Tubular reactor eddy current detection signal processing method

Abstract

A tubular reactor eddy current detection signal processing method comprises the following steps: acquiring an eddy current signal at a jacketed pipe side through an eddy current detection probe, and associating the eddy current signal with a detection position to construct an eddy current signal space sequence; and on the basis of the eddy current signal space sequence, constructing equivalent background characterization of the jacketed pipe and interlayer interface to background eddy current response. And carrying out local anomaly analysis on the eddy current signal space sequence according to the equivalent background representation so as to identify an eddy current anomaly response component corresponding to the inner tube defect. And eddy current abnormal responses conforming to the electromagnetic disturbance characteristics are screened from the eddy current abnormal response components so as to reconstruct inner pipe defect characteristics. Whether the eddy current signal meets the inner pipe defect response condition or not is judged according to the inner pipe defect characteristics, if yes, the corresponding inner pipe defect position and inner pipe defect information are output, and stable extraction and reliable judgment of the real defect signal of the inner pipe layer are achieved by conducting layered equivalent characterization, interface response separation and defect characteristic reconstruction on the eddy current detection signal.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of eddy current field analysis, and in particular to a method for processing eddy current detection signals of a tubular reactor. BACKGROUND

[0002] In a multi-layer composite pipe wall structure formed by the inner pipe alloy layer and the jacket pipe, only a close fit is formed between the two layers of materials without realizing complete metallurgical bonding, and the interlayer interface constitutes a sudden boundary in the electromagnetic sense, and the eddy current will obviously redistribute in space and distort in form during propagation. Under this condition, the local eddy current disturbance caused by the micro-cracks in the inner pipe layer and the eddy current field redistribution caused by the interlayer interface itself present highly similar and superimposed response characteristics on the signal layer, making it difficult to form a stable and distinguishable defect characteristic pattern.

[0003] At present, the existing technical solutions usually implicitly assume that the pipe wall to be detected can be equivalent to a single homogeneous conductor when performing eddy current signal feature enhancement and continuity determination, without independently modeling or constraining the interface effect introduced by the multi-layer structure, so that it is difficult to effectively distinguish the signal changes caused by real crack defects from the eddy current distortion caused by the inherent electromagnetic boundary of the structure in the composite pipe wall scenario, thereby leading to systematic uncertainty in the defect identification results.

[0004] To this end, the present application provides a method for processing eddy current detection signals of a tubular reactor, which realizes stable extraction and reliable discrimination of real defect signals in the inner pipe layer by layering equivalent characterization of eddy current detection signals, separating interface responses, and reconstructing defect features. SUMMARY

[0005] The present application provides a method for processing eddy current detection signals of a tubular reactor, which solves the technical problem that it is difficult to effectively distinguish the signal changes caused by real crack defects from the eddy current distortion caused by the inherent electromagnetic boundary of the structure in the composite pipe wall scenario in the prior art, thereby leading to systematic uncertainty in the defect identification results.

[0006] The present application adopts the following technical solutions.

[0007] The present application discloses a method for processing eddy current detection signals of a tubular reactor, which comprises:

[0008] An eddy current detection probe is used to acquire eddy current signals on the jacket pipe side, and the eddy current signals are associated with the detection position to construct an eddy current signal space sequence;

[0009] Based on the eddy current signal space sequence, an equivalent background representation of the background eddy current response of the jacket pipe and the interlayer interface is constructed;

[0010] performing local anomaly analysis on the eddy current signal space sequence according to the equivalent background representation to identify an eddy current abnormal response component corresponding to the inner tube defect;

[0011] screening the eddy current abnormal response meeting the electromagnetic disturbance feature from the eddy current abnormal response component to reconstruct an inner tube defect feature;

[0012] judging whether the eddy current signal meets an inner tube defect response condition according to the inner tube defect feature, and if so, outputting a corresponding inner tube defect position and inner tube defect information.

[0013] Further, the eddy current signal on the side of the jacket pipe is acquired by the eddy current detection probe, and the eddy current signal is associated with a detection position to construct an eddy current signal space sequence, including:

[0014] establishing a jacket pipe detection coordinate system on the surface of the jacket pipe, and determining a scanning path according to the jacket pipe detection coordinate system, while setting a sampling period of the eddy current signal and a plurality of sampling points on the side of the jacket pipe;

[0015] based on the scanning path and the set sampling period and the plurality of sampling points, setting the eddy current detection probe on the surface of the jacket pipe to acquire an initial eddy current signal sequence with a time stamp;

[0016] wherein the initial eddy current signal sequence contains an eddy current amplitude component and an eddy current phase component corresponding to each sampling time.

[0017] Further, the eddy current signal on the side of the jacket pipe is acquired by the eddy current detection probe, and the eddy current signal is associated with a detection position to construct an eddy current signal space sequence, including:

[0018] selecting a baseline window from the initial eddy current signal sequence, and calculating a corresponding baseline intensity in the baseline window to perform amplitude normalization processing on the initial eddy current signal sequence according to the baseline intensity to obtain a baseline-corrected eddy current signal sequence;

[0019] discretizing the scanning path into a plurality of detection positions, aligning the detection positions with the time stamp, and aggregating a plurality of sampling points at the same detection position to construct the eddy current signal space sequence;

[0020] wherein the baseline intensity is a complex intensity mean value of the eddy current amplitude component and the eddy current phase component corresponding to the baseline window.

[0021] Further, the equivalent background representation of the background eddy current response of the jacket pipe and the interlayer interface is constructed based on the eddy current signal space sequence, including:

[0022] unfolding the spatial sequence of the eddy current signals into a sequence of monotonous positions according to the scanning path, and segmenting the sequence of monotonous positions according to a set background window length, while setting an overlapping section between adjacent background windows;

[0023] calculating a steady-state representative value and a signal fluctuation degree corresponding to the aggregated eddy current signals in each background window, and determining a background eddy current response candidate quantity of the jacket pipe and the interlayer interface and a steady-state score corresponding to the background eddy current response candidate quantity according to the steady-state representative value and the signal fluctuation degree;

[0024] retaining a corresponding background window as a background anchor point when the steady-state score exceeds a set threshold, and sorting the background anchor points according to the window center positions, so as to output the equivalent background representation through segmental continuity interpolation.

[0025] Further, the local anomaly analysis on the spatial sequence of the eddy current signals according to the equivalent background representation to identify the eddy current abnormal response component corresponding to the inner pipe defect comprises:

[0026] subtracting the spatial sequence of the eddy current signals through the equivalent background representation to obtain an abnormal residual signal dominated by local anomalies, and constructing a local analysis window for a detection position corresponding to the abnormal residual signal;

[0027] calculating an abnormal residual mean value and a residual fluctuation in the local analysis window, and calculating a local concentration score according to the abnormal residual mean value and the residual fluctuation, so as to filter out abnormal candidate points based on a comparison result of the local concentration score and a set minimum residual threshold.

[0028] Further, the local anomaly analysis on the spatial sequence of the eddy current signals according to the equivalent background representation to identify the eddy current abnormal response component corresponding to the inner pipe defect further comprises:

[0029] aggregating abnormal candidate points with a gap not exceeding a set gap into a same abnormal segment, and calculating a spatial width and an abnormal segment energy of the abnormal segment, so as to eliminate abnormal segments with a spatial width not lower than the background window length, the abnormal segment energy being a sum of square values of abnormal residuals in all positions in the abnormal segment;

[0030] screening the abnormal segments according to a preset separation strategy, and performing consistency verification on the screened abnormalities, so as to determine the eddy current abnormal response component corresponding to the inner pipe defect according to abnormal segments passing the consistency verification.

[0031] Further, the filtering of the eddy current abnormal response corresponding to the electromagnetic disturbance feature from the eddy current abnormal response component to reconstruct the inner pipe defect feature comprises:

[0032] extracting eddy current signals aggregated in the abnormal segment and position coordinates of the eddy current signals from the eddy current signal space sequence according to the start and end positions corresponding to the abnormal segment, and extracting a background eddy current response corresponding to the abnormal segment from the equivalent background representation, to construct an abnormal segment alignment data packet;

[0033] Based on the abnormal segment alignment data packet, the background eddy current response is normalized by the abnormal residual signal to construct a relative deviation ratio, and the difference of the relative deviation ratio corresponding to the equidistant positions on both sides of the center point of the eddy current peak in the abnormal segment is calculated respectively to obtain the internal and external consistency score corresponding to the abnormal segment.

[0034] The abnormal segment with the internal and external consistency score not lower than the set threshold is screened out, and the internal tube defect feature is reconstructed according to the maximum value of the relative deviation ratio in the abnormal segment and the relative deviation ratio exceeding the length of the eddy current peak segment.

[0035] Further, the internal tube defect feature is reconstructed according to the internal tube defect feature to judge whether the eddy current signal satisfies the internal tube defect response condition, if yes, the corresponding internal tube defect position and internal tube defect information are output, including:

[0036] The internal tube defect features with a distance not exceeding a set threshold in the jacket pipe detection coordinate system are merged into a same defect candidate unit, and the spatial concentration of the defect candidate unit is calculated to construct a defect comprehensive score and a defect level parameter corresponding to the defect candidate unit according to the spatial concentration;

[0037] When the defect comprehensive score reaches a set defect confirmation threshold, the defect candidate unit is determined as an internal tube defect, and the defect level corresponding to the internal tube defect is divided according to the defect level parameter, and a confidence label of the internal tube defect is introduced;

[0038] The internal tube defect position is the spatial center position of the defect candidate unit in the jacket pipe detection coordinate system, and the internal tube defect information is the defect level and the confidence label of the internal tube defect.

[0039] The second aspect of the present application discloses a pipe reactor eddy current detection signal processing device for realizing the pipe reactor eddy current detection signal processing method of any one of the first aspect, and the device comprises:

[0040] An eddy current space construction module is configured to acquire an eddy current signal on the side of the jacket pipe through an eddy current detection probe, and associate the eddy current signal with a detection position to construct an eddy current signal space sequence.

[0041] A background eddy current representation module is configured to construct an equivalent background representation of a background eddy current response of the jacket pipe and the interlayer interface based on the eddy current signal space sequence.

[0042] an eddy current anomaly recognition module configured to perform local anomaly analysis on the eddy current signal spatial sequence according to the equivalent background representation to identify an eddy current anomaly response component corresponding to the inner tube defect;

[0043] an inner tube defect reconstruction module configured to filter the eddy current anomaly response that meets the electromagnetic disturbance feature from the eddy current anomaly response component to reconstruct an inner tube defect feature;

[0044] an inner tube defect determination module configured to determine whether the eddy current signal meets an inner tube defect response condition according to the inner tube defect feature, and output the corresponding inner tube defect position and inner tube defect information if yes.

[0045] A third aspect of the present application discloses a terminal, comprising a processor and a storage medium.

[0046] The storage medium is configured to store instructions.

[0047] The processor is configured to operate according to the instructions to perform the steps of the method of the first aspect.

[0048] A fourth aspect of the present application discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of the first aspect.

[0049] Compared with the prior art, the present application has the following beneficial effects:

[0050] (1) In the present application, the eddy current detection probe is arranged on the outer surface of the jacket pipe in the running or shutdown detection state of the tubular reactor, and the eddy current detection signal is collected, and then the collected signal is synchronously associated with the actual detection position of the probe on the surface of the jacket pipe, to form an eddy current signal spatial sequence reflecting the comprehensive electromagnetic response of the jacket pipe, the interlayer interface and the inner tube. Subsequently, the continuous variation characteristics of the eddy current signal in a larger spatial scale are analyzed, the background eddy current response dominated by the jacket pipe and the interface is identified, and the corresponding equivalent background representation is constructed to describe the structural electromagnetic influence that cannot be avoided under the jacket pipe detection condition, to provide a reliable baseline for the abnormal separation of the inner tube defect, and to provide a basis for effectively distinguishing the signal change caused by the real crack defect from the eddy current distortion caused by the structural inherent electromagnetic boundary.

[0051] (2) The application analyzes the local abnormal change in the spatial sequence of the eddy current signal on the basis of the response characterization of the jacket pipe and the interface eddy current background, identifies the abnormal response component which still shows local concentration and spatial scale obviously smaller than the background change scale under the detection condition of the jacket pipe side, and regards the identified abnormal response as the eddy current disturbance possibly caused by the local defect of the inner pipe, so that the preliminary separation of the eddy current response related to the inner pipe defect is realized. Then, the spatial distribution characteristics, intensity change law and relative relationship with the jacket pipe background response of the eddy current response in the detection coordinate system of the jacket pipe are further analyzed, the abnormal response meeting the electromagnetic disturbance characteristics of the inner-to-outer transmission is retained, the pseudo abnormality caused by the surface state or uneven interlayer adhesion of the jacket pipe is suppressed, so that the eddy current signal characteristic description dominated by the real defect of the inner pipe is reconstructed, the signal change caused by the real crack defect can be effectively distinguished from the eddy current distortion caused by the structural inherent electromagnetic boundary, and the accuracy of the defect identification result is further improved. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.

[0053] Figure 1 is a flow schematic diagram of a pipe reactor eddy current detection signal processing method provided by the application.

[0054] Figure 2 is a structural schematic diagram of a pipe reactor eddy current detection signal processing device provided by the application.

[0055] Through the above drawings, the specific embodiments of the present application have been shown, and will be described in more detail hereinafter. These drawings and the written description are not intended to limit the scope of the inventive concept in any way, but to illustrate the inventive concept to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0056] The exemplary embodiments will be described in detail herein with reference to the attached drawings. The following description is made with reference to the accompanying drawings in which like reference numerals represent like elements, unless the context of use indicates otherwise. The following description of exemplary embodiments is not representative of all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0057] As shown in Figure 1 in one embodiment, a pipe reactor eddy current detection signal processing method comprises the following steps:

[0058] In step S110, the eddy current signal on the side of the jacket pipe is acquired by the eddy current detection probe, and the eddy current signal is associated with the detection position to construct an eddy current signal space sequence.

[0059] In some embodiments, the present application provides a pipe reactor eddy current detection signal processing method, and step S110 specifically includes the following steps:

[0060] In step S111, a jacket pipe detection coordinate system is established on the surface of the jacket pipe, and a scanning path is determined according to the jacket pipe detection coordinate system, and a sampling period of the eddy current signal is set to acquire a plurality of sampling points on the side of the jacket pipe.

[0061] In step S112, the eddy current detection probe is arranged on the surface of the jacket pipe based on the scanning path and the set sampling period and the plurality of sampling points to acquire an initial eddy current signal sequence with a time stamp.

[0062] In the initial eddy current signal sequence, the eddy current amplitude component and the eddy current phase component corresponding to each sampling time are included.

[0063] In some embodiments, the present application provides a pipe reactor eddy current detection signal processing method, and step S110 specifically further includes the following steps:

[0064] In step S113, a baseline window is selected from the initial eddy current signal sequence, and a baseline intensity corresponding to the baseline window is calculated to perform amplitude normalization processing on the initial eddy current signal sequence according to the baseline intensity to obtain a baseline-corrected eddy current signal sequence.

[0065] In step S114, the scanning path is discretized into a plurality of detection positions, the detection positions are aligned with the time stamps, and a plurality of sampling points at the same detection position are aggregated to construct an eddy current signal space sequence.

[0066] The baseline intensity is a complex intensity average of the eddy current amplitude component and the eddy current phase component corresponding to the baseline window.

[0067] In specific embodiments, the present application provides a pipe reactor eddy current detection signal processing method, and when the pipe reactor adopts a multi-layer composite pipe wall structure composed of an inner pipe alloy layer and a jacket pipe, the electromagnetic boundary mutation is introduced due to the non-metallurgical bonding interface between the layers, which causes the real defect response and the structure inherent eddy current distortion in the eddy current detection signal to be highly coupled and difficult to distinguish. By performing layered equivalent characterization, interface response separation and defect feature reconstruction on the eddy current detection signal, stable extraction and reliable discrimination of the real defect signal of the inner pipe layer are realized, including steps 1-5:

[0068] In step 1, the synchronous acquisition and spatial correlation construction of the jacket pipe side eddy current detection signal are performed.

[0069] In the tube reactor operation or shutdown detection state, the eddy current detection probe is arranged on the outer surface of the jacket pipe, and the eddy current detection signal is collected by scanning along the axial or circumferential direction of the pipe body. Then, the collected signal is synchronized and associated with the actual detection position of the probe on the surface of the jacket pipe, to form an eddy current signal space sequence reflecting the comprehensive electromagnetic response of the jacket pipe-interlayer interface-inner pipe, including the following sub-steps:

[0070] Sub-step 1.1, jacket pipe side eddy current detection working condition and scanning reference setting.

[0071] Specifically, a unified jacket pipe side detection coordinate system is established on the outer surface of the jacket pipe, the axial direction of the pipe body is taken as the axial reference, the jacket pipe circumferential development angle is taken as the circumferential reference, and the scanning path is determined as one of axial scanning or circumferential scanning or a combination of the two. For axial scanning, the scanning step is set to a fixed value and is limited within the achievable engineering range, with a value range of 0.5-5 mm, which is used to balance the spatial resolution of the defect response and the scanning efficiency; for circumferential scanning, the circumferential angle step is set, with a value range of 0.5-5 degrees. The sampling period is determined by the sampling frequency of the data acquisition hardware and the scanning speed, and multiple sampling points are required within any scanning step to ensure the formation of a continuous signal sequence. At the same time, in order to ensure that the detection results of different batches can be aligned, a physical repeatable positioning reference mark point is selected on the surface of the jacket pipe as the zero point before each detection starts, and the axial coordinate and circumferential angle of the reference mark are recorded as the global reference.

[0072] Sub-step 1.2, synchronization acquisition of jacket pipe side eddy current excitation and original signal.

[0073] Specifically, the eddy current detection probe is arranged on the outer surface of the jacket pipe, and the probe posture is kept consistent and the coupling state is kept stable during scanning. When the eddy current detection signal is synchronously collected, the original amplitude and phase at each sampling time are combined into an original sampling point, and a time stamp of the same time is attached to the original sampling point. In order to enable subsequent amplitude consistency processing and energy quantization of the signal, the original sampling point needs to be converted into a unified expression of amplitude scalar and phase scalar, and the complex intensity index of the sampling point is calculated, and finally the jacket pipe side original eddy current signal sequence carrying the time stamp is output.

[0074] The complex intensity index is used to unify the amplitude and phase on one engineering quantity, which is convenient for subsequent establishment of a space sequence, and the expression is: ; In the formula, is the complex intensity index at the time, and the unit is the normalized unit of the detection output; is the complex intensity index at the time, and the unit is the normalized unit of the detection output; ​The vortex amplitude component collected at the moment is derived from the amplitude channel output of the acquisition instrument; For The vortex phase component at the moment is derived from the phase channel output of the acquisition instrument; For the phase reference value, used to convert the phase difference into the relative change amount, the value is the average value of the phase in the starting section of the current detection; For the phase weight coefficient, used to adjust the contribution of the phase difference to the composite intensity, the value range is 0.1-10.

[0075] Substep 1.3, baseline correction and amplitude consistency processing of the sampling point on the jacket pipe side.

[0076] Specifically, due to the slight change of the coupling state and the difference of the jacket pipe surface state on the jacket pipe side, amplitude consistency processing is also needed to avoid overall drift in subsequent spatial correlation. First, select a stable area in the starting section of the detection as a baseline window, calculate the average value of the composite intensity index in the baseline window as the baseline intensity, and perform baseline deduction and amplitude normalization processing on the original vortex signal sequence to obtain the dimensionless standardized intensity sequence, and then output the vortex signal sequence after baseline correction, so that the signals of different scanning sections are under comparable amplitude benchmarks.

[0077] Substep 1.4, synchronization correlation of the jacket pipe position and the vortex signal and generation of the spatial sequence.

[0078] Specifically, the scanning process is discretized into a plurality of position points, each position point is assigned a spatial position number, and the position number is aligned with the sampling timestamp. For axial scanning, the axial position of the jacket pipe is obtained by accumulating the starting axial coordinate and the step distance; for circumferential scanning, the circumferential angle position is obtained by accumulating the circumferential zero point and the angle step distance. Then, the multiple sampling points in the same position number are aggregated to obtain the representative signal value of the position, and the aggregation method adopts the weighted combination of mean value and peak value, which can not only retain the peak response caused by local defects, but also avoid the complete dominance of incidental fluctuations.

[0079] Substep 1.5, spatial sequence integrity verification and output solidification.

[0080] Specifically, the spatial sequence is subjected to integrity verification, including whether the position points are continuous, whether there are missing position sections, and whether the aggregated signal has an abnormal saturated section. If a missing position section is found, the spatial number of the missing section is recalculated according to the scanning reference and marked as a to-be-supplemented scanning section; if an abnormal saturated section is found, it is marked as a coupling abnormality or a probe off-surface risk section for shielding processing in subsequent modeling separation. Finally, the spatial sequence is saved in a fixed field order, which includes the jacket pipe axial position, the jacket pipe circumferential angle position, the aggregated signal value and the quality mark.

[0081] Step S120, based on the vortex signal space sequence, the equivalent background characterization of the background vortex response of the jacket pipe and the interlayer interface is constructed.

[0082] In some embodiments, the present application provides a pipe reactor vortex detection signal processing method, and step S120 specifically comprises the following steps:

[0083] Step S121, the vortex signal space sequence is unfolded into a monotonous position sequence according to the scanning path, and the monotonous position sequence is segmented according to the set background window length, and an overlapping section is set between adjacent background windows.

[0084] Step S122, the steady-state representative value and the signal fluctuation degree corresponding to the aggregated vortex signal in each background window are calculated, and the background vortex response candidate quantity of the jacket pipe and the interlayer interface and the steady-state score corresponding to the background vortex response candidate quantity are determined according to the steady-state representative value and the signal fluctuation degree.

[0085] Step S123, when the steady-state score exceeds the set threshold value, the corresponding background window is retained as a background anchor point, and the background anchor points are sorted according to the window center positions, so as to output the equivalent background characterization through the segment continuity interpolation.

[0086] In a specific embodiment, the present application provides a pipe reactor vortex detection signal processing method, and step 2, the equivalent background characterization of the jacket pipe and the interlayer interface to the vortex response. For the vortex signal space sequence output by step 1, the continuous change characteristics of the signal in a larger space scale are analyzed, the stable response generated by the jacket pipe geometry, material properties and interlayer fitting state is recognized as the background vortex response dominated by the jacket pipe and the interface, and the corresponding equivalent background characterization is constructed to describe the inevitable structural electromagnetic influence under the jacket pipe side detection condition, including the following sub-steps:

[0087] Sub-step 2.1, large-scale continuity window construction of the jacket pipe side space sequence.

[0088] Specifically, the space sequence is unfolded into a monotonous position number sequence according to the scanning path, and the position number is segmented with a fixed background window length, and the background window length is recommended to be in the range of 20-200 position points, which is used to ensure that the space scale covered by the window is obviously larger than the space scale of the typical defect response, so that the dominant change in the window is derived from the jacket pipe geometry, material properties or interlayer fitting state. Then, in order to avoid that the window boundary cuts off the continuous structure response, an overlapping section needs to be set between adjacent windows, and the overlapping ratio is in the range of 10%-50%, and the center position number of each window is recorded for the definition and interpolation of the background curve.

[0089] Sub-step 2.2: Extraction of candidate background responses within the window and construction of steady-state criteria.

[0090] Specifically, for each background window, the steady-state representative value and fluctuation degree of the aggregated signal within the window are calculated. Based on the steady-state representative value and the fluctuation degree of the aggregated signal within the window, it is determined whether the background window is suitable as a candidate sample for the background response of the jacket tube and interface. The steady-state representative value is obtained by fusing the mean and median to reduce the interference of a few local anomalies within the window on the representative value; the fluctuation degree of the window is obtained by variance normalization to characterize whether the background is smooth. Then, a steady-state score is introduced to filter out background windows that obviously contain local anomalies, so that the background candidate quantity reflects only the jacket tube and interface effects as much as possible.

[0091] In this embodiment, the expression for the steady-state representative value is: ; In the formula, For the first The steady-state representative value of each background window is used as a background candidate value; For the first The average value of the aggregated signal within each background window; For the first The median of the aggregated signal values ​​within each background window; This is the mean weighting coefficient, with a value ranging from 0.4 to 0.8, used to balance the average trend and the ability to resist anomalies.

[0092] The expression for the degree of window fluctuation is: ; In the formula, For the first The normalized window fluctuation of each background window; For the first The variance of the aggregated signal values ​​within each background window; To stabilize the term and prevent the denominator from being too small, which could lead to an amplification effect, the value ranges from 0.001 to 0.1.

[0093] Finally, the window steady-state score can be determined based on the degree of window fluctuation, expressed as: ; In the formula, For the first The window stability score of each background window ranges from 0 to 1, with a larger value indicating a more stable state.

[0094] Sub-step 2.3: Construction and continuous output of the equivalent background curve of the jacketed tube and interface.

[0095] Specifically, background windows are selected based on steady-state scores. When a steady-state score exceeds a preset threshold, the background window is retained as a background anchor point. This preset threshold ranges from 0.6 to 0.9. Next, the retained background anchor points are sorted by their center position numbers within their background windows, and piecewise continuous interpolation is used to construct a background curve covering the entire scan path, ensuring that each jacket tube location receives a corresponding background response value. Furthermore, to prevent the background curve from excessively following local fluctuations, a maximum rate of change constraint is set on the variation amplitude of adjacent background anchor points during interpolation, ensuring slow and continuous background changes. The recommended maximum rate of change is within the range of 0.05-0.5 normalized units per 10 locations. Finally, the background curve is output as a combination of location numbers and background values, aligned with the spatial sequence output in step 1, forming an equivalent background representation that can be directly used.

[0096] Specifically, when the window stability score of a background window is greater than or equal to a set threshold, the background anchor point of the background window is marked with a value of 1; otherwise, it is marked with a value of 0.

[0097] Sub-step 2.4: Background characterization quality verification and referenceable output solidification.

[0098] Specifically, the background characterization undergoes quality verification, including three checks:

[0099] (1) Whether the background curve is spatially continuous and without breaks. If there are breaks, backtrack to sub-step 2.1 to expand the overlap window ratio and regenerate the anchor points;

[0100] (2) Whether the background curve shows unreasonable high-frequency jitter. If it fluctuates greatly repeatedly within a small number of adjacent positions, it indicates that the anchor point screening threshold is too low or the maximum rate of change constraint is too wide. It is necessary to increase the steady-state threshold or tighten the maximum rate of change.

[0101] (3) Whether the background curve is too flat, which makes it impossible to reflect the slow changes in the geometry or bonding state of the jacket tube. If the overall change is less than the expected minimum amplitude threshold, it means that the background window length is too large, resulting in loss of details, and the background window length needs to be reduced.

[0102] Finally, after the verification is passed, the background representation is solidified and output. The solidified fields include the location number, background response value, anchor mark, and verification mark.

[0103] Step S130: Perform local anomaly analysis on the spatial sequence of eddy current signals based on the equivalent background characterization to identify the eddy current anomaly response components corresponding to the inner tube defects.

[0104] In some embodiments, the eddy current detection signal processing method for a tubular reactor provided by the present invention includes the following steps in step S130:

[0105] Step S131: The spatial sequence of eddy current signals is canceled by equivalent background characterization to obtain the abnormal residual signal dominated by local anomalies, and a local analysis window is constructed for the detection position corresponding to the abnormal residual signal.

[0106] Step S132: Calculate the mean and fluctuation of the abnormal residuals within the local analysis window, and calculate the local concentration score based on the mean and fluctuation of the abnormal residuals. Then, select abnormal candidate points based on the comparison results between the local concentration score and the set minimum residual threshold.

[0107] In some embodiments, the eddy current detection signal processing method for a tubular reactor provided by the present invention further includes the following steps in step S130:

[0108] Step S133: Aggregate the abnormal candidate points with adjacent positions not exceeding the set gap into the same abnormal segment, and calculate the spatial width and energy of the abnormal segment to remove abnormal segments with a spatial width not less than the length of the background window. The energy of the abnormal segment is the sum of the squared values ​​of the abnormal residuals at all positions within the abnormal segment.

[0109] Step S134: The abnormal segments are screened according to the preset separation strategy, and the consistency of the screened abnormal judgments is checked, so as to determine the eddy current abnormal response component corresponding to the inner tube defect based on the abnormal segments that pass the consistency check.

[0110] In a specific embodiment, the present invention provides a method for processing eddy current detection signals in a tubular reactor. Step 3 involves separating the eddy current anomaly response to defects in the inner tube. Based on the characterization of the eddy current background response of the jacketed tube and interface obtained in step 2, local anomaly changes in the spatial sequence of the eddy current signal are analyzed. Anomaly response components that still exhibit local concentration and have a spatial scale significantly smaller than the background change scale under the detection conditions on the jacketed tube side are identified. These anomaly responses are considered to be eddy current disturbances possibly caused by local defects in the inner tube, thereby achieving preliminary separation of the eddy current response related to inner tube defects. This includes the following sub-steps:

[0111] Sub-step 3.1: Construction of the abnormal residual signal after background cancellation.

[0112] Specifically, in the scenario of detecting defects in the inner tube on the jacketed tube side, the eddy current response caused by the interface between the jacketed tube and the interlayer exhibits a slow and continuous change, while the disturbance caused by the inner tube defect usually manifests as a localized concentrated change. Therefore, it is necessary to use background characterization to cancel the original spatial sequence to obtain an anomalous residual signal dominated by local anomalies. In addition, to reduce the uneven impact of residuals being masked when the background amplitude is large and amplified when the background amplitude is small, the anomalous residuals are expressed using a fusion of difference and proportion, so that the anomalous residuals reflect both absolute and relative deviations.

[0113] The expression for the abnormal residual is: ; In the formula, For the first The abnormal residuals at each jacketed pipe location point are dimensionless quantities. For the first Aggregated signal values ​​at each jacket tube location point; For the first Background response values ​​at each jacketed pipe location point.

[0114] Sub-step 3.2, Local concentration measurement and screening of outlier candidate points.

[0115] Specifically, when defects in the inner tube are transmitted to the detection end of the jacketed tube, they typically manifest as a rapid increase and subsequent decrease in residuals over a short distance, i.e., local concentration. Therefore, a local analysis window needs to be constructed for each location point. The length of the analysis window should be significantly shorter than the aforementioned background window length, with a recommended range of 5-30 location points. For the residuals within the window, the degree of increase in the residual at that point relative to the average residual and window fluctuation is calculated to form a local concentration score. This score is then combined with a minimum residual threshold to screen for abnormal candidate points. Simultaneously, to avoid mistaking occasional single-point spikes for defects, the local concentration score and residual intensity of the candidate points must both reach a set threshold.

[0116] The expression for the local concentration score is as follows: ; In the formula, For the first Local concentration score of each jacketed pipe location point; For the first The average value of the absolute value of the residuals within the local window at the center of each jacket location point is used to characterize the local background level. It is the standard deviation of the absolute values ​​of the residuals within the same local window, used to characterize the scale of local fluctuations.

[0117] In this embodiment, during the screening of anomalous candidate points, if the absolute value of the anomalous residual is greater than or equal to a set residual intensity threshold (ranging from 0.2 to 3) and the local concentration score is greater than or equal to a set local concentration threshold (ranging from 1 to 5), then the anomalous candidate point is marked with a value of 1; otherwise, it is marked with a value of 0. The final output set of anomalous candidate points includes candidate location number, candidate residual intensity, and candidate local concentration score, which are used to subsequently form anomalous segments and determine whether their spatial scale is smaller than the background change scale.

[0118] Sub-step 3.3: Anomalous candidate points are aggregated into anomalous fragments and spatial scale is determined.

[0119] Specifically, the response to inner tube defects typically manifests as anomalies at several consecutive locations, rather than isolated single points. Therefore, candidate points with adjacent location numbers not exceeding a preset gap are aggregated into the same anomalous segment. This preset gap ranges from 1 to 3 location points to accommodate the continuity of the defect response under different scan steps. Then, for each anomalous segment, its spatial width and energy index are calculated to determine whether the corresponding anomalous segment belongs to anomalies whose spatial scale is significantly smaller than the background change scale. The spatial width is equal to the difference between the segment's start and end location numbers plus one; the energy index characterizes the overall perturbation intensity of the segment, calculated using the sum of squared residuals, emphasizing both peak value and spatial width. Subsequently, the segment width is compared with the background window length in step 2: if the segment width is much smaller than the background window length, the segment is more consistent with local perturbations caused by inner tube defects; if the segment width is close to or greater than the background scale, it is more likely a residual caused by slow changes in the jacketed tube or interface bonding state, and should be downweighted or eliminated.

[0120] Sub-step 3.4: Preliminary separation and output solidification of abnormal responses related to internal tube defects.

[0121] Specifically, a preliminary separation strategy based on scale priority, energy constraint, and morphological consistency is applied to the set of anomalous segments. First, segments are filtered by relative scale ratio, retaining those with a ratio less than a preset threshold (range 0.05-0.3) to ensure that the retained anomalous segments are significantly smaller than the background variation scale in spatial scale. Next, a minimum threshold is set for the energy index of the anomalous segments to avoid mistaking extremely small local fluctuations for defects. This minimum energy threshold ranges from 0.5-50 and is adjusted in conjunction with the scan step size and residual intensity threshold. Finally, a consistency check is performed on the morphology of the anomalous segments. The residuals within the anomalous segments are required to exhibit a single-peak pattern of rising, peaking, and falling near the peak point. Long plateaus or multiple repetitive peaks with large intervals are more likely to be complex responses caused by changes in interface adhesion and are marked as segments to be reviewed rather than directly output as defective segments. After separation, the retained anomalous segments and their position indices in the original spatial sequence are solidified and output to obtain a set of flow anomaly responses, including the start and end position numbers of the retained anomalous segments, the peak residual of the segments, the segment energy, the relative scale ratio of the segments, and the reference indices of the corresponding original aggregated signals and background signals. This facilitates subsequent fine-grained defect feature reconstruction by tracing back the original eddy current signals and equivalent background representations.

[0122] Step S140: Select eddy current abnormal responses that conform to the characteristics of electromagnetic disturbance from the eddy current abnormal response components in order to reconstruct the characteristics of the inner tube defect.

[0123] In some embodiments, the eddy current detection signal processing method for a tubular reactor provided by the present invention includes the following steps in step S140:

[0124] Step S141: Extract the aggregated eddy current signals and their position coordinates from the eddy current signal spatial sequence according to the start and end positions of the abnormal segments, and extract the background eddy current response corresponding to the abnormal segments from the equivalent background representation to construct an abnormal segment aligned data packet.

[0125] Step S142: Based on the abnormal segment alignment data packet, the background eddy current response is normalized by the abnormal residual signal to construct the relative deviation ratio, and the difference of the relative deviation ratio corresponding to the equidistant positions on both sides of the eddy current peak center point in the abnormal segment is calculated to obtain the inward and outward consistency score corresponding to the abnormal segment.

[0126] Step S143: Select abnormal segments with inward and outward consistency scores not lower than a set threshold, and reconstruct the inner tube defect features based on the maximum relative deviation ratio within the abnormal segments and the length of segments with relative deviation ratios exceeding the eddy current peak.

[0127] In a specific embodiment, the present invention provides a method for processing eddy current detection signals in a tubular reactor. Step 4 involves reconstructing the inner tube defect features based on the detection conditions on the jacketed tube side. For the eddy current anomaly response set obtained in step 3, the spatial distribution characteristics, intensity variation patterns, and relative relationship with the jacketed tube background response of the eddy current anomaly response in the jacketed tube detection coordinate system are further analyzed. Only anomaly responses conforming to the electromagnetic disturbance characteristics transmitted from the inside out are retained, while pseudo-anomalies caused by the surface condition of the jacketed tube or uneven interlayer bonding are suppressed. This reconstructs a description of eddy current signal features dominated by real inner tube defects, including the following sub-steps:

[0128] Sub-step 4.1: Align the abnormal fragment backtracking with the relative response on the jacket side.

[0129] Specifically, for each anomalous segment, the aggregated signal and corresponding position coordinates within the segment's coverage area are extracted back from the original spatial sequence according to its start and end position numbers. Simultaneously, the background response within the same range is extracted from the equivalent background representation, forming a point-by-point alignment sequence. Then, to facilitate comparison of the morphological consistency of different segments in the jacket-side detection coordinate system, the eddy current signal of each anomalous segment needs to be normalized and located using peak point alignment. During the normalization and location process, the point with the largest absolute value of the residual within the segment is determined as the peak center point. A fixed-length analysis interval is then expanded outwards from this point, with a recommended interval length of 5-50 position points to ensure complete coverage of the rise, peak, and fall processes. Finally, the output aligned data packet fields include the jacket position number, axial position, circumferential angle, aggregated signal, background response, anomalous residual, and peak center point number.

[0130] Sub-step 4.2: Pass the relative response consistency judgment from the inside out.

[0131] Specifically, when detecting defects in the inner tube on the jacketed tube side, the electromagnetic disturbance from the inner tube typically exhibits the following characteristics: the deviation from the background is strongest at the center of the segment and gradually decreases towards both sides, with a stable relative proportional relationship between the deviation and the background intensity. Changes in the surface condition of the jacketed tube (e.g., minor scratches on the outer surface, changes in oxide scale thickness) are more likely to cause local abrupt changes in the original polymerization signal, but the relative proportional relationship between this deviation and the background response is unstable and often accompanied by asymmetrical abrupt changes at the segment edges. Therefore, it is necessary to construct a relative deviation ratio sequence and evaluate its symmetry and attenuation within the segment. This relative deviation ratio uses the residual to renormalize the background amplitude, used to characterize the response intensity that can be compared across segments.

[0132] In the alignment coordinates of the abnormal segments, the relative deviation ratio difference between equidistant positions on both sides of the peak center point is calculated as the center of symmetry to form a symmetry loss. The smaller this symmetry loss, the more it conforms to the characteristics of propagation from the inside out and diffusion and attenuation to both sides. Finally, the formed symmetry loss is mapped to an inside-out consistency score to facilitate threshold determination and sorting. The inside-out consistency score sequence is output, including the inside-out consistency score and pass mark for each abnormal segment, to screen out false anomalies caused by changes in the surface condition of the jacket tube or uneven interlayer bonding.

[0133] Sub-step 4.3: False anomaly suppression and defect-dominant feature reconstruction.

[0134] Specifically, firstly, anomalies are filtered based on inside-out consistency scores, retaining those with scores no lower than a set threshold. This threshold ranges from 0.2 to 0.7, used to accommodate response differences under different jacket thicknesses and scanning step distances. Next, for the retained anomalies, their defect-dominant features are further engineered and reconstructed: the defect-dominant intensity is the maximum value of the relative deviation ratio within the segment; the defect-dominant width is the length of a continuous segment whose relative deviation ratio exceeds a certain proportion of the segment's peak value; and the defect-dominant energy is the sum of the squares of the relative deviation ratios, reflecting both intensity and range. Finally, to prevent pseudo-anomalies with extremely small residuals but accidentally good symmetry from passing the screening, dual constraints of intensity and energy lower limits are introduced. The intensity lower limit ranges from 0.2 to 3, and the energy lower limit ranges from 0.5 to 50. This results in the output of a reconstructed set of inner tube defect-dominant eddy current features, which outputs defect-dominant intensity, defect-dominant width, defect-dominant energy, and a defect confidence flag for each retained segment.

[0135] Sub-step 4.4: Solidify the feature set output and bind it to the traceable index.

[0136] Specifically, the features of each retained segment are fixedly output in a fixed field order and bound to a backtracking index, enabling subsequent steps to trace back from the feature set to the original aggregated signal and background response, and supporting engineering verification. Therefore, the final output fixed fields include the axial position of the jacket, the circumferential angular position of the jacket, the dominant defect intensity, the dominant defect width, the dominant defect energy, the inside-outside consistency score, the segment start and end position numbers, and the peak center point number. If the same spatial location is identified as a defect candidate in both the axial and circumferential scanning paths, the features of that spatial location are merged and marked to avoid duplicate counting.

[0137] Step S150: Determine whether the eddy current signal meets the inner tube defect response condition based on the inner tube defect characteristics. If so, output the corresponding inner tube defect location and inner tube defect information.

[0138] In some embodiments, the eddy current detection signal processing method for a tubular reactor provided by the present invention includes the following steps in step S150:

[0139] Step S151: Merge inner tube defect features with spacing not exceeding a set threshold within the jacketed tube detection coordinate system into the same defect candidate unit, and calculate the spatial concentration of the defect candidate unit to construct the defect comprehensive score and defect level parameters corresponding to the defect candidate unit based on the spatial concentration.

[0140] Step S152: When the comprehensive defect score reaches the set defect confirmation threshold, the defect candidate unit is determined to be an inner tube defect, and the defect level corresponding to the inner tube defect is divided according to the defect level parameter. At the same time, the confidence level label of the inner tube defect is introduced.

[0141] Among them, the inner tube defect location is the spatial center position of the defect candidate unit in the jacket tube detection coordinate system, and the inner tube defect information is the defect level and confidence level label of the inner tube defect.

[0142] In a specific embodiment, the present invention provides a method for processing eddy current detection signals in a tubular reactor. Step 5 involves determining and outputting inner tube defects under jacket-side detection conditions. Based on the eddy current signal feature set formed in step 4, the spatial concentration, stability, and relative intensity of abnormal features along the jacket-side detection path are comprehensively determined. When the inner tube defect response conditions are met, the corresponding inner tube defect location and defect severity information are output. This enables reliable detection of actual inner tube defects under conditions where eddy current detection can only be performed on the jacket-side. The method includes the following sub-steps:

[0143] Sub-step 5.1: Merge the defect candidate units of the feature set into spatial concentration calculations.

[0144] Specifically, firstly, candidates that are close together in the jacket tube coordinate system are merged into the same defect candidate unit. The merging radius ranges from 1 to 10 mm (axial distance) and 1 to 10 degrees (circumferential angular distance) to accommodate different scanning step distances and probe sizes. Then, for each candidate unit, its spatial concentration is calculated. This spatial concentration characterizes whether the defect response forms a stable, locally clustered aggregation along the jacket tube detection path. The spatial concentration is expressed as the ratio of the number of associated candidates to the diffusion radius; the more associated candidates and the smaller the diffusion radius, the higher the spatial concentration.

[0145] Sub-step 5.2: Construction of a joint score for stability and relative strength.

[0146] Specifically, under the jacketed tube side inspection conditions, a comprehensive score is constructed for each candidate unit. The score is jointly determined by spatial concentration, the dominant intensity of the largest defect within the candidate unit, and the average inside-out consistency score within the candidate unit. Simultaneously, a defect severity level parameter is constructed using a combination of defect dominant energy and intensity. Each weight in the combination is set as an adjustable constant to facilitate tuning under different jacketed tube thicknesses and inner tube materials. Then, the ratio of energy to peak intensity is used to reflect the extensibility of the defect response, distinguishing the performance differences between sharp, small defects and larger defects under the jacketed tube side inspection conditions, thus forming the defect severity level parameter.

[0147] Sub-step 5.3: Defect confirmation and judgment and result output.

[0148] Specifically, a defect confirmation threshold is set for each candidate unit. When the comprehensive defect score reaches the threshold, it is confirmed as an inner tube defect. The recommended threshold range is 1-20. Simultaneously, defect levels are classified according to a defect severity level parameter. For example, the defect level parameter falling into different intervals can be mapped to three levels: minor, moderate, and severe. The interval boundaries can be adjusted within the range of 0.1-10. The defect location is taken as the spatial center position of the corresponding candidate unit, and the axial and circumferential angular positions of the jacket tube are output as locatable coordinates on the jacket tube side. A confidence flag is included in the output, which is determined by the margin of the comprehensive defect score relative to the threshold, indicating the reliability of the output defect result.

[0149] In this embodiment, when the comprehensive defect score of a candidate unit is greater than or equal to the set defect confirmation threshold constant (the value range is 1-20), the candidate unit is marked as an inner tube defect with a value of 1, indicating that it is confirmed as an inner tube defect; otherwise, it is marked as a value of 0, indicating that it is not confirmed as an inner tube defect.

[0150] The present invention provides a tubular reactor eddy current detection signal processing device. The tubular reactor eddy current detection signal processing device described below and the tubular reactor eddy current detection signal processing method described above can be referred to in correspondence.

[0151] like Figure 2 As shown, in one embodiment, a tubular reactor eddy current detection signal processing device includes an eddy current space construction module, a background eddy current characterization module, an eddy current anomaly identification module, an inner tube defect reconstruction module, and an inner tube defect determination module.

[0152] The eddy current space construction module is used to acquire eddy current signals on the jacket side through the eddy current detection probe and associate the eddy current signals with the detection position to construct a spatial sequence of eddy current signals.

[0153] The background eddy current characterization module is used to construct an equivalent background characterization of the jacketed tube and interlayer interface response to background eddy current based on the spatial sequence of eddy current signals.

[0154] The eddy current anomaly identification module is used to perform local anomaly analysis on the spatial sequence of eddy current signals based on the equivalent background characterization, so as to identify the eddy current anomaly response components corresponding to the inner tube defects.

[0155] The inner tube defect reconstruction module is used to filter out eddy current abnormal responses that conform to electromagnetic disturbance characteristics from the eddy current abnormal response components in order to reconstruct the inner tube defect characteristics.

[0156] The inner tube defect determination module is used to determine whether the eddy current signal meets the inner tube defect response conditions based on the inner tube defect characteristics. If so, it outputs the corresponding inner tube defect location and inner tube defect information.

[0157] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0158] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0159] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0160] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0161] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0162] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0163] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method of processing a vortex detection signal of a tubular reactor, characterized in that, The method comprises: acquiring eddy current signals on the side of the jacket pipe through an eddy current detection probe, and associating the eddy current signals with detection positions to construct an eddy current signal space sequence; based on the eddy current signal space sequence, constructing an equivalent background representation of the background eddy current response of the jacket pipe and the interlayer interface; performing local anomaly analysis on the eddy current signal space sequence according to the equivalent background representation to identify an eddy current anomaly response component corresponding to the inner pipe defect; screening the eddy current anomaly response that meets the electromagnetic disturbance characteristics from the eddy current anomaly response component to reconstruct the inner pipe defect characteristics; judging whether the eddy current signals meet the inner pipe defect response condition according to the inner pipe defect characteristics, and if so, outputting the corresponding inner pipe defect position and inner pipe defect information.

2. The tubular reactor eddy current detection signal processing method of claim 1, wherein, The method comprises: establishing a jacket pipe detection coordinate system on the surface of the jacket pipe, and determining a scanning path according to the jacket pipe detection coordinate system, while setting a sampling period of the eddy current signals and a plurality of sampling points on the side of the jacket pipe; based on the scanning path and the set sampling period and a plurality of sampling points, setting the eddy current detection probe on the surface of the jacket pipe to acquire an initial eddy current signal sequence with time stamps; wherein the initial eddy current signal sequence contains eddy current amplitude components and eddy current phase components corresponding to each sampling time.

3. The tubular reactor eddy current detection signal processing method of claim 2, wherein, The method further comprises: selecting a baseline window from the initial eddy current signal sequence, and calculating a corresponding baseline intensity in the baseline window to perform amplitude normalization processing on the initial eddy current signal sequence according to the baseline intensity, to obtain a baseline-corrected eddy current signal sequence; discretizing the scanning path into a plurality of detection positions, aligning the detection positions with time stamps, and aggregating a plurality of sampling points at the same detection position to construct the eddy current signal space sequence; wherein the baseline intensity is a complex intensity mean value of the eddy current amplitude components and the eddy current phase components corresponding to the baseline window.

4. The tubular reactor eddy current detection signal processing method of claim 3, wherein, The method further comprises: expanding the eddy current signal space sequence into a monotonous position sequence according to the scanning path, and segmenting the monotonous position sequence according to a set background window length, while setting an overlap section between adjacent background windows; calculating a steady-state representative value and a signal fluctuation degree corresponding to the aggregated eddy current signals in each background window, and determining a background eddy current response candidate of the jacket pipe and the interlayer interface and a steady-state score corresponding to the background eddy current response candidate according to the steady-state representative value and the signal fluctuation degree; when the steady-state score exceeds a set threshold, retaining the corresponding background window as a background anchor point, and sorting the background anchor points according to the window center positions to output the equivalent background representation through segmented continuity interpolation.

5. The tubular reactor eddy current detection signal processing method of claim 1, wherein, The local anomaly analysis on the eddy current signal spatial sequence according to the equivalent background characterization is to identify the eddy current abnormal response component corresponding to the inner tube defect, and comprises: The eddy current signal spatial sequence is offset by the equivalent background characterization to obtain an abnormal residual signal dominated by local anomalies, and a local analysis window is constructed for a detection position corresponding to the abnormal residual signal; The abnormal residual mean value and residual fluctuation corresponding to the local analysis window are calculated, and a local concentration score is calculated according to the abnormal residual mean value and residual fluctuation, so as to screen out abnormal candidate points based on the comparison result of the local concentration score and a set minimum residual threshold.

6. The tubular reactor eddy current detection signal processing method of claim 5, wherein, The local anomaly analysis on the eddy current signal spatial sequence according to the equivalent background characterization is to identify the eddy current abnormal response component corresponding to the inner tube defect, and further comprises: Abnormal candidate points with a gap of no more than a set gap are aggregated into the same abnormal segment, and the spatial width and abnormal segment energy of the abnormal segment are calculated to eliminate abnormal segments with a spatial width lower than the background window length, and the abnormal segment energy is the sum of the square values of the abnormal residuals of all positions in the abnormal segment; The abnormal segments are screened according to a preset separation strategy, and consistency verification is performed on the screened abnormalities, so as to determine the eddy current abnormal response component corresponding to the inner tube defect according to the abnormal segments that pass the consistency verification.

7. The tubular reactor eddy current detection signal processing method of claim 6, wherein, The eddy current abnormal response corresponding to the electromagnetic disturbance feature is screened from the eddy current abnormal response component to reconstruct the inner tube defect feature, and comprises: The aggregated eddy current signals in the abnormal segment and the position coordinates of the eddy current signals are extracted from the eddy current signal spatial sequence according to the start and end positions of the abnormal segment, and the background eddy current response corresponding to the abnormal segment is extracted from the equivalent background characterization to construct an abnormal segment alignment data packet; Based on the abnormal segment alignment data packet, the background eddy current response is normalized by the abnormal residual signal to construct a relative deviation ratio, and the difference between the relative deviation ratios of the equidistant positions on both sides of the eddy current peak center point in the abnormal segment is calculated to obtain the inner-outer consistency score corresponding to the abnormal segment; Abnormal segments with an inner-outer consistency score not lower than a set threshold are screened, and the inner tube defect feature is reconstructed according to the maximum relative deviation ratio in the abnormal segment and the relative deviation ratio exceeding the length of the eddy current peak segment.

8. The tubular reactor eddy current detection signal processing method of claim 7, wherein, The eddy current signal is determined to meet the inner tube defect response condition according to the inner tube defect feature, and if so, the corresponding inner tube defect position and inner tube defect information are output, and comprises: Inner tube defect features with a distance of no more than a set threshold in the jacket tube detection coordinate system are merged into the same defect candidate unit, and the spatial concentration of the defect candidate unit is calculated to construct a defect comprehensive score and a defect level parameter corresponding to the defect candidate unit according to the spatial concentration; When the defect comprehensive score reaches a set defect confirmation threshold, the defect candidate unit is determined to be an inner tube defect, and the defect level corresponding to the inner tube defect is divided according to the defect level parameter, and a confidence label of the inner tube defect is introduced. The inner tube defect position is a spatial center position of the defect candidate unit in a jacket pipe detection coordinate system, and the inner tube defect information is a defect level and a confidence mark of the inner tube defect. 9.A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is configured to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-8.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the steps of the method according to any one of claims 1-8.

Images