A microstructure damage-based long-time creep life prediction method for high-temperature alloy

Abstract

The application provides a high-temperature alloy long-time creep life prediction method based on microstructure damage, and relates to the technical field of gas turbine hot end component materials and application. The application is aimed at the high-temperature alloy used for heavy gas turbine turbine blades, considers the creep performance deterioration of the high-temperature alloy due to microstructure degradation under near-service conditions, combines the damage state of the microstructure with the macroscopic creep life prediction method based on the typical damage state structure by systematically analyzing the microstructure, and finally realizes the prediction of the residual creep life of any damage structure. The prediction of the creep life of the application is based on the modified theta projection method of the damage state structure, and the microstructure damage of the alloy to be evaluated is considered, the result is more real and reliable, and the application of the high-temperature alloy on the heavy gas turbine turbine blade has important guiding significance.

Description

Technical Field

[0001] This invention relates to the field of materials and applications for hot-end components of gas turbines, particularly high-temperature alloy materials for turbine blades. Specifically, it relates to a method for predicting the long-term creep life of a nickel-based cast high-temperature alloy. Background Technology

[0002] Heavy-duty gas turbines, as the "heart" of large industrial equipment, are not only undeniably a major national asset but also a reflection of a nation's advanced industrial strength. Turbine blades are the core hot-end components of heavy-duty gas turbines. During service, turbine blades are constantly exposed to the scouring environment of high-temperature gas, experiencing thermal stress from uneven temperature fields, thermal fatigue from frequent start-stop cycles, and centrifugal force from high-speed rotation. Besides surface damage caused by wear, spalling, and internal cracks, internal metallurgical damage is inevitable, mainly manifested as γ′ phase coarsening, carbide degradation and decomposition, grain boundary morphology changes, and brittle phase formation. This microstructural degradation directly leads to the degradation of the blade's high-temperature mechanical properties and a reduction in service life. Therefore, accurately predicting the remaining creep life of blades based on their microstructural damage state is of great significance for determining blade overhaul schedules and ensuring service safety.

[0003] Chinese patent CN105628511A proposes a modified θ-projection method for predicting the creep life of high-temperature alloys. This method mainly predicts the creep curve under different temperature and stress conditions by fitting the constant load creep curve of the standard state alloy. Chinese patent CN110411851A proposes a creep life prediction method combining microstructure and macroscopic creep deformation. This method first uses microstructure to infer the service conditions of DZ125 high-temperature alloy turbine blades, and then uses the modified θ-projection method to calculate the equivalent creep strain of the blade under these service conditions. By comparing this with the maximum allowable creep deformation of the blade, the remaining creep life is obtained. Although this method considers the influence of microstructure degradation on creep life, the modified θ-projection method is still based on the creep curve of the standard state microstructure. Problems remain regarding the possible changes in the creep mechanism of damaged structures and the prediction of creep life under low stress.

[0004] Based on the above background, it is necessary to develop a method for predicting the remaining creep life of long-life alloys that takes into account both the changes in creep behavior caused by microstructural damage and the differences in creep mechanisms under low stress levels.

[0005] Currently, existing technical solutions mainly target turbine blades and alloy materials for aero-engines, with a typical overhaul cycle of 500 hours and a maximum service life of 1500 hours. The creep test time under constant load conditions in the laboratory can be equal to the service life, therefore the accuracy of the corrected θ projection method established using the standard state alloy creep curve is relatively high. Furthermore, the stress experienced by aero-engine turbine blades is relatively high, and there is essentially no difference between high-stress and low-stress creep mechanisms. However, for heavy-duty gas turbine blade materials, the typical service life reaches tens of thousands of hours, and the entire service life is basically in the second stage of creep. The overall service stress is much lower than the creep stress in short-term laboratory experiments, thus potentially leading to fundamental differences in creep mechanisms. In addition, the corrected θ projection method based on the creep curve of standard state alloy materials under high stress levels in the laboratory does not consider the influence of microstructure changes on creep behavior during long-term creep, resulting in significant deviations when predicting long-life creep curves under low stress, failing to meet practical needs. Summary of the Invention

[0006] The purpose of this invention is to address the aforementioned problems by providing a method for predicting the long-term creep life of high-temperature alloys based on microstructural damage. Specifically, for high-temperature alloys used in heavy-duty gas turbine blades, considering the degradation of creep performance due to microstructural deterioration under near-service conditions, this invention systematically analyzes the microstructure and combines the damage state of the microstructure with a macroscopic creep life prediction method based on typical damaged microstructures. This ultimately enables the prediction of the remaining creep life of any damaged microstructure, and is applicable to the long-term creep life prediction of nickel-based cast high-temperature alloys used in heavy-duty gas turbine blades.

[0007] The specific steps are as follows:

[0008] 1) For high-temperature alloy materials in standard heat-treated state, we first established a dataset of microstructure and simulated service conditions by conducting variable cross-section endurance interruption experiments at different temperatures and stresses for different durations, combined with quantitative characterization of microstructure. Based on the dataset, we then established service condition evaluation model A and service time evaluation model B, respectively.

[0009] 2) Analyze the microstructure and simulated service conditions dataset in 1), and combine it with the service conditions of high-temperature alloys to select the severe microstructure damage state S under a certain high stress condition at a typical service temperature as the specific damage state.

[0010] 3) Establish a modified q-projection model for a specific damage state S and obtain the following formula.

[0011]

[0012] Where ε is the creep strain and t is the creep time.

[0013] logθ i =a i +b i σ+c i T+d i σT(i=1-5)#(Formula 2)

[0014] T is the creep temperature, σ is the creep stress, and parameter a i b i c i d i These are constants that are only related to the material.

[0015] 4) For alloys whose remaining creep life needs to be evaluated after a period of use t1, the service temperature T1 and stress σ1 are evaluated using service condition evaluation model A in 1), and then θ is calculated under the conditions of temperature T1 and stress σ1 by substituting them into formula 2 in 3). i 1 (i = 1 - 5), then θ i Substitute 1(i=1-5) into formula 1 in 3) to calculate the time t0 required for a specific damage state S to continue deforming to the specified macroscopic creep deformation amount ε0;

[0016] 5) Using the service time assessment model B in 1), assess the equivalent damage time t2 required for the standard state alloy to reach a specific damage state S under T1 and σ1 conditions. t2-t1 is the time t3 required for the alloy to be assessed to reach the specific damage state S. Calculate the sum of t3 and t0 as the creep remaining life t4 of the alloy to be assessed.

[0017] Furthermore, the variable cross-section sustained interruption test described in step 1) has a temperature range of 850-950℃, a stress range of 0-360MPa, and a time range of 500-5000h.

[0018] Furthermore, the dataset mentioned in step 1) includes the susceptibility temperature T, susceptibility stress σ, susceptibility interruption time t, and dendritic trunk γ′ phase volume fraction V. f , γ′ phase raft assembly perfection Ω, γ phase channel width W.

[0019] Further, the specific damage state S mentioned in step 2) refers to a severe damage state in which the volume fraction of the γ′ phase differs from that of the standard heat-treated state by less than 5%, the perfection of the γ′ phase raft is not less than 0.35, or the width of the γ channel is not less than 200 nm.

[0020] Furthermore, the modified q-projection model described in step 3) is obtained through at least four creep curves of a specific damage state S. The four curves refer to the curves of deformation to 1% creep deformation or fracture under different stress conditions in the temperature range of 850-950℃.

[0021] Furthermore, the deformation ε0 specified in step 4) must be within the strain range of the creep curve of the modified q-projection model constructed under the specific damage state S.

[0022] Furthermore, the time t3 mentioned in step 5) indicates that the state to be evaluated has not yet reached the specific damage state S when t3>0, and that the state to be evaluated has exceeded (below) the specific damage state S when t3<0.

[0023] Furthermore, the high-temperature alloy is a nickel-based cast high-temperature alloy for gas turbine blades.

[0024] This invention targets high-temperature alloys used in heavy-duty gas turbine turbine blades. Considering the degradation of creep performance due to microstructure deterioration under near-service conditions, the invention combines the damage state of the microstructure with a macroscopic creep life prediction method based on typical damaged microstructures after systematic analysis of the microstructure, ultimately enabling the prediction of the remaining creep life of any damaged microstructure.

[0025] The advantage of this invention is that when predicting the remaining creep life of an alloy that has undergone microstructural damage after a certain period of use, it not only takes into account the degradation of the microstructure, but also the corresponding macroscopic creep deformation, so that the remaining creep life can be predicted using the macroscopic creep deformation as the control value.

[0026] Compared with existing technologies, the creep life prediction of this invention is based on the modified θ projection method of the damaged state structure and takes into account the microstructural damage of the alloy under evaluation. The results are more realistic and reliable, and more suitable for evaluating long-service alloy materials with low service stress and macroscopic creep deformation as the failure control value. For long-life gas turbine blades, even with very small macroscopic creep deformation, the remaining creep life of the material under certain conditions to a specified amount of deformation can be predicted relatively accurately through microstructural degradation. This has important guiding significance for the application of high-temperature alloys in heavy-duty gas turbine blades. Attached Figure Description

[0027] Figure 1 This is a flowchart of the method for predicting remaining creep life.

[0028] Figure 2 This is a schematic diagram of the dissection scheme for a variable cross-section long-term specimen.

[0029] Figure 3 This is the service condition assessment model A.

[0030] Figure 4 This is Service Time Assessment Model B. Detailed Implementation

[0031] The present invention will now be further described with reference to the accompanying drawings.

[0032] Example

[0033] For directionally solidified UGTC47 alloy used in heavy-duty gas turbine blades, a long-term creep life prediction based on microstructural damage is proposed. (Flowchart shown) Figure 1 Follow these steps:

[0034] 1. Near-service temperature-dependent creep-out tests were conducted on directionally solidified UGTC47 alloy after standard heat treatment. The temperature range was 850-950℃, the stress range was 0-360MPa, and the time range was 500-5000h. Drawings of the creep-out specimens are shown below. Figure 2 .

[0035] 2. The dendritic trunk microstructure of the alloy in the transverse and longitudinal sections after the experiment was observed and quantitatively characterized. A dataset was established relating the experimental temperature, stress, time and microstructure parameters, as shown in Table 1.

[0036] Table 1. Composition of the service conditions-microtissue dataset

[0037]

[0038] 3. The volume fraction of the γ′ phase (V) in the dataset f A machine learning model A is established using the following parameters as input parameters: γ′ phase raft perfection (Ω), γ phase channel width (W), and time (t), and temperature (T) and stress (σ) as output parameters. Figure 3 (As shown); The model was built using the Spyder development environment under Anaconda, by calling the open-source machine learning library sklearn, and selecting two models: random forest and artificial neural network. The root mean square error (RMSE) values ​​of the two models were compared, and the artificial neural network model had a smaller RMSE; therefore, the artificial neural network model was chosen to build model A.

[0039] 4. The data includes temperature (T), stress (σ), and γ′ phase volume fraction (V). f A machine learning model B is established using the γ′ phase raft perfection (Ω) and γ phase channel width (W) as input parameters and damage time (t) as output parameter. Figure 4 (As shown). Using the same method as in step 3, the open-source machine learning library sklearn was called, and two models were selected: Random Forest and Artificial Neural Network. The root mean square error (RMSE) values ​​of the two models were compared. The RMSE value of the Artificial Neural Network model was smaller, so the Artificial Neural Network model was selected to build model B.

[0040] 5. Referring to the operating conditions of heavy-duty gas turbines, state S corresponding to 900℃ / 240MPa / 500h is selected as the specific damage state from the dataset established in step 2, and the volume fraction V of the γ′ phase at the dendrite trunk center is quantitatively characterized. f0 =60.3%, γ′ phase raft perfection Ω0=0.339, γ phase channel width W0=233nm.

[0041] 6. After pre-damaging the standard heat-treated directional solidified UGTC47 alloy test bar at 900℃ / 240MPa for 500h, further creep tests were carried out under different temperature and stress conditions to obtain four creep fracture curves under four temperature and stress conditions.

[0042] Input the strain and time data from the creep curves under four temperature and stress conditions into Origin software, select the software's built-in formula fitting function, and obtain θ by fitting according to Formula 1. i The parameters (i = 1-5) are shown in Table 2.

[0043]

[0044] Where ε is the creep strain and t is the creep time.

[0045] Table 2 Parameters of θi (i=1-5) for directional solidification of UGTC47 alloy test bars

[0046] 850 250 0.0500 0.3505 8.7076 0.0001 0.3157 850 440 113.0340 0.0003 0.0923 0.0344 0.0063 950 150 0.0284 0.0047 0.6336 0.0006 0.2150 950 210 0.4890 0.1568 0.3162 0.0141 0.0070

[0047] Substitute the data from Table 2 into Formula 2 and solve the system of equations to obtain a set of material constants a. i b i c i d i (i = 1-5), as shown in Table 3. Thus, the q-projection model of the directionally solidified UGTC47 alloy based on the damaged state S is established.

[0048] logθ i =a i +b i σ+c i T+d i σT(i=1-5)#(Formula 2)

[0049] T is the creep temperature, σ is the creep stress, and θ1, θ2, θ3, θ4, and θ5 are the fitted data from Table 2.

[0050] Table 3 Material constants for directional solidification UGTC47 alloy test bars

[0051] 1 -14.9090 -0.0072 0.0108 0.0001 2 84.8617 -0.3639 -0.0957 0.0004 3 28.8769 -0.0559 -0.0298 0.0001 4 -23.1209 -0.0409 0.0175 0.0001 5 -9.4851 0.1258 0.0131 -0.0001

[0052] 7. For the alloy to be evaluated that has been used for t1 = 3000h, its microstructure parameters were quantitatively characterized. The volume fraction of γ′ phase at the dendrite trunk center was 59.1%, the perfection of γ′ phase rafts was 0.275, and the width of γ phase channels was 216nm. Using Model A to evaluate its service conditions, its service temperature was found to be 847℃ and its service stress to be 217MPa.

[0053] 8. Using Formula 2 established in step 6, calculate θ1, θ2, θ3, θ4, and θ5 at 847℃ and 217MPa as 0.0521, 0.0017, 0.6947, 2.7319, and 0.4180, respectively. Then, Formula 1 can be written as:

[0054]

[0055] 9. Using the above formula, the time required for the deformation of a specific damage state S to the macroscopic creep deformation ε = 1% is calculated to be t0 = 2073h.

[0056] 10. Using Model B, the damage time t2 required for the standard alloy to reach a specific damage state S under conditions of 847℃ / 217MPa is 4038h. Therefore, the damage time required for the alloy to be evaluated to reach the specific damage state S under conditions of 847℃ / 217MPa is t3 = t2 - t1 = 1038h.

[0057] 11. Therefore, the remaining creep life of the alloy to be evaluated is t4 = t0 + t3 = 3111h.

[0058] 12. According to the data, the time required for a 1% macroscopic creep deformation in a standard heat-treated, directionally solidified UGTC47 alloy at 850℃ / 217MPa is t. 0实际 =7056h, therefore the creep remaining life of the alloy to be evaluated, which has been in service for 3000h, is 4056h. Thus, the accuracy of the above model prediction reaches 70%.

[0059] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for predicting the long-term creep life of high-temperature alloys based on microstructural damage, characterized in that, Specifically, the following steps are included: (1) For high-temperature alloy materials in the standard heat-treated state, firstly, through variable cross-section endurance interruption experiments at different temperatures and stresses for different durations, combined with quantitative characterization of microstructure, a dataset of microstructure and simulated service conditions was established. Based on the dataset, service condition evaluation model A and service time evaluation model B were established respectively. The dataset includes endurance temperature T, endurance stress σ, endurance interruption time t, and dendritic trunk γ′ phase volume fraction V. f , γ′ phase raft assembly perfection Ω, γ phase channel width W; (2) Analyze the microstructure and simulated service conditions dataset in 1), and combine it with the service conditions of high temperature alloys. Select the severe microstructure damage state under a certain high stress condition at a typical service temperature as the specific damage state S. The specific damage state S refers to the severe damage state in which the volume fraction of γ′ phase differs from the standard heat treatment state by less than 5%, the perfection of γ′ phase raft is not less than 0.35, or the width of γ channel is not less than 200nm. (3) Establishing the correction of a specific damage state S The projection method model yields the following formula: ； Where ε is the creep strain and t is the creep time; ； T is the creep temperature, σ is the creep stress, and parameter a i b i c i d i These are constants that are only related to the material. (4) For alloys whose remaining creep life needs to be evaluated after a period of use t1, the service temperature T1 and stress σ1 are evaluated using the service condition evaluation model A in 1), and θ is calculated under the conditions of temperature T1 and stress σ1 by substituting it into formula 2 in 3). i 1 (i=1-5), then θ i Substitute 1(i=1-5) into formula 1 in 3) to calculate the time t0 required for a specific damage state S to continue deforming to the specified macroscopic creep deformation amount ε0; (5) Using the service time assessment model B in 1), assess the equivalent damage time t2 required for the standard state alloy to reach a specific damage state S under T1 and σ1 conditions. t2-t1 is the time t3 required for the alloy to be assessed to reach a specific damage state S. Calculate the sum of t3 and t0 as the creep remaining life t4 of the alloy to be assessed.

2. The prediction method according to claim 1, characterized in that, The variable cross-section sustained interruption test described in step 1) has a temperature range of 850-950℃, a stress range of 0-360MPa, and a time range of 500-5000h.

3. The prediction method according to claim 1, characterized in that, The correction described in step 3) The projection method model is obtained through at least four creep curves of a specific damage state S. The four curves refer to the curves of deformation to 1% creep deformation or fracture under different stress conditions in the temperature range of 850-950℃.

4. The prediction method according to claim 1, characterized in that, The deformation ε0 specified in step 4) is the correction constructed in a specific damage state S. Within the strain range of the creep curve in the projection method model.

5. The prediction method according to claim 1, characterized in that, The time t3 mentioned in step 5) means that when t3>0, it indicates that the state to be evaluated has not yet reached the specific damage state S, and when t3<0, it indicates that the state to be evaluated has exceeded the specific damage state S.

6. The prediction method according to claim 1, characterized in that, The high-temperature alloy mentioned is a nickel-based cast high-temperature alloy for gas turbine blades.

Images