Phased array composite material ultrasonic flaw detector control method and system

Abstract

The invention provides a phased array composite material ultrasonic flaw detector control method and system, and relates to the technical field of ultrasonic detection, and the technical scheme is characterized in that multi-point distance data between a phased array probe and a detected curved surface of a composite material component is acquired; calculating local geometric parameters of a measured curved surface below the phased array probe based on the multi-point distance data; the actual transit time of ultrasonic echoes from a characteristic reflector in the composite material component is obtained; according to the calculated local geometric parameters and the obtained actual transit time of the ultrasonic echoes, the equivalent propagation sound velocity for representing propagation of the ultrasonic waves in the composite material component is determined; and calculating excitation delay time for each array element of the phased array probe according to the calculated local geometric parameters and the determined equivalent propagation sound velocity. The phased array composite material ultrasonic flaw detector control method and the phased array composite material ultrasonic flaw detector control system provided by the invention have the advantages of improving the detection sensitivity and the accuracy of defect evaluation.

Description

Technical Field

[0001] This application relates to the field of ultrasonic testing technology, and more specifically, to a control method and system for a phased array composite material ultrasonic flaw detector. Background Technology

[0002] Composite materials, with their superior physical properties such as high specific strength, high specific modulus, corrosion resistance, and high designability, have become the preferred material for manufacturing load-bearing structural components with complex shapes, and are widely used in aerospace, wind power generation, rail transportation, and other fields. These structural components often have significant, irregular curved shapes, such as aircraft fuselage sections, wings, turbine engine blades, wind turbine blades, and high-pressure hydrogen storage tanks. Effective non-destructive testing of these complex components, especially the accurate assessment of potential internal manufacturing defects (such as porosity, inclusions, delamination, and debonding) or in-service damage (such as impact damage and fatigue cracks), is a key technical aspect in ensuring their structural integrity and service safety.

[0003] Phased array ultrasonic testing technology, due to its unique capabilities in electronic beam deflection, scanning, and focusing, has become a widely used technique in such inspection tasks. By precisely controlling the excitation time and relative phase of multiple piezoelectric elements within the probe, a phased array system can synthesize an ultrasonic beam with a specific propagation direction and focal length within the material. When inspecting composite material plates with flat surfaces, assuming the sound velocity in the material is known and uniformly distributed, the ultrasonic beam can be accurately focused to a predetermined depth within the material using established geometric acoustic principles and preset delay rules, thereby achieving high detection sensitivity and defect resolution.

[0004] However, the detection process becomes exceptionally complex when the object of inspection changes from a planar surface to a curved composite material with intricate variations. First, when the probe couples with the curved workpiece, the actual acoustic path from different array elements to the material surface and the refraction angle of the ultrasonic waves after entering the material change significantly due to the curvature. This differs considerably from the ideal situation based on the planar assumption. If the focusing rules designed for planar workpieces are still applied, the actual focal point of the ultrasonic beam within the material will deviate from the preset position, and the focal spot size will become diffuse, resulting in blurred ultrasonic imaging and potentially leading to missed detections of critical defects or misjudgments of their dimensions.

[0005] To address the adverse effects of curved surface geometry, some phased array flaw detection systems have introduced compensation mechanisms based on surface geometric parameters. Operators need to input approximate geometric information about the surface of the workpiece before inspection, such as the local radius of curvature. The system adjusts the delay rule based on this input geometric information to correct the propagation path of the sound beam. However, in practical applications, many composite material components do not have regular spherical or cylindrical surfaces, but rather complex and continuously varying free-form surfaces, such as the wing-body blending region of an aircraft or the torsional transition section of a large blade. In these areas, a single radius of curvature parameter cannot accurately describe the local geometry beneath the probe. If the inspection system can only handle simplified surface models, or if the geometric parameters input by the operator deviate from the actual local geometry, the effectiveness of the sound beam focusing compensation will be significantly reduced. Especially when the probe continuously scans along complex surfaces, the local curvature beneath the probe may change with each new position. If the compensation parameters cannot be precisely adjusted accordingly, the focusing accuracy will continuously fluctuate.

[0006] This problem is further exacerbated by the acoustic properties of the composite material itself. The sound velocity within a composite material is not always uniform and constant. Factors such as the material's layup design, variations in fiber orientation, local differences in resin content, porosity distribution, and inconsistent local curing levels can all contribute to spatial non-uniformity of sound velocity. This non-uniformity further disrupts the propagation path of the sound beam within the material, making it difficult to achieve ideal focusing effects using methods that rely solely on external geometry for compensation. When the sound beam passes through regions of varying sound velocity, both its propagation direction and speed change, similar to light propagating through media with different refractive indices, further intensifying the shift and dispersion of the focal point.

[0007] In applications requiring high-precision quantitative assessment of defects, such as evaluating the initiation and propagation of microcracks or accurately measuring the size and depth of delamination defects, the requirements for beam focusing accuracy are even more stringent. Poor focusing not only reduces the amplitude of the defect echo signal but also distorts its contour in the ultrasonic image, directly affecting the accuracy of subsequent structural integrity assessments and remaining life predictions.

[0008] Furthermore, when it is necessary to inspect regions at different depths within a material, phased array systems typically adjust the focusing depth dynamically. If the compensation control method cannot accurately and in real-time handle the combined effects of surface curvature and potential sound velocity inhomogeneities on different focusing depths, then the advantages of dynamic focusing cannot be fully realized. For example, when inspecting a thick-walled, curved pressure vessel, it is necessary to focus on both minute defects near the surface and internal defects in deeper regions; the sound beam needs to maintain good focusing across a wide depth range.

[0009] When a probe continuously scans the surface of a curved composite material to obtain C-scan or B-scan images, the relative position and orientation of the probe to the material surface are constantly changing. Even for a regularly defined curved surface, the local normal direction and curvature parameters of the contact point of the probe at different scanning positions may differ. This means that the ideal focusing delay law needs to be dynamically updated according to the real-time position and orientation of the probe. If the response speed of the compensation control is not fast enough, or the update accuracy is insufficient, the focusing quality will be inconsistent during the scanning process, resulting in alternating areas of clear focus and blurry focus in the final detection imaging results, which seriously affects the reliability and consistency of the detection results. For example, when automatically scanning areas with drastic curvature changes, such as the leading or trailing edge of a large wind turbine blade, the probe moves along a predetermined trajectory. If the beam focusing accuracy cannot be accurately compensated in real time with the changes in the curved surface, detection blind spots or signal distortion may occur in areas with large curvature changes.

[0010] Therefore, in specific scenarios involving phased array ultrasonic scanning of composite material components with complex, continuously varying curvature surfaces, when considering the potential local non-uniformity of sound velocity within the composite material, the real-time requirements for adjusting focusing parameters due to rapid changes in surface geometry, the limitations on the accuracy of the compensation model due to the unknown distribution of actual sound velocity within the material, and the coupling effect of surface effects and sound velocity non-uniformity on the sound beam propagation path, making it difficult to achieve optimal results with separate compensations, and when dynamically adjusting the focusing depth during the scanning process requires the compensation method to calculate and apply focusing rules that adapt to the current probe position, attitude, target depth, and potential sound velocity changes in real time, existing technologies struggle to provide an effective focusing compensation method that can comprehensively, in real time, and adaptively solve the aforementioned problems.

[0011] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0012] The purpose of this application is to provide a control method and system for a phased array composite material ultrasonic flaw detector, which has the advantages of improving detection sensitivity and the accuracy of defect assessment.

[0013] Firstly, this application provides a control method for a phased array composite material ultrasonic flaw detector, the technical solution of which is as follows:

[0014] include:

[0015] Acquire multi-point distance data between the phased array probe and the measured surface of the composite material component;

[0016] Based on multi-point distance data, the local geometric parameters of the measured surface below the phased array probe are calculated.

[0017] Obtain the actual transit time of ultrasonic echoes from characteristic reflectors inside composite material components;

[0018] Based on the calculated local geometric parameters and the actual transit time of the ultrasonic echo, the equivalent propagation velocity of ultrasonic waves in composite material components is determined through iterative correction.

[0019] Based on the calculated local geometric parameters and the determined equivalent propagation velocity, the excitation delay time for each element of the phased array probe is calculated to compensate for the focusing of the phased array ultrasonic beam within the composite material component.

[0020] Furthermore, this application also proposes a step of determining the equivalent propagation velocity of ultrasound in composite material components through iterative correction, based on the calculated local geometric parameters and the obtained actual transit time of the ultrasonic echo:

[0021] The actual transit time of the acquired ultrasonic echo is smoothed to obtain the target transit time;

[0022] Based on the calculated local geometric parameters and target transit time, the equivalent propagation speed of ultrasonic waves in composite material components is determined through iterative correction. When the iterative calculation of the iterative correction reaches the preset resource constraint, the equivalent propagation speed based on the calculation of that iteration is output.

[0023] The equivalent propagation speed of sound output by iterative correction is subjected to variation amplitude limitation processing to obtain the final equivalent propagation speed of sound.

[0024] Furthermore, this application also proposes a step for smoothing the actual transit time of the acquired ultrasonic echo to obtain the target transit time, including:

[0025] The actual transit time series of the acquired ultrasonic echoes is detected to identify transit time step changes in the transit time series and obtain the identified transit time step changes.

[0026] Based on the identified step changes in transit time, the actual transit time sequence of the acquired ultrasonic echo is divided into step change segments corresponding to step changes in transit time and non-step change segments corresponding to non-step changes.

[0027] Smoothing is performed on the transit time data of the non-step change section to obtain smoothed transit time data of the non-step change section.

[0028] The target transit time is generated by combining the transit time data of the smoothed non-step change segment with the transit time data of the step change segment.

[0029] Furthermore, this application also proposes a step for determining the equivalent propagation velocity characterizing ultrasonic waves in composite material components through iterative correction based on the calculated local geometric parameters and the target transit time, including:

[0030] Acquire the ultrasonic echo signals and corresponding actual transit times of multiple characteristic reflectors inside the composite material component;

[0031] For each acquired ultrasonic echo signal, the signal-to-noise ratio (SNR) of the ultrasonic echo signal is evaluated, and the evaluated SNR is obtained.

[0032] For each acquired ultrasonic echo signal, the sensitivity index of the transit time of the ultrasonic echo signal to the change in the equivalent propagation speed of sound is determined, and the determined sensitivity index is obtained.

[0033] Based on the obtained signal-to-noise ratio and the determined sensitivity index, at least one characteristic reflector is selected from multiple characteristic reflectors to obtain the selected characteristic reflector.

[0034] The actual transit time of the selected characteristic reflector is used as the target transit time for iterative correction. Iterative correction is performed based on the solved local geometric parameters and the target transit time to determine the equivalent propagation speed characterizing the ultrasonic wave in the composite material component.

[0035] Furthermore, this application also proposes a step for determining the sensitivity index of the transit time of each acquired ultrasonic echo signal to changes in the equivalent propagation speed of sound, including:

[0036] From the ultrasonic echo signal corresponding to each acquired ultrasonic echo signal, an acoustic characteristic change indicator parameter is extracted to characterize the acoustic characteristic change of the ultrasonic propagation path of the ultrasonic echo signal.

[0037] Based on the geometric information of the ultrasonic propagation path corresponding to each acquired ultrasonic echo signal, the initial sensitivity index corresponding to the ultrasonic echo signal is calculated.

[0038] Based on the extracted acoustic characteristic change indicator parameters and the calculated initial sensitivity index, the initial sensitivity index is adjusted to determine the sensitivity index of the transit time of the ultrasonic echo signal to the change in the equivalent propagation speed.

[0039] Furthermore, this application also proposes a step of extracting acoustic characteristic change indication parameters from the ultrasonic echo signals corresponding to each acquired ultrasonic echo signal, for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal, including:

[0040] For each acquired ultrasonic echo signal, a segmentation process is performed on the ultrasonic echo signal to identify noise-affected segments, signal morphology abnormal segments, and echo overlap segments within the ultrasonic echo signal, thereby obtaining the identification results of the segmentation process.

[0041] Based on the identification results of the segment division process, at least one sub-echo segment whose signal features satisfy the preset criteria is selected from the ultrasonic echo signal to obtain at least one selected sub-echo segment;

[0042] For at least one selected sub-echo segment, preliminary acoustic characteristic parameters are extracted from each selected sub-echo segment to obtain preliminary acoustic characteristic parameters corresponding to each selected sub-echo segment.

[0043] The preliminary acoustic characteristic parameters corresponding to each selected sub-echo segment are extracted from at least one selected sub-echo segment and fused to obtain the data fusion result. Consistency assessment and outlier removal are then performed on the data fusion result to obtain acoustic characteristic change indication parameters for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal.

[0044] Furthermore, this application proposes a step of fusing preliminary acoustic characteristic parameters extracted from at least one selected sub-echo segment, corresponding to each selected sub-echo segment, to obtain the data fusion result, and performing consistency evaluation and outlier removal on the data fusion result to obtain acoustic characteristic change indication parameters for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal. The steps include:

[0045] Receive preliminary acoustic characteristic parameters corresponding to each selected sub-echo segment, extracted from at least one selected sub-echo segment;

[0046] The preliminary acoustic characteristic parameters are fused using a fusion algorithm with computational complexity that meets preset conditions to obtain the fusion result.

[0047] A consistency assessment is performed on the results of data fusion, using a consistency assessment method whose computational complexity meets preset conditions;

[0048] Outlier removal is performed on the data fusion results using an outlier removal method that meets preset computational complexity requirements.

[0049] Monitor computing resource usage or processing time during data fusion, consistency assessment, and outlier removal.

[0050] When the preset resource limit is reached, the current processing result is output as an acoustic characteristic change indication parameter to characterize the acoustic characteristic change of the ultrasonic propagation path of the ultrasonic echo signal.

[0051] Furthermore, this application proposes that the steps for performing a consistency assessment on the results of data fusion include:

[0052] Analyze the variation characteristics of the resulting sequence of data fusion;

[0053] Based on the characteristics of change, identify the segments in the result sequence of data fusion that meet the preset rapid change criteria;

[0054] For segments in the data fusion result sequence that do not meet the preset rapid change criteria, assess their data fluctuations.

[0055] Based on the relationship between data fluctuations and preset noise thresholds, it is determined whether the results of data fusion are consistent.

[0056] Furthermore, this application also proposes a method for calculating the local geometric parameters of the measured surface below the phased array probe based on multi-point distance data, including:

[0057] Receive distance data from multiple points;

[0058] Perform data processing on the multi-point distance data to reduce noise and remove outliers, resulting in processed multi-point distance data;

[0059] The processed multi-point distance data is analyzed to identify the local geometric features of the measured surface, and the identified local geometric features are obtained.

[0060] Based on the identified local geometric features, a computational strategy for solving local geometric parameters is selected, and the selected computational strategy is obtained.

[0061] Based on the selected calculation strategy and the processed multi-point distance data, the local geometric parameters of the measured surface below the phased array probe are calculated.

[0062] Furthermore, this application also proposes a control system for a phased array composite material ultrasonic flaw detector, the system comprising:

[0063] The distance acquisition module is used to acquire multi-point distance data between the phased array probe and the measured surface of the composite material component;

[0064] The geometry calculation module is used to calculate the local geometric parameters of the measured surface below the phased array probe based on multi-point distance data.

[0065] The echo acquisition module is used to acquire the actual transit time of ultrasonic echoes from characteristic reflectors inside composite material components;

[0066] The sound velocity determination module is used to determine the equivalent propagation sound velocity, which characterizes the propagation of ultrasonic waves in composite material components, based on the calculated local geometric parameters and the actual transit time of the ultrasonic echo through iterative correction.

[0067] The delay calculation module is used to calculate the excitation delay time for each element of the phased array probe based on the solved local geometric parameters and the determined equivalent propagation speed, in order to compensate for the focusing of the phased array ultrasonic beam in the composite material component.

[0068] As can be seen from the above, the phased array composite material ultrasonic flaw detector control method and system provided in this application adaptively acquires the local geometric parameters of the tested surface and determines the equivalent propagation speed, thereby accurately calculating the excitation delay time and improving the focusing accuracy of the ultrasonic beam inside the complex curved composite material. It has the advantages of improving detection sensitivity and the accuracy of defect assessment. Attached Figure Description

[0069] Figure 1 A flowchart illustrating a control method for a phased array composite material ultrasonic flaw detector provided in this application.

[0070] Figure 2 This is a schematic diagram of the control system of a phased array composite material ultrasonic flaw detector provided in this application.

[0071] In the diagram: 1. Distance acquisition module; 2. Geometry calculation module; 3. Echo acquisition module; 4. Sound speed determination module; 5. Delay calculation module. Detailed Implementation

[0072] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0073] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0074] When performing phased array ultrasonic scanning inspection on composite material components with complex and continuously varying curvature surfaces, it is difficult to adaptively and in real-time compensate for the beam focusing accuracy of the phased array ultrasonic flaw detector, taking into account the potential local non-uniformity of sound velocity within the composite material, the real-time requirements for adjusting focusing parameters due to rapid changes in surface geometry, the limitations of the accuracy of the compensation model due to the unknown distribution of actual sound velocity within the material, and the coupling effect of surface effects and sound velocity non-uniformity on the sound beam propagation path. This is to ensure that a stable and accurate focusing effect can be obtained throughout the scanning path and at different detection depths.

[0075] For example, suppose we are performing an automated phased array ultrasonic scan on a composite material skin of an aircraft wing with a complex freeform surface. The probe moves along a preset path, with the local curvature and normal direction constantly changing beneath it. Simultaneously, the sound velocity varies in different regions within the skin due to factors such as layup design and curing process. The phased array system needs to calculate the excitation delay time of each element in real time to ensure precise focusing of the ultrasonic beam at a predetermined depth within the material. If calculations are based solely on a simplified surface model or a preset average sound velocity, the actual beam path will deviate from the expected path, leading to focus position shift and focal spot dispersion. During the scan, this deviation dynamically changes with the probe position and the acoustic properties within the material, resulting in unstable focusing quality.

[0076] In this regard, refer to Figure 1 This application proposes a control method for a phased array composite material ultrasonic flaw detector, comprising:

[0077] S110. Obtain multi-point distance data between the phased array probe and the measured surface of the composite material component;

[0078] S120. Based on multi-point distance data, calculate the local geometric parameters of the measured surface below the phased array probe;

[0079] S130. Obtain the actual transit time of the ultrasonic echo from the internal characteristic reflector of the composite material component;

[0080] S140. Based on the calculated local geometric parameters and the actual transit time of the ultrasonic echo, the equivalent propagation velocity of ultrasonic waves in composite material components is determined through iterative correction.

[0081] S150. Based on the calculated local geometric parameters and the determined equivalent propagation velocity, calculate the excitation delay time for each element of the phased array probe to compensate for the focusing of the phased array ultrasonic beam within the composite material component.

[0082] Among them, acquiring multi-point distance data between the phased array probe and the measured surface of the composite material component refers to collecting distance information between multiple spatial points between the phased array probe and the measured surface. This can be achieved using optical measurement technology, mechanical measurement technology, or ultrasonic measurement technology, such as using laser sensors, structured light scanners, or contact sensors on robotic arms.

[0083] Among them, calculating the local geometric parameters of the measured surface below the phased array probe based on multi-point distance data refers to calculating the parameters describing the surface characteristics below the current position of the probe based on the collected multi-point distance data. This can be achieved by using surface fitting algorithms, local curvature calculation methods, or normal vector calculation methods. For example, the least squares method can be used to fit a polynomial surface or calculate the local principal curvature.

[0084] Among them, the actual transit time of the ultrasonic echo from the characteristic reflector inside the composite material component refers to the time it takes for the received ultrasonic wave to be emitted from the probe, propagate through the inside of the material, encounter the characteristic reflector, and return to the probe. It can be determined by detecting the time difference of a specific characteristic point of the ultrasonic echo signal relative to the emission time, such as detecting the peak time or zero-crossing time of the echo envelope.

[0085] Specifically, based on the calculated local geometric parameters and the actual transit time of the ultrasonic echo, the equivalent propagation velocity used to characterize the propagation of ultrasonic waves in composite material components is determined through iterative correction. This refers to calculating a parameter that describes the average propagation velocity of ultrasonic waves along the path based on known geometric path information and measured ultrasonic propagation time. This can be achieved by using an optimization algorithm based on minimizing the error between the calculated transit time and the actual transit time, or by a parameter adjustment process based on a preset model and measured data.

[0086] The calculation of the excitation delay time for each element of the phased array probe based on the calculated local geometric parameters and the determined equivalent propagation speed, in order to compensate for the focusing of the phased array ultrasonic beam in the composite material component, refers to determining an ultrasonic emission delay relative to a reference time for each element of the phased array probe. This delay time can be calculated based on the calculated local geometric parameters and the determined equivalent propagation speed using acoustic propagation principles, such as using the ray tracing method or Huygens' principle.

[0087] The core innovation of this application lies in combining the real-time acquired local surface geometric parameters with the equivalent propagation velocity determined based on the measured echo of the internal characteristic reflector, and adaptively determining the equivalent velocity using an iterative correction method. This allows for comprehensive compensation of the influence of complex surface geometry and the non-uniformity of sound velocity within the material on the focusing of the ultrasonic beam, achieving the effect of adjusting the focusing law in real time and accurately during the scanning process and maintaining a stable focusing effect.

[0088] Specifically, first, the system acquires multi-point distance data of the measured surface below the probe, reflecting the spatial relationship between the probe and the surface. Next, based on this multi-point distance data, geometric parameters describing the local surface morphology, such as curvature and normal direction, are calculated. Simultaneously, the system receives ultrasonic echo signals from characteristic reflectors within the material and extracts the actual transit times. These actual transit times contain geometric and acoustic information about the ultrasonic wave's propagation path within the material. Then, using the calculated local geometric parameters and the acquired actual transit times, an equivalent propagation velocity is determined through an iterative correction process. This equivalent velocity is a parameter that comprehensively reflects the current local geometry and the material's acoustic properties. Finally, based on the calculated local geometric parameters and the determined equivalent propagation velocity, the excitation delay time of each element of the phased array probe is calculated. These delay times are applied to the probe elements, enabling the emitted ultrasonic beam to be precisely focused at a predetermined position after passing through the surface and the material's interior. The entire process forms a closed loop, allowing the focusing law to be adjusted in real-time according to the probe position, surface changes, and the material's acoustic properties.

[0089] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0090] A laser rangefinder array integrated on a phased array probe is used to acquire distance data in real time from multiple points on the surface of the composite material being measured beneath the probe. The acquired distance data is input into a processing unit, which runs a surface fitting algorithm, such as least-squares quadratic surface fitting, to calculate the radius of curvature and normal direction of the local surface beneath the probe. Simultaneously, the phased array probe emits ultrasonic waves and receives echo signals from pre-defined reflectors within the material (e.g., the bottom surface of the material or a known embedded reflector). The processing unit analyzes these echo signals, determining the actual transit time of the ultrasonic wave by detecting the peak time of the echo envelope. The processing unit then executes an iterative correction algorithm, taking the calculated local surface geometry parameters and the acquired actual transit time as input. This algorithm repeatedly adjusts a sound velocity parameter until the error between the theoretical transit time calculated based on the sound velocity and geometry parameters and the actual transit time is less than a preset threshold, thereby determining the equivalent propagation speed of sound. Finally, using the calculated local geometric parameters and the determined equivalent propagation speed of sound, the excitation delay time of each element of the phased array probe is calculated according to the principle of ray tracing. These delay times are then sent to the transmitting circuit of the flaw detector to control the excitation sequence of each element, thereby achieving focusing compensation of the ultrasonic beam inside the material.

[0091] Through the above-described scheme, this application can adaptively and in real-time compensate for the influence of complex surface geometry and internal sound velocity inhomogeneities on the focusing of phased array ultrasonic beams. This enables stable and accurate focusing results when scanning and inspecting composite material components with complex, continuously varying curvature surfaces, regardless of probe location or depth variations. Consequently, it improves defect detection rates and the accuracy of quantitative assessments.

[0092] Specifically, in some of the solutions described above in this application, an equivalent propagation velocity is determined through iterative correction based on the calculated local geometric parameters and the actual transit time of the ultrasonic echo to compensate for the focusing of the phased array ultrasonic beam within the composite material component. However, in this process, the actual transit time of the ultrasonic echo may contain noise or fluctuations, and directly using it for iterative correction may result in the determination of the equivalent propagation velocity lacking accuracy or stability, affecting the subsequent focusing compensation function. Furthermore, the iterative correction process may be limited by computational resources or time, affecting the reliability of real-time operation or results.

[0093] To address this, this application further proposes a step of determining the equivalent propagation velocity of ultrasound in composite material components through iterative correction based on the calculated local geometric parameters and the actual transit time of the acquired ultrasonic echo. This step includes: performing smoothing processing on the acquired actual transit time of the ultrasonic echo to obtain a target transit time; determining the equivalent propagation velocity of ultrasound in composite material components through iterative correction based on the calculated local geometric parameters and the target transit time, wherein when the iterative calculation of the iterative correction reaches a preset resource constraint condition, the equivalent propagation velocity based on the calculation of that iteration is output; and performing variation amplitude limitation processing on the equivalent propagation velocity output by the iterative correction to obtain the final equivalent propagation velocity.

[0094] Specifically, this scheme aims to address the technical challenge of ensuring that the determined output of the equivalent propagation velocity meets the real-time operation requirements of the system and that its numerical variation is stable in scenarios where local geometric parameters change rapidly and the actual transit time of the ultrasonic echo fluctuates.

[0095] The actual transit time of the acquired ultrasonic echo is a measurement data reflecting the sound propagation characteristics within the material. However, this data may contain discontinuous variations due to fluctuations during signal acquisition. Directly using the transit time containing these fluctuations for subsequent iterative corrections can easily lead to oscillations in the calculated equivalent propagation velocity. Therefore, by smoothing the acquired actual transit time of the ultrasonic echo, for example using mean filtering or median filtering, these acquisition fluctuations can be filtered out or reduced, thus obtaining the target transit time. Compared to the originally acquired transit time, this target transit time exhibits a smoother numerical change. As input for iterative correction, it provides a data foundation for subsequently determining an equivalent propagation velocity with a stable change, contributing to solving the problem of oscillations in the equivalent propagation velocity calculation results due to transit time fluctuations. The core of this step lies in using the calculated local geometric parameters (which characterize the geometric information related to the ultrasonic propagation path) and the target transit time obtained after the smoothing process in the previous step (which characterizes the actual propagation time of the ultrasonic wave under the geometric path and has reduced the influence of fluctuations) to calculate the equivalent propagation speed of sound that can match the theoretical transit time calculated based on the local geometric parameters and sound speed with the target transit time through iterative correction.

[0096] In phased array ultrasonic scanning inspection, especially when the local geometric parameters of the tested component are updated rapidly with changes in the scanning position, there are real-time requirements for the calculation and update rate of the equivalent propagation velocity; that is, the iterative correction process needs to be completed within a limited time. To address this, a constraint is imposed: when the iterative calculation of the correction reaches a preset resource limit (e.g., a preset maximum number of iterations or a maximum allowed computation time), the equivalent propagation velocity calculated in that iteration is output. This constraint ensures that even if the iterative process does not mathematically converge completely to the theoretical optimal solution, an equivalent propagation velocity result for the current detection point can still be obtained while meeting the system's processing time requirements. This avoids the inability to meet the real-time operation requirements of scanning inspection due to excessively long iterative calculation times, contributing to solving the problem of completing the output of the iterative correction process within a limited time.

[0097] Although the preceding steps improve the stability and real-time operability of sound velocity calculation by smoothing the transit time and setting resource constraints in the iterative correction, the equivalent propagation sound velocity output by the iterative correction may still exhibit numerical jumps or short-term oscillations during continuous output due to residual fluctuations in the input data, the truncation effect of the iterative calculation, or the characteristics of the iterative algorithm itself. To ensure that the equivalent propagation sound velocity finally applied to phased array focusing compensation has numerical stability and avoids instability in the focusing function or disturbances to the subsequent control system caused by drastic or frequent changes in sound velocity parameters, it is necessary to perform amplitude limitation processing on the equivalent propagation sound velocity output by the iterative correction. This processing can be, for example, limiting the maximum allowable change between two adjacent updated sound velocity values, or performing further low-pass filtering on the output sound velocity sequence. The aim is to smooth the change curve of sound velocity over time or scan position to obtain the final equivalent propagation sound velocity. This final equivalent propagation sound velocity exhibits continuous and gentle numerical changes, thus ensuring the stability and continuity of the focusing compensation function and contributing to solving the problem of ensuring the stability of the equivalent propagation sound velocity output numerical change.

[0098] Through the above scheme, this application can reduce the impact of noise and fluctuation in the actual transit time of ultrasonic echo on the calculation of equivalent propagation speed, ensure that usable speed results can still be output when computing resources or time are limited, and make the output equivalent propagation speed value change smoothly, thereby improving the accuracy and stability of the phased array ultrasonic beam focusing compensation in composite material components.

[0099] Specifically, some of the solutions described above in this application propose determining the equivalent propagation velocity of ultrasound in composite material components through iterative correction based on the calculated local geometric parameters and the obtained actual transit time of the ultrasonic echo. However, the obtained actual transit time of the ultrasonic echo may be affected by noise, measurement errors, and abrupt changes in the internal structure or geometry of the composite material, resulting in noise, outliers, or actual step changes in the transit time series. If the original actual transit time is used directly for iterative correction, the determined equivalent propagation velocity may be unstable or inaccurate, affecting the subsequent calculation of the excitation delay time and the beam focusing accuracy. Therefore, how to effectively process the obtained actual transit time of the ultrasonic echo to obtain a more stable and accurate target transit time for iterative velocity correction is a technical problem that needs to be solved.

[0100] In response, this application further proposes a step of smoothing the actual transit time of the acquired ultrasonic echo to obtain the target transit time, which includes:

[0101] The actual transit time series of the acquired ultrasonic echoes is detected to identify transit time step changes in the transit time series and obtain the identified transit time step changes.

[0102] Based on the identified step changes in transit time, the actual transit time sequence of the acquired ultrasonic echo is divided into step change segments corresponding to step changes in transit time and non-step change segments corresponding to non-step changes.

[0103] Smoothing is performed on the transit time data of the non-step change section to obtain smoothed transit time data of the non-step change section.

[0104] The target transit time is generated by combining the transit time data of the smoothed non-step change segment with the transit time data of the step change segment.

[0105] Specifically, the operational logic of this scheme is as follows: First, the original ultrasonic echo transit time series is analyzed to distinguish between true signal steps caused by abrupt changes in the acoustic properties of the material and random fluctuations caused by noise and other factors. By detecting and identifying these step changes, it can be determined which data points or segments represent signal features that need to be preserved (step change segments) and which data points or segments mainly contain noise that needs to be suppressed (non-step change segments). Subsequently, for the identified non-step change segments, a smoothing algorithm is applied to effectively filter out random noise, thereby improving the stability of this part of the data. For the identified step change segments, smoothing that might blur or eliminate step features is avoided to preserve their original form. Finally, the smoothed non-step change segment data and the step change segment data that retains the original features are reintegrated to form a target transit time series that both suppresses noise and retains key step information. This target transit time series, used as input for subsequent equivalent propagation sound velocity iterative correction, provides more stable and accurate transit time information compared to directly using the original data, thereby improving the accuracy of sound velocity determination. This scheme, through targeted preprocessing of the original transit time data, provides high-quality input data for subsequent sound velocity determination steps, thus enhancing the performance of the entire flaw detection method.

[0106] As a preferred embodiment, the solution of this application is implemented as follows: In this embodiment, the acquired actual transit time sequence of the ultrasonic echo is input into a processing unit. The processing unit first performs step change detection, specifically by calculating the difference between adjacent data points in the sequence and setting a threshold. When the difference exceeds the threshold, a transit time step change is considered to exist at that location. Based on the detected step change location, the original transit time sequence is divided into multiple segments. Data adjacent to the step change point is classified as step change segments, and the remaining data is classified as non-step change segments. Next, for the data in each non-step change segment, a moving average filter is applied for smoothing. For example, a fixed-length window is used, and the average value of the data within the window is calculated as the smoothing value at the center point. The window slides along the data sequence. After smoothing all non-step change segments, these smoothed data segments are recombine with the original step change segment data according to their order in the original sequence to generate the final target transit time sequence.

[0107] Through the above scheme, this application can effectively handle the random noise and real step changes that exist simultaneously in the actual transit time series of ultrasonic echoes. While suppressing noise, it retains important step characteristics and obtains a more stable and accurate target transit time, thereby improving the accuracy of subsequent equivalent propagation velocity iterative correction, and thus ensuring the accuracy of excitation delay time calculation and beam focusing.

[0108] Specifically, in some of the solutions described above in this application, an equivalent propagation velocity is determined through iterative correction based on the calculated local geometric parameters and the actual transit time of the ultrasonic echo to compensate for the focusing of the phased array ultrasonic beam within the composite material component. However, in this process, the actual transit time of the ultrasonic echo may come from multiple characteristic reflectors inside the composite material component. The signal quality and sensitivity to changes in the equivalent propagation velocity of these characteristic reflectors may differ. If all actual transit times are directly used or simply smoothed for iterative correction, interference from low-quality or insensitive signals may be introduced, affecting the accuracy of the determination of the equivalent propagation velocity, and consequently affecting the calculation of the excitation delay time and the beam focusing compensation effect.

[0109] To address this, this application further proposes a step for determining the equivalent propagation velocity of ultrasound in a composite material component through iterative correction based on the calculated local geometric parameters and the target transit time. This step includes: acquiring the ultrasound echo signals and corresponding actual transit times of multiple characteristic reflectors within the composite material component; evaluating the signal-to-noise ratio (SNR) of each acquired ultrasound echo signal to obtain an evaluated SNR; determining the sensitivity index of the transit time of each acquired ultrasound echo signal to changes in the equivalent propagation velocity to obtain a determined sensitivity index; selecting at least one characteristic reflector from the multiple characteristic reflectors based on the obtained evaluated SNR and the determined sensitivity index to obtain a selected characteristic reflector; using the actual transit time of the selected characteristic reflector as the target transit time for iterative correction, and performing iterative correction based on the calculated local geometric parameters and the target transit time to determine the equivalent propagation velocity of ultrasound in the composite material component.

[0110] Specifically, after acquiring the ultrasonic echo signals and corresponding actual transit times from multiple characteristic reflectors inside the composite material component, this solution does not simply smooth all transit times, but instead evaluates the quality and sensitivity of each echo signal.

[0111] By evaluating the signal-to-noise ratio of each echo signal, reliable echoes with high signal strength and low susceptibility to noise interference can be identified. By determining the sensitivity index of each echo transit time to changes in the equivalent propagation speed of sound, the effectiveness of the echo information for sound speed correction can be understood.

[0112] Subsequently, based on these evaluation results, at least one characteristic reflector with high signal quality and suitable sensitivity to changes in sound velocity was systematically selected from multiple characteristic reflectors. The actual transit time of these selected characteristic reflectors was used as the target for iterative correction and compared with the theoretical transit time calculated based on the solved local geometric parameters. The equivalent propagation speed was adjusted through an iterative process until the theoretical transit time and the target transit time were sufficiently close.

[0113] This strategy, based on signal quality and sensitivity optimization, enables the iterative correction process to utilize the most reliable and effective echo information, thereby improving the accuracy and robustness of equivalent propagation velocity determination. Compared to methods that rely solely on smoothing all echo transit times, this approach better addresses the complexities of varying signal quality from reflectors at different depths or along different paths, as well as varying responses to changes in sound velocity. This allows for more accurate sound velocity determination in the inspection of composite material components with complex surfaces and potential sound velocity inhomogeneities, leading to optimized excitation delay calculations and more accurate beam focusing compensation.

[0114] In some of the solutions described above in this application, an equivalent propagation velocity of ultrasound in composite material components is determined through iterative correction based on the actual transit time of the acquired ultrasonic echo. Furthermore, it proposes to improve the accuracy of equivalent propagation velocity determination by evaluating the signal-to-noise ratio of the ultrasonic echo signal and determining its sensitivity index to changes in the equivalent propagation velocity. At least one characteristic reflector is selected from multiple characteristic reflectors, and its actual transit time is used as the target transit time for iterative correction. However, simply determining the sensitivity index of transit time to changes in the equivalent propagation velocity may not fully consider the actual acoustic characteristic changes (such as attenuation, scattering, signal distortion, etc.) encountered by ultrasound along its complex propagation path within the composite material. These acoustic characteristic changes affect the quality and stability of the ultrasonic echo, thus affecting its actual response to changes in sound velocity. Consequently, the calculated sensitivity index may not accurately reflect the reliability or effectiveness of the echo for sound velocity determination, potentially affecting the optimization effect of characteristic reflector selection and limiting the accuracy of equivalent propagation velocity determination.

[0115] In response, this application further proposes a step for determining the sensitivity index of the transit time of each acquired ultrasonic echo signal to changes in the equivalent propagation speed of sound, comprising: extracting an acoustic characteristic change indication parameter from the ultrasonic echo signal corresponding to each acquired ultrasonic echo signal to characterize changes in acoustic characteristics of the ultrasonic propagation path; calculating an initial sensitivity index corresponding to the ultrasonic echo signal based on the geometric information of the ultrasonic propagation path corresponding to each acquired ultrasonic echo signal; and adjusting the initial sensitivity index according to the extracted acoustic characteristic change indication parameter and the calculated initial sensitivity index to determine the sensitivity index of the transit time of the ultrasonic echo signal to changes in the equivalent propagation speed of sound.

[0116] Among them, the acoustic characteristic change indicator parameter indicates the quantitative value of the acoustic environment changes experienced by ultrasound waves along a specific propagation path. It can be extracted by analyzing signal characteristics such as amplitude attenuation, spectral component changes, pulse width broadening, phase distortion, or signal-to-noise ratio of the ultrasound echo signal. For example, the ratio of the peak amplitude of the echo signal to the peak amplitude of the transmitted signal can be calculated as an attenuation indicator, or the center frequency shift of the echo signal can be calculated as an indicator of spectral changes caused by scattering or absorption.

[0117] The geometric information of the ultrasonic propagation path refers to the spatial shape and dimensional data of the complete path of the ultrasonic wave from the probe array element, through the coupling layer, into the composite material, propagating to the characteristic reflector, and returning. It can be characterized by the local geometric parameters of the probe and the component surface, as well as the depth or position information of the characteristic reflector within the material.

[0118] The initial sensitivity index refers to the theoretical sensitivity of the ultrasonic echo transit time to the change of the equivalent propagation speed of sound in the material, calculated solely based on the geometric information of the ultrasonic propagation path under an ideal or simplified acoustic medium model. It can be calculated using the derivative of the ultrasonic propagation path length with respect to the speed of sound or the rate of change of transit time calculated based on a geometric acoustic model.

[0119] Adjusting the initial sensitivity index refers to the process of correcting or weighting the initial sensitivity index calculated solely based on geometric information, according to the actual extracted acoustic characteristic change indicator parameters. This can be achieved using methods such as preset correction functions, lookup tables, machine learning models, or weighted averaging, to ensure that the final determined sensitivity index more accurately reflects the impact of the actual acoustic path on transit time.

[0120] The transit time sensitivity index to changes in equivalent propagation speed refers to the quantitative sensitivity of the ultrasonic echo transit time to changes in the material's equivalent propagation speed, after comprehensively considering the geometric information of the ultrasonic propagation path and the actual acoustic characteristics changes along that path. This index is used to evaluate the reliability or magnitude of the response of a specific echo signal's transit time to changes in sound speed.

[0121] Specifically, this technical solution processes each acquired ultrasonic echo signal. First, parameters are extracted from the corresponding ultrasonic echo signal. These parameters indicate the changes in acoustic characteristics encountered by the ultrasonic wave along a specific propagation path, capturing actual effects such as signal attenuation or distortion. Simultaneously, an initial sensitivity index is calculated based on the geometric information of the ultrasonic propagation path. This index only reflects the theoretical influence of path geometry on the transit time as a function of sound speed, assuming uniform acoustic properties of the material.

[0122] The core of this approach lies in using extracted acoustic characteristic change indicators to adjust the initial sensitivity index, which is based solely on geometric information. This adjustment process modifies the initial index according to the actual acoustic conditions encountered by the ultrasonic pulse. For example, if the acoustic parameters indicate that the signal is severely degraded due to high attenuation or distortion, the adjustment process can reduce the calculated sensitivity, reflecting that the transit time measured from such a degraded signal may be unreliable or the response to changes in the true sound velocity may be inaccurate.

[0123] Conversely, a clear signal may lead to fewer adjustments, allowing geometric sensitivity to dominate. By combining geometric influences and actual acoustic path conditions, this scheme determines a final sensitivity index that more accurately represents the actual reliability and sensitivity of the transit time of a specific echo signal to changes in the material's equivalent propagation velocity of sound under real-world conditions. This optimized sensitivity index provides a more reliable basis for subsequently selecting the most suitable characteristic reflector for sound velocity determination. Combining acoustic characterization with geometric calculations and subsequent adjustments allows the system to adapt to the complexity and non-uniformity of composite materials and their bending geometry, a significant challenge in existing technologies. This improved sensitivity calculation directly supports the selection process in the preceding steps, enabling the system to select the most reliable echo signal for iterative sound velocity correction, thereby improving the accuracy and stability of the overall sound velocity determination process.

[0124] In some of the solutions described above in this application, an equivalent propagation velocity of sound, characterizing the propagation of ultrasound in composite material components, is determined through iterative correction based on the calculated local geometric parameters and the actual transit time of the acquired ultrasonic echo. The excitation delay time is then calculated based on this velocity and the local geometric parameters to compensate for the focusing of the ultrasonic beam. To improve the accuracy of the equivalent propagation velocity determination, at least one characteristic reflector needs to be selected from multiple characteristic reflectors. The selection is based on the evaluated signal-to-noise ratio and the sensitivity index of the transit time to changes in the equivalent propagation velocity. An important step in determining the sensitivity index is to extract acoustic characteristic change indicator parameters that characterize the changes in the acoustic characteristics of the ultrasonic propagation path of the ultrasonic echo signal. However, when performing phased array ultrasonic scanning on composite material components with complex, continuously varying curvature surfaces, the sound velocity within the composite material may exhibit local inhomogeneities, and the surface geometry may change rapidly. These factors can cause the acquired ultrasonic echo signals to be adversely affected by noise, abnormal signal morphology, and echo overlap. This makes it difficult to accurately and reliably extract indication parameters characterizing changes in the acoustic properties of the ultrasonic propagation path directly from the original ultrasonic echo signals, thereby affecting the accuracy of sensitivity index determination and ultimately reducing the accuracy of equivalent propagation velocity calculation and focusing compensation. Therefore, how to robustly extract indication parameters characterizing changes in the acoustic properties of the ultrasonic propagation path from complex, interfered ultrasonic echo signals is a key issue in improving the accuracy and reliability of the entire control method.

[0125] To address this, this application further proposes a step of extracting an acoustic characteristic change indication parameter from the ultrasonic echo signal corresponding to each acquired ultrasonic echo signal, used to characterize the acoustic characteristic change of the ultrasonic propagation path of the ultrasonic echo signal. This step includes: performing segmentation processing on the ultrasonic echo signal corresponding to each acquired ultrasonic echo signal to identify noise-affected segments, signal morphology abnormal segments, and echo overlap segments within the ultrasonic echo signal, obtaining the segmentation processing identification result; and based on the segmentation processing identification result, selecting at least one signal feature from the ultrasonic echo signal that satisfies a preset condition. Based on the sub-echo segments, at least one selected sub-echo segment is obtained; for each selected sub-echo segment, preliminary acoustic characteristic parameters are extracted to obtain preliminary acoustic characteristic parameters corresponding to each selected sub-echo segment; the preliminary acoustic characteristic parameters extracted from the at least one selected sub-echo segment and corresponding to each selected sub-echo segment are fused to obtain the data fusion result, and consistency evaluation and outlier removal are performed on the data fusion result to obtain acoustic characteristic change indication parameters for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal.

[0126] Among them, segmentation processing refers to analyzing the ultrasound echo signal, dividing it into different time or sample intervals, and classifying or labeling each interval.

[0127] The noise-affected section refers to the part of the signal whose main component is random noise, which can be identified using methods based on signal amplitude thresholds, spectral analysis, or statistics.

[0128] Abnormal signal morphology segments refer to the parts where the signal waveform differs significantly from the normal echo signal waveform. These segments can be identified using methods such as waveform correlation analysis, template matching, or feature extraction and classification.

[0129] The echo overlap section refers to the part where multiple independent echo signals superimpose in time. It can be identified by methods such as signal deconvolution, blind source separation, or signal modeling based on prior knowledge.

[0130] Preset criteria refer to the standards used to evaluate the signal quality or applicability of sub-echo sections, which may include, but are not limited to, signal-to-noise ratio thresholds, waveform similarity thresholds, energy concentration thresholds, or the range of values ​​for specific characteristic parameters.

[0131] Sub-echo segments refer to one or more continuous or discontinuous time intervals selected from a complete ultrasound echo signal. These intervals are considered to contain useful echo information and have relatively high signal quality.

[0132] Preliminary acoustic characteristic parameters refer to quantitative indicators extracted from a single sub-echo segment to describe the signal characteristics of that segment. These parameters may include peak amplitude, integral energy, dominant frequency, bandwidth, phase information, transit time (such as first wave arrival time, peak time), or waveform shape parameters.

[0133] Data fusion refers to the integration and processing of similar data obtained from multiple sources or measurements to obtain more accurate and reliable comprehensive results. It can be achieved using methods such as average calculation, median filtering, weighted averaging, Kalman filtering, or Bayesian estimation.

[0134] Consistency assessment refers to the analysis of a set of data or a data sequence to determine whether there are significant inconsistencies or abnormal fluctuations within it. It can be achieved using statistical methods (such as standard deviation and analysis of variance), trend analysis, or model-based predictions compared with actual data.

[0135] Outlier removal refers to identifying and removing outliers that are significantly different from other data in a dataset to clean up the dataset. This can be achieved using methods based on statistical distance (such as Z-score, IQR), cluster analysis, or model prediction error.

[0136] Acoustic characteristic change indicator parameter refers to the final quantitative index obtained after processing and fusion, which can stably and accurately reflect the changes in acoustic characteristics of ultrasound along the propagation path. This parameter is used for subsequent sensitivity index adjustment.

[0137] Specifically, this method first analyzes the raw ultrasonic echo signal, decomposing it into different segments and identifying those affected by noise, waveform anomalies, or echo overlap. Since the raw signal may contain a large amount of interference, directly extracting parameters from it would introduce errors. The identification results clearly identify which parts of the signal are unreliable, providing guidance for subsequent processing. Next, based on the identification results, sub-echo segments with better signal quality and meeting preset criteria are selected from the raw signal. This selection process avoids severely interfered signal segments, ensuring the high validity of the raw data used for parameter extraction.

[0138] Then, for each selected sub-echo segment, preliminary acoustic characteristic parameters are extracted. Since these parameters are extracted from preferred signal segments, their initial accuracy is relatively high. Finally, the preliminary parameters extracted from multiple preferred sub-echo segments are fused to comprehensively utilize information from different reliable signal segments and smooth random fluctuations.

[0139] Furthermore, consistency assessment and outlier removal are performed on the fusion results to identify and eliminate any residual outliers that may occur during the fusion process. This ultimately yields an indicator parameter that characterizes changes in the acoustic properties of the ultrasonic propagation path, after multiple purification and comprehensive processing. This parameter exhibits higher accuracy and robustness compared to parameters extracted directly from the original signal. Through this step-by-step processing and data optimization approach, this method can effectively extract reliable acoustic property change indicator parameters from complex and disturbed ultrasonic echo signals. The accurate extraction of this parameter further improves the accuracy of determining the sensitivity index of transit time to changes in equivalent propagation velocity. This makes the process of selecting characteristic reflectors for iterative correction of equivalent propagation velocity based on signal-to-noise ratio and sensitivity index more reliable, ultimately improving the accuracy of equivalent propagation velocity calculation and excitation delay time calculation. This contributes to more accurate beam focusing compensation in the detection of complex curved composite material components.

[0140] In some of the solutions described above in this application, preliminary acoustic characteristic parameters extracted from at least one selected sub-echo segment are subjected to data fusion, consistency assessment, and outlier removal to obtain acoustic characteristic change indication parameters for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal. However, in this process, the data fusion, consistency assessment, and outlier removal steps may consume a lot of computational resources and time. In phased array ultrasonic scanning scenarios with high real-time requirements, accurate acoustic characteristic change indication parameters may not be obtained in a timely manner, affecting the real-time performance and accuracy of subsequent sound velocity determination and focus delay calculation.

[0141] To address this, this application further proposes a step of fusing preliminary acoustic characteristic parameters extracted from at least one selected sub-echo segment, corresponding to each selected sub-echo segment, to obtain a data fusion result, and performing consistency evaluation and outlier removal on the data fusion result to obtain an acoustic characteristic change indication parameter for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal. The steps include: receiving preliminary acoustic characteristic parameters extracted from at least one selected sub-echo segment, corresponding to each selected sub-echo segment; fusing the preliminary acoustic characteristic parameters using a fusion algorithm with computational complexity meeting preset conditions to obtain a data fusion result; performing a consistency evaluation on the data fusion result using a consistency evaluation method with computational complexity meeting preset conditions; removing outlier data points from the data fusion result using an outlier removal method with computational complexity meeting preset conditions; monitoring computational resource usage or processing time during the data fusion, consistency evaluation, and outlier removal processes; and outputting the current processing result as an acoustic characteristic change indication parameter for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal when preset resource limitations are reached.

[0142] Specifically, this solution aims to address the problem of how to optimize the algorithm to meet real-time requirements when the system's processing capacity or time window is limited, and a large number of preliminary acoustic characteristic parameters need to be fused, evaluated, and outliers removed.

[0143] First, preliminary acoustic characteristic parameters extracted from the selected sub-echo segments are received; these parameters serve as the foundational input for subsequent processing. Next, these preliminary parameters are fused. By employing a fusion algorithm with computational complexity meeting preset conditions, computational overhead can be controlled while ensuring a certain level of fusion effectiveness, resulting in preliminary data fusion results. Then, a consistency assessment is performed on the data fusion results using a consistency assessment method with computational complexity meeting preset conditions to quickly check the stability and reliability of the fusion results. Finally, outlier removal is performed on the data fusion results using an outlier removal method with computational complexity meeting preset conditions to quickly eliminate outliers and further purify the data.

[0144] Throughout the data fusion, consistency assessment, and outlier removal processes, the system continuously monitors computational resource usage and processing time. Once preset resource limits are reached, the system immediately outputs the current processing result as an indicator of acoustic characteristic changes. This mechanism ensures that even under conditions of limited computational resources or time windows, the system can provide a timely output of usable acoustic characteristic parameters, thereby maintaining the real-time performance of subsequent sound velocity determination and focus delay calculations.

[0145] By employing computationally limited algorithms and introducing resource monitoring and conditional output mechanisms, this scheme maximizes the stability and accuracy of acoustic characteristic indicator parameters while ensuring real-time performance. This provides a more reliable input for subsequent determination of equivalent propagation sound velocity, thereby improving the focusing accuracy and detection efficiency of phased array ultrasonic scanning.

[0146] Specifically, some of the solutions described above in this application propose performing a consistency assessment on the data fusion results to determine the reliability of the fusion results. However, after data fusion of the preliminary acoustic characteristic parameters extracted from different sub-echo segments, the fusion result sequence may be affected by noise, local material property changes, or signal quality fluctuations, resulting in it not being completely stationary and possibly exhibiting normal fluctuations or local rapid changes. Simple consistency assessment methods struggle to distinguish these different types of changes, potentially misjudging normal fluctuations as inconsistencies or ignoring genuine abnormal changes, thereby affecting the accuracy of subsequent outlier removal and the final acoustic characteristic change indicator parameters. Especially in scenarios requiring rapid processing to meet real-time requirements, designing a method that can accurately distinguish different types of changes and robustly determine consistency is a challenge.

[0147] In this regard, this application further proposes a step for performing a consistency assessment on the results of data fusion, which includes:

[0148] Analyze the variation characteristics of the resulting sequence of data fusion;

[0149] Based on the characteristics of change, identify the segments in the result sequence of data fusion that meet the preset rapid change criteria;

[0150] For segments in the data fusion result sequence that do not meet the preset rapid change criteria, assess their data fluctuations.

[0151] Based on the relationship between data fluctuations and preset noise thresholds, it is determined whether the results of data fusion are consistent.

[0152] Among them, the change characteristics refer to the attributes that describe the changes of data sequence values ​​with their index or time, such as the rate of change, the magnitude of change, or a specific change pattern. These can be achieved by calculating the difference, derivative, and slope of the sequence, or by analyzing the local maximum, minimum, and average values ​​of the sequence.

[0153] Preset rapid change criteria refer to rules or standards used to define what changes in a data sequence are considered rapid changes. They can be defined by setting a change rate threshold, an amplitude jump threshold, or identifying specific waveform patterns.

[0154] Data volatility refers to the degree of dispersion or instability of a data sequence within a certain range. It can be assessed by calculating statistical indicators such as the standard deviation, variance, mean absolute deviation, or the difference between the maximum and minimum values ​​of the data within that range.

[0155] A preset noise threshold is a pre-defined numerical limit used to distinguish between acceptable fluctuations caused by noise and unacceptable fluctuations that indicate inconsistency. It can be determined by statistical analysis based on the system noise level or by experimental calibration.

[0156] Specifically, the scheme first analyzes the overall or local variation attributes of the data fusion result sequence to obtain information about how the sequence evolves with scanning location or time. Based on these variation attributes, the system identifies regions in the sequence that exhibit drastic and rapid numerical changes, which may correspond to rapid changes in the acoustic properties of the actual materials. For the remaining regions in the sequence that do not belong to the rapid change type, the scheme further quantifies the degree of fluctuation in their internal data, which mainly reflects residual noise or slight instability. Finally, the quantified data fluctuation is compared with a pre-set noise tolerance limit. If the fluctuation is within the noise threshold range, the data in that segment is considered consistent; otherwise, other inconsistencies may exist.

[0157] This step-by-step processing method distinguishes between rapid changes caused by actual physical variations and fluctuations caused by noise, avoiding misjudging the former as inconsistencies while accurately assessing the impact of the latter, thus improving the accuracy of consistency judgment. This consistency assessment process operates on the preliminary acoustic characteristic parameter sequence extracted and fused from the sub-echo segments, providing a more reliable basis for subsequent outlier removal and helping to obtain more accurate acoustic characteristic change indication parameters, thereby supporting the determination of the equivalent propagation speed.

[0158] Specifically, in some of the solutions mentioned above in this application, it is proposed to calculate the local geometric parameters of the measured surface below the phased array probe based on multi-point distance data, so as to provide a basis for subsequent determination of equivalent propagation speed and calculation of excitation delay time. However, the original multi-point distance data may contain noise and outliers, and the local geometric features of complex surfaces are diverse. Directly calculating based on the original data may lead to inaccurate local geometric parameters, which in turn affects the accuracy of subsequent focusing compensation.

[0159] In this regard, this application further proposes that this step includes:

[0160] Receive distance data from multiple points;

[0161] Perform data processing on the multi-point distance data to reduce noise and remove outliers, resulting in processed multi-point distance data;

[0162] The processed multi-point distance data is analyzed to identify the local geometric features of the measured surface, and the identified local geometric features are obtained.

[0163] Based on the identified local geometric features, a computational strategy for solving local geometric parameters is selected, and the selected computational strategy is obtained.

[0164] Based on the selected calculation strategy and the processed multi-point distance data, the local geometric parameters of the measured surface below the phased array probe are calculated.

[0165] Data processing refers to the preprocessing of the original multi-point distance data, which can be achieved by using filtering algorithms (such as median filtering and moving average filtering) or outlier detection and removal methods based on statistical principles (such as methods based on standard deviation, interquartile range, or cluster analysis).

[0166] Among them, analyzing the processed multi-point distance data to identify the local geometric features of the measured surface refers to judging the surface shape of the current region by calculating or evaluating the spatial distribution characteristics of the processed data points. This can be achieved by calculating local curvature, detecting edges (e.g., gradient-based or difference-based methods), or detecting corner points (e.g., feature-based or template-matching methods).

[0167] Among them, the selection of a computational strategy for solving local geometric parameters refers to determining the most suitable algorithm or model for solving the parameters of the identified local geometric features. This can be achieved by using rule-based judgment logic or machine learning classifiers.

[0168] The computational strategy refers to the specific algorithm or mathematical model used to extract or fit local geometric parameters from the processed multi-point distance data. It can be implemented by least squares fitting (e.g., plane fitting, quadratic surface fitting), straight line or curve fitting based on edge detection results, or parametric computation methods for specific geometric features (e.g., cylinder, sphere).

[0169] Specifically, the scheme first receives raw multi-point distance data acquired from a probe or external sensor. This data reflects the spatial position of the measured surface below the probe. Due to interference or errors during the acquisition process, this raw data may contain noise or outliers, which can affect the accuracy of subsequent calculations if used directly. Therefore, data processing of the raw data is necessary. By reducing noise and removing outliers, higher-quality processed data is obtained, laying the foundation for accurate calculations. Next, the processed data is analyzed to identify the geometric features of the current local region, such as whether it is relatively flat, smoothly curved, or has sharp edges or corners. Identifying local geometric features is necessary because different surface shapes require different mathematical methods to accurately describe their parameters. Based on the identified local geometric features, the most suitable calculation strategy is selected. For example, a surface fitting strategy is chosen for smooth surfaces, and a parameterization strategy based on edge detection is chosen for edge regions. Selecting an appropriate calculation strategy is essential, as it ensures that the optimal solution method is used for the current local surface shape. Finally, based on the selected calculation strategy and utilizing the processed, high-quality multi-point distance data, a specific solution process is executed to obtain the local geometric parameters of the measured surface below the phased array probe, such as local curvature and normal direction. Through this series of steps, even with poor raw data quality or complex and varied surface morphology, the local geometric parameters can be accurately and robustly determined. These accurate local geometric parameters are key inputs for subsequent determination of the equivalent propagation velocity and calculation of the excitation delay time, thereby enabling the entire phased array ultrasonic flaw detector control method to effectively compensate for surface effects and sound velocity non-uniformity, and improve beam focusing accuracy.

[0170] Secondly, referring to Figure 2 This application proposes a control system for a phased array composite material ultrasonic flaw detector, the system comprising:

[0171] Distance acquisition module 1 is used to acquire multi-point distance data between the phased array probe and the measured surface of the composite material component;

[0172] The geometry calculation module 2 is used to calculate the local geometric parameters of the measured surface below the phased array probe based on multi-point distance data.

[0173] Echo acquisition module 3 is used to acquire the actual transit time of ultrasonic echoes from characteristic reflectors inside composite material components.

[0174] The sound velocity determination module 4 is used to determine the equivalent propagation sound velocity, which characterizes the propagation of ultrasonic waves in composite material components, based on the calculated local geometric parameters and the actual transit time of the ultrasonic echo through iterative correction.

[0175] The delay calculation module 5 is used to calculate the excitation delay time for each element of the phased array probe based on the solved local geometric parameters and the determined equivalent propagation speed, in order to compensate for the focusing of the phased array ultrasonic beam in the composite material component.

[0176] By adaptively acquiring the local geometric parameters of the tested surface and determining the equivalent propagation velocity of sound, the excitation delay time can be accurately calculated, thereby improving the focusing accuracy of the ultrasonic beam within the complex curved composite material. This has the advantages of improving detection sensitivity and the accuracy of defect assessment.

[0177] Furthermore, in some preferred embodiments, the phased array composite material ultrasonic flaw detector control system proposed in this application can operate any one of the steps in the above method.

[0178] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A control method for a phased array composite material ultrasonic flaw detector, characterized in that, include: Acquire multi-point distance data between the phased array probe and the measured surface of the composite material component; Based on the multi-point distance data, the local geometric parameters of the measured surface below the phased array probe are calculated. Obtain the actual transit time of the ultrasonic echo from the internal characteristic reflector of the composite material component; Based on the calculated local geometric parameters and the obtained actual transit time of the ultrasonic echo, the equivalent propagation velocity of the ultrasonic wave in the composite material component is determined through iterative correction. Based on the calculated local geometric parameters and the determined equivalent propagation speed, the excitation delay time for each element of the phased array probe is calculated to compensate for the focusing of the phased array ultrasonic beam within the composite material component.

2. The control method for a phased array composite material ultrasonic flaw detector according to claim 1, characterized in that, The step of determining the equivalent propagation velocity of ultrasound in the composite material component by iterative correction based on the calculated local geometric parameters and the obtained actual transit time of the ultrasonic echo includes: The actual transit time of the acquired ultrasonic echo is smoothed to obtain the target transit time; Based on the calculated local geometric parameters and the target transit time, the equivalent propagation speed of the ultrasonic wave in the composite material component is determined by iterative correction. When the iterative calculation of the iterative correction reaches the preset resource limit, the equivalent propagation speed based on the iterative calculation is output. The equivalent propagation speed of sound output by the iterative correction is subjected to a variation amplitude limitation process to obtain the final equivalent propagation speed of sound.

3. The control method for a phased array composite material ultrasonic flaw detector according to claim 2, characterized in that, The step of smoothing the actual transit time of the acquired ultrasonic echo to obtain the target transit time includes: The actual transit time sequence of the acquired ultrasonic echo is detected to identify the transit time step change in the transit time sequence and obtain the identified transit time step change. Based on the identified step change in transit time, the acquired actual transit time sequence of ultrasonic echo is divided into a step change segment corresponding to the step change in transit time and a non-step change segment corresponding to the non-step change. Smoothing processing is performed on the transit time data of the non-step change segment to obtain smoothed transit time data of the non-step change segment. The target transit time is generated by combining the transit time data of the smoothed non-step change segment with the transit time data of the step change segment.

4. The control method for a phased array composite material ultrasonic flaw detector according to claim 2, characterized in that, The step of determining the equivalent propagation velocity of ultrasonic waves in the composite material component through iterative correction based on the calculated local geometric parameters and the target transit time includes: Acquire the ultrasonic echo signals and corresponding actual transit times of each of the multiple characteristic reflectors inside the composite material component; For each acquired ultrasonic echo signal, the signal-to-noise ratio (SNR) of the ultrasonic echo signal is evaluated, and the evaluated SNR is obtained. For each acquired ultrasonic echo signal, the sensitivity index of the transit time of the ultrasonic echo signal to the change in the equivalent propagation speed of sound is determined, and the determined sensitivity index is obtained. Based on the obtained signal-to-noise ratio and the determined sensitivity index, at least one feature reflector is selected from the plurality of feature reflectors to obtain the selected feature reflector; The actual transit time of the selected characteristic reflector is used as the target transit time for the iterative correction, and the iterative correction is performed based on the calculated local geometric parameters and the target transit time for the iterative correction to determine the equivalent propagation speed characterizing the ultrasonic wave propagating in the composite material component.

5. The control method for a phased array composite material ultrasonic flaw detector according to claim 4, characterized in that, The step of determining the sensitivity index of the transit time of each acquired ultrasonic echo signal to changes in the equivalent propagation speed includes: From the ultrasonic echo signals corresponding to each of the acquired ultrasonic echo signals, an acoustic characteristic change indication parameter is extracted to characterize the acoustic characteristic change of the ultrasonic propagation path of the ultrasonic echo signal. Based on the geometric information of the ultrasonic propagation path corresponding to each acquired ultrasonic echo signal, the initial sensitivity index corresponding to the ultrasonic echo signal is calculated. Based on the extracted acoustic characteristic change indication parameters and the calculated initial sensitivity index, the initial sensitivity index is adjusted to determine the sensitivity index of the transit time of the ultrasonic echo signal to the change in the equivalent propagation speed.

6. The control method for a phased array composite material ultrasonic flaw detector according to claim 5, characterized in that, The step of extracting acoustic characteristic change indication parameters from the ultrasonic echo signals corresponding to each of the acquired ultrasonic echo signals to characterize the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal includes: For each acquired ultrasonic echo signal, a segmentation process is performed on the ultrasonic echo signal to identify noise-affected segments, signal morphology abnormal segments, and echo overlap segments within the ultrasonic echo signal, thereby obtaining the identification result of the segmentation process. Based on the identification results of the segment division process, at least one sub-echo segment whose signal features satisfy the preset criteria is selected from the ultrasonic echo signal to obtain at least one selected sub-echo segment; For at least one selected sub-echo segment, preliminary acoustic characteristic parameters are extracted from each selected sub-echo segment to obtain preliminary acoustic characteristic parameters corresponding to each selected sub-echo segment. The preliminary acoustic characteristic parameters extracted from at least one selected sub-echo segment and corresponding to each selected sub-echo segment are fused to obtain the data fusion result. Consistency assessment and outlier removal are performed on the data fusion result to obtain acoustic characteristic change indication parameters for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal.

7. The control method for a phased array composite material ultrasonic flaw detector according to claim 6, characterized in that, The step of fusing the preliminary acoustic characteristic parameters extracted from at least one selected sub-echo segment, corresponding to each selected sub-echo segment, to obtain a data fusion result, and performing consistency evaluation and outlier removal on the data fusion result to obtain an acoustic characteristic change indication parameter for characterizing the acoustic characteristic changes of the ultrasonic propagation path of the ultrasonic echo signal, includes: Receive the preliminary acoustic characteristic parameters extracted from at least one selected sub-echo segment, corresponding to each selected sub-echo segment; The preliminary acoustic characteristic parameters are fused using a fusion algorithm with computational complexity meeting preset conditions to obtain the fusion result. A consistency assessment is performed on the results of the data fusion, using a consistency assessment method with computational complexity that meets preset conditions; Outlier removal is performed on the data fusion result using an outlier removal method with computational complexity meeting preset conditions. During the execution of the data fusion, the consistency assessment, and the removal of outlier data points, monitor the usage of computing resources or the processing time. When the preset resource limit is reached, the current processing result is output as an acoustic characteristic change indication parameter to characterize the acoustic characteristic change of the ultrasonic propagation path of the ultrasonic echo signal.

8. The control method for a phased array composite material ultrasonic flaw detector according to claim 7, characterized in that, The step of performing a consistency assessment on the results of the data fusion includes: Analyze the variation characteristics of the resulting sequence of the data fusion; Based on the aforementioned change characteristics, identify segments in the data fusion result sequence that meet the preset rapid change criteria; For segments in the data fusion result sequence that do not meet the preset rapid change criterion, their data fluctuations are evaluated. Based on the relationship between the data fluctuations and the preset noise threshold, it is determined whether the data fusion results are consistent.

9. The control method for a phased array composite material ultrasonic flaw detector according to claim 1, characterized in that, The step of calculating the local geometric parameters of the measured surface below the phased array probe based on the multi-point distance data includes: Receive the multi-point distance data; Data processing is performed on the multi-point distance data to reduce noise and remove outliers, resulting in processed multi-point distance data; The processed multi-point distance data is analyzed to identify the local geometric features of the measured surface, and the identified local geometric features are obtained. Based on the identified local geometric features, a calculation strategy for solving the local geometric parameters is selected, and the selected calculation strategy is obtained. Based on the selected calculation strategy and the processed multi-point distance data, the local geometric parameters of the measured surface below the phased array probe are calculated.

10. A control system for a phased array composite material ultrasonic flaw detector, characterized in that, The system includes: The distance acquisition module is used to acquire multi-point distance data between the phased array probe and the measured surface of the composite material component; The geometry calculation module is used to calculate the local geometric parameters of the measured surface below the phased array probe based on the multi-point distance data. An echo acquisition module is used to acquire the actual transit time of ultrasonic echoes from the internal characteristic reflectors of the composite material component; The sound velocity determination module is used to determine the equivalent propagation sound velocity, which characterizes the propagation of ultrasonic waves in the composite material component, based on the calculated local geometric parameters and the actual transit time of the acquired ultrasonic echo through iterative correction. The delay calculation module is used to calculate the excitation delay time for each element of the phased array probe based on the calculated local geometric parameters and the determined equivalent propagation speed, in order to compensate for the focusing of the phased array ultrasonic beam in the composite material component.

Images