Defect inversion method for thick-walled steel pipe pile foundation broadband acoustic tomography imaging

Abstract

The application discloses a thick-wall steel pipe pile foundation broadband acoustic tomography defect inversion method and relates to the technical field of engineering foundation structure detection. The method comprises the following steps: collecting a continuous echo time sequence formed by the propagation of broadband acoustic waves in a steel pipe pile, and synchronously extracting an energy concentration change mark at the end of the continuous echo time sequence; performing time density analysis on the continuous echo time sequence around the energy concentration change mark, expanding the continuous enhanced section in time sequence to form a surrounding propagation candidate section, and writing a circumferential residence position record in the corresponding position of the surrounding propagation candidate section. The application identifies the surrounding locking section and generates an adjustment starting point, actively breaks the coincidence relationship between the frequency change rhythm and the circumferential propagation rhythm, avoids energy concentration misjudgment from the source, simultaneously eliminates the circumferential repeated propagation interference through time dislocation, frequency offset and segmented energy release, and improves the reliability and accuracy of tomography and defect inversion.

Description

Technical Field

[0001] This invention relates to the field of engineering foundation structure testing technology, specifically to a broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations. Background Technology

[0002] Broadband acoustic tomography defect inversion for thick-walled steel pipe pile foundations refers to the process of deploying acoustic excitation and receiving devices on the outer surface or inside the pile body to emit broadband acoustic waves covering multiple frequency bands, allowing the sound waves to propagate through the steel pipe pile wall and its internal medium. Due to potential internal defects such as cracks, inclusions, corrosion thinning, voids, and incomplete penetration, different frequency components will exhibit varying response differences during propagation, including changes in propagation speed, energy attenuation, phase shift, and scattering anomalies. By collecting multi-path, multi-angle acoustic response data, an acoustic propagation parameter matrix is ​​constructed, and the spatial distribution of the propagation medium is reconstructed using tomography principles to obtain two-dimensional or three-dimensional images of the acoustic parameters inside the pile. Based on this, defect inversion calculations are then used to establish a correspondence between acoustic anomaly areas and the actual physical defect types, locations, sizes, and morphologies, thereby achieving quantitative identification and visual representation of internal defects in thick-walled steel pipe pile foundations.

[0003] The existing technology has the following shortcomings: In existing technologies, during broadband acoustic tomography inspection of thick-walled steel pipe pile foundations, when acoustic wave components in the transmitted signal, located in the mid-to-high frequency range and with wavelengths matching the steel pipe wall thickness, repeatedly reflect between the inner and outer walls of the pipe and propagate continuously along the circumferential direction, they easily form stable circumferential propagation paths locally. This causes the energy of this frequency component to continuously superimpose in a fixed area. The detection system will continuously receive echo signals with significantly enhanced amplitude. Since existing tomographic reconstruction methods mainly rely on echo intensity and attenuation characteristics for defect identification, they are prone to misjudging the energy enhancement areas caused by abnormal propagation paths as large-area structural defect areas, which can lead to the entire pile being deemed unusable.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations, comprising the following steps: The continuous echo time series formed by broadband sound waves propagating inside steel pipe piles was collected, and the energy concentration change marker was synchronously extracted at the end of the continuous echo time series to record the phenomenon of continuous local energy enhancement. Temporal density analysis was performed on the continuous echo time series around the energy concentration change markers. The continuous enhancement segments were unfolded in time sequence to form circumferential propagation candidate segments, and circumferential dwell position records were written at the corresponding positions of the circumferential propagation candidate segments. Based on the circumferential dwell position record, the propagation direction in the continuous echo time series is compared, and repeated time segments that repeatedly return along the same circumferential direction are extracted to form a circumferential locking segment. The locking start time is marked on the circumferential locking segment. Based on the locked start time, a retrospective analysis of the changes in the earlier part of the continuous echo time series is performed to locate the time interval where the frequency change rhythm coincides with the circumferential propagation rhythm, and an adjustment start point is generated at the corresponding position in the time interval. The timing of subsequent launches is gradually adjusted around the starting point. Unequal frequency offsets are introduced before each launch, and energy intensity is released in segments. The phenomenon of circumferential repetitive propagation is eliminated by continuously offsetting the timing.

[0007] Preferably, the steps for acquiring continuous echo time series and generating energy concentration change markers are as follows: Maintain continuous sampling, sample the echo signal point by point at fixed time intervals, assign a unique time coordinate to each sampling moment to form a continuous echo time sequence, and determine the last time coordinate of the continuous echo time sequence and establish a time reference point at the end of the transmission cycle. Based on the last time coordinate of the continuous echo time series, the amplitude values ​​of adjacent time coordinates are compared point by point forward to determine the amplitude increase points and merge them to form amplitude increase segments. When the amplitude increase segment connects with the last time coordinate of the continuous echo time series, an energy concentration change mark is generated and the start and end time coordinates are recorded. Around the energy concentration change marker, read the time coordinates and amplitude values ​​corresponding to the increasing amplitude segments to form an energy change time band, and record the duration and amplitude change range; The time range containing energy concentration change markers is segmented and encapsulated, maintaining the consistent order of the time coordinates, and an index association is established.

[0008] Preferably, the steps for performing time density analysis on continuous echo time series around energy concentration change markers are as follows: The analysis start and end time coordinates are determined by marking the time coordinates corresponding to the energy concentration change. The corresponding time points are read and the time intervals are counted to form a time interval sequence. At the same time, the number of amplitude increase points is accumulated to form an energy density statistical sequence. Based on the time interval sequence and energy density statistical sequence, continuously enhanced segments are identified, the start time coordinates and end time coordinates are recorded, and they are expanded in chronological order to form a set of continuously enhanced segments; By comparing the duration of continuous enhancement segments with the circumferential propagation reference time, continuous enhancement segments with consistent duration and time interval are combined to form candidate segments for circumferential propagation. Write circumferential dwell position records within the time range corresponding to the candidate propagation segments, and record the number of continuously enhanced segments and time interval data.

[0009] Preferably, the formation of the circumferential propagation candidate segment is based on the fact that the duration of the continuous enhancement segment is consistent with the circumferential propagation reference time and the time interval between adjacent continuous enhancement segments is consistent, and the start time coordinate, end time coordinate and number of continuous enhancement segments are recorded within the time range corresponding to the circumferential propagation candidate segment.

[0010] Preferably, the steps for comparing the propagation direction in a continuous echo time series based on the circumferential dwell position record are as follows: Extract time segments within the time range corresponding to the circumferential dwell position records, read the amplitude data within the time segments, and determine the propagation direction identifier based on the peak position and time coordinate order; Adjacent time segments are compared based on the propagation direction identifier, and time segments with the same propagation direction identifier and time interval corresponding to the circumferential propagation reference time are combined to form a repeating time segment sequence; By integrating the repeated time segment sequences, determining the start and end time coordinates, a surrounding locking segment is formed; The starting time coordinate of the first time segment in the repeating time segment sequence is used as the locking start time, and the locking start time is written to the corresponding time position around the locking segment.

[0011] Preferably, the propagation direction indicator is determined according to the arrangement order of the wave peak positions along the time coordinate within the time segment, the time interval between each time segment in the repeated time segment sequence corresponds to the circumferential propagation reference time, and the propagation direction indicator remains consistent within the surrounding locked section.

[0012] Preferably, the steps for retrospective analysis of the changes in the earlier part of the continuous echo time series based on the locked start time are as follows: Using the lock start time as the time reference point, read the continuous echo time series forward to form the time interval before the lock start time, extract the amplitude data point by point and record the time difference between the peak time coordinate and the adjacent peak time coordinate to form a frequency change rhythm sequence. The frequency variation rhythm sequence is compared with the circumferential propagation reference time. The time difference and the rhythm overlap time segment corresponding to the circumferential propagation reference time are extracted and the rhythm overlap segment set is formed in chronological order. Integrate the set of rhythmically overlapping segments and determine the start and end time coordinates to form the rhythmically overlapping time interval; The starting time coordinate of the time interval where the rhythms overlap is used as the adjustment starting point, and the adjustment starting point is written into the corresponding time position of the continuous echo time series.

[0013] Preferably, the steps for gradually adjusting the subsequent launch timing rhythm around the adjustment starting point are as follows: Using the time coordinates corresponding to the starting point as the initial reference, the original launch time intervals are rearranged, and a continuous staggered structure of launch time rhythm is formed by alternating the time offset. Under the continuous staggered structure of the launch time rhythm, the launch frequency is shifted at unequal intervals before each launch, and the frequency shift direction and the time shift direction are kept to change alternately. During the implementation of unequal interval frequency shift, the energy intensity of a single transmission process is released in segments, and the start time and duration of each energy release stage are adjusted to form a non-repetitive structure. The continuous echo time series corresponding to the adjustment starting point is continuously collected, and the circumferential repetitive propagation phenomenon is eliminated based on the changes in the repeated time segments in the continuous echo time series.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention, by performing temporal density expansion, propagation direction comparison, and lock-in start time backtracking on continuous echo time series, can identify the circumferential locking segment before the continuous superposition of circumferential energy enters the imaging reconstruction stage. By generating an adjustment start point to actively change the subsequent transmission time rhythm, it breaks the overlap between the frequency change rhythm and the circumferential propagation rhythm from the source, thereby avoiding the misjudgment of local energy concentration areas as large-scale structural anomaly areas, improving the authenticity and stability of defect identification results, and reducing the risk of misjudgment.

[0015] This invention achieves continuous offset trajectories in the time, frequency, and energy dimensions of the transmitted signal by gradually adjusting the transmission time rhythm, unequal frequency offsets, and segmented energy intensity at the starting point. This disrupts the rhythm conditions of the fixed propagation path, gradually dissipates the circumferential repetitive propagation phenomenon, and thus obtains a continuous echo time series free from the interference of surround locking. This provides a uniform propagation basis for subsequent tomographic reconstruction and defect inversion, improving the reliability of locating and determining the size of internal defects in thick-walled steel pipe pile foundations. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of the broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations according to the present invention. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0019] This invention provides, for example Figure 1 The broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations shown includes the following steps: The continuous echo time series formed by broadband sound waves propagating inside steel pipe piles was collected, and the energy concentration change marker was synchronously extracted at the end of the continuous echo time series to record the phenomenon of continuous local energy enhancement. To ensure that the acquisition of continuous echo time series and the generation of energy concentration change markers have a clear temporal positioning basis and a complete energy change recording process, the following steps are followed: After the broadband acoustic wave transmission is completed, continuous sampling is maintained. Starting from the initial moment when the broadband acoustic wave enters the steel pipe pile and propagates, the echo signal is sampled point by point at fixed time intervals. The echo amplitude corresponding to each sampling moment is arranged in chronological order to form a continuous echo time sequence covering the entire propagation cycle. During continuous sampling, each sampling moment is assigned a unique time coordinate, so that each amplitude data in the continuous echo time sequence corresponds to a unique time position. When the current transmission cycle ends, the time coordinate corresponding to the last sampling moment is determined as the last time coordinate of the continuous echo time sequence, and a time reference point is established at the last time coordinate of the continuous echo time sequence to keep the last time coordinate of the continuous echo time sequence synchronized with the end time of the transmission cycle. After the time reference point is established, all amplitude data from the starting time coordinate to the last time coordinate of the continuous echo time sequence are kept in a continuous storage state, so that the continuous echo time sequence forms a complete time axis structure, providing a time continuity basis for the subsequent generation of energy concentration change markers.

[0020] After establishing a time reference point at the last time coordinate of the continuous echo time series, the amplitude changes in the continuous echo time series are scanned point by point backward from the last time coordinate. The amplitude difference between two adjacent time coordinates in the continuous echo time series is compared point by point. When the amplitude value corresponding to the later time coordinate is greater than the amplitude value corresponding to the earlier time coordinate, the time coordinate is marked as an amplitude increase point. When multiple consecutive time coordinates are marked as amplitude increase points, these consecutive time coordinates are merged into an amplitude increase segment. When the end time coordinate of the amplitude increase segment is the same as or adjacent to the last time coordinate of the continuous echo time series, the amplitude increase segment is determined as an energy increase segment directly connected to the last time coordinate of the continuous echo time series. After determining the energy increase segment, an energy concentration change marker is generated at the last time coordinate of the continuous echo time series, and the energy concentration change marker is associated with the start and end time coordinates of the energy increase segment, so that the energy concentration change marker contains complete time range information.

[0021] After generating energy concentration change markers and completing time range association records, the continuous echo time series is processed by expanding the energy distribution using the energy concentration change markers as the central time nodes. Specifically, the end time coordinate corresponding to the energy concentration change marker is used as the starting point, and the time coordinates and corresponding amplitude values ​​of the energy-increasing segments in the continuous echo time series are read point by point backward, forming energy change time bands in chronological order. Within each energy change time band, the amplitude values ​​corresponding to each time coordinate are continuously arranged to maintain the continuity of the energy change time band along the time axis while preserving the original order of amplitude change trends. After the energy change time band is formed, the time coordinates of the energy concentration change markers are bound to the end time coordinates within the energy change time band, and the duration and amplitude change range of the energy change time band are recorded in the energy concentration change markers. This ensures that the energy concentration change markers not only indicate a single time point but also encompass the complete time interval and amplitude change range of the local energy enhancement phenomenon.

[0022] After binding the energy concentration change markers to the energy change time bands, the continuous echo time series containing the energy concentration change markers is segmented and encapsulated. Specifically, all time data from the start time coordinate of the energy change time band to the end time coordinate of the continuous echo time series is divided into independent time periods, and the arrangement order of the time coordinates within each independent time period is kept consistent with the original arrangement order of the continuous echo time series. Within each independent time period, the energy concentration change markers are locked to their corresponding time coordinate positions, so that the positions of the energy concentration change markers in the continuous echo time series maintain a fixed mapping relationship. After location locking is completed, an index is established between the independent time period containing the energy concentration change marker and the original continuous echo time series, so that the subsequent time density analysis around the energy concentration change marker can be directly located to the corresponding time interval. Through the above continuous sampling, incremental segment identification, time band expansion and segment encapsulation process, the continuous echo time series formed by broadband sound waves propagating inside the steel pipe pile is generated and recorded with energy concentration change markers at the same time as the acquisition is completed. This provides a continuous time basis and energy change basis for subsequent circumferential dwell position recording and surrounding locked segment identification based on the energy concentration change markers.

[0023] Temporal density analysis was performed on the continuous echo time series around the energy concentration change markers. The continuous enhancement segments were unfolded in time sequence to form circumferential propagation candidate segments, and circumferential dwell position records were written at the corresponding positions of the circumferential propagation candidate segments. After completing the time binding of energy concentration change markers with continuous echo time series, time density analysis can be carried out around the energy concentration change markers, gradually forming candidate propagation segments and corresponding circumferential dwell position records. The specific implementation process is as follows: Using the time coordinates corresponding to the energy concentration change markers as the central time reference point, an analysis start time coordinate extending forward and a fixed analysis end time coordinate are determined in the continuous echo time series. The analysis end time coordinate is the time coordinate corresponding to the energy concentration change markers, and the analysis start time coordinate is obtained by moving forward point by point along the time axis until the amplitude change in the continuous echo time series changes from an increasing state to a non-increasing state. After determining the analysis start time coordinates, all time points between the analysis start time coordinates and the analysis end time coordinates are read one by one, forming a continuous time segment set in chronological order. Within the continuous time segment set, the time intervals of each time segment are statistically analyzed segment by segment, and the time difference between adjacent time segments is recorded as time interval data, which are then arranged in chronological order to form a time interval sequence. After forming the time interval sequence, the number of amplitude increase points occurring within a unit time length is accumulated segment by segment to form an energy density statistical sequence that corresponds one-to-one with the time interval sequence. This makes the analysis interval present a correspondence between the time interval distribution and the energy density distribution along the time axis, thereby establishing a time density basic structure around the energy concentration change markers.

[0024] After the time interval sequence and energy density statistical sequence are formed, the numerical changes in the energy density statistical sequence are examined segment by segment along the time axis, starting from the analysis start time coordinate. When the energy density values ​​in multiple consecutive time segments remain in an increasing state, the range of the consecutive time segments is defined as a continuous enhancement segment, and the start and end time coordinates of the continuous enhancement segment are recorded. When the energy density values ​​in subsequent time segments are in a non-increasing state, the previous continuous enhancement segment is closed, and the search continues to find the next combination of time segments that meets the continuous increasing condition. Through the above segment-by-segment identification process, multiple continuous enhancement segments are formed within the analysis interval. After multiple continuous enhancement segments are formed, they are sorted and arranged in ascending order according to the start time coordinates of each continuous enhancement segment, so that all continuous enhancement segments form a sequential unfolding structure in the time axis direction. Inside the sequential unfolding structure, the original time boundaries of each continuous enhancement segment are kept unchanged, and the sorting result is kept consistent with the original time order of the continuous echo time series, thus forming a set of continuous enhancement segments unfolded in time order.

[0025] After the set of continuous enhancement segments is formed, the time length of each continuous enhancement segment is compared with the time period corresponding to a single circumferential propagation in the continuous echo time series. Specifically, the time length corresponding to a complete circumferential propagation in the continuous echo time series is selected as the circumferential propagation reference time, and this circumferential propagation reference time is compared with the time length of each continuous enhancement segment one by one. When the time lengths of multiple continuous enhancement segments are consistent with the circumferential propagation reference time, and the time interval between adjacent continuous enhancement segments is also consistent with the circumferential propagation reference time, these continuous enhancement segments that meet the conditions are combined into an overall time band in chronological order, and this overall time band is defined as a circumferential propagation candidate segment. After the circumferential propagation candidate segment is formed, the start time coordinate, end time coordinate, and the number of continuous enhancement segments contained within the circumferential propagation candidate segment are recorded, so that the circumferential propagation candidate segment has a clear boundary and internal structure on the time axis.

[0026] After the candidate propagation segments are determined, circumferential dwell position records are written within the time range corresponding to the candidate propagation segments in the continuous echo time series. Specifically, a circumferential dwell start record is established at the start time coordinate of the candidate propagation segment, and a circumferential dwell end record is established at the end time coordinate of the candidate propagation segment. Within the candidate propagation segment, the time position number of each continuous enhancement segment is recorded sequentially according to the time order of the continuous enhancement segments. The circumferential dwell position records explicitly record the number of continuous enhancement segments included in the candidate propagation segment, the start and end time coordinates of each continuous enhancement segment, and the time interval data between adjacent continuous enhancement segments, ensuring the circumferential dwell positions are accurate. The recording fully reflects the temporal distribution characteristics of the circumferential propagation candidate segments in the continuous echo time series. After completing the writing of the circumferential dwell position record, a one-to-one correspondence is maintained between the circumferential dwell position record and the continuous echo time series. This allows for direct location of the corresponding time position of the circumferential propagation candidate segment when comparing propagation directions and extracting repeated time segments. Through the above-mentioned time density statistics, continuous enhancement segment identification, time period comparison combination, and circumferential dwell position record writing process, the continuous echo time series around the energy concentration change marker is used to form circumferential propagation candidate segments with circumferential propagation characteristics, providing clear temporal and positional basis for subsequent determination of the circumferential locking segment.

[0027] Based on the circumferential dwell position record, the propagation direction in the continuous echo time series is compared, and repeated time segments that repeatedly return along the same circumferential direction are extracted to form a circumferential locking segment. The locking start time is marked on the circumferential locking segment. After establishing a time correspondence between the circumferential dwell position records and the continuous echo time series, the propagation direction in the continuous echo time series can be compared segment by segment. Furthermore, repeating time segments that repeatedly return along the same circumferential direction can be extracted, thereby forming a circumferential locking segment and marking the locking start time. The specific implementation process is as follows: Based on each time range marked in the circumferential dwell position record, the original time segments within the corresponding time range are extracted one by one from the continuous echo time series, while maintaining the arrangement order of the time coordinates within the time segments consistent with the original order of the continuous echo time series. After extraction, the amplitude change process within each time segment is fully unfolded, and the amplitude data is read point by point from the start time to the end time according to the time coordinates. The peak positions, trough positions and their corresponding time coordinates formed during the time progression are recorded. After the amplitude unfolding is completed, the propagation direction identifier is determined according to the correspondence between the time sequence of peak appearance and the direction of time progression. When the peak positions appear sequentially from front to back along the time axis, it is defined as the clockwise propagation direction identifier. When the peak positions correspond to a backward trend from back to front along the time axis, it is defined as the counterclockwise propagation direction identifier. The propagation direction identifier is written into the record entry of the corresponding time segment, so that each time segment carries clear propagation direction information.

[0028] After propagation direction identifiers are generated for all time segments, the propagation direction identifiers between adjacent time segments are compared one by one in order from early to late according to the time coordinates. When the propagation direction identifier of the current time segment is consistent with that of the previous time segment, the time interval between the two time segments is further calculated and compared with the reference time for single circumferential propagation. When the time interval is consistent with the reference time for single circumferential propagation, the current time segment and the previous time segment are grouped into the same repeating time segment sequence, and the time segment arrangement order is recorded within the repeating time segment sequence. If the propagation direction identifiers are consistent but the time interval is inconsistent with the reference time for single circumferential propagation, the extension of the current repeating time segment sequence is stopped, and a new comparison process is started from the current time segment. Through the above segment-by-segment comparison and time interval correspondence process, the time segments that repeatedly return along the same circumference in the continuous echo time series are extracted one by one and multiple repeating time segment sequences are formed.

[0029] After forming multiple repeating time segment sequences, each repeating time segment sequence is integrated as a whole. Specifically, the start time coordinate of the first time segment in the repeating time segment sequence is recorded as the sequence start time, the end time coordinate of the last time segment in the repeating time segment sequence is recorded as the sequence end time, and the entire time range from the sequence start time to the sequence end time is defined as a continuous time segment. Within this continuous time segment, the propagation direction indicators of all time segments are checked one by one in chronological order to confirm that the propagation direction indicators remain consistent throughout the continuous time segment, and to confirm that the time interval values ​​between adjacent time segments maintain a correspondence with the reference time of a single circumferential propagation throughout the continuous time segment. When the above conditions are continuously met throughout the continuous time segment, the continuous time segment is determined as the orbital locking segment, and the time range of the orbital locking segment in the continuous echo time sequence is recorded independently.

[0030] After determining the surrounding locking segment, the locking start time is marked on the surrounding locking segment. Specifically, the starting time coordinate of the first repeated time segment within the surrounding locking segment is determined as the locking start time, and the locking start time is written into the time record entry of the surrounding locking segment. At the same time, a time mapping relationship is established between the locking start time and the corresponding time range in the circumferential dwell position record, so that the locking start time can be traced back to the initial time position that formed the surrounding propagation candidate segment. After completing the marking of the locking start time, the overall time range of the surrounding locking segment, the number of repeated time segments, the propagation direction identifier, and the locking start time are uniformly bound to the corresponding time position of the continuous echo time series, so that the continuous echo time series forms a surrounding locking segment record structure with consistent propagation direction and periodic return characteristics in the time axis direction. Through the above process of determining the propagation direction identifier, comparing segment by segment, constructing the repeated time segment sequence, integrating the continuous time segment, and marking the locking start time, the accurate time positioning of the phenomenon of repeated return along the same circumferential direction is achieved, and a complete time basis is provided for subsequent frequency change rhythm backtracking analysis around the locking start time.

[0031] Based on the locked start time, a retrospective analysis of the changes in the earlier part of the continuous echo time series is performed to locate the time interval where the frequency change rhythm coincides with the circumferential propagation rhythm, and an adjustment start point is generated at the corresponding position in the time interval. After the locking segment and the locking start time have been clearly written into the continuous echo time series, the changes in the preceding segment of the continuous echo time series can be traced back segment by segment around the locking start time. This allows for the precise location of the time interval where the frequency change rhythm coincides with the circumferential propagation rhythm, and the generation of the adjustment start point at the corresponding position. The specific implementation process is as follows: Using the locked start time as the time reference point for backtracking analysis, all time data before the locked start time is read point by point along the time axis in the continuous echo time series until the starting time coordinate of the continuous echo time series is reached, thus forming the time interval before the locked start time. Within the time interval before the locked start time, the amplitude data corresponding to each time coordinate is read point by point in chronological order, and the oscillation period of the amplitude data in the time axis direction is recorded segment by segment. Specifically, when the amplitude data changes from an upward trend to a downward trend, a peak time coordinate is determined, and when the amplitude data changes from a downward trend to an upward trend, a trough time coordinate is determined. The time difference between the peak time coordinate and the adjacent peak time coordinate is recorded in chronological order, thus forming a complete frequency change rhythm sequence. Through the above point-by-point reading and time difference recording, the frequency change rhythm in the time interval before the locked start time forms a continuous arrangement structure in the time axis direction.

[0032] After forming the frequency variation rhythm sequence, the frequency variation rhythm sequence is compared segment by segment with the circumferential propagation rhythm already recorded in the surrounding locked section. Specifically, each time difference value in the frequency variation rhythm sequence is read sequentially backward from the locking start time, and each time difference value is compared with the circumferential propagation reference time. When a time difference value is consistent with the circumferential propagation reference time in numerical value, the start time coordinate and end time coordinate corresponding to the time difference value are recorded as a rhythm overlap time segment. When multiple adjacent time differences are consistent with the circumferential propagation reference time, the time segments corresponding to these adjacent time differences are merged into a continuous rhythm overlap segment in chronological order. Through the above segment-by-segment comparison and continuous merging process, the correspondence between the frequency variation rhythm and the circumferential propagation rhythm in the time axis direction is clearly defined.

[0033] After the rhythmic overlap segments are formed, boundary integration processing is performed on them. Specifically, the starting time coordinate corresponding to the first time difference of each consecutive rhythmic overlap segment is determined as the starting boundary of the segment, and the ending time coordinate corresponding to the last time difference of the consecutive rhythmic overlap segment is determined as the ending boundary of the segment. The entire time range between the starting boundary and the ending boundary is defined as the complete rhythmic overlap time interval. Within the complete rhythmic overlap time interval, the correspondence between the frequency variation rhythm sequence and the circumferential propagation reference time remains unchanged, and the time position of the complete rhythmic overlap time interval in the continuous echo time sequence is recorded independently. Through the above boundary integration processing, the time interval where the frequency variation rhythm and the circumferential propagation rhythm overlap form a clear time boundary in the continuous echo time sequence.

[0034] After determining the complete rhythm overlap time interval, an adjustment starting point is generated at the corresponding position within that time interval. Specifically, the starting boundary time coordinate of the complete rhythm overlap time interval is determined as the adjustment starting point, and an adjustment starting point marker is written at the corresponding time position in the continuous echo time series. Simultaneously, a time correlation is established between the adjustment starting point and the locking start time, ensuring that the adjustment starting point is located before the locking start time and directly corresponds to the state of overlap between the frequency change rhythm and the circumferential propagation rhythm. After marking the adjustment starting point, the time coordinate of the adjustment starting point, the corresponding frequency change rhythm information, and the corresponding circumferential propagation rhythm information are uniformly recorded in the continuous echo time series, making the adjustment starting point the time reference benchmark for the subsequent gradual misalignment adjustment of the transmission time rhythm. Through the above-mentioned backtracking reading, rhythm comparison, boundary integration, and adjustment starting point generation process, accurate analysis of the changes before the locking start time is achieved, and the location of the time interval where the frequency change rhythm and the circumferential propagation rhythm overlap and the generation of the adjustment starting point are completed.

[0035] The timing of subsequent launches is gradually staggered around the starting point. Unequal interval frequency shifts are introduced before each launch, and energy intensity is released in segments. The phenomenon of circumferential repetitive propagation is eliminated by continuously shifting the timing. After the adjustment starting point has been written into the continuous echo time series and established a time correspondence with the locked start time, the subsequent transmission time rhythm can be gradually staggered around the adjustment starting point. Simultaneously, unequal interval frequency offsets are introduced before each transmission, and segmented energy intensity is released, thereby forming a continuous offset structure in the time, frequency, and energy dimensions to eliminate the circumferential repetitive propagation phenomenon. The specific implementation process is as follows: Using the time coordinate corresponding to the adjustment starting point as the starting reference for replanning the launch timing rhythm, the original launch time intervals are rearranged in the continuous launch cycles after the adjustment starting point. Specifically, the fixed launch time interval value used in the continuous launch cycles before the adjustment starting point is read and used as the reference time interval. In the first launch after the adjustment starting point, the launch start time is shifted forward by a first time offset based on the reference time interval. In the second launch, the launch start time is shifted backward by a second time offset based on the first launch time. In the third launch, the launch start time is shifted forward by a third time offset based on the second launch time, so that the time interval value between adjacent launches alternates around the reference time interval. During continuous launch, each time offset is different from the previous time offset value, and the direction of the time offset alternates during continuous launch, so that the subsequent launch timing rhythm forms a continuous misaligned trajectory in the time axis direction, thereby breaking away from the original time structure that maintains a fixed correspondence with the circumferential propagation rhythm.

[0036] After gradually adjusting the timing of the launch, an unequal frequency offset is applied to the launch frequency before each launch. Specifically, in the first launch after the adjustment start point, the original launch frequency is increased by a first frequency offset; in the second launch, the frequency is decreased by a second frequency offset; and in the third launch, a third frequency offset is added to the frequency of the second launch. This ensures that the frequency value changes with each launch, and there is no fixed difference between the frequency offset values. During the frequency offset, the direction of the frequency offset is alternated with the direction of the launch time offset. That is, when the launch time is shifted forward, the frequency shifts upward, and when the launch time is shifted backward, the frequency shifts downward, creating an alternating structure between the time offset and the frequency offset in the continuous launch cycle. Through this unequal frequency offset process, the launched signal forms a continuous trajectory in the frequency dimension, thereby breaking the overlap between the original frequency change rhythm and the circumferential propagation rhythm.

[0037] While implementing gradual staggered adjustment of the launch timing and unequal interval frequency offset, the energy intensity of each launch process is segmented for release. Specifically, a single launch process is divided into three continuous energy release phases. The first energy release phase outputs the first energy intensity, the second energy release phase outputs the second energy intensity, and the third energy release phase outputs the third energy intensity, so that the energy intensity in a single launch process exhibits a segmented change structure over time. During continuous launches, the start time of the first energy release phase of each launch is rearranged according to the time offset, so that the start time of the same energy release phase changes in different launch cycles. At the same time, the duration of the second and third energy release phases alternates in different launch cycles, so that the energy release forms a non-repetitive structure in the time axis direction. Through the above segmented energy intensity release processing, the launched signal forms a multi-dimensional continuous staggered relationship with the time and frequency offsets in the energy dimension.

[0038] After completing the gradual misalignment adjustment of the launch time rhythm, unequal interval frequency shift, and segmented energy release intensity, continuous echo time series after the adjustment start point are continuously collected, and the new continuous echo time series is compared with the surrounding locked segment formed before the adjustment start point. When the new continuous echo time series no longer contains repeating time segments arranged according to the single circumferential propagation reference time period, it indicates that the original circumferential repetitive propagation phenomenon has been resolved. Throughout the resolution process, the launch time offset, frequency offset, and energy release phase start time are kept to change gradually, so that subsequent launch cycles form a continuous offset structure in the time, frequency, and energy dimensions. Through the above-mentioned gradual misalignment adjustment process around the adjustment start point, the launch time rhythm, which originally had a fixed overlap with the circumferential propagation rhythm, gradually deviates from the original rhythm path, thereby resolving the circumferential repetitive propagation phenomenon and providing a continuous echo time series basis unaffected by surrounding locking for subsequent defect inversion.

[0039] This invention, by performing temporal density expansion, propagation direction comparison, and lock-in start time backtracking on continuous echo time series, can identify the circumferential locking segment before the continuous superposition of circumferential energy enters the imaging reconstruction stage. By generating an adjustment start point to actively change the subsequent transmission time rhythm, it breaks the overlap between the frequency change rhythm and the circumferential propagation rhythm from the source, thereby avoiding the misjudgment of local energy concentration areas as large-scale structural anomaly areas, improving the authenticity and stability of defect identification results, and reducing the risk of misjudgment.

[0040] This invention achieves continuous offset trajectories in the time, frequency, and energy dimensions of the transmitted signal by gradually adjusting the transmission time rhythm, unequal frequency offsets, and segmented energy intensity at the starting point. This disrupts the rhythm conditions of the fixed propagation path, gradually dissipates the circumferential repetitive propagation phenomenon, and thus obtains a continuous echo time series free from the interference of surround locking. This provides a uniform propagation basis for subsequent tomographic reconstruction and defect inversion, improving the reliability of locating and determining the size of internal defects in thick-walled steel pipe pile foundations.

[0041] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for defect inversion of thick-walled steel pipe pile foundation broadband acoustic tomography, characterized in that, Includes the following steps: The continuous echo time series formed by broadband sound waves propagating inside the steel pipe pile was collected, and the energy concentration change marker was synchronously extracted at the end of the continuous echo time series. Temporal density analysis was performed on the continuous echo time series around the energy concentration change markers. The continuous enhancement segments were unfolded in time sequence to form circumferential propagation candidate segments, and circumferential dwell position records were written at the corresponding positions of the circumferential propagation candidate segments. Based on the circumferential dwell position record, the propagation direction in the continuous echo time series is compared, and repeated time segments that repeatedly return along the same circumferential direction are extracted to form a circumferential locking segment. The locking start time is marked on the circumferential locking segment. Based on the locked start time, a retrospective analysis of the changes in the earlier part of the continuous echo time series is performed to locate the time interval where the frequency change rhythm coincides with the circumferential propagation rhythm, and an adjustment start point is generated at the corresponding position in the time interval. The timing of subsequent launches is gradually staggered around the starting point. Unequal interval frequency shifts are introduced before each launch, and energy intensity is released in segments. The phenomenon of circumferential repetitive propagation is eliminated by continuously shifting the timing. The steps for gradually staggering the subsequent launch timing around the adjustment starting point are as follows: Using the time coordinates corresponding to the starting point as the initial reference, the original launch time intervals are rearranged, and a continuous staggered launch time rhythm structure is formed by alternating the time offset. Under the continuous staggered structure of the launch time rhythm, the launch frequency is shifted at unequal intervals before each launch, and the frequency shift direction and the time shift direction are kept to change alternately. During the implementation of unequal interval frequency shift, the energy intensity of a single transmission process is released in segments, and the start time and duration of each energy release stage are adjusted to form a non-repetitive structure. The continuous echo time series corresponding to the adjustment starting point is continuously collected, and the circumferential repetitive propagation phenomenon is eliminated based on the changes in the repeated time segments in the continuous echo time series.

2. The broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations according to claim 1, characterized in that, The steps for acquiring continuous echo time series and generating energy concentration change markers are as follows: Maintain continuous sampling, sample the echo signal point by point at fixed time intervals, assign a unique time coordinate to each sampling moment to form a continuous echo time sequence, and determine the last time coordinate of the continuous echo time sequence and establish a time reference point at the end of the transmission cycle. Based on the last time coordinate of the continuous echo time series, the amplitude values ​​of adjacent time coordinates are compared point by point forward to determine the amplitude increase points and merge them to form amplitude increase segments. When the amplitude increase segment connects with the last time coordinate of the continuous echo time series, an energy concentration change mark is generated and the start and end time coordinates are recorded. Around the energy concentration change marker, read the time coordinates and amplitude values ​​corresponding to the amplitude increasing segment to form an energy change time band, and record the duration and amplitude change range; The time range containing energy concentration change markers is segmented and encapsulated, maintaining the consistent order of the time coordinates, and an index association is established.

3. The broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations according to claim 2, characterized in that, The steps for performing time density analysis on continuous echo time series based on energy concentration variation markers are as follows: The analysis start and end time coordinates are determined by marking the time coordinates corresponding to the energy concentration change. The corresponding time points are read and the time intervals are counted to form a time interval sequence. At the same time, the number of amplitude increase points is accumulated to form an energy density statistical sequence. Based on the time interval sequence and energy density statistical sequence, continuously enhanced segments are identified, the start time coordinates and end time coordinates are recorded, and they are expanded in chronological order to form a set of continuously enhanced segments; By comparing the duration of continuous enhancement segments with the circumferential propagation reference time, continuous enhancement segments with consistent duration and time interval are combined to form candidate segments for circumferential propagation. Write circumferential dwell position records within the time range corresponding to the candidate segments of the circumferential propagation, and record the number of continuously enhanced segments and the time interval data.

4. The broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations according to claim 3, characterized in that, The formation of the circumferential propagation candidate segment is based on the consistency of the continuous enhancement segment time length with the circumferential propagation reference time and the consistency of the time interval between adjacent continuous enhancement segments. The start time coordinate, end time coordinate and number of continuous enhancement segments are recorded within the time range corresponding to the circumferential propagation candidate segment.

5. The broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations according to claim 3, characterized in that, The steps for comparing the propagation direction in a continuous echo time series based on circumferential dwell position records are as follows: Extract time segments within the time range corresponding to the circumferential dwell position records, read the amplitude data within the time segments, and determine the propagation direction identifier based on the peak position and time coordinate order; Adjacent time segments are compared based on the propagation direction identifier, and time segments with the same propagation direction identifier and time interval corresponding to the circumferential propagation reference time are combined to form a repeating time segment sequence; By integrating the repeated time segment sequences, determining the start and end time coordinates, a surrounding locking segment is formed; The starting time coordinate of the first time segment in the repeating time segment sequence is used as the locking start time, and the locking start time is written to the corresponding time position around the locking segment.

6. The broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations according to claim 5, characterized in that, The propagation direction indicator is determined according to the order of the wave peak positions along the time coordinate within the time segment. The time interval between each time segment in the repeating time segment sequence corresponds to the circumferential propagation reference time, and the propagation direction indicator remains consistent within the surrounding locked section.

7. The broadband acoustic tomography defect inversion method for thick-walled steel pipe pile foundations according to claim 5, characterized in that, The steps for retrospective analysis of the early stages of continuous echo time series changes based on the locked start time are as follows: Using the lock start time as the time reference point, read the continuous echo time series forward to form the time interval before the lock start time, extract the amplitude data point by point and record the time difference between the peak time coordinate and the adjacent peak time coordinate to form a frequency change rhythm sequence. The frequency variation rhythm sequence is compared with the circumferential propagation reference time. The time difference and the rhythm overlap time segment corresponding to the circumferential propagation reference time are extracted and the rhythm overlap segment set is formed in chronological order. Integrate the set of rhythmically overlapping segments and determine the start and end time coordinates to form the rhythmically overlapping time interval; The starting time coordinate of the time interval where the rhythms overlap is used as the adjustment starting point, and the adjustment starting point is written into the corresponding time position of the continuous echo time series.

Images