A method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition

Abstract

The present invention belongs to the field of non-destructive testing, and discloses a method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition. On the basis of full data acquisition, multiple gates are set to extract ultrasonic signals, each gate forms an image respectively, and defect classification is performed according to the image display. The present invention is based on the full data acquisition method, and all structural information is detected and collected at one time. For multi-layer adhesive structures, the imaging display gates are subsequently set twice to respectively realize the internal quality assessment of the stiffeners and skins in the composite wall panels and the skin-stiffener bonding area. The detectable defect types include: delamination, debonding, and inclusions; the detectable defect depth covers the entire part depth; and the internal quality of different structures is displayed separately as needed. The image analysis is simple, the analysis object is clearer, the analysis results are easier to understand, and it is more friendly to the internal quality analysis of complex structures. It has a wider adaptability, improves the detection efficiency, and maintains the stability of defect assessment.

Description

Technical Field

[0001] The invention relates to the field of nondestructive testing, and in particular to a method for ultrasonic signal extraction and defect classification of composite material parts based on full data acquisition. Background Art

[0002] Composite stiffened panel structures are widely used in wing skin panels, fuselage panels, vertical tails and horizontal tails. Their internal quality is closely related to the safety of aircraft service. Defect detection of such parts is the focus of attention in the composite manufacturing process. Ultrasonic testing is currently the most effective and widely used non-destructive testing technology for composite materials. Its testing process mainly includes three steps: (a) ultrasonic signal acquisition; (b) ultrasonic signal extraction; (c) classification of typical defects.

[0003] Common ultrasonic signal acquisition modes include peak acquisition and full data acquisition. Peak acquisition means that the system only collects information about the highest amplitude signal within a range. Full data acquisition retains all ultrasonic signals within the time domain. Generally speaking, the defects inside metal parts are usually small-sized defects relative to the width of the ultrasonic beam, that is, the structural wave (bottom wave) of the part exists stably in any area. Therefore, the signal acquisition mode can be set to peak acquisition during the automated detection process. Its characteristic is that an imaging gate that changes with the position of the structural wave (that is, the thickness of the part) is set between the surface wave and the structural wave (such as the bottom wave), and only the peak information of the defect signal between the surface wave and the structural wave (that is, inside the part) is extracted for detection imaging and defect evaluation. The defect size in composite materials is often large. In the defective area, the ultrasonic beam is "blocked" by the defect and cannot be transmitted to the bottom surface of the part, resulting in the disappearance of the structural wave (bottom wave) of the part, and it is impossible to set a follow-up gate such as metal part detection for defect signal extraction and imaging detection evaluation.

[0004] The full data acquisition mode of ultrasonic testing retains ultrasonic signals in the entire time domain and does not require the setting of a signal acquisition gate. It is not affected by any factors such as changes in part thickness, the presence of part structural waves, signal amplitude and position, etc., thus avoiding erroneous and missed signal acquisition, which is beneficial to the subsequent analysis and extraction of defect features. It is an ideal method for composite parts testing.

[0005] But there are still problems: the signals obtained from full data collection are numerous, containing a large amount of redundant and irrelevant information. It is necessary to develop an ultrasonic signal extraction and defect classification method. First, the signals that are beneficial to the internal quality assessment of the parts should be extracted, which can not only show more structural information, but also promote defect imaging display, and classify each defect to improve the overall efficiency of the internal quality assessment of the parts. Summary of the invention

[0006] The present invention proposes a method for ultrasonic signal extraction and defect classification of composite material parts based on full data acquisition, that is, on the basis of full data acquisition, multiple gates are set to extract ultrasonic signals, each gate forms an image respectively, and defect classification is performed according to the image display. It is suitable for composite material reinforced wall panel parts, and typical part structures include laminated structures and plate-to-plate adhesive structures. It mainly solves the ultrasonic detection and defect evaluation of composite material reinforced wall panel parts.

[0007] The technical solution of the present invention is as follows: a method for ultrasonic signal extraction and defect classification of composite material parts based on full data acquisition. On the basis of full data acquisition, multiple gates are set to extract ultrasonic signals, each gate forms an image respectively, and defect classification is performed according to the image display.

[0008] The composite material part is a composite material reinforced wall panel, including: skin A, film B, and rib C; the signals collected by the probe D at different positions are different; when the probe D is at position E, the signal includes surface wave 1, material noise wave 2, skin bottom wave 3, and defect wave 4; when the probe D is at position F, the signal includes surface wave 1, material noise wave 2, defect wave 4, film wave 5, and rib bottom wave 6.

[0009] The full data acquisition is to collect all signals in the full time domain; time is multiplied by the speed of sound to obtain depth information, and the amplitude information and depth information of the defect signal are obtained at the same time, and are displayed as an image.

[0010] The multiple gates include gate A7, gate B8, gate C9 and gate D10 respectively; the starting position of gate A7 is located before surface wave 1, and the ending position is located after surface wave 1. The threshold of gate A7 is higher than the signal amplitude of material noise wave 2, which can be 2 to 3 times the signal amplitude of material noise wave 2; the signal feature extraction method in gate A7 is: extracting the depth information of the first wave peak higher than the threshold of gate A7.

[0011] The starting position of gate B8 is located at θ1 after surface wave 1, and the ending position is located at θ2 after surface wave 1. The threshold of gate B8 is twice the average amplitude of the material noise wave 2 signal. The signal feature extraction method in gate B8 is: extracting the amplitude information and depth information of the highest amplitude signal in gate B8; wherein, θ1 is 3 times the single layer thickness of the composite material prepreg, and θ2 is 5 times the single layer thickness of the composite material prepreg.

[0012] The starting position of gate C9 is located at θ2 after surface wave 1, and the ending position is located at 1.0 mm after the bottom wave 6 of the reinforcing rib lying edge; the threshold of gate C9 is three times the average level τ of the signal amplitude of material noise wave 2; the signal feature extraction method in gate C9 is: extracting the amplitude information and depth information of the first wave peak higher than the threshold of gate C9.

[0013] The starting position of the gate D10 is located at θ2 after the surface wave 1, and the ending position is 1.0mm after the bottom wave 6 of the reinforcing rib lying edge; the threshold of the gate D10 is twice the average level τ of the signal amplitude of the material noise wave 2; the signal feature extraction method in the gate D10 is: extracting the amplitude information and depth information of the maximum wave peak in the gate D10.

[0014] In the defect classification, it is assumed that only one defect exists at a certain position of the composite material part, and more than one defect does not exist at the same time, and the minimum defect size to be detected is required to be larger than the probe size;

[0015] An ultrasonic signal contains two parts of information, one is the amplitude information of the signal (i.e. the height of the signal), and the other is the position information of the signal (i.e. the depth of the signal). The amplitude information of the ultrasonic signal can be used to form an amplitude C-scan image; the depth information of the ultrasonic signal can be used to form a depth C-scan image. The height of the defect signal in the part can be seen from the amplitude C-scan image, and the depth of the defect signal in the part can be seen from the depth C-scan image.

[0016] Each gate generates an image separately, and forms an amplitude C-scan image and a depth C-scan image respectively, and the defect size and area are measured through the amplitude C-scan image and the depth C-scan image;

[0017] Analyze each C-scan image in turn; T A is the thickness of the skin A corresponding to the position of the probe D, T C is the total thickness of the skin A, film B and rib C corresponding to the position of the probe D, τ is the average amplitude of the material noise wave 2 signal, A VB is the average amplitude level of the film wave (5) at the skin-stiffener interface; based on the above parameters, analyze the gates one by one to determine the defect category.

[0018] The gate B8 analyzes its amplitude C-scan image A8 and depth C-scan image T8:

[0019] For A8 ≥ 8τ and T8 ≥ 28 mm 2 The area is characterized as a delamination defect in the skin;

[0020] For 8τ>A8≥4τ and T8≥28mm 2 The area is characterized as a defect display of inclusions in the skin;

[0021] The gate C9 analyzes its amplitude C-scan image A9 and depth C-scan image T9:

[0022] For T9<T A And the area with A9 ≥ 8τ is qualitatively characterized as a delamination defect display within the skin;

[0023] For T9<TA And the area where 8τ>A9≥3τ is qualitatively characterized as the display of inclusion defects in the skin;

[0024] For T9 = T A , if the probe is in the "E position", the area is characterized as a defect-free area;

[0025] For T9 = T A If the probe is in the "F position", the area is identified as a skin-long stringer debonding defect area;

[0026] The gate D10 analyzes the amplitude of the C scan image A 10 , depth C scan image T 10 :

[0027] For T 10 <T A And A 10 The area ≥8τ is qualitatively characterized as a delamination defect display within the skin;

[0028] For T 10 <T A And 8τ>A 10 The area ≥4τ is qualitatively characterized as inclusion defect display in the skin;

[0029] For T 10 =T A =T9 and A 10 =A VB The area is characterized as a defect display of inclusions on the lying side of the long stringer;

[0030] For T 10 =T A =T9 and A 10 The area with a value of >8τ is characterized by debonding defects at the skin-string interface;

[0031] For T 10 >T A And A 10 The area with a value of ≥8τ is characterized as a delamination defect inside the lying side of the long stringer;

[0032] For T 10 >T A And 8τ>A 10 The area with a value of ≥4τ is characterized as a defect display inside the lying edge of the long stringer;

[0033] For T 10 =T C And T9<T A The area is characterized as a defect display of inclusions in the skin;

[0034] For T 10 =T CAnd T9 = T A And A 10 The area ≥6τ is characterized as defect-free area;

[0035] For T 10 =T C And T9 = T A And A 10 The area of ​​<6τ is characterized as a defect display of inclusion on the horizontal side of the long stringer.

[0036] The beneficial effects of the present invention are as follows: based on the full data collection method, all structural information can be collected in one detection. For multi-layer adhesively bonded structures, the imaging display gates are subsequently set twice to respectively realize the internal quality evaluation of the stiffeners and skins in the composite wall panels and the skin-stiffener bonding area. The detectable defect types include: delamination, debonding, and inclusions; the detectable defect depth covers the entire part depth; and the internal quality of different structures can be displayed separately as needed. The image analysis is simple, the analysis object is clearer, the analysis results are easier to understand, and the internal quality analysis of complex structures is more friendly. The process is clear and simple, and it is applicable to the same or other similar structures. It has a wide adaptability, can improve the detection efficiency, and maintain the stability of defect evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1(a) is a schematic diagram of full data acquisition; Figure 1(b) is a schematic diagram of acquisition signals at position one; Figure 1(c) is a schematic diagram of acquisition signals at position two;

[0038] Figure 2(a) is a schematic diagram of the gate setting at position one;

[0039] Figure 2(b) is a schematic diagram of the gate setting at position two.

[0040] A is the skin, B is the film, C is the rib, D is the probe, E is position one, and F is position two; 1 is the surface wave, 2 is the material noise wave, 3 is the skin bottom wave, 4 is the defect wave, 5 is the film wave, 6 is the rib lying bottom wave, 7 is gate A, 8 is gate B, 9 is gate C, and 10 is gate D. DETAILED DESCRIPTION

[0041] The typical structure of composite reinforced wall panels includes: skin A, film B, and reinforcement C. Full data collection is implemented during the automated detection process, and all signals within the full time domain are collected. Figure 1(a)-Figure 1(c) As shown, the signals collected by probe D at different positions are different. When the probe is at position 1 E, the signals include surface wave 1, material noise wave 2, skin bottom wave 3, and defect wave 4; when the probe is at position 2 F, the signals include surface wave 1, material noise wave 2, defect wave 4, film wave 5, and reinforcing rib lying bottom wave 6. Time multiplied by the speed of sound equals distance (depth).

[0042] The ultrasonic testing parameters are set according to the testing specifications, including the time baseline range, probe frequency, filter, gain, delay distance, etc. Depth gain compensation is performed according to the thickness variation range of the composite wall panel parts, so that the bottom reflection signal of the defect-free area in different thickness areas can reach about 80% FSH, and the full data acquisition mode is set.

[0043] 1. Ultrasonic signal extraction:

[0044] Ultrasonic signals are extracted by setting a gate, and the gate settings include the starting position, the ending position, the gate threshold, and the signal feature extraction method within the gate, as shown in Figure 2(a) and Figure 2(b).

[0045] 1.1. Set gate A7. The starting position of gate A7 is before surface wave 1, and the ending position is after surface wave 1. The gate threshold is slightly higher than the signal amplitude of material noise wave 2. The signal feature extraction method in the gate is set as follows: the depth information of the first wave peak higher than the gate threshold.

[0046] The function of gate A7 is to extract the depth information of surface wave 1 signal, which is then used as the depth zero reference of the gate for defect feature extraction to avoid the change of distance between the probe and the part surface due to part profiling and shape curvature during automated inspection, thereby causing all signals to move and wander, leading to disorder in subsequent defect feature extraction;

[0047] 1.2. Set gate B8. The starting position of gate B8 is located at θ1 after surface wave 1, and the ending position is located at θ2 after surface wave 1. The gate threshold is about twice the average amplitude of the material noise wave 2 signal. The signal feature extraction method in the gate is set as: the amplitude and depth information of the highest amplitude signal in the gate. Among them, θ1 is 3 times the single layer thickness of the composite material prepreg, and θ2 is 5 times the single layer thickness of the composite material prepreg.

[0048] Because fiber-reinforced resin-based composite materials are anisotropic materials, the ultrasonic attenuation caused by material attenuation is large. In order to improve the ability to detect far-surface defects of thick parts, it is often necessary to perform depth-gain compensation on the material attenuation that is positively correlated with the depth, which causes the amplitude of the noise signal of the near-surface material noise wave 2 to increase, affecting the evaluation of near-surface defects. Gate B8 is used to separately monitor the internal quality of the 3-5 layers of prepreg near the surface of the part to avoid the adverse effects of the near-surface noise signal of the material noise wave 2 on the feature extraction of other defect signals.

[0049] 1.3. Set gate C9. The starting position of gate C9 is located at θ2 after surface wave 1, and the ending position is located 1.0 mm after the bottom wave of the maximum thickness of the part (i.e., the bottom wave of the 6th reinforcing rib lying edge). The gate threshold is approximately three times the average level τ of the signal amplitude of material noise wave 2. The signal feature extraction method in the gate is set as follows: the amplitude information and depth information of the first peak above the gate threshold.

[0050] The calculation formula of τ is as shown in formula (1).

[0051] τ=∑τ i ÷n (1)

[0052] Where i = 1, 2, 3, ..., n

[0053] 1.4. Set gate D10. The starting position of gate D10 is located at θ2 after surface wave 1, and the ending position is located 1.0 mm after the bottom wave of the maximum thickness of the part (i.e., the bottom wave 6 of the lying edge of the reinforcement rib). The gate threshold is approximately twice the average level τ of the signal amplitude of material noise wave 2. The signal feature extraction method in the gate is set as: the amplitude information and depth information of the maximum peak in the gate.

[0054] 2. Defect classification:

[0055] Based on the above ultrasonic signal extraction method, a defect classification method is proposed, which involves common internal defects in composite panel parts, including delamination, debonding, and inclusion defects. The delamination and inclusion defects can be evaluated inside the skin and the lying edge of the long stringer, and the debonding defects can be evaluated at the skin-long stringer bonding interface.

[0056] The present invention assumes that there is only one defect in a certain position of a part, not more than one defect at the same time, and requires that the minimum defect size that can be detected is larger than the probe size (generally Φ6mm). The specific defect classification method is:

[0057] First, each gate generates an image separately, and forms an amplitude C-scan image and a depth C-scan image respectively. Each image can measure the defect size and area.

[0058] Then, each C-scan image is analyzed in turn. In the following description, T A is the thickness of the skin A corresponding to the position of the D probe, T C is the total thickness of the skin A, film B and the long stringer side C corresponding to the position of the D probe, τ is the average amplitude of the material noise signal, A VB is the average level of film B-wave amplitude at the skin-stringer interface;

[0059] A VB The calculation formula is as shown in formula (2).

[0060] A VB =∑AB i ÷n (2)

[0061] Where i = 1, 2, 3, ..., n

[0062] Analyze gates AD one by one:

[0063] (1) For gate B8, analyze its amplitude C-scan image A8 and depth C-scan image T8:

[0064] ①For A8≥8τ and T8≥28mm 2 The area is characterized as a delamination defect in the skin;

[0065] ②For 8τ>A8≥4τ and T8≥28mm 2 The area is characterized as a defect display of inclusions in the skin;

[0066] (2) For gate C9, analyze its amplitude C-scan image A9 and depth C-scan image T9:

[0067] ①For T9<T A And the area with A9 ≥ 8τ is qualitatively characterized as a delamination defect display within the skin;

[0068] ②For T9<T A And the area where 8τ>A9≥3τ is qualitatively characterized as the display of inclusion defects in the skin;

[0069] ③For T9=T A , if the probe is in the "E position", the area is characterized as a defect-free area;

[0070] ④For T9=T A If the probe is in the "F position", the area is identified as a skin-long stringer debonding defect area;

[0071] (3) For gate D10, analyze its amplitude C scan image A 10 , depth C scan image T 10 :

[0072] ①For T 10 <T A And A 10 The area ≥8τ is qualitatively characterized as a delamination defect display within the skin;

[0073] ②For T 10 <T A And 8τ>A 10 The area ≥4τ is qualitatively characterized as inclusion defect display in the skin;

[0074] ③For T 10 =T A =T9 and A10 =A VB The area is characterized as a defect display of inclusions on the lying side of the long stringer;

[0075] ④For T 10 =T A =T9 and A 10 The area with a value of >8τ is characterized by debonding defects at the skin-string interface;

[0076] ⑤For T 10 >T A And A 10 The area with a value of ≥8τ is characterized as a delamination defect inside the lying side of the long stringer;

[0077] ⑥For T 10 >T A And 8τ>A 10 The area with a value of ≥4τ is characterized as a defect display inside the lying edge of the long stringer;

[0078] ⑦For T 10 =T C And T9<T A The area is characterized as a defect display of inclusions in the skin;

[0079] ⑧For T 10 =T C And T9 = T A And A 10 The area ≥6τ is characterized as defect-free area;

[0080] ⑨For T 10 =T C And T9 = T A And A 10 The area of ​​<6τ is characterized as a defect display of inclusion on the horizontal side of the long stringer.

[0081] Take the inspection of a composite wall panel as an example. The wall panel is a skin + long stringer structure. The skin is partially bonded with long strings. The skin thickness ranges from 2mm to 5mm, the long stringer thickness ranges from 2mm to 3mm, and the overall thickness range of the part is 2mm to 8mm. A comparison test block with the same structure and thickness as the part is used to adjust the detection sensitivity. Multiple artificial defects of different depths are embedded in the comparison test block, and the size of the artificial defects meets the part acceptance requirements. The specific implementation steps are:

[0082] Step 1: Use the automated reflection method to ultrasonically inspect composite wall panel parts, use a comparison test block to adjust the system detection sensitivity, set the ultrasonic detection parameters and mechanical scanning parameters, and set the data collection mode to full data collection;

[0083] Step 2: Set 4 defect feature extraction gates, namely Gate A7, Gate B8, Gate C9, Gate D10, and set the gate's starting position, end position, gate threshold, and gate defect feature extraction method:

[0084] Step 2.1: Set gate A7, with the starting position at (surface wave -1mm), the ending position at (surface wave +1mm), the gate threshold at 10% FSH, and the signal feature extraction method within the gate is the depth information of the first peak above the gate threshold;

[0085] Step 2.2: Set gate B8, with the starting position at (surface wave + θ1) and the ending position at (surface wave + θ2), the gate threshold is 20% FSH, and the signal feature extraction method within the gate is set to the amplitude and depth information of the highest amplitude signal within the gate;

[0086] Step 2.3: Set gate C9, with the starting position at (surface wave + θ2), the ending position at (part maximum thickness bottom wave + 1.0 mm), the gate threshold at 30% FSH, and the signal feature extraction method within the gate set to the amplitude information and depth information of the first wave peak above the gate threshold;

[0087] Step 2.4: Set the gate D10, with the starting position at (surface wave + θ2), the ending position at (part maximum thickness bottom wave + 1.0mm), the gate threshold at 20% FSH, and the signal feature extraction method within the gate set to the amplitude information and depth information of the maximum peak within the gate;

[0088] Step 3: First test the comparison block, then test the parts;

[0089] Step 4: Compare and analyze each C-scan image to qualitatively and quantitatively identify the defects.

Claims

1. A method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition, characterized in that: On the basis of full data collection, multiple gates are set to extract ultrasonic signals, each gate forms an image, and defects are classified according to the image display.

2. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to claim 1 is characterized in that: The composite material part is a composite material reinforced wall panel, comprising: a skin (A), a film (B), and a reinforcing rib (C); different signals are collected when the probe (D) is located at different positions; when the probe (D) is located at position one (E), the signal includes a surface wave (1), a material noise wave (2), a skin bottom wave (3), and a defect wave (4); when the probe (D) is located at position two (F), the signal includes a surface wave (1), a material noise wave (2), a defect wave (4), a film wave (5), and a reinforcing rib lying bottom wave (6).

3. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to claim 2 is characterized in that: The full data acquisition is to collect all signals in the full time domain; time is multiplied by the speed of sound to obtain depth information, and the amplitude information and depth information of the defect signal are obtained at the same time, and are displayed as an image.

4. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to claim 3 is characterized in that: The multiple gates include gate A (7), gate B (8), gate C (9) and gate D (10); the starting position of gate A (7) is located before the surface wave (1), and the ending position is located after the surface wave (1); the threshold of gate A (7) is higher than the signal amplitude of the material noise wave (2); the signal feature extraction method in gate A (7) is: extracting the depth information of the first wave peak higher than the threshold of gate A (7).

5. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to claim 4 is characterized in that: The starting position of the gate B (8) is located at θ1 after the surface wave (1), and the ending position is located at θ2 after the surface wave (1). The threshold of the gate B (8) is twice the average amplitude of the material noise wave (2) signal. The signal feature extraction method in the gate B (8) is: extracting the amplitude and depth information of the highest amplitude signal in the gate B (8); wherein θ1 is 3 times the single layer thickness of the composite material prepreg, and θ2 is 5 times the single layer thickness of the composite material prepreg.

6. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to claim 4 is characterized in that: The starting position of the gate C (9) is located at θ2 after the surface wave (1), and the ending position is located at 1.0 mm after the bottom wave (6) of the reinforcing rib lying side; the threshold of the gate C (9) is three times the average level τ of the signal amplitude of the material noise wave (2); the signal feature extraction method in the gate C (9) is: extracting the amplitude information and depth information of the first wave peak that is higher than the threshold of the gate C (9).

7. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to claim 4 is characterized in that: The starting position of the gate D (10) is located at θ2 after the surface wave (1), and the ending position is 1.0 mm after the bottom wave (6) of the reinforcing rib; the threshold of the gate D (10) is twice the average level τ of the signal amplitude of the material noise wave (2); the signal feature extraction method in the gate D (10) is: extracting the amplitude information and depth information of the maximum wave peak in the gate D (10).

8. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to any one of claims 4 to 7, characterized in that: In the defect classification, it is assumed that there is only one defect at a certain position of the composite material part, and there should not be more than one defect at the same time, and the minimum defect size to be detected is required to be larger than the probe size; Each gate generates an image separately, and forms an amplitude C-scan image and a depth C-scan image respectively, and the defect size and area are measured through the amplitude C-scan image and the depth C-scan image; Analyze each C-scan image in turn; T A is the thickness of the skin (A) corresponding to the position of the probe (D), T C is the total thickness of the skin (A), film (B) and rib (C) corresponding to the position of the probe (D), τ is the average amplitude of the material noise wave (2), A VB is the average amplitude level of the film wave (5) at the skin-stiffener interface; based on the above parameters, analyze the gates one by one to determine the defect category.

9. The method for ultrasonic signal extraction and defect classification of composite parts based on full data acquisition according to claim 8, characterized in that: The gate B (8) analyzes its amplitude C-scan image A8 and depth C-scan image T8: For A8 ≥ 8τ and T8 ≥ 28 mm 2 The area is characterized as a delamination defect in the skin; For 8τ>A8≥4τ and T8≥28mm 2 The area is characterized as a defect display of inclusions in the skin; The gate C (9) analyzes its amplitude C scan image A9 and depth C scan image T9: For T9<T A And the area with A9 ≥ 8τ is qualitatively characterized as a delamination defect display within the skin; For T9<T A And the area where 8τ>A9≥3τ is qualitatively characterized as the display of inclusion defects in the skin; For T9 = T A , if the probe is in "E position", the area is characterized as a defect-free area; For T9 = T A If the probe is in the "F position", the area is identified as a skin-reinforcement debonding defect area; the gate D (10) is analyzed for its amplitude C scanning image A 10 , depth C scan image T 10 : For T 10 <T A And A 10 The area ≥8τ is qualitatively characterized as a delamination defect display within the skin; For T 10 <T A And 8τ>A 10 The area ≥4τ is qualitatively characterized as inclusion defect display in the skin; For T 10 =T A =T9 and A 10 =A VB The area is characterized as a defect display of reinforcement lying edge inclusion; For T 10 =T A =T9 and A 10 The area with a value of >8τ is characterized as a debonding defect at the skin-stiffener interface; For T 10 >T A And A 10 The area with a value of ≥8τ is characterized as a delamination defect inside the lying edge of the reinforcement; For T 10 >T A And 8τ>A 10 The area ≥4τ is characterized as inclusion defect display inside the lying edge of the reinforcement; For T 10 =T C And T9<T A The area is characterized as a defect display of inclusions in the skin; For T 10 =T C And T9 = T A And A 10 The area ≥6τ is characterized as defect-free area; For T 10 =T C And T9 = T A And A 10 The area of ​​<6τ is characterized as a defect display of reinforcement edge inclusion.

Images