Hybrid Fast Prediction Method for Fatigue Life of SFRP Based on Stiffness

Abstract

A hybrid fast prediction method for the fatigue life of SFRP based on stiffness, which respectively obtains the first S-N curve of the target structure according to the relationship between the life of the reference structure and the first-cycle damage, and obtains the second S-N curve of the target structure according to the relationship between the life of the reference structure and the relative stiffness degradation rate in the stable section, and quickly obtains the S-N curve of the target structure by fitting the two. The present invention can quickly predict the first S-N curve of the target structure (with high accuracy in the low-life region) only through an S-N curve of a reference structure and the stress-strain curve of the target structure, and can quickly predict the second S-N curve of the target structure (with high accuracy in the high-life region) only through an S-N curve of a reference structure and the relative stiffness degradation rate in the stable section of the target structure, and can quickly and accurately predict the S-N curve of any structure through very few tests and simulations.

Description

Technical Field

[0001] The present invention relates to a technology in the field of composite materials, and specifically to a hybrid fast prediction method for stress-life (S-N) curves based on the stiffness degradation of short fiber reinforced composites (SFRP) during cyclic loading. Background Art

[0002] The fiber structure of SFRP has complex distribution characteristics, resulting in different macroscopic fatigue properties of the material everywhere (inhomogeneous and anisotropic). The existing S-N curve prediction methods are divided into two categories: experimental methods and simulation methods. Among them, the material-level stress fatigue test can only obtain the S-N curve under a specific structure and cannot meet the requirements of component-level fatigue analysis. Although the computational mesomechanics simulation method can theoretically predict the S-N curve under any structure, there are many problems such as large modeling difficulty, long calculation period, and difficult acquisition of fatigue simulation parameters, making it difficult to be applied to component-level fatigue analysis. Summary of the Invention

[0003] Aiming at the deficiencies that the existing prediction methods cannot accurately predict the fatigue life in the low-cycle region, cannot consider the volatility of the microstructure and thus cannot predict the fatigue life of SFRP with arbitrary structures, and cannot balance the diversity of prediction structures and the high prediction cost, the present invention proposes a hybrid fast prediction method for the fatigue life of SFRP based on stiffness. By making full use of the structure-independent relationship between the first-cycle damage and the relative stiffness degradation rate in the stable section during cyclic loading of SFRP materials and the life, the first S-N curve of the target structure (with high accuracy in the low-life region) can be quickly predicted only through an S-N curve of a reference structure and the stress-strain curve of the target structure. The second S-N curve of the target structure (with high accuracy in the high-life region) can be quickly predicted only through an S-N curve of a reference structure and the relative stiffness degradation rate in the stable section of the target structure, and the S-N curve of the target structure can be quickly obtained by fitting the two, and the S-N curve of any structure can be quickly and accurately predicted through very few experiments and simulations.

[0004] The present invention is realized through the following technical solutions:

[0005] The present invention relates to a hybrid fast prediction method for the fatigue life of SFRP based on stiffness. The first S-N curve of the target structure is obtained according to the relationship between the life of the reference structure and the first-cycle damage, and the second S-N curve of the target structure is obtained according to the relationship between the life of the reference structure and the relative stiffness degradation rate in the stable section, and the S-N curve of the target structure is quickly obtained by fitting the two.

[0006] Both the reference structure and the target structure are SFRP with specific fiber microstructures.

[0007] The described damage refers to the ratio of the difference between the initial stiffness and the slope of the line connecting the current point and the origin on the stress-strain curve (current stiffness) to the initial stiffness.

[0008] The relationships between the described life and the first-cycle damage, and between the life and the relative stiffness degradation rate in the stable stage are independent of the structure, that is, these relationships are not affected by the specific fiber microstructure, meaning that the relationship obtained from one structure can be directly extended to other structures.

[0009] The described relative stiffness degradation rate in the stable stage refers to the first derivative value when the second derivative (approximate) of the relative stiffness degradation curve is stably at 0.

[0010] Technical effects

[0011] The present invention can quickly predict the S-N curve from the stress-strain curve; extract the initial damage and the relative stiffness degradation rate in the stable stage from the stiffness evolution curve for fatigue life prediction in the low-cycle and high-cycle regions respectively. Compared with the prior art, it can quickly obtain the fatigue life of SFRP materials under any microstructure, avoiding a large number of fatigue tests; and can achieve high-precision prediction of fatigue life in both the low-cycle and high-cycle regions. Description of the drawings

[0012] Figure 1 is the microscopic fiber structure of the short fiber composite material;

[0013] Figure 2 is the schematic flow chart of the present invention;

[0014] Figure 3 is the schematic diagram of the first-cycle damage and the relative stiffness degradation rate in the stable stage;

[0015] Figure 4 is the schematic diagram of the effect of the embodiment. Detailed implementation manners

[0016] As Figure 1 shown, this embodiment relates to SFRP with two different fiber structures. The left side is the reference structure G1, Figure 1 and the right side is the target structure G2. Both are loaded in the horizontal direction. Except for the S-N curve of the target structure, other mechanical properties are known quantities.

[0017] As Figure 2 shown, a hybrid rapid prediction method for the S-N curve of SFRP stiffness degradation involved in this embodiment includes the following steps:

[0018] The first step: Select a point (N1, S1) on the S-N curve of the reference structure G1 obtained from the test, and calculate the damage at the stress level of S1 as the first-cycle damage according to the quasi-static loading stress-strain curve of G1 obtained from the test Wherein: E0 is the initial stiffness of the structure, and E is the current stiffness calculated from the stress-strain curve when the stress level of the structure is S, as shown in Figure 3 (a).

[0019] Step 2: Repeat Step 1 multiple times, and obtain the life-damage relationship f(N, D)=0 between the life N independent of the structure and the first-cycle damage D through the data fitting method.

[0020] Step 3: For the target structure G2, select the stress level S2, calculate the damage D2 at the stress level S2 according to the quasi-static loading stress-strain curve of G2 obtained from the test, and predict the target structure life N2 according to the life-damage relationship f(N, D)=0; repeat multiple times and then fit to obtain the first S-N curve SN1 of the target structure G2.

[0021] Step 4: Select a point (N1, S1) from the S-N curve of the reference structure G1 obtained from the test, and collect the relative stiffness degradation curve shown in Figure 3 (b) of the reference structure G1 in the cyclic loading test at the stress level S1 and obtain the relative stiffness degradation rate in the stable section.

[0022] The cyclic loading test mentioned above refers to: a repeated loading-unloading test using a constant stress waveform, i.e., a sine wave with determined peak and valley values.

[0023] The relative stiffness degradation curve is obtained in the following way: in the cyclic loading test, record the serial number of the cycle number as the horizontal axis and the maximum relative stiffness in this cycle as the vertical axis, and plot the relative stiffness degradation curve.

[0024] The relative stiffness degradation rate in the stable section mentioned above refers to: the first derivative value when the second derivative of the relative stiffness degradation curve starts to stabilize at 0.

[0025] Step 5: Repeat Step 4 multiple times, and obtain the relationship g(N, V)=0 between the life N independent of the structure and the relative stiffness degradation rate V in the stable section through the data fitting method.

[0026] Step 6: After collecting the relative stiffness degradation curve of the target structure G2 obtained in the cyclic loading test at the stress level S2 and obtaining the relative stiffness degradation rate in the stable section, substitute it into the relationship between the life and the relative stiffness degradation rate in the stable section obtained in Step 5 to obtain the target structure life N2; repeat multiple times and fit to obtain the S-N curve SN2 of the target structure G2.

[0027] Step 7: Find the intersection point of the curve SN1 and the curve SN2, select the points on SN1 to the left of the intersection point and the points on SN2 to the right of the intersection point, and perform fitting through the least squares method to obtain the final asFigure 4 The S-N curve shown

[0028] Through specific actual experiments, such as Figure 4 shown, the dashed line is the result of the single fast prediction method, which can control the maximum error of the predicted life within 5 times. The thin solid line is the result of the hybrid prediction method, and the mean square error of the predicted S-N curve is reduced by more than 80%.

[0029] Compared with the prior art, the present invention utilizes the structural independence of the relationship between the initial damage, the relative stiffness degradation rate of the stable section, and the microstructure to achieve fast prediction of the S-N curve under any microstructure. Fast prediction methods of the S-N curve based on the initial damage and the relative stiffness degradation rate of the stable section are respectively proposed to achieve fast prediction of the fatigue life. By combining the prediction advantages of the two methods in the low-cycle region and the high-cycle region, the prediction accuracy of the S-N curve is further improved.

[0030] The above specific implementation can be locally adjusted by those skilled in the art in different ways without departing from the principles and purposes of the present invention. The protection scope of the present invention is subject to the claims and is not limited by the above specific implementation, and each implementation within its scope is subject to the present invention.

Claims

1. A hybrid fast prediction method for the S-N curve of SFRP stiffness degradation, characterized in that The first S-N curve of the target structure is obtained based on the relationship between the life of the reference structure and the first-cycle damage, and the second S-N curve of the target structure is obtained based on the relationship between the life of the reference structure and the relative stiffness degradation rate in the stable section. The S-N curve of the target structure is quickly obtained by fitting the two curves; Both the reference structure and the target structure are SFRPs with specific fiber microstructures; The damage mentioned refers to: the initial stiffness minus the slope of the line connecting the current point on the stress-strain curve to the origin, that is, the ratio of the difference in the current stiffness to the initial stiffness; Both the relationship between the life and the first-cycle damage and the relationship between the life and the relative stiffness degradation rate in the stable section are independent of the structure, that is, this relationship is not affected by the specific fiber microstructure, that is, the relationship obtained through one structure can be directly extended to other structures; The relative stiffness degradation rate in the stable section refers to: the first derivative value when the second derivative of the relative stiffness degradation curve stabilizes at 0.

2. The hybrid rapid prediction method of the S-N curve with SFRP stiffness degradation according to claim 1, characterized in that specifically Including: Step 1: Select a point (N1, S1) on the S-N curve of the reference structure G1 obtained from the test, and calculate the damage at the stress level of S1 as the first-cycle damage according to the quasi-static loading stress-strain curve of G1 obtained from the test. Where: E0 is the initial stiffness of the structure, and E is the current stiffness calculated from the stress-strain curve at the stress level S of the structure. Step 2: Repeat Step 1 multiple times, and obtain the life-damage relationship f(N, D)=0 with structure independence through the data fitting method; Step 3: For the target structure G2, select the stress level S2, calculate the damage D2 at the stress level S2 according to the quasi-static loading stress-strain curve of G2 obtained through the test, and predict the target structure life N2 according to the life-damage relationship f(N, D)=0; Repeat multiple times to fit and obtain the first S-N curve SN1 of the target structure G2; Step 4: Select a point (N1, S1) from the S-N curve of the reference structure G1 obtained through the test, collect the relative stiffness degradation curve obtained in the cyclic loading test of the reference structure G1 at the stress level S1, and obtain the relative stiffness degradation rate in the stable section; Step 5: Repeat Step 4 multiple times, and obtain the relationship g(N, V)=0 between the life and the relative stiffness degradation rate in the stable section with structure independence through the data fitting method; Step 6: After collecting the relative stiffness degradation curve obtained in the cyclic loading test of the target structure G2 at the stress level S2 and obtaining the relative stiffness degradation rate in the stable section, substitute it into the relationship between the life and the relative stiffness degradation rate obtained in Step 5 to obtain the target structure life N2; Repeat multiple times to fit and obtain the S-N curve SN2 of the target structure G2; Step 7: Find the intersection point of the curve SN1 and the curve SN2, select the points on SN1 to the left of the intersection point and the points on SN2 to the right of the intersection point, and obtain the final S-N curve through least squares fitting.

3. The hybrid fast prediction method for the S-N curve of SFRP stiffness degradation according to claim 2, characterized in that The cyclic loading test mentioned refers to: a repeated loading-unloading test using a constant stress waveform, that is, a sine wave with determined peak and valley values.

4. The S-N curve hybrid rapid prediction method for SFRP stiffness degradation according to claim 2, characterized in that, The relative stiffness degradation curve is obtained in the following way: In the cyclic loading test, record the serial number of the cycle count as the horizontal axis and the maximum relative stiffness in this cycle as the vertical axis, and plot the relative stiffness degradation curve.

5. The S-N curve hybrid fast prediction method for SFRP stiffness degradation according to claim 2, characterized in that, The relative stiffness degradation rate in the stable section refers to: the first derivative value when the second derivative of the relative stiffness degradation curve begins to stabilize at 0.

Images