Single crystal turbine blade crystal orientation defect detection system based on adaptive optics

Abstract

The invention discloses a single crystal turbine blade crystal orientation defect detection system based on adaptive optics, particularly relates to the field of nondestructive testing, and is used for solving the problem that thermal damage and signal consistency are difficult to consider in acoustic imaging of complex structural parts. Through partition energy threshold setting, dynamic pulse planning, photoacoustic processing and unified coordinate cross-subarea data synthesis, local thermal damage caused by excessive concentration of energy is effectively avoided in the material detection process, and the requirement for collecting high-bandwidth signals can be met in a large range. The signal-to-noise ratio and the detection efficiency are improved by regulating and controlling pulse energy and a scanning path, the amplitude comparability and the image coherence among regions are ensured through signal normalization correction and subregion image set splicing, high-resolution and comprehensive-coverage acoustic mapping is output, the crystal orientation defects of the single crystal turbine blade are positioned and evaluated, the detection precision is improved, the measurement period is shortened, and the method is suitable for large-scale popularization and application. And the method has an applicable value for aero-engine parts requiring rapid and high-precision nondestructive detection.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing, and more specifically, to a defect detection system for the crystal orientation of single-crystal turbine blades based on adaptive optics. Background Technology

[0002] In the aerospace field, high-bandwidth photoacoustic detection is commonly used to assess the service status of single-crystal turbine blades. However, increasing laser pulse energy to expand the sampling frequency band can easily lead to localized overheating or micro-area melting, affecting subsequent test results and component performance. Traditional methods that reduce pulse energy to minimize the risk of thermal damage often result in weak signals being masked by background noise, insufficient spectral feature extraction, and decreased detection efficiency. These contradictions are even more pronounced in multi-region scanning tasks, necessitating the coordinated advancement of zoned energy allocation strategies and dynamic path planning to balance large-scale high-throughput detection with high-bandwidth signal extraction.

[0003] Balancing high bandwidth with high scanning efficiency is the core challenge of this type of detection process. It is necessary to fully consider the impact of surface temperature accumulation on the pulse energy safety threshold in the process control stage, and ensure the integrity of normalization correction and the consistency of cross-sub-region data fusion in the algorithm processing stage, so as to achieve fast and accurate global diagnosis of single-crystal turbine blades.

[0004] To address the aforementioned problems, a technical solution is provided. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a defect detection system for the crystal orientation of single-crystal turbine blades based on adaptive optics. Through partitioned energy threshold setting, dynamic pulse strategy planning, online photoacoustic signal processing, and unified coordinate cross-sub-region data synthesis, this system effectively avoids localized thermal damage caused by excessive energy concentration during material inspection. It also accommodates the acquisition requirements of high-bandwidth signals over a wide range. By adjusting pulse energy and scanning path, the system collaboratively improves the signal-to-noise ratio and detection efficiency. Furthermore, signal normalization correction and sub-region atlas stitching achieve amplitude comparability between different regions and overall image coherence, outputting a high-resolution and fully covered acoustic mapping. Finally, through image segmentation, spectral analysis, and coordinate transformation, the system accurately identifies, locates, and evaluates crystal orientation defects in single-crystal turbine blades. This significantly shortens the measurement cycle while improving detection accuracy, making it applicable to aero-engine components requiring rapid and high-precision non-destructive testing, thus addressing the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A defect detection system for the crystal orientation of single-crystal turbine blades based on adaptive optics includes: a threshold region mapping module, a pulse modulation tracing module, a frequency amplitude modulation module, and a fusion mapping and integration module.

[0008] Threshold Zone Delineation Module: This module undertakes the thermal diffusion analysis and photoacoustic sensitivity determination of the blade surface. It divides the entire blade into several sub-regions, calculates the pulse energy upper limit and irradiation number threshold for each sub-region, and checks the heat accumulation of the previous scan before entering the sub-region, dynamically adjusting the pulse distribution strategy. The output includes sub-region coordinates, thermal diffusion parameters, pulse upper limit and number information to the pulse modulation tracer module.

[0009] Pulse modulation and tracing module: Generates pulse trigger timing and stepping path based on the energy threshold and scanning range of each sub-region, calculates heat estimation results in real time and updates actual pulse energy in each scan, and finally outputs the echo sequence and dynamic energy record of each sub-region to the frequency amplitude modulation module.

[0010] Frequency amplitude normalization module: performs pre-filtering, frequency domain conversion and envelope extraction on pulse echo, distinguishes between broadband and narrowband components, and performs amplitude normalization correction based on the actual pulse energy calculation results of each sub-region. The output includes local acoustic atlas and pulse energy markers to the fusion and stitching module.

[0011] The fusion and stitching module performs geometric registration based on the sub-region coordinates and overlapping boundary features. It achieves cross-sub-region amplitude consistency through amplitude correction and stitches them together under unified coordinates to form the acoustic mapping of the entire blade, thus obtaining the blade acoustic distribution map after cross-sub-region fusion.

[0012] In a preferred embodiment, the threshold area delineation module includes the following:

[0013] S1.1, based on blade geometry and material roughness and thermal diffusivity The blade surface is divided into multiple detection sub-regions, forming a region set. And record the material property matrix for each sub-region. ;

[0014] S1.2, through the thermal diffusion equation Calculate each sub-region Surface temperature distribution, where For ambient temperature, For laser pulse energy, For material density, For the specific heat capacity of the material, The thermal diffusivity;

[0015] S1.3, using formula Compute sub-region Photoacoustic sensitivity ,in For laser conversion efficiency, For pulse power, The absorption coefficient is... The laser incident depth;

[0016] S1.4, considering both thermal diffusion and photoacoustic sensitivity, sets an upper limit for pulse energy and a threshold for the number of irradiations, based on the optimization objective function: Compute sub-region upper limit of pulse energy and maximum number of irradiations ,in Indicates the cumulative pulse energy. The maximum permissible temperature rise threshold;

[0017] S1.5, each time a sub-region is entered Previously, based on the cumulative thermal diffusion of the preceding sub-region Adjust the current energy allocation, dynamically correct the laser power and scanning frequency, and ensure that the pulse energy limit is met. and maximum number of irradiations Require.

[0018] In a preferred embodiment, the pulse signature tracing module includes the following:

[0019] S2.1, based on the sub-region list and the corresponding pulse energy limit With maximum number of irradiations Initialize the parameter table of the dynamic trigger module; for entering the sub-region Pre-reading the preorder sub-region Temperature accumulation It also verifies the impact on the current maximum allowable energy and updates the current sub-region pulse energy threshold. Make it satisfy: ,in For material density, For specific heat capacity, For sub-region thermal diffusivity, This represents the estimated total scan time.

[0020] S2.2, Based on the determined pulse energy threshold of the current sub-region, arrange the triggering time and power of each pulse in the triggering plan using a schedule, denoted as the pulse sequence. ; Compare this sequence with the scan speed of the motion control unit and step distance Alignment is performed so that the pulse position at any given time is within the sub-region. Inside: ,in For sub-region Length in the scanning direction.

[0021] In a preferred embodiment, the pulse signature tracing module further includes the following:

[0022] S2.3, If thermal accumulation increases during the scanning process, dynamically calculate the pulse power correction amount. And increase the step size accordingly. This ensures that the cumulative energy per unit area is below the safety threshold defined by the threshold zone planning module, while also guaranteeing that the maximum number of irradiations is completed before the end of the sub-region. The following constraints apply: By coordinating the adjustment of laser power and step size in the middle of the scan, the risk of thermal damage is reduced without sacrificing the photoacoustic signal intensity, while ensuring scanning efficiency.

[0023] S2.4, in sub-region After performing a pulse scan, the cumulative temperature rise corresponding to the echo signal is calculated; the results are then... The energy of the pulse that has been executed is recorded as a heat transfer message and passed to subsequent sub-regions. The planning module allows it to assess actual energy limitations: ,in This represents the temperature rise increment function based on pulse energy and the thermal diffusion characteristics of the material.

[0024] In a preferred embodiment, the frequency amplitude modulation module includes the following:

[0025] S3.1, Obtain the time-domain echo sequence for each sub-region. Simultaneously read the pulse energy and irradiation count markers used for scanning the corresponding sub-region. Align the data according to the timeline and attach a time-series index. This forms a unified format that can be used for preprocessing.

[0026] S3.2, performs a series of high-pass and low-pass filtering operations on the time-domain echo to first remove background interference and over-range low-frequency drift, and then combines an adaptive truncation method to remove transient pulse spikes; the filtering result is denoted as... : ,in and These represent high-pass and low-pass filtering operations, respectively.

[0027] In a preferred embodiment, the frequency amplitude modulation module further includes the following:

[0028] S3.3, Perform a Fast Fourier Transform on the filtered time-domain echo to obtain the spectrum. The amplitude and phase information are extracted to distinguish between broadband and narrowband components. Envelope extraction is performed in parallel using Hilbert transform or full-wave rectification to obtain the envelope function. And based on threshold comparison, the main echo band and the noise band are separated: ,in Indicates Fourier transform, Represents the Hilbert transform;

[0029] S3.4, Calculate and obtain the actual pulse energy: ,in, For sub-region The Sub-pulse power; For the first The duration of the next pulse; For sub-region The total number of pulses;

[0030] Combined with real pulse energy and the upper limit of sub-region pulse energy Normalize the waveform amplitude and phase; define the correction factor as... Multiply it by the spectrum The amplitude portion is adjusted according to preset rules to compensate for the difference in detection intensity caused by energy fluctuations, denoted as... .

[0031] In a preferred embodiment, the frequency amplitude modulation module further includes the following:

[0032] S3.5, the corrected spectrum and envelope function Converted to acoustic image format, with sub-region coordinates and pulse energy information Packaged together into a sub-region acoustic atlas .

[0033] In a preferred embodiment, the image blending and stitching module includes the following:

[0034] S4.1, Obtain the acoustic atlas of all sub-regions. and its mapping sub-region coordinates Pulse energy information marker Integrate this data into a unified structure to ensure that each sub-region is aligned under the same coordinate reference;

[0035] S4.2, for adjacent sub-regions By matching the scan position recorded in the pulse modulation tracer module with the features of the overlapping area, a boundary correspondence is established: Then, acoustic features are matched within the corresponding overlapping areas to fine-tune the relative positions of the sub-regions, obtaining the optimal translation or rotation matrix. Apply this transformation matrix to the sub-region The acoustic atlas is mapped to the sub-regions. Alignment position and direction.

[0036] In a preferred embodiment, the fusion and stitching module further includes the following:

[0037] S4.3, Since the pulse energy in different sub-regions may differ due to thermal diffusion or photoacoustic sensitivity requirements, amplitude unification processing is required for the merged image; let the correction factor be: ,in This represents the average value of the maximum permissible pulse energy for all sub-regions.

[0038] In a preferred embodiment, the fusion and stitching module further includes the following:

[0039] S4.4, Acoustic atlas of all registered and amplitude-corrected sub-regions The images are stitched together according to a unified coordinate order. Blank or unscanned areas are filled using nearest neighbor interpolation to obtain a complete acoustic distribution map of the blades. ;

[0040] The Canny edge detection algorithm is used to segment the acoustic distribution map of the blade to extract the boundary features of the abnormal area. The acoustic signal of the abnormal area is analyzed by fast Fourier transform to distinguish the physical properties of the defect. The defect is located to the specific geometric area of ​​the blade by coordinate transformation. The impact of the defect on the mechanical performance of the blade is evaluated according to the preset defect size threshold and intensity change rate.

[0041] The technical effects and advantages of the adaptive optics-based single-crystal turbine blade crystal orientation defect detection system of this invention are as follows:

[0042] This invention effectively avoids localized thermal damage caused by excessive energy concentration during material testing by setting partitioned energy thresholds, dynamic pulse strategy planning, online photoacoustic signal processing, and unified coordinate cross-sub-region data synthesis. It can also meet the acquisition requirements of high-bandwidth signals over a wide range. By adjusting pulse energy and scanning path, it can improve the signal-to-noise ratio and detection efficiency. Furthermore, by using signal normalization correction and sub-region atlas stitching, it can achieve amplitude comparability between different regions and overall image coherence. It outputs acoustic mapping with both high resolution and comprehensive coverage. Finally, through image segmentation, spectrum analysis, and coordinate transformation, it can accurately identify, locate, and evaluate crystal orientation defects in single-crystal turbine blades. While improving detection accuracy, it can significantly shorten the measurement cycle and has applicable value for aero-engine components that require rapid and high-precision non-destructive testing. Attached Figure Description

[0043] Figure 1This is a schematic diagram of the structure of the single-crystal turbine blade crystal orientation defect detection system based on adaptive optics according to the present invention. Detailed Implementation

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0045] Example 1: Figure 1 The present invention provides a defect detection system for the crystal orientation of a single-crystal turbine blade based on adaptive optics, comprising: a threshold region mapping module, a pulse modulation tracing module, a frequency amplitude modulation module, and a fusion and stitching module;

[0046] Threshold Zone Delineation Module: This module undertakes the thermal diffusion analysis and photoacoustic sensitivity determination of the blade surface. It divides the entire blade into several sub-regions, calculates the pulse energy upper limit and irradiation number threshold for each sub-region, and checks the heat accumulation of the previous scan before entering the sub-region, dynamically adjusting the pulse distribution strategy. The output includes sub-region coordinates, thermal diffusion parameters, pulse upper limit and number information to the pulse modulation tracer module.

[0047] Pulse modulation and tracing module: Generates pulse trigger timing and stepping path based on the energy threshold and scanning range of each sub-region, calculates heat estimation results in real time and updates actual pulse energy in each scan, and finally outputs the echo sequence and dynamic energy record of each sub-region to the frequency amplitude modulation module.

[0048] Frequency amplitude normalization module: performs pre-filtering, frequency domain conversion and envelope extraction on pulse echo, distinguishes between broadband and narrowband components, and performs amplitude normalization correction based on the actual pulse energy calculation results of each sub-region. The output includes local acoustic atlas and pulse energy markers to the fusion and stitching module.

[0049] The fusion and stitching module performs geometric registration based on sub-region coordinates and overlapping boundary features. Amplitude correction ensures consistency across sub-region amplitudes, and the resulting modules stitch together the sub-regions to form the acoustic map of the entire blade, obtaining a fused acoustic distribution map of the blade. The surface of a single-crystal turbine blade is complex and diverse, and the detection effect is affected by factors such as thermal diffusion characteristics and photoacoustic sensitivity thresholds during laser pulse excitation. To avoid localized surface ablation or thermal damage due to high energy accumulation, and to ensure high-resolution signals are obtained in different regions, the upper limit of pulse energy and the number of beam irradiations must be reasonably set according to the regional characteristics. The threshold region delineation module aims to divide the region and set the energy threshold, providing fundamental support for dynamic pulse strategies.

[0050] The threshold area delineation module includes the following:

[0051] S1.1, Regional Division:

[0052] Due to the complexity of the blade's geometry and the uneven distribution of material properties, different regions exhibit significantly different sensitivities to heat diffusion and laser response. Therefore, the surface needs to be segmented to match the characteristic requirements of different regions. This is based on the blade's geometry and material roughness. and thermal diffusivity The blade surface is divided into multiple detection sub-regions, forming a region set. And record the material property matrix for each sub-region. Regional division enables granular management of subsequent laser parameter adjustments, making energy distribution more targeted, effectively reducing the risk of thermal damage and improving detection accuracy.

[0053] S1.2, Calculation of thermal diffusion parameters:

[0054] The thermal diffusion generated by laser pulse excitation has a significant impact on surface temperature rise. Accurate assessment of the thermal diffusion range helps avoid melting caused by localized overheating. This can be achieved through the thermal diffusion equation. Calculate each sub-region Surface temperature distribution, where For ambient temperature, For laser pulse energy, For material density, For the specific heat capacity of the material, This is the thermal diffusivity. The calculation results determine which regions have slower thermal diffusion and require lower pulse energy. Through thermal diffusivity calculations, the temperature rise limit for each region can be clearly defined, preventing accumulated heat from exceeding the material's tolerance threshold and providing accurate data for energy threshold setting.

[0055] S1.3, Photoacoustic sensitivity threshold evaluation:

[0056] The photoacoustic signal response in different regions is significantly affected by the laser energy conversion efficiency and the material absorption coefficient. Photoacoustic sensitivity analysis can ensure that the pulse energy meets the minimum excitation requirement. The formula is used... Compute sub-region Photoacoustic sensitivity ,in For laser conversion efficiency, For pulse power, The absorption coefficient is... The laser incident depth is defined. The minimum excitation energy requirement for each region is set based on the sensitivity assessment results. Sensitivity assessment ensures that the photoacoustic signal intensity meets detection requirements even under low-energy excitation conditions, while avoiding surface damage to the material due to excessively high energy.

[0057] S1.4, Pulse energy and frequency threshold settings:

[0058] By comprehensively considering both thermal diffusion and photoacoustic sensitivity, the upper limit of pulse energy and the threshold of irradiation frequency are precisely set to ensure a balance between thermal safety and signal sensitivity. Based on the optimized objective function: Compute sub-region upper limit of pulse energy and maximum number of irradiations .in, Indicates the cumulative pulse energy. This represents the maximum permissible temperature rise threshold. Taking into account thermal diffusion and sensitivity requirements, the energy and frequency parameters are adjusted to create a threshold table. Precise setting of pulse energy and frequency thresholds can prevent thermal damage and maintain signal quality, providing a scientific basis for dynamic scanning strategies.

[0059] S1.5, heat accumulation and energy threshold matching:

[0060] When entering each region, the thermal accumulation of the preceding sub-region needs to be comprehensively considered to ensure that the pulse excitation conditions do not violate the established safety threshold. This applies to each entry into a sub-region. Previously, based on the cumulative thermal diffusion of the preceding sub-region Adjust the current energy allocation, dynamically correct the laser power and scanning frequency, and ensure that the pulse energy limit is met. and maximum number of irradiations Require.

[0061] Specifically, the dynamic correction analysis process for the cumulative thermal diffusion is as follows:

[0062] 1) Based on the calculation method of thermal diffusion parameters, the aforementioned sub-regions are calculated and obtained. Surface temperature distribution, i.e., temperature accumulation ,in For the preceding sub-region in time Temperature accumulation; The laser pulse energy of the preceding sub-region; is the thermal diffusivity of the preceding subregion.

[0063] Accumulated temperature of the preceding sub-region Substitute the current subregion The pulse energy safety threshold formula ensures that the following conditions are met: Thus, the laser pulse energy of the current sub-region is obtained. Adjustment range: Calculate the upper limit of laser pulse energy in the current sub-region to prevent heat accumulation from exceeding the material's allowable temperature rise. .

[0064] 2) Calculate the photoacoustic sensitivity of the current sub-region based on the photoacoustic sensitivity formula. ,in, The current photoacoustic sensitivity of the sub-region; Laser energy conversion efficiency; This represents the current laser power. The material absorption coefficient; The laser incident depth is denoted as .

[0065] If the current sub-region's photoacoustic sensitivity is less than the set minimum sensitivity threshold Therefore, the laser power needs to be increased within the energy safety range. ,satisfy: When adjusting the laser power, the photoacoustic sensitivity requirements should also be taken into account to ensure that the amplitude and frequency characteristics of the acquired signal are within the effective range.

[0066] 3) Based on the maximum number of irradiations This requires calculating the laser scanning step size and frequency for the current sub-region to achieve uniform energy distribution and avoid local hotspots. Assume the current scanning range is... Step size is Then the scanning frequency Expressed as: ,in This is a threshold for the number of irradiations; For scanning speed; The area of ​​the sub-region.

[0067] If the thermal accumulation height of the preceding sub-region meets the standard, then the scanning frequency needs to be appropriately reduced and the step size increased within the current sub-region to reduce the pulse accumulation energy per unit area, satisfying the following: By dynamically adjusting the scanning frequency and step size, the pulse energy of the current sub-region is ensured to be evenly distributed, avoiding local hot spots.

[0068] 4) Based on the above adjustments, the current sub-region The dynamic correction output is:

[0069] laser power : ;

[0070] Scan frequency : ;

[0071] Ensure that the comprehensive adjustment of laser power and scanning frequency meets the requirements of thermal diffusion and sensitivity without affecting the quality of the detection signal.

[0072] By dynamically adjusting the laser power and scanning frequency, a balance between signal sensitivity and detection efficiency can be ensured while avoiding local thermal damage, providing high-quality photoacoustic data input for subsequent data processing and global fusion.

[0073] Dynamically adjusting laser parameters can adapt to changes in heat accumulation, improve laser power utilization efficiency, and further optimize thermal safety and detection results.

[0074] The threshold region delineation module provides a precise energy allocation strategy for single-crystal turbine blade detection through progressive processing logic including region division, thermal diffusion parameter calculation, photoacoustic sensitivity assessment, pulse threshold setting, and dynamic matching. This method not only ensures a balance between high-bandwidth signal acquisition and thermal safety but also lays the data foundation for subsequent dynamic scanning and signal processing.

[0075] In the threshold region planning module, pulse energy upper limits and maximum number of irradiations are set for each sub-region, and the safe excitation range and sensitivity requirements are obtained through thermal diffusion analysis and photoacoustic sensitivity assessment. The pulse modulation tracing module aims to plan dynamic pulse scheduling and scanning paths according to the zone energy thresholds, so that the laser triggering and motion control unit can work together precisely to ensure that the pulse energy meets the sub-region limits and maintains effective photoacoustic detection.

[0076] The pulse diagnosis module includes the following:

[0077] S2.1, Read and initialize pulse energy conditions:

[0078] Based on the sub-region list and the corresponding pulse energy limit With maximum number of irradiations Initialize the parameter table of the dynamic trigger module. For entering the sub-region... Pre-reading the preorder sub-region Temperature accumulation It also verifies the impact on the current maximum allowable energy and updates the current sub-region pulse energy threshold. Make it satisfy: ,in For material density, For specific heat capacity, For sub-region thermal diffusivity, This is the estimated total scan duration. The results from the threshold region delineation module are read, and the thermal safety boundary of the current sub-region is combined with sensitivity requirements to obtain the real-time allowable pulse energy range, laying the foundation for subsequent scheduling.

[0079] S2.2, Timing planning and laser triggering synchronization:

[0080] Based on the determined pulse energy threshold of the current sub-region, the triggering time and power of each pulse are arranged in a schedule within the triggering plan, and this is denoted as the pulse sequence. Compare this sequence with the scan speed of the motion control unit. and step distance Alignment is performed so that the pulse position at any given time is within the sub-region. Inside: ,in For sub-region The length in the scanning direction allows the pulse energy to be linked to the spatial range and step parameters of the current sub-region, positioning the pulse trigger in the appropriate area. This reduces ineffective excitation and ensures the laser operates with optimal energy within the target area.

[0081] S2.3, Combined adjustment of energy and stride:

[0082] If thermal accumulation increases during the scanning process, the pulse power correction amount is dynamically calculated. And increase the step size accordingly. This ensures that the cumulative energy per unit area is below the safety threshold defined by the threshold zone planning module, while also guaranteeing that the maximum number of irradiations is completed before the end of the sub-region. The following constraints apply: By coordinating the adjustment of laser power and step size in the middle of the scan, the risk of thermal damage can be minimized without sacrificing the photoacoustic signal intensity, while ensuring scanning efficiency.

[0083] S2.4, in sub-region After performing a pulse scan, the cumulative temperature rise corresponding to the echo signal is calculated. The results are then... The energy of the pulse that has been executed is recorded as a heat transfer message and passed to subsequent sub-regions. The planning module allows it to assess actual energy limitations: ,in This represents the temperature rise increment function based on pulse energy and the material's thermal diffusion characteristics. Heat estimation is the core of coherent sub-intervals; only by accurately grasping the temperature rise level of the previous stage can reasonable planning be made for the next stage. Ensuring that measured heat data is transmitted after each sub-interval provides a safety reference for subsequent stages such as signal processing in S3.

[0084] The pulse modulation tracing module reads the energy upper limit and irradiation number limit from the threshold region planning module and monitors the heat accumulation in the preceding sub-region in real time. It dynamically arranges the laser triggering sequence, pulse power, and scanning step size to ensure efficient detection path planning while meeting safety constraints. This method avoids thermal damage caused by excessive pulse energy concentration and provides a stable and comprehensive data foundation for subsequent S3 photoacoustic signal processing.

[0085] In the pulse modulation tracing module, based on sub-region pulse energy limitations and scanning strategies, pulse echo time-domain data for each sub-region has been obtained, and the corresponding energy parameter labels have been recorded using a heat estimation method. The frequency amplitude modulation module needs to perform online processing and noise suppression on this part of the time-domain data. Through frequency domain transformation, envelope extraction, and normalization correction, it obtains the separation results of broadband and narrowband components and outputs a local acoustic atlas with pulse energy label information, preparing for subsequent data fusion in S4.

[0086] The frequency amplitude modulation module includes the following:

[0087] S3.1, Read and organize the time-domain echo data:

[0088] Obtain the time-domain echo sequence for each sub-region Simultaneously read the pulse energy and irradiation count markers used for scanning the corresponding sub-region. To ensure a smooth transition to subsequent processing, the data is aligned along the timeline and a time-series index is added. This creates a unified format for preprocessing. The multiple raw waveforms obtained after scanning are integrated into a standardized data structure and mapped one-to-one with their pulse energy information, facilitating subsequent frequency domain analysis and amplitude correction.

[0089] S3.2, Pre-filtering and signal denoising:

[0090] A series of high-pass and low-pass filtering operations are performed on the time-domain echo to first remove background interference and over-range low-frequency drift, and then an adaptive truncation method is used to remove transient pulse spikes. The filtering result is denoted as... : ,in and These represent high-pass and low-pass filtering operations, respectively. Dual filtering eliminates noise interference, pulse spikes, and baseline drift, preserves key photoacoustic echo characteristics, and creates cleaner input conditions for frequency domain conversion.

[0091] S3.3, Parallel execution of frequency domain transformation and envelope extraction:

[0092] Perform a Fast Fourier Transform on the filtered time-domain echo to obtain the spectrum. The amplitude and phase information are extracted to distinguish between broadband and narrowband components. Containment extraction is performed in parallel using Hilbert transform or full-wave rectification to obtain the envelope function. The main echo segment and noise segment are separated based on threshold comparison. ,in Indicates Fourier transform, This represents the Hilbert transform. Combined time-domain and frequency-domain analysis is used to identify broadband defect characteristics and narrowband material resonances, increasing the distinguishability of various defect morphologies and providing a visual basis for subsequent amplitude correction.

[0093] S3.4, Amplitude and Phase Normalization Correction:

[0094] Calculate and obtain the actual pulse energy: ,in, For sub-region The Sub-pulse power; For the first The duration of the next pulse; For sub-region The total number of pulses. Combined with actual pulse energy. and the upper limit of sub-region pulse energy Normalization correction is applied to the waveform amplitude and phase. The correction factor is defined as follows: Multiply it by the spectrum The amplitude portion is adjusted according to preset rules to appropriately adjust the phase drift in order to compensate for the difference in detection intensity caused by energy fluctuations, denoted as... Within each sub-region, real-time heat accumulation may reduce pulse output, affecting amplitude and phase. This can be addressed by introducing a normalization correction factor. This ensures that broadband and narrowband features remain comparable across different sub-intervals.

[0095] S3.5 generates a local acoustic atlas with energy markers:

[0096] The corrected spectrum and envelope function Converted to an acoustic image format (e.g., time-frequency mapping or two-dimensional amplitude projection), and sub-region coordinates and pulse energy information Packaged together into a sub-region acoustic atlas This atlas contains broadband and narrowband analysis results and amplitude correction information, providing positioning and energy references for the next step of cross-sub-region fusion.

[0097] This ensures that the subsequent fusion and splicing module can fully utilize energy markers and amplitude normalization information when splicing in a unified coordinate system, maintaining the coherence and comparability of the acoustic distribution of the entire blade.

[0098] The frequency amplitude modulation module performs pre-filtering and denoising, frequency domain transformation, and envelope extraction on the time-domain echo data acquired by the pulse modulation tracer module. It then performs amplitude and phase correction based on pulse energy differences, finally outputting a local acoustic atlas with energy markers. This processing logic not only distinguishes between broadband and narrowband defect signals but also lays a high-quality spectral foundation for subsequent S4 cross-sub-region data fusion, ensuring the comparability of detection results even under different energy conditions in each sub-region.

[0099] In the frequency amplitude normalization module, the pulse echoes collected from each sub-region have undergone pre-filtering, frequency domain transformation, and amplitude normalization correction to finally obtain local acoustic atlases. Each atlas is bound to its sub-region coordinates and pulse energy parameters. The fusion and stitching module needs to perform spatial geometric registration and amplitude difference correction on these atlases, and stitch them together under a unified coordinate reference to form the acoustic mapping of the entire turbine blade, providing a visual basis for judging the overall health of the blade.

[0100] The fusion and integration module includes the following:

[0101] S4.1, Read the sub-region acoustic atlas and coordinate information:

[0102] Get the acoustic atlas of all sub-regions and its mapping sub-region coordinates Pulse energy information marker Integrate this data into a unified structure to ensure that each sub-region is aligned under the same coordinate reference.

[0103] S4.2, Geometric registration and connection of adjacent sub-region boundaries:

[0104] For adjacent sub-regions By matching the scan position recorded in the pulse modulation tracer module with the features of the overlapping area, a boundary correspondence is established: Then, within the corresponding overlapping area, acoustic features (such as the main echo time period or the main frequency peak) are matched to fine-tune the relative positions of the sub-regions, thereby obtaining the optimal translation or rotation matrix. Apply this transformation matrix to the subregion. The acoustic atlas is mapped to the sub-regions. Alignment position and orientation. By comparing features of overlapping areas, the relative coordinate deviation of adjacent sub-regions is calibrated to ensure seamless connection of sub-region acoustic images at the boundaries during mapping, avoiding misalignment or uneven overlap.

[0105] S4.3, Amplitude Difference Correction and Pulse Energy Merging Marker:

[0106] Because pulse energy can vary in different sub-regions due to thermal diffusion or photoacoustic sensitivity requirements, amplitude normalization processing is needed for the merged image. Let the correction factor be: ,in This represents the average of the maximum permissible pulse energy across all sub-regions. A correction factor is multiplied by the amplitude component of each pixel in the sub-region acoustic atlas to suppress overall intensity differences caused by energy inconsistencies. When merging marker information, pulse energy information markers are retained for subsequent checks.

[0107] When performing cross-sub-region mapping fusion, amplitude uniformization prevents certain areas from being significantly brighter or darker due to high energy, ensuring the consistency and comparability of the imaging. Merging marker information allows subsequent users to trace the energy parameters of each region.

[0108] S4.4, stitching together to generate a global acoustic map:

[0109] Acoustic atlas of all registered and amplitude-corrected sub-regions The images are stitched together according to a unified coordinate order. Blank or unscanned areas can be filled using nearest neighbor interpolation to obtain a complete acoustic distribution map of the blades. In this figure, the pulse energy marker layer is retained to view the energy status of each region during data acquisition.

[0110] The acoustic information of each sub-region is synthesized into an overall mapping under a unified coordinate system, providing continuously distributed visualization results for inspection personnel or subsequent algorithms, thus completing the presentation of the entire image level.

[0111] The fusion and stitching module utilizes the local atlas and pulse energy markers output by the frequency amplitude modulation module. Through geometric registration and amplitude correction, it seamlessly stitches together the acoustic images of all sub-regions into a complete blade acoustic map. This method avoids image defects caused by inconsistencies in energy or coordinates between sub-regions and maintains high resolution and comparability globally, providing an effective reference for the acoustic health assessment of large-area turbine blades.

[0112] In the acoustic distribution map of turbine blades, crystal orientation defects appear as anomalous regions in the acoustic image. Specifically, crystal orientation defects in single-crystal turbine blades, such as grain boundary dislocations, crystal orientation deviations, or microcracks, can cause disturbances in the sound wave propagation path or abnormal changes in intensity, thus exhibiting specific morphological characteristics on the distribution map. For example, grain boundary defects may appear as linear or band-shaped high-intensity or low-intensity regions, reflecting enhanced reflection or scattering of sound waves at the grain boundaries; crystal orientation deviations may appear as gradually changing amplitude anomalies, indicating a deviation in the sound wave propagation direction; microcracks may appear as isolated bright spots or dark spots, indicating local sound wave blockage or attenuation. By quantitatively analyzing the morphology, size, location, and intensity distribution of these anomalous regions, the type, spatial location, and severity of the defects can be accurately determined.

[0113] The detection process relies on the high-resolution characteristics of the blade acoustic distribution map, enabling the system to identify micron-level crystal orientation defects. In practice, the acoustic distribution map of the blade is first segmented, and the Canny edge detection algorithm is used to identify the boundary features of abnormal regions, extracting the contour information of crystal orientation defects in the acoustic image, such as the edge shape of grain boundary defects or microcracks. Next, spectral analysis of the acoustic signal is performed on each extracted abnormal region. Specifically, the time-domain signal is converted to the frequency domain using a Fast Fourier Transform (FFT), and the power spectral density of the acoustic signal within the abnormal region is calculated to analyze its spectral components. For example, grain boundary defects exhibit energy concentration in the high-frequency range (10-20MHz) due to strong reflection or scattering of sound waves at the grain boundary; while microcracks show significant energy attenuation in the low-frequency range (1-5MHz) because sound waves are blocked at the crack. These spectral characteristics allow for a clear distinction of the physical properties of the defects. Finally, using a pre-established unified coordinate system of the blade geometry model and acoustic distribution map, the spatial location of the abnormal area is mapped to the specific geometric area of ​​the blade through coordinate transformation, accurately calibrating the three-dimensional coordinates of the defect. For example, a grain boundary defect is located at the blade root at a distance of (0,0,0) from the reference point (x=5mm, y=10mm, z=2mm). Simultaneously, the system makes a judgment based on preset defect evaluation criteria, such as:

[0114] Defect size threshold: The length of the linear region of a grain boundary defect exceeds 0.5 mm, or the area of ​​a bright spot of a microcrack exceeds 0.1 mm²;

[0115] Defect intensity variation rate: The ratio of the acoustic signal intensity in the abnormal area to that in the surrounding normal area exceeds 1.5 times.

[0116] By comprehensively analyzing the size and strength change rate of defects, the degree of their impact on the mechanical properties of the blade can be determined. For example, if a grain boundary defect has a length of 0.8 mm and a strength ratio of 2.0, it may significantly reduce the fatigue resistance of the blade, thereby assessing the safety and service life of the blade.

[0117] By leveraging adaptive optics to enhance signal quality and the global coverage of acoustic mapping, the sensitivity and accuracy of crystal orientation defect detection are ensured. At the same time, rapid data processing and image generation significantly shorten the detection cycle, meeting the requirements for efficient and high-precision non-destructive testing of aero-engine components. This provides reliable technical support for the quality assurance and safe operation of single-crystal turbine blades.

[0118] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0119] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0120] It should be noted that, in this document, the use of relational terms such as "first" and "second" is merely to distinguish one entity or operation from another, and does not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0121] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A defect detection system for the crystal orientation of a single-crystal turbine blade based on adaptive optics, characterized in that, include: The module includes a threshold area delineation module, a pulse tone tracking module, a frequency amplitude modulation module, and a fusion and splicing module. Threshold Zone Planning Module: It undertakes the thermal diffusion analysis and photoacoustic sensitivity determination of the blade surface. By dividing the entire blade into several sub-regions, it calculates the pulse energy upper limit and irradiation number threshold for each sub-region, and checks the heat accumulation of the previous scan before entering the sub-region, and dynamically adjusts the pulse distribution strategy. The output includes sub-region coordinates, thermal diffusion parameters, pulse upper limit and number of pulses information to the pulse modulation tracer module; Pulse modulation and tracing module: Generates pulse trigger timing and stepping path based on the energy threshold and scanning range of each sub-region, calculates heat estimation results in real time and updates actual pulse energy in each scan, and finally outputs the echo sequence and dynamic energy record of each sub-region to the frequency amplitude modulation module. Frequency amplitude normalization module: performs pre-filtering, frequency domain conversion and envelope extraction on pulse echo, distinguishes between broadband and narrowband components, and performs amplitude normalization correction based on the actual pulse energy calculation results of each sub-region. The output includes local acoustic atlas and pulse energy markers to the fusion and stitching module. The fusion and stitching module performs geometric registration based on the sub-region coordinates and overlapping boundary features. It achieves cross-sub-region amplitude consistency through amplitude correction and stitches them together under unified coordinates to form the acoustic mapping of the entire blade, thus obtaining the blade acoustic distribution map after cross-sub-region fusion.

2. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 1, characterized in that: The threshold area delineation module includes the following: S1.1, based on blade geometry and material roughness and thermal diffusivity The blade surface is divided into multiple detection sub-regions, forming a region set. And record the material property matrix for each sub-region. ; S1.2, through the thermal diffusion equation Calculate each sub-region Surface temperature distribution, where For ambient temperature, For laser pulse energy, For material density, For the specific heat capacity of the material, The thermal diffusivity; S1.3, using formula Compute sub-region Photoacoustic sensitivity ,in For laser conversion efficiency, For pulse power, The absorption coefficient is... The laser incident depth; S1.4, considering both thermal diffusion and photoacoustic sensitivity, sets an upper limit for pulse energy and a threshold for the number of irradiations, based on the optimization objective function: Compute sub-region upper limit of pulse energy and maximum number of irradiations ,in Indicates the cumulative pulse energy. The maximum permissible temperature rise threshold; S1.5, each time a sub-region is entered Previously, based on the cumulative thermal diffusion of the preceding sub-region Adjust the current energy allocation, dynamically correct the laser power and scanning frequency, and ensure that the pulse energy limit is met. and maximum number of irradiations Require.

3. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 2, characterized in that: The pulse diagnosis module includes the following: S2.1, based on the sub-region list and the corresponding pulse energy limit With maximum number of irradiations Initialize the parameter table of the dynamic trigger module; for entering the sub-region Pre-reading the preorder sub-region Temperature accumulation It also verifies the impact on the current maximum allowable energy and updates the current sub-region pulse energy threshold. Make it satisfy: ,in For material density, For specific heat capacity, For sub-region thermal diffusivity, This represents the estimated total scan time. S2.2, Based on the determined pulse energy threshold of the current sub-region, arrange the triggering time and power of each pulse in the triggering plan using a schedule, denoted as the pulse sequence. ; Compare this sequence with the scan speed of the motion control unit and step distance Alignment is performed so that the pulse position at any given time is within the sub-region. Inside: ,in For sub-region Length in the scanning direction.

4. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 3, characterized in that: The pulse diagnosis module also includes the following: S2.3, If thermal accumulation increases during the scanning process, dynamically calculate the pulse power correction amount. And increase the step size accordingly. This ensures that the cumulative energy per unit area is below the safety threshold defined by the threshold zone planning module, while also guaranteeing that the maximum number of irradiations is completed before the end of the sub-region. The following constraints apply: By coordinating the adjustment of laser power and step size in the middle of the scan, the risk of thermal damage is reduced without sacrificing the photoacoustic signal intensity, while ensuring scanning efficiency. S2.4, in sub-region After performing a pulse scan, the cumulative temperature rise corresponding to the echo signal is calculated; the results are then... The energy of the pulse that has been executed is recorded as a heat transfer message and passed to subsequent sub-regions. The planning module allows it to assess actual energy limitations: ,in This represents the temperature rise increment function based on pulse energy and the thermal diffusion characteristics of the material.

5. The defect detection system for the crystal orientation of a single-crystal turbine blade based on adaptive optics according to claim 4, characterized in that: The frequency amplitude modulation module includes the following: S3.1, Obtain the time-domain echo sequence for each sub-region. Simultaneously read the pulse energy and irradiation count markers used for scanning the corresponding sub-region. Align the data according to the timeline and attach a time-series index. This forms a unified format that can be used for preprocessing. S3.2, perform serial high-pass filtering and low-pass filtering on the time-domain echo to first remove background interference from the equipment and low-frequency drift due to over-range, and then combine the adaptive truncation method to remove transient pulse spikes. The filtering result is denoted as : ,in and These represent high-pass and low-pass filtering operations, respectively.

6. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 5, characterized in that: The frequency amplitude modulation module also includes the following: S3.3, Perform a Fast Fourier Transform on the filtered time-domain echo to obtain the spectrum. The amplitude and phase information are extracted to distinguish between broadband and narrowband components. Envelope extraction is performed in parallel using Hilbert transform or full-wave rectification to obtain the envelope function. And based on threshold comparison, the main echo band and the noise band are separated: ,in Indicates Fourier transform, Represents the Hilbert transform; S3.4, Calculate and obtain the actual pulse energy: ,in, For sub-region The Sub-pulse power; For the first The duration of the next pulse; For sub-region The total number of pulses; Combined with real pulse energy and the upper limit of sub-region pulse energy Normalize the waveform amplitude and phase; define the correction factor as... Multiply it by the spectrum The amplitude portion is adjusted according to preset rules to compensate for the difference in detection intensity caused by energy fluctuations, denoted as... .

7. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 6, characterized in that: The frequency amplitude modulation module also includes the following: S3.5, the corrected spectrum and envelope function Converted to acoustic image format, with sub-region coordinates and pulse energy information Packaged together into a sub-region acoustic atlas .

8. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 7, characterized in that: The fusion and integration module includes the following: S4.1, Obtain the acoustic atlas of all sub-regions. and its mapping sub-region coordinates Pulse energy information marker Integrate this data into a unified structure to ensure that each sub-region is aligned under the same coordinate reference; S4.2, for adjacent sub-regions By matching the scan position recorded in the pulse modulation tracer module with the features of the overlapping area, a boundary correspondence is established: Then, acoustic features are matched within the corresponding overlapping areas to fine-tune the relative positions of the sub-regions, obtaining the optimal translation or rotation matrix. Apply this transformation matrix to the sub-region The acoustic atlas is mapped to the sub-regions. Alignment position and direction.

9. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 8, characterized in that: The fusion and assembly module also includes the following: S4.3 Since the pulse energy in different sub-regions may differ due to thermal diffusion or photoacoustic sensitivity requirements, the amplitude of the merged image needs to be uniformized. Let the correction factor be: ,in This represents the average value of the maximum permissible pulse energy for all sub-regions.

10. The defect detection system for crystal orientation of single-crystal turbine blades based on adaptive optics according to claim 9, characterized in that: The fusion and assembly module also includes the following: S4.4, Acoustic atlas of all registered and amplitude-corrected sub-regions The images are stitched together according to a unified coordinate order. Blank or unscanned areas are filled using nearest neighbor interpolation to obtain a complete acoustic distribution map of the blades. ; The Canny edge detection algorithm is used to segment the acoustic distribution map of the blade to extract the boundary features of the abnormal area. The acoustic signal of the abnormal area is analyzed by fast Fourier transform to distinguish the physical properties of the defect. The defect is located to the specific geometric area of ​​the blade by coordinate transformation. The impact of the defect on the mechanical performance of the blade is evaluated according to the preset defect size threshold and intensity change rate.

Images