitude Fdv, which represents the overall offset intensity, is obtained by summing all the comprehensive offset values ​​in the set. The damage deviation magnitude Fdv is compared with the preset damage criterion threshold Thr: When the damage deviation magnitude Fdv > the damage criterion threshold Thr, the damage state along with the damage deviation magnitude Fdv is encapsulated into the damage result set Drs; When the damage deviation amplitude Fdv ≤ the damage criterion threshold Thr, the no-damage state is encapsulated into the damage result set Drs; The damage result set Drs is output as the result of the wire rope damage assessment.

Images