WO2022149297A1

WO2022149297A1 - Fatigue life prediction method, fatigue life prediction device, fatigue life prediction program, and recording medium

Info

Publication number: WO2022149297A1
Application number: PCT/JP2021/024825
Authority: WO
Inventors: 壽雄高城; 敬宜村上
Original assignee: 株式会社寿ホールディングス
Priority date: 2021-01-08
Filing date: 2021-06-30
Publication date: 2022-07-14
Also published as: TW202244476A; JP6944736B1; JP2022107231A

Abstract

This fatigue life prediction method includes: acquiring fatigue test data including test results from a fatigue test on a subject carried out under a plurality of conditions, initial defect dimensions of a test piece used in the fatigue test, and test conditions for the fatigue test; calculating, on the basis of the fatigue test data, material constants for calculating the amount of development of a crack per one application of stress load, on the assumption that a crack develops from the initial defect when a stress load equal to or greater than a fatigue limit is applied to the subject and that the fatigue limit decreases according to the dimensions of the defect and the crack developing from the defect; and predicting, on the basis of the amount of development of the crack per one application of stress load calculated using the material constants, the fatigue life when a certain stress is applied on the subject.

Description

Fatigue life prediction method, fatigue life prediction device, fatigue life prediction program and storage medium

The present disclosure relates to a fatigue life prediction method, a fatigue life prediction device, a fatigue life prediction program, and a storage medium.

When evaluating fatigue fracture of an object such as a metal material, fatigue life and fatigue limit are often important evaluation factors, and various studies have been conducted. For example, in Patent Document 1, regarding fatigue life, based on the stress distribution inside a member without cracks, the relationship between crack depth and creep contribution, and the relationship between creep contribution and the values of parameters C and m of the Paris law, A method of estimating the growth behavior of a crack that grows in a member is shown. Further, for example, in Patent Document 2, as a method of predicting the fatigue limit of a metal material having minute defects with respect to the fatigue limit, a method of predicting a tensile fatigue limit and a shear fatigue limit using the latent fatigue crack length of the metal material. It is shown.

International Publication No. 2014/033927 Japanese Unexamined Patent Publication No. 2018-185274

However, there was room for improvement in terms of accuracy in predicting the fatigue life of an object assuming that the stress to be applied fluctuates.

The present disclosure has been made in view of the above, and an object of the present disclosure is to provide a technique capable of predicting fatigue life with higher accuracy.

In order to achieve the above object, the fatigue life prediction method according to one embodiment of the present disclosure includes test results by a fatigue test under a plurality of conditions for an object, initial defect dimensions of a test piece used in the fatigue test, and the fatigue. Fatigue test data including test conditions related to the test are acquired, cracks grow from initial defects when a stress load exceeding the fatigue limit is applied to the object, and cracks grow from defects and defects. On the premise that the fatigue limit decreases according to the dimensions of the cracks, the material constant for calculating the amount of crack growth for one stress load is calculated based on the fatigue test data. , Predicting fatigue life based on the amount of crack growth when a certain stress is applied to the object using the material constants.

According to the above fatigue life prediction method, the amount of crack growth for one stress load can be estimated by specifying the material constant for calculating the crack growth amount for one stress load. By utilizing this, it is possible to predict the fatigue life when a certain stress is applied to the test piece of the object. Here, when a stress load exceeding the fatigue limit is applied to the object, the crack grows from the initial defect, and the fatigue limit decreases according to the size of the defect and the crack grown from the defect. Since it is a premise, the growth of defects (cracks) due to stress load and the accompanying decrease in fatigue limit are taken into consideration, so the net crack growth driving force in the repeatedly applied load is properly grasped and fatigue is achieved. Since the life is predicted, more accurate prediction is realized.

Here, the prediction of the fatigue life is to calculate the fatigue limit based on the defect of the object and the size of the crack extending from the defect, and once when the load is larger than the calculated fatigue limit. It includes calculating the amount of crack growth with respect to stress loading and calculating the number of stress loads until the size of the crack that has grown from the defect and the defect reaches the limit defect size.

With the above configuration, when a certain stress is applied, the amount of crack growth for one stress load when the load is larger than the calculated fatigue limit is calculated, and the dimensions of the defect and the crack propagated from the defect are calculated. The number of stress loads until the limit defect dimension is reached is calculated. Therefore, the fatigue life can be predicted with high accuracy regardless of the magnitude of the stress load on the object.

Here, the calculation of the material constant is an embodiment in which C ^* , m ^* and n ^* in the following mathematical formula (A) for calculating the crack growth amount Δ√area for one stress load are calculated. Can be.

[However, Δ√area is the amount of crack growth for one stress load, σ is the stress amplitude of the load, _σw is the fatigue limit when the load is applied, and √area is the dimension of the defect. .. ]

With the above configuration, it is possible to specify a material constant that can appropriately calculate the amount of crack growth for one stress load, and using this material constant, predict the fatigue life with high accuracy. can do.

In predicting the fatigue life, the stress amplitude applied to the object can be constant or two or more steps.

According to the above configuration, the magnitude of the load on the object is not particularly limited. Therefore, the fatigue life can be predicted with high accuracy under any condition when the stress amplitude to be applied is constant and when there are two or more stages.

The fatigue life predictor according to one embodiment of the present disclosure includes test results of a fatigue test on an object under a plurality of conditions, initial defect dimensions of the test piece used in the fatigue test, and test conditions related to the fatigue test. Fatigue test result acquisition unit that acquires fatigue test data including With the material constant calculation unit that calculates the material constant for calculating the amount of crack growth for one stress load based on the fatigue test data, assuming that the fatigue limit decreases according to the dimensions. Includes a fatigue life prediction unit that predicts fatigue life when a certain stress is applied to the object based on the amount of crack growth for one stress load calculated using the material constants. .. According to the above-mentioned life prediction device, the same effect as the above-mentioned life prediction method is obtained.

The fatigue life prediction program according to one embodiment of the present disclosure includes test results by a fatigue test under a plurality of conditions on an object, initial defect dimensions of a test piece used in the fatigue test, and test conditions related to the fatigue test. Fatigue test data including fatigue test data is obtained, cracks grow from initial defects when a stress load exceeding the fatigue limit is applied to the object, and the size of the defects and cracks grown from the defects On the premise that the fatigue limit is lowered, the material constant for calculating the amount of crack growth for one stress load is calculated based on the fatigue test data, and the material constant is used. Based on the calculated amount of crack growth for one stress load, the computer system is made to predict the fatigue life when a certain stress is applied to the object. According to the above-mentioned life prediction program, the same effect as that of the above-mentioned life prediction method is obtained.

The storage medium according to one embodiment of the present disclosure includes test results of a fatigue test on an object under a plurality of conditions, initial defect dimensions of the test piece used in the fatigue test, and test conditions related to the fatigue test. Fatigue test data is acquired, cracks grow from initial defects when a stress load exceeding the fatigue limit is applied to the object, and fatigue grows according to the dimensions of the defects and the cracks that grow from the defects. Based on the fatigue test data, the material constant for calculating the amount of crack growth for one stress load is calculated on the premise that the limit is lowered, and the material constant is calculated using the material constant. A computer that stores a fatigue life prediction program that predicts the fatigue life when a certain stress is applied to the object based on the amount of crack growth for one stress load and causes the computer system to execute. It is a readable storage medium. According to the above-mentioned storage medium, the same effect as the above-mentioned life prediction method is obtained.

According to the present disclosure, a technique capable of predicting fatigue life with higher accuracy is provided.

FIG. 1 is a schematic view of a fatigue life prediction device according to one embodiment. FIG. 2 is a flowchart illustrating an example of a fatigue life prediction method. FIG. 3 is a flowchart illustrating an example of a fatigue life prediction method using material constants included in the fatigue life prediction method. FIG. 4 is a flowchart illustrating an example of a method for calculating material constants included in the fatigue life prediction method. FIG. 5 is a flowchart illustrating an example of a subroutine for obtaining N _fpred included in FIG. FIG. 6 is a diagram showing an example of the hardware configuration of the fatigue life prediction device. FIG. 7 is a diagram for explaining the evaluation result of fatigue life prediction by the fatigue life prediction device.

Hereinafter, the mode for carrying out the present disclosure will be described in detail with reference to the attached drawings. In the description of the drawings, the same elements are designated by the same reference numerals, and duplicate description will be omitted.

[Fatigue life predictor]
First, with reference to FIG. 1, a schematic configuration of the fatigue life prediction device 1 according to the embodiment will be described. The fatigue life prediction device 1 shown in FIG. 1 is a device that predicts the fatigue life when a predetermined stress is applied to an object.

The type of object for which the fatigue life is predicted is not particularly limited, but industrial products, structures, etc. and industrial materials that can be used for members of industrial products, structures, etc. can be the main objects. Further, it is considered that the fatigue life prediction device 1 can be used to predict the life of industrial materials, especially metals, with high accuracy. The above method can be applied to non-metals as long as the material can obtain an SN curve.

The fatigue life predicted by the fatigue life prediction device 1 is the number of repeated stress loads until the object is repeatedly stressed when the object is repeatedly stressed. Generally, when a stress of a predetermined value or more is applied to an object, the object is destroyed at a stage exceeding a certain number of times. The fatigue life prediction device 1 has a function of predicting the number of repetitions of stress load until this fracture occurs.

Fatigue life is generally predicted by creating an SN curve based on the results of fatigue tests. At this time, usually, an SN curve is created by applying the results of a plurality of fatigue tests acquired with a constant stress amplitude to a known model. However, when the stress fluctuates, the stress smaller than the fatigue limit based on the above SN curve may contribute to the growth of cracks. Therefore, the fatigue life can be estimated based on the SN curve. It is known that there is room for improvement.

The fatigue life prediction device 1 according to the present embodiment focuses on the crack growth behavior inside the material, which is the process of fatigue, and formulates the behavior based on mechanical grounds to determine the fatigue life under variable stress. Highly accurate prediction can be achieved. Although the details will be described later, in the fatigue life prediction device 1, when a stress exceeding the fatigue limit is applied to the object, minute cracks (initial defects) existing in the object grow (grow). Based on the premise, the fatigue life is predicted by evaluating the growth of cracks due to repeated load, considering the decrease in fatigue limit due to the increase in cracks. To this end, first, three types of material constants, which are parameters used for predicting fatigue life, are calculated from fatigue test data under a plurality of conditions relating to the object. Then, using the calculated three types of material constants, the amount of crack growth when a certain stress is applied to the object is calculated, and the crack dimensions (initial defects and cracks that have grown from the initial defects) are calculated. Predict fatigue life by updating (dimensions), continuing to perform the next assessment based on the fatigue limit according to the updated new crack dimensions, and identifying the stage at which the fatigue crack dimensions exceed the limit. ..

(Functional part of fatigue life prediction device)
Each part of the fatigue life prediction device 1 will be described with reference to FIG. As shown in FIG. 1, the fatigue life prediction device 1 includes a fatigue test data acquisition unit 11 (fatigue test result acquisition unit), a prediction condition acquisition unit 12, a material constant calculation unit 13, a fatigue life prediction unit 14, and a result output unit. 15 (output unit) and a data storage unit 16 are included.

The fatigue test data acquisition unit 11 has a function of acquiring data related to the fatigue test result of the object whose fatigue life is predicted. The data related to the fatigue test results acquired by the fatigue test data acquisition unit 11 is the data of general fatigue test results, and specifically, "a stress of a constant amplitude for a test piece having a defect of a certain size". These are the test conditions and their results when the test related to "the number of repetitions until breakage by applying a load" was performed. Details will be described later, but multiple of the above test results will be obtained and used for fatigue life prediction.

The prediction condition acquisition unit 12 has a function of acquiring information specifying conditions for fatigue life prediction in the fatigue life prediction device 1. The information that specifies the conditions to be acquired by the prediction condition acquisition unit 12 includes, for example, the Vickers hardness (HV) of the object, the final crack size regarded as fracture, and the information necessary for calculating the above three types of material constants. Numerical ranges and increments of candidate material constants can be mentioned. Further, as the information specifying the conditions to be acquired by the prediction condition acquisition unit 12, for example, as the information necessary for predicting the fatigue life after calculating the material constant, the initial value of the defect dimension in the object and the fatigue life prediction The increment of the number of repetitions of the load and the upper limit value when performing the above are mentioned. The above information is necessary when the material constant is calculated and the fatigue life is predicted by the method described later, but depending on the actual calculation method, only a part of the above information may be used. However, information not included in the above may be used. Therefore, the type of information acquired by the prediction condition acquisition unit 12 can be changed according to a specific method for predicting fatigue life.

The material constant calculation unit 13 has a function of calculating three types of material constants based on the information acquired by the fatigue test data acquisition unit 11 and the prediction condition acquisition unit 12. The method of calculating the material constant will be described later.

The fatigue life prediction unit 14 has a function of predicting the fatigue life of an object by using the material constant calculated by the material constant calculation unit 13. The method of predicting fatigue life will be described later.

The result output unit 15 has a function of outputting the result of the fatigue life prediction obtained by the processing of the fatigue life prediction unit 14. The output method is not particularly limited, but known methods such as file output, screen output, and return of values to other programs can be used. When outputting the result, for example, the material constant calculated by the material constant calculation unit 13 may be output together.

The data storage unit 16 has a function of storing the information acquired by the fatigue test data acquisition unit 11 and the prediction condition acquisition unit 12 and the information required for the processing performed by each of the above units. It also has a function of storing the results obtained by the processing in the material constant calculation unit 13 and the fatigue life prediction unit 14.

[Fatigue life prediction method]
Next, a fatigue life prediction method by the fatigue life prediction device 1 will be described with reference to FIGS. 2 to 5.

First, in the fatigue life prediction device 1, fatigue test data is acquired by the fatigue test data acquisition unit 11 (step S01). This process may be performed, for example, by the operator (user) of the fatigue life prediction device 1 operating the fatigue life prediction device 1. Further, the test results acquired by the fatigue test device designated in advance may be sequentially transmitted to the fatigue life prediction device 1, and the fatigue test data may be sequentially acquired by the fatigue life prediction device 1. May be.

Next, the fatigue test data prediction condition acquisition unit 12 acquires information that specifies conditions related to a series of processes related to fatigue life prediction in the fatigue life prediction device 1 (step S02). This process may be performed, for example, by the operator (user) of the fatigue life prediction device 1 operating the fatigue life prediction device 1. The order of step S01 and step S02 is not particularly limited, and for example, step S02 may be performed first, or step S01 and step S02 may be performed at the same time. Further, the information for designating the conditions related to the fatigue life prediction may be in a state of being acquired and held in advance by the prediction condition acquisition unit 12 of the fatigue life prediction device 1.

Next, the material constant calculation unit 13 of the fatigue life prediction device 1 first calculates three material constants (C ^* , m ^* , n ^* ) based on the above information (step S03). The method of calculating the material constant is described in FIGS. 3 and 4, but the details will be described later.

Next, the fatigue life prediction unit 14 of the fatigue life prediction device 1 predicts the fatigue life using the result calculated by the material constant calculation unit 13 (step S04). The method for predicting fatigue life is described in FIG. 5, but details will be described later.

Next, the result output unit 15 of the fatigue life prediction device 1 outputs the result of the fatigue life prediction obtained by the processing of the fatigue life prediction unit 14 (step S05). At the time of output, conversion processing or the like for converting the prediction result into a state suitable for output may be performed.

(Basic concept of fatigue life prediction)
Before explaining the detailed procedure, the basic idea of the method described in this embodiment will be described.

First, we use the idea that if there is a dimensional defect in the material, the fatigue limit σ _w (stress amplitude) corresponding to the defect is determined. The fatigue limit at this time can be predicted by using the following mathematical formula (1), for example, in the case of a metal material and a stress ratio R = -1. Here, the stress ratio R is defined by the ratio of the minimum value and the maximum value of the cyclic stress. This formula (1) is described in Takanori Murakami, "Metal Fatigue: Effects of MicroDefects and Occlusions," Yokendo, 1993. The method for calculating σ _w is not limited to the mathematical formula (1). For example, when R = -1, or when this method is applied to a non-metal material, it is conceivable to predict σ _W by another method.

In this way of thinking, stress below the fatigue limit does not contribute to fatigue. In addition, if a load larger than the fatigue limit is applied, fatigue will be given to the object. That is, it can be said that σ / σ _w -1 is a mechanical quantity that contributes to fatigue.

Further, the above mathematical formula (1) shows that a defect exists in the material and the fatigue limit is determined by the dimension thereof. On the other hand, when a load with a stress (stress amplitude) equal to or higher than the fatigue limit is applied to the object, cracks are generated from the previously existing defects in the object. Therefore, the size of the defect including the crack increases as the number of repeated stress loads increases. According to the mathematical formula (1), it can be seen that the fatigue limit σ _w gradually decreases according to the size of the defect (defect size). That is, even if the stress load is the same, the value of the mechanical quantity σ / σ _w -1, which contributes to fatigue, gradually increases as the internal defect increases.

As described above, it can be said that the mechanical quantity σ / σ _w -1 becomes the driving force (crack growth driving force) for expanding the defect (crack). At this time, the crack growth state using the above mechanical quantity can be described as the mathematical formula (2).

In the formula (2), a corresponds to the size of the defect √area, da / dN is the amount of growth of the crack a per one cycle of the load, and C ^* , m ^* , and n ^* are material constants. .. It may be assumed that n ^* is usually 1. The above mathematical formula (2) is numerically integrated, and the place where the value of the crack dimension a becomes a sufficiently large value (generally about 1 mm or more) is predicted as a substantial fatigue life.

When the above method is used, even if the value of the stress applied to the object changes during the repetition, the growth and expansion of the crack can be tracked. Therefore, it is possible to predict the fatigue life of an object even under conditions where the load fluctuates.

(Fatigue life prediction method)
The method of predicting the fatigue life performed in step S04 will be described with reference to FIG. Here, it is assumed that the above-mentioned material constants C ^* , m ^* , and n ^* are obtained by using the method described later.

First, input the parameters used for predicting the fatigue life (step S11). As shown in FIG. 3, the parameters shown in Table 1 below are prepared for predicting fatigue life.

Of the above settings, the initial defect dimension √area ₀ is the dimension of the initial defect contained in the material of the object and is defined by the square root of the projected area where the defect is projected onto the plane perpendicular to the principal stress. For example, a method of measuring using an optical method, an X-ray CT, or the like can be considered. It may also be set using the method described in Y. Murakami, "Metal Fatigue: Effects of MicroDefects and Enclosures," Yokendo, 1993.

Further, the critical crack size √area _crit may change depending on the material properties of the object, but may be set to 1 mm, for example, in consideration of the process from crack growth to fracture. Since the number of repetitions required from 1 mm to the final fracture is relatively small compared to the total number of repetitions, such a setting is also possible. Further, the upper limit N _stop of the number of repetitions N may be set large (for example, ¹⁰⁷ ) in consideration of the expiration date of the object and the repetition period of stress. Further, when the stress is a function σ _a (N) by N, the accuracy of predicting the fatigue life is improved by setting the increment ΔN according to the interval of fluctuation, and the calculation time is shortened. It can be compatible.

Next, the initial value is set (step S12). As the initial value, N = 0 is set, and the initial defect dimension √area ₀ is selected as √area.

Next, σ in cycle N and fatigue limit σ _w in √area are calculated (steps S13 to S16). Specifically, first, in order to set the cycle N, N = N + ΔN is set (S13), and then it is confirmed that N is less than the upper limit N _stop (S14), and the calculation is continued (S14-YES). In), the stress amplitude σ corresponding to the cycle N (in the case of a function of N, σa (N) is set (S15), and then the fatigue limit σ _w is calculated based on the equation (1). (S16). At this time, F = 1.43 (in the case of surface defects) or F = 1.56 (in the case of internal defects) can be selected depending on the position of the defect or the crack extending from the defect, but generally. Uses F = 1.43, which is the safe side.

When N becomes equal to or greater than the upper limit N _stop (S14-NO), the process is terminated and the result of non-breaking is output (step S17).

After calculating the fatigue limit σ _w , it is confirmed whether or not the stress σ applied as a load is larger than the fatigue limit σ _w (step S18). When the stress σ applied as a load is equal to or less than the fatigue limit σ _w (S18-NO), the crack does not grow due to the repetition of the stress, and the cycle N is newly set and the above procedure (S13 to S18) is performed. The same study is performed on the stress applied next by repeating. On the other hand, when the stress σ applied as a load is larger than the fatigue limit σ _w (S18-YES), it is assumed that the crack grows due to the repetition of the stress, and the amount of crack growth is calculated (step S19). Specifically, based on the above formula (2), the crack growth amount per repetition is calculated using the formula (3), and the new crack size is calculated using the formula (4).

After calculating the new crack size √area, it is confirmed whether or not the new crack size √area is larger than the limit crack size √area _crit (step S20). When the new crack size √area is less than or equal to the limit crack size √area _crit (S20-NO), it is determined that the object has not yet broken, a new cycle N is set, and the above procedure (S13 to S20) is repeated. .. On the other hand, when the new crack size √area is larger than the limit crack size √area _crit (S20-YES), it is determined that the object has broken, and the number of repeated breaks N _f is set to the current number of cycles N (step S21). Then, the fracture repetition number N _f is output as a prediction result of the fatigue life (step S22). This completes a series of processes related to the prediction of fatigue life.

(Calculation method of material constant)
The method of calculating the material constant performed in step S03 will be described with reference to FIGS. 4 and 5. FIG. 4 is a flowchart relating to one step included in the flowchart shown in FIG. As described above, the three material constants to be calculated are C ^* , m ^* , and n ^* .

When calculating the material constant, the mathematical formula used in the above-mentioned fatigue life prediction method is used. Specifically, using arbitrary C ^* , m ^* and n ^* , the predicted value N _fpred of the number of repeated fractures N _f is calculated based on the measurement conditions included in the results of the fatigue test. The difference between this predicted value N _fpred and the measured value N _fexp of the number of fracture repetitions N _f obtained by the fatigue test is calculated, and the combination of C ^* , m ^* , and n ^* that minimizes this difference is obtained. By doing so, C ^* , m ^* and n ^* can be specified. The method for specifying the optimum material constants C ^* , m ^* , n ^* such that the difference between the predicted value N _fpred and the measured value N _fexp is minimized is not particularly limited, but in the following embodiment, the formula (5) A method for specifying the C ^* , m ^* , and n ^* that minimizes the objective function O by setting the objective function O shown in the above will be described. In formula (5), M is the number of sets of experimental data, and E _i is the difference E in the logarithms of N _fpred and N _fpred in the i-th set of experimental data. E _i is defined by the mathematical formula (6).

For example, iterative calculation can be used as a method for specifying C ^* , m ^* and n ^* that minimize the objective function O. Specifically, there is a method of brute force calculation of O for C ^* , m ^* and n ^* within a preset range. In addition, various commonly used optimization algorithms can be used.

4 and 5 show an example of a procedure for concretely executing the above method.

First, the parameters used for calculating the material constants are input (step S31). As shown in FIG. 4, the parameters shown in Table 2 below are prepared for calculating the material constants.

The number of data of fatigue test results is 1 to M, but when obtaining three material constants, at least three different test data (conditions are different from each other) are required (i ≧ 3). ). However, in reality, it can be assumed that the material constant n ^* is 1, and in that case, the test data may be two.

Next, the initial value is set (step S32). As initial values, the minimum values C ^* _min , m ^* _min , and n ^* _min are set for C ^* , m ^* , and n ^* , respectively. Further, as an example of the initial value of the objective function O, _Default = 1000 is set. As described above, since the procedure is to specify the material constants that can minimize the objective function, a sufficiently large value is set as the initial value. Note that _Update refers to the latest minimum value of the objective function, and is used as a reference when performing repeated calculations in the procedure described below.

Next, first, with i = 1 (step S33), the predicted value N _fpred of the number of repeated fractures N _f is calculated (step S34).

The details of step S34, that is, the method of calculating the predicted value N _fpred of the number of repeated fractures N _f , is shown in FIG. 5, but is the same as the procedure shown in schematic FIG. 3, and will be briefly described.

First, input the parameters used for predicting the fatigue life (step S51). Here, a part of the parameters prepared in step S31 is input.

Next, the fatigue limit σ _w is calculated based on the mathematical formula (1) using the conditions used in the fatigue test (parameters input above) (step S52). As a result, the predicted fatigue limit based on the fatigue test data is calculated. After calculating the fatigue limit σ _w , it is confirmed whether or not the load stress σ _exp (i) used in the fatigue test is larger than the fatigue limit σ _w (step S53). When the load stress σ _exp (i) is equal to or less than the fatigue limit σ _w (S53-NO), it is predicted that cracks will not _grow due to this load stress σ _exp (i). A notification is output indicating that the data cannot be used in the calculation of the fatigue life prediction (step S54). In this case, since the predicted value N _fpred cannot be calculated, an error message or a warning may be output.

On the other hand, when the load σ _exp (i) used in the fatigue test is larger than the fatigue limit σ _w (S53-YES), the amount of crack growth is calculated (steps S55 to 57). Specifically, N = 0 is set as the initial value, the initial defect dimension √area ₀ is selected as √area (S55), then N = N + ΔN (S56), the predicted fatigue limit σ _w , and one repetition. The crack growth amount Δ√area and the new crack size √area are calculated (S57). These calculation methods are the same as steps S16 and S18 shown in FIG. However, the following mathematical formula (7) is used to calculate the crack growth amount Δ√area. The formula (7) corresponds to the formula (3), except that the load σ _exp (i) used in the test is used as the stress σ.

After calculating the new crack size √area, it is confirmed whether or not the new crack size √area is larger than the limit crack size √area _crit (step S58). When the new crack dimension √area is less than or equal to the limit crack dimension √area _crit (S58-NO), it is determined that the object has not yet broken, a new cycle N is set, and the above procedure (S56 to S57) is repeated. .. On the other hand, when the new crack size √area is larger than the limit crack size √area _crit (S58-YES), it is determined that the object has broken, and the predicted value N _fpred of the number of repeated breaks N _f is set to the current number of cycles N. (Step S59). Then, the predicted value N _fpred is output as the predicted result of the fatigue life (step S60). In this series of procedures, the predicted value N _fpred based on the fatigue test data can be obtained.

Returning to FIG. 4, when the predicted value N _fpred is obtained, the difference E (i) between the test result and the predicted value is calculated based on the above mathematical formula (6) (step 35). This series of processes is repeated while increasing i by +1 from i = 1 to M (steps S36 and 37). For all fatigue test results (i = 1 to M), when the calculation of the difference E (i) between the test result and the predicted value is completed (S36-YES), it is substituted for the objective function O shown in the formula (5). Then, the calculation result is calculated (S38). Then, the calculation result of the objective function O is compared with _Update (step S39). When the calculation result of the current objective function O is smaller than the minimum value _Update of the previous objective function (S39-YES), the _Update is changed (updated) to the calculation result of the current objective function O, and the result is changed (updated). The current C ^* , m ^* and n ^* used in the calculation of the objective function O are changed (updated) as C ^* _updated , m ^* _updated and n ^* updated (step _S40 ). If the calculation result of the objective function O is equal to or greater than the minimum value of the objective function before that ( _{S39-NO), the update} _or the like is not updated (S40).

After that, the above processing (S33 to S40) is repeated while changing C ^* , m ^* and n ^* . Specifically, first, with m ^* and n ^* fixed at the minimum values m ^* _min and n ^* _min , the calculation is repeated until C ^* _{max is} reached while changing C ^* in increments of ΔC ^* (step S41). , 42). As a result, it is possible to specify the C ^* _update used when calculating the minimum value _Update with m ^* and n ^* fixed. After that, by performing the same calculation for m ^* (steps S43 to S45) and n ^* (steps S46 to S48), when the minimum value _Update is calculated with the other material constants fixed to the minimum value. The m ^* _update and n ^* _update used can be calculated.

By performing this series of iterative calculations, the smallest objective function within the range of m ^* _min ≤ m ^* ≤ m ^* _max , C ^* _min ≤ C ^* ≤ C ^* _max , n ^* _min ≤ n ^* ≤ n ^* _max . Since the optimum C ^* _update , m ^* _update and n ^* _update that give O are obtained, these are output as the calculation results of C ^* , m ^* and n ^* (step S49).

By the above procedure, C ^* , m ^* and n ^* can be calculated using the fatigue test results.

[Hardware configuration]
The hardware configuration of the fatigue life prediction device 1 will be described with reference to FIG. FIG. 6 is a diagram showing an example of the hardware configuration of the fatigue life prediction device 1. The fatigue life predictor 1 includes one or more computers 100. The computer 100 includes a CPU (Central Processing Unit) 101, a main storage unit 102, an auxiliary storage unit 103, a communication control unit 104, an input device 105, and an output device 106. The fatigue life prediction device 1 is composed of one or a plurality of computers 100 composed of these hardware and software such as a program.

When the fatigue life prediction device 1 is composed of a plurality of computers 100, these computers 100 may be connected locally or may be connected via a communication network such as the Internet or an intranet. By this connection, one fatigue life prediction device 1 is logically constructed.

The CPU 101 executes an operating system, an application program, and the like. The main storage unit 102 is composed of a ROM (Read Only Memory) and a RAM (Random Access Memory). The auxiliary storage unit 103 is a storage medium composed of a hard disk, a flash memory, or the like. The auxiliary storage unit 103 generally stores a larger amount of data than the main storage unit 102. At least a part of the material constant calculation unit 13 and the fatigue life prediction unit 14 is realized by the auxiliary storage unit 103. The communication control unit 104 is composed of a network card or a wireless communication module. At least a part of the fatigue test data acquisition unit 11, the prediction condition acquisition unit 12, and the result output unit 15 may be realized by the communication control unit 104. The input device 105 includes a keyboard, a mouse, a touch panel, a microphone for voice input, and the like. For example, at least a part of the prediction condition acquisition unit 12 may be realized by the input device 105. The output device 106 includes a display, a printer, and the like. At least a part of the result output unit 15 is realized by the output device 106. For example, the output device 106 may display the result of fatigue life prediction on a display or the like.

The auxiliary storage unit 103 stores the program 110 (fatigue life prediction program) and data necessary for processing in advance. The program 110 causes the computer 100 to execute each functional element of the fatigue life prediction device 1. The program 110 executes, for example, the processes related to steps S01 to S04 described above in the computer 100. For example, the program 110 is read by the CPU 101 or the main storage unit 102, and operates at least one of the CPU 101, the main storage unit 102, the auxiliary storage unit 103, the communication control unit 104, the input device 105, and the output device 106. For example, the program 110 reads and writes data in the main storage unit 102 and the auxiliary storage unit 103.

The program 110 may be provided after being recorded on a tangible storage medium such as a CD-ROM, a DVD-ROM, or a semiconductor memory. The program 110 may be provided as a data signal via a communication network.

[Effectiveness of fatigue life prediction results]
The results of verifying the effectiveness of the fatigue life prediction result by the fatigue life prediction device 1 are shown below.

First, a test piece of carbon steel S45C (HV = 176) was prepared as an object, and two experimental data were obtained by a fatigue test. The experimental data is as follows.
Experimental data 1: σ = 270MPa, N _f = 30,407
Experimental data 2: σ = 235MPa, N _f = 120, 264
The experimental data 1 and 2 are shown in a diagram (SN curve diagram) showing the correspondence between the fracture repetition number Nf and the stress amplitude σ shown in FIG. 7. The initial defect dimension √area ₀ in the test piece of the object was 92 μm.

Next, using the above two experimental data, the material constants C ^* , m ^* and n ^* were calculated by the above-mentioned calculation method of the material constants C ^* , m ^* and n ^* . However, since it was assumed that n ^* = 1 from past experimental data and dimensional analysis, C ^* and m ^* were actually calculated as follows.
C ^* = ^10-3.4
m ^* = 2.8

Next, using the above material constants, the fatigue life of the object (test piece) in the experiment simulating the following fluctuating load was predicted. The initial defect dimension √area ₀ was set to 92 μm.

(Test conditions)
Step 1: Stress load up to N = 23,400 at σ = 270 MPa Step 2: After completion of step 1, stress load until fracture at σ = 185 MPa

As a result of the calculation based on the above-mentioned fatigue life prediction method, it was predicted that the size √area of the crack generated from the initial defect was 251 μm at the time when the stress loading in step 1 was completed. The fatigue limit at this time was predicted to be 169 MPa. From this result, it was estimated that the load stress in step 2 was lower than the fatigue limit in the initial state and higher than the fatigue limit at the end of the stress load in step 1. Further, the predicted value of fatigue life, which is the predicted value of the number of stress loads until fracture in step 2 (the predicted value of the number of stress loads in steps 1 and 2), was N _fpred = 564,700.

On the other hand, when the fatigue life was actually measured using the test piece under the above test conditions, the fatigue life was N _fexp = 585,830. The breakdown was 23,400 (experimental conditions) in step 1 and 562,430 in step 2.

Comparing the above predicted value N _fpred with the measured value N _fexp , the relative error was 4%. From this result, it can be said that the fatigue life could be predicted with sufficiently high accuracy.

FIG. 7 shows the above test results. In FIG. 7, the estimated fatigue limit value σ _W = 199 MPa calculated from the conditions in the initial state is shown as T1, and the estimated fatigue limit value σ _W = 169 MPa at the time of switching from step 1 to step 2 is set as T2. Shows.

[Action]
According to the fatigue life prediction method by the fatigue life prediction device 1 described above, the material constant for calculating the crack growth amount for one stress load is specified, so that the crack growth amount for one stress load is specified. Can be estimated. By utilizing this, it is possible to predict the fatigue life when a certain stress is applied to the object. Here, when a stress load exceeding the fatigue limit is applied to the object, the crack grows from the initial defect, and the fatigue limit decreases according to the size of the defect and the crack grown from the defect. Since it is a premise, the growth of defects (cracks) due to stress loading and the accompanying decrease in fatigue limit are taken into consideration. Therefore, the fatigue life is predicted by appropriately grasping the net crack growth driving force in the stress repeatedly applied, and the prediction with higher accuracy is realized.

As mentioned above, it has been pointed out that there is room for improvement in the estimation of fatigue limit and fatigue life based on the SN curve. In this regard, a new interpretation is added to the SN curve, for example, the Modified Miner's rule, which extends the area destroyed by a finite number of iterations below the fatigue limit to form a design curve. That was being considered. However, these methods are logical for associating the experimental results with the SN curve, and there is still room for improvement in that the mechanical basis is not sufficient.

On the other hand, in the above-mentioned fatigue life prediction method, when a stress load exceeding the fatigue limit is applied to the object, cracks grow from the initial defect and the size of the defect and the crack grown from the defect. Assuming that the fatigue limit is lowered, the amount of crack growth for one stress load is calculated, and the fatigue life is predicted based on this result. In this way, by capturing the change in the object when stress is applied to the object and predicting the fatigue life, it is possible to predict the fatigue life with high accuracy unlike the conventional method. In particular, the above-mentioned fatigue life prediction method predicts the fatigue life after recognizing that the size of the crack increases when the object is repeatedly subjected to stress load, and as a result, the fatigue limit of the object decreases. Is very different from the conventional method. In the conventional method, the increase of cracks during repeated stress loading is not taken into consideration. On the other hand, in the above-mentioned fatigue life prediction method, the decrease in the fatigue limit due to the increase in the size of the crack is grasped moment by moment, and the result is reflected in the prediction of the fatigue life. Therefore, the above fatigue life prediction method not only makes it possible to predict the fatigue life with higher accuracy, but also makes it possible to predict the fatigue life even when the magnitude of stress changes in the middle. It has become.

In addition, the above method can significantly reduce the number of fatigue tests required to predict fatigue life. According to the method described in the above embodiment, the number of material constants can be three (or two by adding assumptions), so that the number of fatigue tests may be at least three (or two). Conventionally, it has been considered to perform multiple fatigue tests to create an SN curve that is more in line with the actual situation, but it was desired to increase the number of fatigue tests in consideration of accuracy. In addition, when trying to calculate the fatigue life more accurately, it is conceivable to perform a fatigue test in which stress is actually applied using a test piece, but it is necessary to repeat the fatigue test every time the applied stress is changed. .. On the other hand, the method described in the above embodiment is advantageous over the conventional method in that the fatigue life can be predicted accurately from less fatigue test data. In addition, even if the stress to be applied is changed as a condition for predicting fatigue life, the above method can be easily recalculated, so that it can be applied to various conditions. It is advantageous for the method of.

In addition, when predicting the fatigue life, the fatigue limit is calculated based on the dimensions of the defect of the object and the cracks that have propagated from the defect, and for one stress load when the load is larger than the calculated fatigue limit. Calculation of the amount of crack growth and calculation of the number of stress loads until the size of the crack that has grown from the defect reaches the limit defect size are performed. Specifically, when a certain stress is applied and the load is larger than the calculated fatigue limit, the amount of crack growth for one stress load is calculated, while the load stress is higher than the calculated fatigue limit. When it is small, the amount of crack growth for one stress load is set to 0, and the number of stress loads until the defect and the crack propagated from the defect reach the limit defect dimension is calculated. Therefore, the fatigue life can be predicted with high accuracy regardless of the magnitude of the stress load.

Further, to calculate the material constant, C ^* , m ^* and n ^* in the above-mentioned mathematical formula (3) for calculating the crack growth amount Δ√area for one stress load are calculated. With such a configuration, it is possible to specify a material constant that can appropriately calculate the amount of crack growth for one stress load, and using this material constant, the fatigue life can be achieved with high accuracy. Can be predicted.

Also, in predicting fatigue life, the stress amplitude applied to the object is constant or two or more steps. As described above, in the method described in this embodiment, the magnitude of the load is not particularly limited. Therefore, the fatigue life can be predicted with high accuracy under any condition when the stress amplitude to be applied is constant and when there are two or more stages. In particular, it is clear from the above evaluation of effectiveness that even if the stress amplitude applied to the object is two or more steps, it can be predicted with high accuracy.

[Modification example]
As described above, the present disclosure is not necessarily limited to the above-described embodiment, and various changes can be made without departing from the gist thereof.

In the above embodiment, the case where the fatigue life prediction device 1 is configured by one computer system has been described, but it may be configured by a plurality of computer systems.

In the above embodiment, it is assumed that the operator directly operates the fatigue life prediction device 1 to perform the above-mentioned processing. However, the fatigue life prediction device 1 and the fatigue life prediction method according to the above embodiment may be provided as a service via the Web or the like accessible to, for example, an unspecified number of users.

1 ... Fatigue life prediction device, 11 ... Fatigue test data acquisition unit, 12 ... Prediction condition acquisition unit, 13 ... Material constant calculation unit, 14 ... Fatigue life prediction unit, 15 ... Result output unit, 16 ... Data storage unit.

Claims

Obtaining fatigue test data including test results by fatigue test under multiple conditions for an object, initial defect dimensions of the test piece used in the fatigue test, and test conditions related to the fatigue test.
It is assumed that the crack grows from the initial defect when a stress load equal to or higher than the fatigue limit is applied to the object, and that the fatigue limit decreases according to the size of the defect and the crack developed from the defect. As a result, the material constant for calculating the amount of crack growth for one stress load is calculated based on the fatigue test data.
Using the material constants, predicting the fatigue life based on the amount of crack growth when a certain stress is applied to the object.
Fatigue life prediction methods, including.
Predicting the fatigue life is
To calculate the fatigue limit based on the defect size of the object,
To calculate the amount of crack growth for one stress load when the load is larger than the calculated fatigue limit,
To calculate the number of stress loads until the dimensions of the defects and cracks propagated from the defects reach the limit defect dimensions.
The fatigue life prediction method according to claim 1.
The calculation of the material constant is to calculate C * , m * and n * in the following formula (A) for calculating the crack growth amount Δ√area for one stress load, claim 1. Alternatively, the fatigue life prediction method according to 2.

[However, Δ√area is the amount of crack growth for one stress load, σ is the stress amplitude of the load, σw is the fatigue limit when the load is applied, and √area is the dimension of the defect. .. ]
The fatigue life prediction method according to any one of claims 1 to 3, wherein the stress amplitude applied to the object is constant or has two or more stages in predicting the fatigue life.
Fatigue test result acquisition unit that acquires fatigue test data including the test results of the fatigue test under multiple conditions for the object, the initial defect dimensions of the test piece used in the fatigue test, and the test conditions related to the fatigue test. When,
It is assumed that the crack grows from the initial defect when a stress load equal to or higher than the fatigue limit is applied to the object, and that the fatigue limit decreases according to the size of the defect and the crack developed from the defect. As a material constant calculation unit for calculating the material constant for calculating the amount of crack growth for one stress load based on the fatigue test data.
A fatigue life prediction unit that predicts the fatigue life when a certain stress is applied to the object based on the amount of crack growth for one stress load calculated using the material constants.
Fatigue life predictor, including.
Obtaining fatigue test data including test results by fatigue test under multiple conditions for an object, initial defect dimensions of the test piece used in the fatigue test, and test conditions related to the fatigue test.
It is assumed that the crack grows from the initial defect when a stress load equal to or higher than the fatigue limit is applied to the object, and that the fatigue limit decreases according to the size of the defect and the crack developed from the defect. As a result, the material constant for calculating the amount of crack growth for one stress load is calculated based on the fatigue test data.
Predicting the fatigue life when a certain stress is applied to the object based on the amount of crack growth for one stress load calculated using the material constants.
Fatigue life prediction program that causes a computer system to execute.
Obtaining fatigue test data including test results by fatigue test under multiple conditions for an object, initial defect dimensions of the test piece used in the fatigue test, and test conditions related to the fatigue test.
It is assumed that the crack grows from the initial defect when a stress load equal to or higher than the fatigue limit is applied to the object, and that the fatigue limit decreases according to the size of the defect and the crack developed from the defect. As a result, the material constant for calculating the amount of crack growth for one stress load is calculated based on the fatigue test data.
Predicting the fatigue life when a certain stress is applied to the object based on the amount of crack growth for one stress load calculated using the material constants.
A computer-readable storage medium that stores a fatigue life prediction program that causes a computer system to run.