Method for quantitatively describing morphological characteristics of beta spots in TC17 titanium alloy cast ingot

Abstract

The invention relates to a method for quantitatively describing morphological characteristics of beta spots in a TC17 titanium alloy cast ingot, and belongs to the technical field of metal material detection. The method comprises the following steps: A, preparing and treating a sample; b, image analysis and feature extraction; and C, fractal dimension calculation. The fractal dimension is innovatively introduced to quantitatively characterize the beta spot defect in the solidification structure of the TC17 titanium alloy cast ingot. Compared with a traditional qualitative or semi-quantitative method, the fractal dimension can reflect complex morphological characteristics of the beta spots more carefully and comprehensively, and the quantitative statistical analysis problem caused by the characteristics of a large number of beta spots, emission, irregular shapes and the like is effectively solved. Through the quantitative analysis of the fractal dimension on the beta spot morphology, the dependence on artificial experience judgment is obviously reduced, the error generated by subjective recognition is reduced, and the refined and objective comparison of the morphology characteristics of different beta spots is realized.

Description

Technical Field

[0001] This invention relates to a method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots, belonging to the field of metal material testing technology. Background Technology

[0002] Near-β type titanium alloy TC17 (Ti-5Al-2Sn-2Zr-4Mo-4Cr) is widely used in key rotating components of aero-engines (such as compressor discs and drums) due to its excellent high strength, fracture toughness, and hardenability. The performance and safety of these components are highly dependent on the quality of the raw material ingots.

[0003] β spots are a typical harmful metallurgical defect in near-β titanium alloy ingots. They are essentially caused by the segregation of alloying elements (especially β-stabilizing elements such as Cr and Mo) during solidification, leading to a significant decrease in the β-phase transformation point in localized areas. During subsequent hot working (such as forging) and heat treatment, these segregated areas easily form coarse β-grain structures (i.e., β spots), severely impairing the material's fatigue performance, fracture toughness, and uniformity of mechanical properties.

[0004] Currently, the detection and evaluation of β-spot defect morphology in bars and forgings mainly rely on traditional low-magnification inspection methods, such as standards GB / T 5168-2020 and GJB / T 1538A-2008. However, there is no unified standard for the detection of β-spots in ingots. Furthermore, due to the large differences in the morphology of β-spot defects in ingots, and their often numerous, scattered, and irregular shapes, existing conventional methods have significant limitations in accurately quantifying and statistically analyzing their morphological characteristics: First, there is a lack of precise quantitative characterization methods; existing assessments mostly rely on qualitative or semi-quantitative observation, making it difficult to capture the complex morphological characteristics of β-spots in detail and comprehensively, and failing to provide unified quantitative parameters. Second, they are highly subjective and prone to introducing errors, relying excessively on visual inspection and experience-based judgment, resulting in insufficient objectivity and reliability of the analysis results, and making it difficult to achieve refined comparisons between different β-spots. Third, it is difficult to establish quantitative correlations, and it is impossible to establish quantitative relationships with known solidification models, which restricts a deeper understanding of the formation law of β-spots and the verification of segregation theory models.

[0005] Patent CN113624793A discloses a method for determining the presence of β-spot defects in near-β titanium alloys. This method first establishes the relationship between the primary α-phase content and heat treatment temperature in a standard sample. Then, it observes the sample to be tested. If a suspected area with a primary α-phase content below 5% and a large size is found, the chemical composition of that area is further analyzed to calculate its actual β-phase transformation point. The theoretical value of the α-phase content is then corrected using a formula. Finally, whether it is a β-spot is determined based on whether the corrected α-phase content is below 5%. The core of this method is to avoid misjudgments caused by normal fluctuations in micro-area composition through compositional correction. Its judgment is essentially based on chemical composition and phase transformation behavior, rather than the geometric morphology of the defect. While CN113624793A can determine the presence or absence of β-spots, it cannot provide any quantitative description of the morphological characteristics (such as complexity, sharpness, and uniformity of distribution) of existing β-spots. Morphological characteristics are likely directly related to the severity of the defect and its potential for damage to the part. Without quantitative morphological indicators, it is difficult to perform more refined classification, evaluation, and process traceability of defects.

[0006] On the other hand, fractal theory, as a powerful mathematical tool for describing complex and irregular shapes in nature, has been attempted to be applied in the field of materials science to quantify microstructure. For example, patent CN104197858A discloses a method for quantitatively describing the solidification microstructure characteristics of continuously cast steel billets. This method exposes the solidification microstructure by hot pickling the cross-section of the continuously cast billet, and then calculates the fractal dimension of different regions such as the chilled layer, columnar crystal region, and equiaxed crystal region using the box method or perimeter-area method, thereby quantifying the complexity of the grain morphology and attempting to correlate it with the internal quality of the billet. CN104197858A focuses on the solidified grains of continuously cast steel billets, whose morphological characteristics (such as columnar crystals and equiaxed crystals) are fundamentally different from the β spots formed by segregation in titanium alloy ingots in terms of formation mechanism, scale, and morphology. β spots typically exhibit diffuse distribution on the matrix and extremely irregular boundary "island" or "cloud"-like features, with a degree of irregularity far exceeding that of ordinary solidified grains. The effectiveness and accuracy of directly applying methods used for steel to titanium alloy β spots are unknown.

[0007] Existing methods cannot provide a unified, quantitative morphology parameter to establish a direct bridge between solidification process parameters (such as cooling rate) and the final β-spot morphology characteristics, which limits the ability to actively control the β-spot morphology and thus suppress its harmfulness by optimizing the solidification process.

[0008] Therefore, there is an urgent need in the field for a new method that can accurately and quantitatively characterize the irregular morphology of β spots in titanium alloy ingots to make up for the shortcomings of existing technologies. Summary of the Invention

[0009] The purpose of this invention is to provide a method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots.

[0010] To achieve the objective of this invention, the method includes the following steps:

[0011] A. Sample preparation and processing: The sample was cut from the TC17 titanium alloy ingot to be tested, and the sample was subjected to heat treatment and surface processing. Then, its solidification structure and β-spot morphology were exposed by chemical etching, and the macroscopic morphology image of the sample was obtained.

[0012] B. Image analysis and feature extraction: Identify and extract the morphological parameters of the β spots from the macroscopic morphological image, the morphological parameters including the perimeter P and area A of each β spot;

[0013] C. Fractal dimension calculation: Based on the perimeter P and area A of each β spot, the fractal dimension D of the β spot morphology is calculated using a fractal geometry algorithm. The calculation satisfies the requirement of goodness of fit R² ≥ 0.9. The fractal dimension D is used as an index to quantitatively characterize the complexity of the β spot morphology.

[0014] In one specific embodiment, the heat treatment in step A includes:

[0015] First stage of heat treatment: The sample is heated to 20~50°C above the α+β / β phase transformation point of the titanium alloy, held at that temperature for 1~3 hours, and then cooled to room temperature;

[0016] Second heat treatment: The sample after the first heat treatment is heated to 25-15°C below the α+β / β phase transformation point of the titanium alloy, held at that temperature for 0.5-1.5 hours, and then cooled.

[0017] In one specific embodiment, the cooling rate during the first stage of heat treatment is ≤1.5℃ / min.

[0018] In one specific embodiment, the cooling in the second heat treatment stage is water cooling.

[0019] In one specific embodiment, the chemical etching uses a mixed solution of hydrofluoric acid, concentrated nitric acid, and water as the etchant, with a volume ratio of hydrofluoric acid: concentrated nitric acid: water = 1:1-3:40-100. Preferably, the method further includes rinsing the sample after chemical etching and taking a picture using a high-definition digital camera to obtain the macroscopic morphology image. The mass fraction of the concentrated nitric acid is preferably 65%-68%.

[0020] In one specific embodiment, the chemical etching time is 1 to 5 hours.

[0021] In one specific embodiment, the surface processing in step A involves removing the oxide layer and oxygen-permeable layer from the detection surface of the sample, processing it to meet the detection standards for low-magnification samples.

[0022] In one specific implementation, step B, image analysis and feature extraction, includes:

[0023] b1) Based on the obtained macroscopic morphology image, select a specific region for analysis. The specific region contains typical β-spot morphology and solidification structure. The number of grains in the specific region is not less than 10, and the number of β-spots is not less than 15.

[0024] b2) By image processing of the morphology of β spots and coagulated tissue in the specific region, the β spots in the region are identified and the morphological information of the β spots is clarified;

[0025] In preferred step b2), the image processing is implemented in conjunction with computer recognition technology.

[0026] In one specific implementation, the fractal dimension calculation in step C includes:

[0027] c1) Take the natural logarithms of the perimeter P and area A of each β spot, lnP and lnA, respectively;

[0028] c2) Using lnA as the abscissa and lnP as the ordinate, perform linear fitting on lnP and lnA to obtain the fitted line lnP=S. lnA+b;

[0029] c3) Based on the slope S of the fitted line, use the formula D=2 S calculates the fractal dimension D of the β-pattern morphology.

[0030] In one specific implementation, if the goodness of fit R² in step C is less than 0.9, then steps B to C are repeated to appropriately expand or shrink the specific region in order to improve the self-similarity of the β-pattern morphology within the region, until the goodness of fit R² is greater than or equal to 0.9.

[0031] Beneficial effects:

[0032] 1. This invention innovatively introduces fractal dimension to quantitatively characterize β-spot defects in the solidification structure of TC17 titanium alloy ingots. Compared with traditional qualitative or semi-quantitative methods, fractal dimension can more meticulously and comprehensively reflect the complex morphological characteristics of β-spots, effectively overcoming the difficulties in quantitative statistical analysis caused by the large number, dispersion, and irregular shape of β-spots.

[0033] 2. By using fractal dimension to quantify the morphology of β spots, this invention significantly reduces the reliance on human experience and judgment, reduces errors caused by subjective identification, and achieves a refined and objective comparison of morphological features among different β spots.

[0034] 3. Easy to integrate with automation technology: This method can be closely integrated with digital image processing technology, and is expected to achieve rapid and automatic identification and rating of β spots, thereby improving detection efficiency.

[0035] 4. Based on the precise quantification of β-spot morphology using fractal dimension, this invention can better establish the quantitative correlation between ingot solidification process control parameters and the β-spot defect formation process. Specifically, this invention reveals the following key correlations: the closer the fractal dimensions, the higher the self-similarity of the β-spot morphology, and the more similar their formation stages; the larger the fractal dimension, the higher the complexity of the β-spot morphology, and the higher the tendency for defect occurrence. This finding provides a quantitative and reliable guiding basis for optimizing ingot solidification processes and achieving effective prediction, evaluation, and suppression of β-spot defects. Attached Figure Description

[0036] Figure 1 Low-magnification microstructure of the core of a TC17 titanium alloy ingot sample;

[0037] Figure 2 for Figure 1 The image shows the morphology of the β-spot within a selected area (8mm × 8mm) of the ingot core.

[0038] Figure 3 for Figure 2 The β-pattern image of the specific region shown is processed and identified by a computer.

[0039] Figure 4 The figure shows the linear fitting results of calculating the fractal dimension of β-spot morphology in a specific region using the perimeter-area method;

[0040] Figure 5 This is a β-spot morphology image of the lower center of the ingot in process group A of Example 3;

[0041] Figure 6 This is an image showing the identification results of the β spot located at the lower center of the ingot in process group A of Example 3;

[0042] Figure 7 The image shows the β-spot morphology at the same location after the ingot of process group B in Example 3 was optimized.

[0043] Figure 8 This is a diagram showing the identification results of β spots at the same location after the ingot in process group B of Example 3 was optimized. Detailed Implementation

[0044] To achieve the objective of this invention, the method includes the following steps:

[0045] A. Sample preparation and processing: The sample was cut from the TC17 titanium alloy ingot to be tested, and the sample was subjected to heat treatment and surface processing. Then, its solidification structure and β-spot morphology were exposed by chemical etching, and the macroscopic morphology image of the sample was obtained.

[0046] B. Image analysis and feature extraction: Identify and extract the morphological parameters of the β spots from the macroscopic morphological image, the morphological parameters including the perimeter P and area A of each β spot;

[0047] C. Fractal dimension calculation: Based on the perimeter P and area A of each β spot, the fractal dimension D of the β spot morphology is calculated using a fractal geometry algorithm. The calculation satisfies the requirement of goodness of fit R² ≥ 0.9. The fractal dimension D is used as an index to quantitatively characterize the complexity of the β spot morphology.

[0048] In one specific embodiment, the heat treatment in step A includes:

[0049] First stage of heat treatment: The sample is heated to 20~50°C above the α+β / β phase transformation point of the titanium alloy, held at that temperature for 1~3 hours, and then cooled to room temperature;

[0050] Second heat treatment: The sample after the first heat treatment is heated to 25-15°C below the α+β / β phase transformation point of the titanium alloy, held at that temperature for 0.5-1.5 hours, and then cooled.

[0051] In one specific embodiment, the cooling rate during the first stage of heat treatment is ≤1.5℃ / min.

[0052] In one specific embodiment, the cooling in the second heat treatment stage is water cooling.

[0053] In one specific embodiment, the chemical etching uses a mixed solution of hydrofluoric acid, concentrated nitric acid, and water as the etchant, with a volume ratio of hydrofluoric acid: concentrated nitric acid: water = 1:1-3:40-100. Preferably, the method further includes rinsing the sample after chemical etching and taking a picture using a high-definition digital camera to obtain the macroscopic morphology image. The mass fraction of the concentrated nitric acid is preferably 65%-68%.

[0054] In one specific embodiment, the chemical etching time is 1 to 5 hours.

[0055] In one specific embodiment, the surface processing in step A involves removing the oxide layer and oxygen-permeable layer from the detection surface of the sample, processing it to meet the detection standards for low-magnification samples.

[0056] In one specific implementation, step B, image analysis and feature extraction, includes:

[0057] b1) Based on the obtained macroscopic morphology image, select a specific region for analysis. The specific region contains typical β-spot morphology and solidification structure. The number of grains in the specific region is not less than 10, and the number of β-spots is not less than 15.

[0058] b2) By image processing of the morphology of β spots and coagulated tissue in the specific region, the β spots in the region are identified and the morphological information of the β spots is clarified;

[0059] In preferred step b2), the image processing is implemented in conjunction with computer recognition technology.

[0060] In one specific implementation, the fractal dimension calculation in step C includes:

[0061] c1) Take the natural logarithms of the perimeter P and area A of each β spot, lnP and lnA, respectively;

[0062] c2) Using lnA as the abscissa and lnP as the ordinate, perform linear fitting on lnP and lnA to obtain the fitted line lnP=S. lnA+b;

[0063] c3) Based on the slope S of the fitted line, use the formula D=2 S calculates the fractal dimension D of the β-pattern morphology.

[0064] In one specific implementation, if the goodness of fit R² in step C is less than 0.9, then steps B to C are repeated to appropriately expand or shrink the specific region in order to improve the self-similarity of the β-pattern morphology within the region, until the goodness of fit R² is greater than or equal to 0.9.

[0065] In one specific embodiment, the method includes:

[0066] 1) Based on the detection requirements of β spots in titanium alloy ingots, ingot samples of appropriate size are cut for subsequent low-magnification observation. The ingot samples are then subjected to heat treatment, which includes:

[0067] i) Heat the sample to 20-50°C above the α+β / β phase transition point, keep it heated for 1-3 hours after heat penetration, and then cool it to room temperature in the furnace at a controlled cooling rate of no more than 1.5°C / min.

[0068] ii) Heat the sample to 25-15°C below the α+β / β phase transition point, heat it thoroughly, and then hold it at that temperature for 0.5-1.5 hours. After removing it from the furnace, immediately quench it in water.

[0069] 2) The oxide layer and oxygen-permeable layer of the sample detection surface are removed by processing. The detection surface is processed to meet the detection standards for low-magnification samples in GB / T5168-2020. The solidification structure and morphology of the titanium alloy ingot are obtained by chemical etching. The etchant is a mixed solution of hydrofluoric acid, concentrated nitric acid and distilled water with a volume ratio of hydrofluoric acid: concentrated nitric acid: water = 1:1~3:40~100. The etching time is 1~5 hours. After etching, the sample is rinsed with running water and photographed using a high-definition digital camera to obtain the morphology of the solidification structure and β spots of the ingot. The mass fraction of concentrated nitric acid is 65%~68%.

[0070] 3) Based on the obtained solidification structure morphology diagram of the ingot, select a specific region containing typical β-spot morphology and solidification structure. The β-spot morphology in the region is required to have a high degree of self-similarity, and the region contains a sufficient number of β-spots and grains, wherein the number of grains is not less than 10 and the number of β-spots is not less than 15.

[0071] 4) By processing images of the morphology of β spots and coagulated tissue within the region, β spots within the region are identified, and the morphological information of β spots is clarified;

[0072] 5) Calculate the fractal dimension of the β-pattern morphology within the region using the perimeter-area method in fractal theory, and count the number of β-patterns (n) and the perimeter (P, P1, P2, P3...P) of each β-pattern. n ) and the corresponding areas (A, respectively A1, A2, A3...A n );

[0073] 6) Calculate the fractal dimension. Based on the perimeter P and area A of each β patch within the region, take the natural logarithm lnP (lnP1, lnP2, lnP3...lnP) respectively. n ) and lnA (lnA1, lnA2, lnA3...lnA n Linear fitting is performed using the least squares method:

[0074] lnP=S*lnA+b

[0075] Required goodness of fit R 2 ≥0.9, where the coefficient S is:

[0076]

[0077] In the formula and The values ​​are the averages of lnP and lnA, respectively. Finally, the fractal dimension D of the β-pattern morphology in the measurement area is obtained using the formula D=2*S.

[0078] 7) If R 2<0.9, repeat steps 4) to 7), appropriately expanding or shrinking the specific region to improve the self-similarity of the β-pattern morphology within the region, until the fitting degree R is obtained. 2 ≥0.9, the fractal dimension D of the β-pattern morphology in the region is obtained.

[0079] The specific embodiments of the present invention will be further described below with reference to examples, but the present invention is not limited to the scope of the embodiments described herein.

[0080] Example 1

[0081] 1) Sample Preparation. A TC17 titanium alloy ingot with a diameter of 240 mm was selected as the experimental material. A low-magnification sample with a thickness of approximately 20 mm was cut along the longitudinal section of the ingot, and the α+β / β phase transformation temperature of the alloy was measured to be 895℃. The low-magnification sample underwent two-stage heat treatment. The first stage heat treatment: the heat treatment furnace was heated to 930℃, the low-magnification sample was placed in the furnace, and after thorough heating, it was held at that temperature for 3 hours, followed by cooling to room temperature with the furnace at a cooling rate not exceeding 1.5℃ / min. The second stage heat treatment: the heat treatment furnace was heated to 875℃, the low-magnification sample was placed in the furnace, and after thorough heating, it was held at that temperature for 1 hour, and immediately water-quenched after removal from the furnace.

[0082] The surface of the low-magnification sample after heat treatment was surface-treated by machining to remove approximately 3 mm of oxide and oxygen-permeable layers, and then processed according to GB / T 5168-2020 standard to meet the requirements for low-magnification testing. The treated low-magnification sample was then immersed in an etching solution with a ratio of hydrofluoric acid: concentrated nitric acid: water = 1:2:80 for 2.5 hours (concentrated nitric acid mass fraction was 68%), followed by rinsing with water, and the surface was photographed to obtain the results shown below. Figure 1 The low-magnification image shows the morphology of the coagulated tissue in the core of the sample. From... Figure 1 β spots can be observed to be numerous, scattered, and irregularly shaped in the core of the ingot.

[0083] 2) Image Acquisition and Processing. From the coagulated tissue images obtained above, a specific region of 8mm × 8mm was selected for analysis. This region needed to contain a sufficient number of grains and β spots, and the morphology of the β spots within the region should have a high degree of similarity. Figure 2 The morphology of the β-spots within this specific region is shown. Computer image recognition processing is then performed on the β-spots within this specific region. Figure 3 The image in the middle is the computer-processed image, and the white area represents the β spots within the identified region. The perimeter (P) and area (A) of a total of 40 β spots within this region were statistically analyzed, and the specific data are shown in Table 1.

[0084] Table 1. Statistics on the perimeter (P) and area (A) of β spots in selected regions within TC17 titanium alloy ingots.

[0085] 3) Fractal dimension calculation. The fractal dimension of the β-pattern morphology within the region was calculated using the perimeter-area method in *Fractal Theory and Its Applications* (Science Press, 2011, edited by Zhu Hua and Ji Cuicui). A point plot was drawn with lnA as the abscissa and lnP as the ordinate. Linear fitting was performed on the data points in the plot, and the fitting results are shown below. Figure 4 As shown. The slope S of the linear fitting line is calculated using the least squares method: Finally, the fractal dimension of the β-pattern morphology in this region was calculated using the formula D=2*S, which is D=1.0913, and the goodness of fit R²=0.9299.

[0086] This embodiment successfully achieved quantitative characterization of the β-spot morphology in TC17 titanium alloy ingots using the method described above, thus enabling a quantitative description of its complexity and self-similarity. This lays the foundation for subsequent research on the relationship between β-spot morphology and the solidification process and defect formation.

[0087] Example 2

[0088] This embodiment aims to verify the objectivity, accuracy, and repeatability of the method of the present invention in quantitative characterization of β-spot morphology by comparing it with traditional visual evaluation methods, especially its ability to assess morphological complexity and self-similarity.

[0089] 1) Sample Selection and Preparation. A TC17 titanium alloy ingot sample with typical β-spot morphology was selected as a reference standard sample. This standard sample exhibited moderate β-spot morphology complexity and moderate self-similarity. Three other groups of TC17 titanium alloy ingot samples were selected, labeled as Sample A, Sample B, and Sample C, respectively. Sample A showed a high degree of similarity to the reference standard sample in terms of β-spot morphology; Sample B exhibited lower β-spot morphology complexity and low self-similarity; Sample C showed higher β-spot morphology complexity and high self-similarity. All samples were processed according to the sample preparation and processing method described in Example 1 to obtain clear visual images of the β-spots.

[0090] 2) Traditional Visual Assessment. Five titanium alloy testing experts with over five years of experience were invited to visually compare the β-pattern morphology of samples A, B, and C with the standard sample. The experts' task was to assess the complexity and self-similarity of the β-pattern morphology of each test sample compared to the standard sample. An evaluation scale of 5 points was used, with complexity and self-similarity ranging from 1 to 5 points. The standard sample's β-pattern morphology complexity and self-similarity were both assigned a score of 3. The expert evaluation results are shown in Table 2.

[0091] Table 2. Experts' visual evaluation results of the test samples and standard samples.

[0092] As can be seen from Table 2, even experienced experts have certain differences in their assessment of the complexity and self-similarity of the same test sample. In particular, the standard deviation of the self-similarity rating results is relatively large, indicating that the traditional visual method is highly subjective and has poor repeatability.

[0093] 3) The β-spot morphology of the standard sample and samples A, B and C was quantitatively characterized according to the image analysis and feature extraction and fractal dimension calculation method in Example 1. The calculated fractal dimension D value and its goodness of fit R² are shown in Table 3.

[0094] Table 3. β-spot fractal dimension D and its goodness of fit R calculated by the method of this invention. 2

[0095] As shown in Table 3, the method of this invention, when performing multiple detections on β spots in the same sample, yielded highly consistent fractal dimension D values ​​and goodness-of-fit R² values ​​with minimal standard deviation, indicating that the method possesses extremely high objectivity and repeatability. Comparison reveals that sample A's D value (average 1.0849) and R² (average 0.9500) are closest to the standard sample (average D 1.0909, average R² 0.9530), sample B has the lowest D value (average 0.9864) and R² (average 0.9077), while sample C has the highest D value (average 1.1588) and R² (average 0.9885). This largely matches the preset sample morphological characteristics.

[0096] 4) Results Analysis. As demonstrated in Example 2, the method of this invention, by introducing fractal dimension, accurately and quantitatively characterizes the complexity and self-similarity of β-spot morphology, effectively eliminating errors caused by human subjectivity in traditional visual methods. This provides objective, repeatable, and discriminative quantitative results. This method can accurately reflect subtle differences in β-spot morphology, providing a reliable quantitative basis for standardized classification, quality control, and comparison with reference standards for β-spots.

[0097] Example 3

[0098] This embodiment aims to verify the guiding role of the method of the present invention in optimizing the smelting process and suppressing β-spot defects by comparing the changes in the fractal dimension of β-spots under different smelting processes.

[0099] 1) Initial melting and β-spot morphology analysis (process group A)

[0100] a) Sample preparation and process retrospection. A TC17 alloy ingot with a diameter of 120 mm and a weight of approximately 16 kg was selected as the experimental material. A β-spot appeared near the lower center of the ingot, the morphology of which is shown in the attached figure. Figure 5 As shown in Table 4, the main smelting process parameters of this ingot are traced back to the present.

[0101] Table 4 Main smelting process parameters for ingots in process group A

[0102] b) Calculation of β-spot fractal dimension. Following the sample preparation and processing method of Example 1, the locations of β-spots on the ingot were processed. Subsequently, the region was divided into partitions according to the method of this invention, and the β-spot fractal dimension D of each partition was calculated through image processing. The processed image is attached. Figure 6 As shown in Table 5, the calculation results are as follows.

[0103] Table 5. Fractal Dimension D of β-spot on Ingot of Process Group A

[0104] c) Results Analysis and Judgment. Table 5 shows that the fractal dimension D value (average 1.3891) of the β spots in the lower part of the ingot (c1~c4) is higher than that in the upper part (a1~a4) (average 1.2002), indicating that the morphological complexity and self-similarity of the β spots in the lower region are relatively higher. Further judgment suggests that the location of these β spots corresponds to the transition stage from arc initiation to stable melting during the smelting process. Based on the beneficial effect of this invention, "the larger the fractal dimension, the higher the complexity of the β spot morphology, and the higher the tendency for defects to occur," regions with lower D values ​​(such as the upper a1~a4) indicate a lower tendency for β spot defect formation in these regions.

[0105] 2) Melting process optimization and verification (Process Group B)

[0106] a) Process Parameter Adjustment. Based on the above analysis and judgment, in order to optimize the melting process and suppress β-spot defects, another TC17 alloy ingot of the same specifications was melted (process group B). Adjustments were made to the process parameters during the transition from arc ignition to stable melting. The main measures included reducing the melting current and melting voltage, and further reducing the arc-stabilizing current. The main process parameters are shown in Table 6.

[0107] Table 6 Main smelting process parameters for ingots in process group B

[0108] b) Optimized β-spot morphology analysis. Following the sample preparation and processing method in Example 1, the same location on the ingot of process group B was processed and observed. The results showed that the number of β-spots at the same location was significantly reduced, and the morphology was as shown in the attached figure. Figure 7 As shown in Table 7, calculations of the fractal dimension at each location revealed that the overall fractal dimension D was lower than that of process group A.

[0109] Table 7. Fractal Dimension of β-spot on Ingot of Process Group B

[0110] c) Results Analysis. Comparing the β-spot morphology and fractal dimension D value of process group A and process group B, the optimized process group B ingot showed a significant reduction in β-spot defects at the same location, and the overall fractal dimension D value was also lowered. This indicates that the quantitative analysis of β-spot morphology parameters using the method of this invention can effectively guide the adjustment of smelting process parameters, thereby successfully suppressing the formation of β-spot defects. This embodiment fully demonstrates the practicality and guiding value of the method of this invention in smelting process optimization.

Claims

1. A method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots, characterized in that, Includes the following steps: A. Sample preparation and processing: The sample was cut from the TC17 titanium alloy ingot to be tested, and the sample was subjected to heat treatment and surface processing. Then, its solidification structure and β-spot morphology were exposed by chemical etching, and the macroscopic morphology image of the sample was obtained. B. Image analysis and feature extraction: Identify and extract the morphological parameters of the β spots from the macroscopic morphological image, the morphological parameters including the perimeter P and area A of each β spot; C. Fractal Dimension Calculation: Based on the perimeter P and area A of each β-spot, the fractal dimension D of the β-spot morphology is calculated using a fractal geometry algorithm. This calculation satisfies the goodness-of-fit R² ≥ 0. The requirement of 0.9 is to use the fractal dimension D as an indicator to quantitatively characterize the complexity of the β-pattern morphology.

2. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 1, characterized in that, The heat treatment described in step A includes: First stage of heat treatment: The sample is heated to 20~50°C above the α+β / β phase transformation point of the titanium alloy, held at that temperature for 1~3 hours, and then cooled to room temperature; Second heat treatment: The sample after the first heat treatment is heated to 25-15°C below the α+β / β phase transformation point of the titanium alloy, held at that temperature for 0.5-1.5 hours, and then cooled.

3. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 1, characterized in that, The cooling rate during the first stage of heat treatment is ≤1.5℃ / min.

4. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 1, characterized in that, The cooling process described in the second heat treatment stage uses water cooling.

5. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 1 or 2, characterized in that, The chemical etching process uses a mixed solution of hydrofluoric acid, concentrated nitric acid, and water as the etchant, with a volume ratio of hydrofluoric acid: concentrated nitric acid: water = 1:1-3:40-100. Preferably, the method further includes rinsing the sample after chemical etching and taking a picture using a high-definition digital camera to obtain the macroscopic morphology image. The mass fraction of the concentrated nitric acid is preferably 65%-68%.

6. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 5, characterized in that, The chemical etching time is 1 to 5 hours.

7. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 1 or 2, characterized in that, Step A involves surface processing to remove the oxide layer and oxygen-permeable layer from the test surface of the sample, processing it to meet the test standards for low-magnification samples.

8. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 1 or 2, characterized in that, Step B, the image analysis and feature extraction, includes: b1) Based on the obtained macroscopic morphology image, select a specific region for analysis. The specific region contains typical β-spot morphology and solidification structure. The number of grains in the specific region is not less than 10, and the number of β-spots is not less than 15. b2) By image processing of the morphology of β spots and coagulated tissue in the specific region, the β spots in the region are identified and the morphological information of the β spots is clarified; In preferred step b2), the image processing is implemented in conjunction with computer recognition technology.

9. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 1 or 2, characterized in that, Step C involves calculating the fractal dimension, including: c1) Take the natural logarithms of the perimeter P and area A of each β spot, lnP and lnA, respectively; c2) Using lnA as the abscissa and lnP as the ordinate, perform linear fitting on lnP and lnA to obtain the fitted line lnP=S. lnA+b; c3) Based on the slope S of the fitted line, use the formula D=2 S calculates the fractal dimension D of the β-pattern morphology.

10. The method for quantitatively describing the morphological characteristics of β spots in TC17 titanium alloy ingots according to claim 8, characterized in that, If the goodness of fit R² in step C is less than 0.9, then repeat steps B to C, appropriately expanding or shrinking the specific region to improve the self-similarity of the β-pattern morphology within the region, until the goodness of fit R² is greater than or equal to 0.9.

Images