JP7487248B2 - Fatigue damage level identification device and method for identifying fatigue damage level - Google Patents

Description

The present invention relates to a fatigue damage level determination device and a fatigue damage level determination method.

Various devices and methods have been researched for determining the remaining life of materials. For example, Japanese Patent Application Laid-Open No. 2006-250683 (Patent Document 1) discloses a system for determining the initiation point of fatigue damage that may occur in a test specimen by applying a repeated load to the test specimen and whether or not the fatigue damage will progress, the fatigue damage determination system being composed of at least an infrared camera, an imaging device that creates a dissipated energy image from the amount of temperature change captured by the infrared camera, and a strain gauge, and the fatigue damage determination system determines whether or not the fatigue damage will progress based on the dissipated energy image while determining the initiation point of fatigue damage that occurs with repeated loads using the strain gauge attached to the test specimen.

JP 2010-223957 A (Patent Document 2) discloses a fatigue limit determination system that includes a vibrator that repeatedly applies a stress amplitude to a measurement object, an infrared thermography device that measures minute temperature changes in the measurement object and obtains a temperature image of the measurement object, and an information processing device having a fast Fourier transform means that processes the temperature image of the measurement object obtained from the infrared thermography device, and that has a process of evaluating the stress concentration factor of the measurement object, a process of measuring dissipated energy, and a process of determining the fatigue limit from the value of the stress concentration factor obtained in the process of evaluating the stress concentration factor and the measurement results obtained from the process of measuring the dissipated energy.

In addition, Japanese Patent Laid-Open Publication No. 2016-024056 (Patent Document 3) discloses a fatigue limit stress specification system that gradually increases the load applied to a measurement object and measures the temperature amplitude of the measurement object generated for each load, and includes a vibration exciter that repeatedly applies a load to the measurement object, an infrared camera that obtains a temperature image of the measurement object, and an information processing device having a Fourier transform means that processes the temperature image of the measurement object obtained from the infrared camera. The information processing device has a dissipated energy measurement process that measures dissipated energy, and a fatigue limit stress specification process that specifies the fatigue limit stress from the measurement results obtained from the dissipated energy measurement process. The dissipated energy measurement process acquires temperature amplitude images of the fundamental frequency component and the second harmonic component of the vibration from the temperature image captured by the infrared camera, and extracts the dissipated energy of a pixel region in which the slope of the load characteristic with respect to the temperature amplitude image of the fundamental frequency component is maximum within the region showing the maximum of the temperature amplitude image of the second harmonic component.

Furthermore, Japanese Patent Application Laid-Open No. 2018-105709 (Patent Document 4) discloses a fatigue limit stress identification system for measuring the fatigue limit stress of a measurement object based on the temperature fluctuation of the measurement object that occurs when the measurement object is vibrated while increasing the load in a stepwise manner, the fatigue limit stress identification system including a vibrator that repeatedly applies each load to the measurement object at a predetermined frequency, an infrared camera that captures a temperature image showing the temperature fluctuation of the measurement object to which the load is applied, and an information processing device that determines the fatigue limit stress of the measurement object based on the temperature image obtained from the infrared camera, the information processing device determines the relationship between the temperature amplitude of the second harmonic component and the temperature amplitude of the fundamental frequency component of the vibration for the measurement object from the temperature image obtained from the infrared camera, fits the relationship with a first approximation line that is a quadratic curve and a second approximation line that is a quadratic curve, and determines the fatigue limit stress of the measurement object based on the intersection of the first approximation line and the second approximation line.

In addition, JP 2019-060901 A (Patent Document 5) discloses a method of repeatedly applying a load to a test specimen until the test specimen breaks, measuring the temperature of the test specimen, determining a change point in the rate of increase in the temperature of the test specimen, calculating the ratio between the time from when the load is started to be applied to the test specimen until the change point appears and the time from when the load is started to be applied to the test specimen until the test specimen breaks, measuring the temperature of a structure made of the same material as the test specimen, detecting a change point in the rate of increase in the temperature of the structure, and estimating the remaining life of the structure from the detected change point.

However, the conventional devices and methods described in Patent Documents 1 to 4 identify fatigue fracture locations and fatigue limit stresses, but are not capable of identifying the degree of fatigue damage to materials. Furthermore, the method of estimating the remaining life of a structure described in Patent Document 5 is not necessarily sufficient in terms of work efficiency and measurement accuracy.

JP 2006-250683 A JP 2010-223957 A JP 2016-024056 A JP 2018-105709 A JP 2019-060901 A

The present invention has been made in consideration of the problems with the above-mentioned conventional technology, and aims to provide a fatigue damage level determination device and a fatigue damage level determination method that are capable of measuring the fatigue damage level of a measurement object with high accuracy.

In order to achieve the above object, the present inventors first studied a method for estimating the remaining life of a structure as described in the above Patent Document 5, and found that the measurement method uses raw temperature data as information on temperature change, which results in insufficient measurement accuracy (this is because raw temperature data is easily affected by various disturbances such as heat conduction and the influence of the surrounding environment, which affects the measurement accuracy). Therefore, as a result of further intensive research by the present inventors to achieve the above object, a repeated load is applied to a measurement object at a predetermined frequency; the temperature change of the measurement object is measured; from the data on the temperature change of the measurement object, the thermoelastic temperature amplitude of the measurement object and the average temperature of the measurement object are calculated; the thermoelastic temperature amplitude is non-dimensionalized by the average temperature to calculate the non-dimensionalized thermoelastic temperature amplitude of the measurement object; the non-dimensionalized thermoelastic temperature amplitude of the measurement object is normalized by the non-dimensionalized thermoelastic temperature amplitude of the measurement object at the beginning of fatigue to calculate the normalized thermoelastic temperature amplitude of the measurement object; By determining the degree of fatigue damage from the relationship between the normalized thermoelastic temperature amplitude and the degree of fatigue damage previously determined for a sample of the same material, it is possible to measure the thermoelastic temperature amplitude of the measurement object, which is less susceptible to the effects of thermal conduction or the surrounding environment (the temperature amplitude of the measurement object at the same frequency as the frequency at which the repeated load is applied, which is determined by performing frequency analysis on the data on the temperature change of the measurement object for a specified number of repetitions), and to determine the degree of fatigue damage using this value. This has led to the discovery that it is possible to measure the degree of fatigue damage of the measurement object with high accuracy, and has led to the completion of the present invention.

That is, the fatigue damage level identifying device of the present invention comprises:
A load applying means for applying a repeated load to the object to be measured at a predetermined frequency;
A temperature measuring means for measuring a temperature change of the measurement object;
A first calculation means for calculating a thermoelastic temperature amplitude and an average temperature of the object to be measured from data of a temperature change of the object to be measured;
A second calculation means for non-dimensionalizing the thermoelastic temperature amplitude by the average temperature to calculate the non-dimensional thermoelastic temperature amplitude of the measurement object;
A third calculation means for normalizing the non-dimensional thermoelastic temperature amplitude of the measurement object by the non-dimensional thermoelastic temperature amplitude at the initial stage of fatigue to calculate a normalized thermoelastic temperature amplitude of the measurement object;
a fatigue damage degree specification means for specifying a fatigue damage degree of the measurement object based on a relationship between a normalized thermoelastic temperature amplitude and a fatigue damage degree obtained in advance using a sample made of the same material as the measurement object;
The present invention is characterized by comprising:

The method for identifying a degree of fatigue damage according to the present invention includes the steps of:
applying a cyclic load to the object to be measured at a predetermined frequency;
Measuring a temperature change of the measurement object;
Calculating a thermoelastic temperature amplitude and an average temperature of the object from the data of the temperature change of the object;
A step of non-dimensionalizing the thermoelastic temperature amplitude with the average temperature to calculate the non-dimensional thermoelastic temperature amplitude of the measurement object;
A step of normalizing the non-dimensional thermoelastic temperature amplitude of the measurement object by the non-dimensional thermoelastic temperature amplitude of the initial fatigue stage to calculate a normalized thermoelastic temperature amplitude of the measurement object;
determining the degree of fatigue damage from a relationship between a normalized thermoelastic temperature amplitude and a degree of fatigue damage previously obtained for a sample made of the same material as the measurement target;
The method is characterized by comprising:

In addition, in the fatigue damage determination device and the fatigue damage determination method of the present invention, it is preferable that the measurement object is made of fiber-reinforced plastic (e.g., carbon fiber-reinforced plastic, etc.).

According to the present invention, it is possible to provide a fatigue damage level identification device and a fatigue damage level identification method that can measure the fatigue damage level of a measurement object with high accuracy. Furthermore, according to the present invention, it is also possible to estimate (identify) the remaining fatigue life of the measurement object based on the fatigue damage level identified by the fatigue damage level identification device and the fatigue damage level identification method.

1 is a schematic diagram showing a preferred example (one embodiment) of the configuration of a measuring unit that can be used in the fatigue damage level identifying device of the present invention. FIG. 1 is a graph showing a schematic diagram of the relationship between the temperature of a measurement object and the number of repeated loads. 3 is a graph showing a schematic diagram of the relationship between frequency and temperature amplitude, which is obtained by performing frequency analysis on time-series temperature fluctuation data (such as the data shown in FIG. 2 ). 1 is a schematic diagram showing a state of a test object in an initial state and a state of the test object in which fatigue has progressed due to repeated load application when a load is repeatedly applied to the test object; FIG. 1 is a graph showing a schematic relationship between an angle θ _f between a fiber and a load direction and a change in the value of “α _T sin θ _f +α _L cos θ _f ”. 10 is a flowchart showing an example of a procedure that can be suitably adopted when determining in advance the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree. 1 is a table showing a preferred example of the relationship between the number of measurements and the specified number of load repetitions. FIG. 1 is an example of a graph showing the relationship between the number of load repetitions (prescribed number: X ₁ to X _n ) for each measurement and the non-dimensional thermoelastic temperature amplitude for each measurement, obtained when two pre-measurement samples are measured using different stress amplitudes for each sample. 13 is an example of a graph showing the relationship between normalized thermoelastic temperature amplitude and life ratio of a preliminary measurement sample. 1 is an example of a graph showing normalized thermoelastic temperature swing versus lifetime ratio for two pre-measurement samples. 10 is a flowchart showing an example of a procedure that can be suitably adopted when identifying a fatigue damage level. FIG. 2 is a schematic diagram showing a surface of a test piece of a measurement object (CFRP) used in Test Example 1. 1 is a table showing the relationship between the number of measurements employed in Test Example 1 and the prescribed number of load repetitions (prescribed numbers: X ₁ to X ₂₅ ) in each measurement. 1 is a graph showing the relationship between the non-dimensional thermoelastic temperature amplitude and the number of repeated loads for test pieces A and B used in Test Example 1. 1 is a graph showing the relationship between normalized thermoelastic temperature amplitude and life ratio for test specimens A and B. 13 is a graph showing the relationship between the dimensionless thermoelastic temperature amplitude and the number of repeated loads for test piece C. 13 is a graph showing the relationship between the actual life ratio D _act and the estimated value D _est of the life ratio of a test specimen C.

Below, a preferred embodiment of the present invention will be described in detail with reference to the drawings. Note that in the following description and drawings, the same or corresponding elements are given the same reference numerals, and duplicate explanations will be omitted.

<A preferred embodiment of a fatigue damage level identification device and a fatigue damage level identification method>
A preferred embodiment of the fatigue damage level identifying device of the present invention will be described below with reference to Fig. 1, but the fatigue damage level identifying device of the present invention is not limited to the embodiment shown below. In addition, since the fatigue damage level identifying device of the present invention can be suitably used when carrying out the fatigue damage level identifying method of the present invention, a preferred embodiment of the fatigue damage level identifying method of the present invention will also be described by explaining a method that can be suitably adopted when identifying fatigue damage level using such an device.

Figure 1 is a schematic diagram showing a preferred example (one embodiment) of the configuration of a measurement unit that can be used in the fatigue damage identification device of the present invention. The measurement unit 1 shown in Figure 1 includes a test piece 10 to be measured, a load application means 11 for applying a cyclic load to the test piece 10 to be measured at a predetermined frequency, and a temperature measurement means 12 for measuring the temperature change of the test piece 10 to be measured. Note that in Figure 1, the arrows indicating both the up and down directions conceptually depict the cyclic load F applied by the load application means 11.

In addition, in the measurement unit 1 shown in FIG. 1, the temperature measurement means 12 is connected to an external computer (an appropriate combination of known peripheral devices such as a CPU, ROM, RAM, etc., necessary to enable the desired calculation) not shown. The external computer to which such a connection is made comprises: a first calculation means (a calculation unit within a computer) configured to be able to calculate (calculate) the thermoelastic temperature amplitude of the measurement object and the average temperature of the measurement object from the temperature change data when information on the temperature change (information (data) on the temperature information) measured by the temperature measurement means 12 is input; and a second calculation means (a calculation unit within a computer) that calculates (calculates) the dimensionless thermoelastic temperature amplitude of the measurement object by non-dimensionalizing the thermoelastic temperature amplitude with the average temperature based on the thermoelastic temperature amplitude and the average temperature calculated by the first calculation means. a calculation unit in the first calculation means); a third calculation means (a calculation unit in a computer) that uses the non-dimensional thermoelastic temperature amplitude of the measurement object calculated by the second calculation means to normalize the non-dimensional thermoelastic temperature amplitude of the measurement object with the non-dimensional thermoelastic temperature amplitude at the beginning of fatigue to calculate the normalized thermoelastic temperature amplitude of the measurement object; and a fatigue damage degree determination means (a calculation unit in a computer) that uses the normalized thermoelastic temperature amplitude of the measurement object calculated by the third calculation means to determine (calculate) the fatigue damage degree of the measurement object from the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree determined in advance using a sample of the same material as the measurement object.

The computer used for such various calculations may be configured to appropriately combine known peripheral devices such as the necessary CPU, ROM, RAM, and programs required for the various calculations (such programs may be recorded in the ROM or may be recorded on a separate recording medium) to enable the desired calculations to be performed, and there are no particular limitations on the specific configuration. For example, a computer equipped with hardware such as a CPU and memory for performing the above-mentioned calculations, and a computer program (software) installed to perform the required calculations may be used. Examples of such CPUs include a central processing unit, processing unit, arithmetic unit, processor, microprocessor, microcomputer, DSP (Digital Signal Processor), etc.

The object to be measured is not particularly limited, but in the embodiment shown in FIG. 1, the object to be measured is made of fiber-reinforced plastic. Examples of such fiber-reinforced plastic include carbon fiber-reinforced plastic, glass fiber-reinforced plastic, natural fiber-reinforced plastic, and recycled fiber-reinforced plastic. When such fiber-reinforced plastic is used as the object to be measured, it is possible to measure the degree of fatigue damage of the object to be measured with even higher accuracy. In this regard, the inventors speculate as follows. That is, first, since the strength of the fiber-reinforced plastic is higher in the fiber direction than in the fiber orthogonal direction, for example, when a load F is applied as shown in FIG. 1, as fatigue progresses, the fibers that are close to parallel to the load direction (vertical direction) will mainly bear the load. On the other hand, carbon fiber-reinforced plastic has different linear expansion coefficients in the fiber direction and the fiber orthogonal direction. Therefore, when a load F is applied to carbon fiber-reinforced plastic, the change in the state (fatigue degree) of the fiber in the material (fiber-reinforced plastic) is particularly noticeable as a change in the non-dimensional thermoelastic temperature amplitude. Here, the non-dimensional thermoelastic temperature amplitude is the thermoelastic temperature amplitude non-dimensionalized at the average temperature. Therefore, the inventors speculate that when measurements (identification) are performed on objects made of such fiber-reinforced plastics, it will be possible to perform measurements with even greater accuracy by utilizing the non-dimensional thermoelastic temperature amplitude.

The shape of the test piece 10 to be measured is not particularly limited, and can be designed appropriately depending on the type of the load application means 11. In addition, when the temperature change during repeated load application is measured using a radiation thermometer as the temperature measurement means 12, the test piece 10 to be measured may be used by applying black body paint to the area on the surface including the area where the temperature is to be measured so that the test piece 10 to be measured has a known emissivity. In this way, applying black body paint to the surface of the test piece to be measured (for example, the fiber-reinforced plastic) makes it possible to perform temperature measurement more efficiently.

The load applying means 11 is not particularly limited as long as it is capable of applying a cyclic load to the test piece 10 to be measured at a predetermined frequency, and a known dynamic fatigue testing machine (e.g., a hydraulic servo type fatigue testing machine that applies a cyclic load by raising and lowering a hydraulic cylinder, etc.) can be used as appropriate.

The temperature measuring means 12 is not particularly limited, and may be any device capable of measuring the temperature change of the object to be measured (temperature change while a repeated load is applied at a predetermined frequency). Any known temperature measuring device capable of measuring temperature can be used as appropriate. As such a temperature measuring means 12, a device capable of measuring temperature without contact (e.g., an infrared camera (infrared thermography camera), a non-contact temperature sensor such as a radiation thermometer) can be suitably used, and an infrared camera is used in the embodiment of FIG. 1. When an infrared camera is used as the temperature measuring means 12, it is possible to detect the amount of infrared radiation emitted from the object according to the temperature of the object, and measure the temperature for each unit area (e.g., for each pixel), and it is possible to measure the temperature change for each pixel of any measurement area (the measurement area may be set appropriately according to the size of the test piece, the type of load applying means 11, the characteristics of the camera, etc.) on the surface of the test piece 10 of the object to be measured. In addition, as in the embodiment shown in FIG. 1, when a device (infrared thermography camera) that uses infrared rays to measure the temperature of the test piece 10 to be measured is used as the temperature measurement means 12, the measurement unit 1 may be placed in a darkroom to make the infrared measurement more accurate.

The measurement unit 1 in the form shown in FIG. 1 is connected to an external computer as described above, and the data on the temperature change of the test piece 10 as the measurement object measured by the temperature measurement means 12 is input using the external computer, which performs various calculations to determine the normalized thermoelastic temperature amplitude, and the degree of fatigue damage can be determined using the normalized thermoelastic temperature amplitude. Here, before explaining the first to third calculation means and the fatigue damage determination means, the reason (principle) why it is possible to determine the degree of fatigue damage using the normalized thermoelastic temperature amplitude will be explained.

When a load is applied to a solid, the temperature changes in the same way as a gas (the phenomenon in which such a temperature change occurs is known as the thermoelastic effect). Therefore, it can be seen that when a repeated load is applied to a material, a temperature change due to the thermoelastic effect occurs at the same frequency as the repeated load. Below, an example will be described in which the test piece 10 to be measured is carbon fiber reinforced plastic (CFRP) and the temperature measuring means 12 is an infrared camera. First, the amplitude ΔT ^* of the temperature change due to the thermoelastic effect of carbon fiber reinforced plastic is calculated using the following formula (1):

(In formula (1), ΔT ^* represents the amplitude (absolute value) of temperature fluctuation due to thermoelastic effect, ρ represents density, C represents the specific heat of the measurement object (CFRP), T _amb represents the ambient temperature, Δσ _L represents the stress amplitude acting in the fiber direction, Δσ _T represents the stress amplitude acting in the direction perpendicular to the fiber, α _L represents the linear expansion coefficient in the fiber direction, and α _T represents the linear expansion coefficient in the direction perpendicular to the fiber.)
(For example, Toshiyuki Noya et al., "Evaluation of Internal Damage in Carbon Fiber Composite Materials by High Resolution Thermoelastic Analysis", Materials (J. Soc. Mat. Sci., Japan), Vol. 49, No. 8, pp. 941-947 (hereinafter, this document will be simply referred to as "Reference 1"). In the present invention, the amplitude of temperature fluctuation (thermoelastic temperature amplitude ΔT, etc.) is used as an absolute value.

In addition, when the repeated load is continuously applied at a predetermined frequency, if the temperature fluctuation of the material under the repeated load condition is measured over a period of a specific number of repetitions (a period for a specific number of repetitions) to obtain temperature fluctuation data, as described above, the temperature fluctuation due to the thermoelastic effect has the same frequency as the frequency of the repeated load. This point will be explained in more detail with reference to FIG. 2. FIG. 2 is an example of a graph that shows a schematic relationship between the temperature of the measurement object and the number of repeated loads. Such a graph can be obtained, for example, by setting the frequency of the repeated load to a predetermined frequency (for example, 10 Hz), setting the frame rate of the infrared camera as the temperature measurement means 12 to a predetermined value (for example, 211 Hz), setting the measurement period to a period during which a measurement is performed for a predetermined number of frames (for example, 4009 frames), and measuring the temperature change of the test piece 10 of the measurement object (Note that in the measurement using an infrared camera, a specific measurement area may be set on the surface of the measurement object 10, and the temperature change between frames may be obtained in time series from the temperature of each pixel of the specific measurement area in the image for each frame). In such a measurement, the measurement time is the number of frames divided by the frame rate (for example, if the frame rate is 211 Hz and the number of frames is 4009, the measurement time is 4009/211 = 19 seconds). The number obtained by multiplying the measurement time by the frequency of the repeated load is the number of load repetitions (for example, when a measurement is performed for 19 seconds and the frequency of the repeated load is 10 Hz, the load is applied 10 times per second, so the number of load repetitions is 19 x 10 = 190 times (cycles)). Here, by arranging the data of all frames during the measurement period in chronological order in correspondence with the number of load repetitions, a graph of the relationship between the temperature of the measurement object and the number of load repetitions as shown in Figure 2 can be obtained. Note that, for each temperature image of each frame, it is desirable to obtain the temperature of all pixels in a specific measurement area, and then obtain the average temperature of all pixels, and adopt the average value as the temperature of the measurement object at the measurement time of that frame. Then, such a graph can confirm that the temperature fluctuation caused by the thermoelastic effect has the same frequency as the frequency of the repeated load.

In this way, since the temperature fluctuation caused by the thermoelastic effect has the same frequency as the frequency of the repeated load, the relationship between frequency and temperature amplitude can be obtained by frequency analysis (e.g., Fourier transform) of the time-series temperature fluctuation data (data as shown in Figure 2). An example of such a relationship is shown in Figure 3. Since the temperature fluctuation caused by the thermoelastic effect basically has the same frequency as the repeated load, the temperature amplitude at the same frequency as the frequency of the repeated load can be obtained from the data (graph) as shown in Figure 3 obtained by frequency analysis of the temperature fluctuation data (time-series data), and the temperature amplitude can be obtained as the thermoelastic temperature amplitude ΔT (absolute value: hereinafter, the notation "absolute value" will be omitted in some cases) of the measured object. In addition, the average temperature of the measured object can be obtained by dividing the sum of the temperatures of the measured object obtained for each image of each frame from the time-series temperature fluctuation data (the temperatures of the measured object obtained for each image of the time series) by the number of frames. The calculation formula for obtaining such an average temperature is the following formula (2):

(In formula (2), T indicates the average temperature of the object to be measured, N indicates the number of frames (total number of frames) in which the temperature fluctuation was measured, and Ti indicates the temperature in a specific frame i (where i is a natural number indicating the frame number (number) and is a value from 1 to N) (in other words, the temperature in the i-th frame).)
It is expressed as:

Here, in FIG. 4, the initial state of the measurement object and the state of the measurement object in which fatigue has progressed due to the repeated load are shown in the case where a load is repeatedly applied to the measurement object. CF in FIG. 4 is a schematic drawing of carbon fibers in the measurement object (fiber reinforced plastic: CFRP in this example). Since fiber reinforced plastic has a higher strength in the fiber direction than in the fiber orthogonal direction, when a repeated load F as shown in FIG. 4 is applied to the test piece 10 (measurement object), as fatigue progresses, the fibers close to parallel to the load direction (vertical direction: F direction) mainly bear the load. On the other hand, carbon fiber reinforced plastic has different linear expansion coefficients in the fiber direction and the fiber orthogonal direction. Then, using the thermoelastic temperature amplitude ΔT (absolute value) obtained based on the above-mentioned temperature fluctuation data measurement (time-series temperature data) and the average temperature T of the object to be measured, the thermoelastic temperature amplitude ΔT is introduced as ΔT ^* in formula (1), the average temperature T of the object to be measured is introduced as the atmospheric temperature _Tamb in formula (1), the stress amplitude in the load direction is Δσ (see FIG. 4 ), and the angle between the fiber and the load direction is θ _f (see FIG. 4 ), and the above formula (1) is transformed to the following formula (3):

(ρ, C, _αL , and _αT in formula (3) are the same as those in formula (1), ΔT represents the thermoelastic temperature amplitude (absolute value) of the object to be measured, T represents the average temperature of the object to be measured, Δσ represents the stress amplitude in the load direction, and _θf represents the angle between the fiber and the load direction.)
can be sought.

On the other hand, by using the data on the linear expansion coefficient of CFRP described in Reference 1 (α _L = -0.4 × 10 ^-6 (1/K), α _T = 30 × 10 ^-6 (1/K)), a graph like that shown in FIG. 5 can be obtained to determine the relationship between the angle θ _f between the fiber and the load direction and the change in the value of "α _T sin θ _f + α _L cos θ _f ". From the graph shown in FIG. 5, it can be seen that the value of "α _T sin θ _f + α _L cos θ _f " decreases as the angle θ _f between the fiber CF and the load direction decreases due to the progression of fatigue. Therefore, under repeated loading conditions with a constant load amplitude, if it is assumed that ρ and C are both constants and that stress changes due to crack generation can be ignored (Δσ is constant), it can be seen from equation (3) that the value of "ΔT/T" gradually decreases as fatigue progresses. From such a consideration, it can be seen that the value of "ΔT/T" obtained by the formula (3) is a physical quantity related to the fatigue damage of the measurement object (CFRP in this embodiment). In the present invention, taking into consideration the relationship between "ΔT/T" and fatigue damage, the thermoelastic temperature amplitude is non-dimensionalized by the average temperature to calculate the non-dimensional thermoelastic temperature amplitude (ΔT/T) of the measurement object, and then the normalized value of the non-dimensional thermoelastic temperature amplitude is obtained and used to specify the fatigue damage degree. In addition, when specifying such a fatigue damage degree, for example, a sample (pre-measurement sample) made of the same material as the measurement object is used, and data regarding the degree of fatigue progression of the sample (pre-measurement sample) and the change in the normalized value of the non-dimensional thermoelastic temperature amplitude (ΔT/T) are obtained in advance, and the fatigue damage degree is measured by comparing with the data. In this way, in the present invention, the fatigue damage degree is specified based on the correlation between the value (normalized value) regarding the non-dimensional thermoelastic temperature amplitude (ΔT/T) of the material and the fatigue damage degree.

The above explains the reason (principle) why it is possible to identify the degree of fatigue damage using values related to the normalized non-dimensional thermoelastic temperature amplitude (the reason for using the normalized non-dimensional thermoelastic temperature amplitude will be described later), but the calculation to determine the non-dimensional thermoelastic temperature amplitude (ΔT/T) used to identify such fatigue damage is performed by a first calculation means and a second calculation means in an external computer connected to the measurement unit 1.

As for the first calculation means and the second calculation means in such an external computer, in order to make it possible to perform the desired calculation based on the data on temperature changes obtained by the measurement unit 1, for example, the first calculation means is configured to calculate the temperature change data for a predetermined number of times from the start of the application of the repeated load obtained by the temperature measurement means 12 (temperature data for all frames for a predetermined number of frames: time-series temperature fluctuation data; however, the period for measuring the temperature fluctuation data is the same period (number of repetitions or number of frames) as the period set when measuring the temperature fluctuation data for each measurement for the measurement sample described below when determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree". A preferred embodiment of the present invention is a calculation unit configured to, when a temperature change data (which must be between 100 and 200° C.) is input, perform frequency analysis of the temperature change data to calculate (calculate) the thermoelastic temperature amplitude ΔT (absolute value) of the object to be measured, and to calculate (calculate) the average temperature T of the object to be measured using the formula (2) based on the temperature change data, and a second calculation unit configured to, when the thermoelastic temperature amplitude ΔT and average temperature T calculated by the first calculation unit are input, non-dimensionalize the thermoelastic temperature amplitude ΔT with the average temperature T to calculate (calculate) the non-dimensional thermoelastic temperature amplitude (ΔT/T) of the object to be measured. When the first calculation unit and the second calculation unit configured in this way are used, it is possible to efficiently calculate the non-dimensional thermoelastic temperature amplitude (ΔT/T). The method of frequency analysis of the time series temperature data (temperature fluctuation data) is not particularly limited, but a preferred method thereof is, for example, Fourier transformation.

In addition, in the present invention, the non-dimensional thermoelastic temperature amplitude (ΔT/T) calculated by the first calculation means and the second calculation means is used to calculate a normalized thermoelastic temperature amplitude (a value obtained by normalizing the non-dimensional thermoelastic temperature amplitude) by a third calculation means (a calculation unit in the computer described above), and the normalized thermoelastic temperature amplitude value is used to determine (calculate) the fatigue damage degree of the measurement object by a fatigue damage degree determination means (a calculation unit in the computer described above) based on the "relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree" calculated in advance using a sample of the same material as the measurement object.

Before describing the third calculation means and the fatigue damage degree determination means, a method that can be suitably adopted as a method for determining in advance the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" and the reason for using the normalized thermoelastic temperature amplitude to determine (assess) the fatigue damage degree will be described with reference to the flowchart shown in FIG. 6. In this specification, a sample (test piece) made of the same material as the measurement object used to determine in advance the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" is sometimes simply referred to as a "pre-measurement sample" for convenience in order to consider it separately from the test piece of the measurement object. In addition, the method (procedure) described with reference to FIG. 6 is an example of a case in which the measurement unit shown in FIG. 1 (when the temperature measurement means 12 in the measurement unit 1 is an infrared camera) and an external computer connected to the measurement unit that enables the necessary calculations are used.

In the method for determining the relationship between the normalized thermoelastic temperature amplitude and the degree of fatigue damage, first, as a preliminary preparation, a pre-measurement sample (test piece) is set in place of the test piece 10 of the measurement object at the position of the measurement unit 1 shown in Figure 1 where the test piece 10 of the measurement object is depicted. Then, in the first step S101, the application of a cyclic load at a predetermined frequency (e.g., 10 Hz) is started to the pre-measurement sample.

Next, the process proceeds to step S102, where each time the number of load repetitions reaches a specified number (any of _X1 to _Xn (see FIG. 7)), the change (temperature fluctuation) in the surface temperature of the sample in time series is measured for a specified period (the specified number of repetitions) starting from the specified number of repetitions, and for each n number of measurements, data on the temperature fluctuation (time-series temperature change) from the start point until the specified period has elapsed is measured (the surface temperature is measured n times in total). In this way, in step S102, each time the number of load repetitions reaches a specified number, the surface temperature of the sample in time series is measured for the specified number of repetitions. This point will be briefly explained below with reference to FIG. 7.

Fig. 7 shows a preferred example of the setting of the prescribed number of load repetitions when performing measurement. In step S102, the prescribed number of load repetitions as shown in Fig. 7 is set in advance, and the number of load repetitions is counted from the start of the application of the repeated load in step S101. The operation of the measurement unit 1 shown in Fig. 1 is controlled so that the measurement of the surface temperature of the sample is started each time the count number reaches any of the prescribed numbers X ₁ to X _n shown in Fig. 7 (note that for such control, an external computer is used to grasp the operating status of the load application means 11 (counting the number of load repetitions) and control the operating status of the temperature measurement means 12 according to the count number). In this way, the measurement of the surface temperature of the sample is started each time the prescribed number of times X ₁ to X _n shown in Fig. 7 (hereinafter, for convenience, the prescribed number itself will be described using the symbols "X ₁ to X _n " in some cases) is reached. Then, for each measurement, the surface temperature of the sample is measured a predetermined number of times ( _{predetermined} period: a predetermined number of frames of measurement by the infrared camera) from the start of surface temperature measurement (the point at _which the specified number of times X1 to Xn is reached), thereby measuring data on changes in the surface temperature of the preliminary measurement sample over time. Note that, as a method of measuring the sample _X1 to _Xn by the infrared camera, for example, when an infrared camera is used as the temperature measuring means 12, the frame rate is set to a predetermined value in advance, the measurement period of each measurement is set to a period from the start of measurement to a predetermined number of frames, and the changes in the surface temperature of the preliminary measurement sample over time are measured over the same period (predetermined number of frames) for each measurement. For example, when the frame rate is set to 211 Hz and the number of frames is set to 4009, the measurement period is 19 seconds (=4009/211) (in this case, when the frequency of the load is 10 Hz, the number of load repetitions during the measurement period is 190 cycles), and in this case, the time-series surface temperature change of the pre-measurement sample is measured (measurement of images of temperature change for 4009 frames) for 19 seconds from the timing when the number of repetitions (prescribed number) of X ₁ to X _n is reached in each measurement. Here, the temperature of the pre-measurement sample for each image of each frame is not particularly limited, but for example, when a specific region of the pre-measurement sample is set as the measurement region (note that in the sample having the form shown in FIG. 12 described in the Example section below, a rectangular part of the surface, region A2, is set as the measurement region), the temperature of each pixel in the specified measurement region is obtained for each image of each frame, the average value of the temperatures of all the pixels is calculated, and the obtained average value of the temperatures of all the pixels is adopted as the temperature of the pre-measurement sample for the image of that frame. Then, by arranging the temperatures of the pre-measurement sample for each frame image in chronological order, it is possible to calculate and obtain the relationship between the number of load repetitions and the temperature (for example, the relationship shown in Fig. 2). Note that such calculations (analysis) can be appropriately performed using an external computer (connected to the measurement unit 1) that has a calculation unit designed to enable such calculations.

In the procedure (flow) shown in Fig. 6, the measurement in step S102 is performed until the preliminary measurement sample breaks. When breakage of the preliminary measurement sample is confirmed, the process proceeds to step S103, where the operation of the load applying means 11 in the measurement unit 1 shown in Fig. 1 is stopped to terminate the repeated load application, and the number of load repetitions counted until breakage of the preliminary measurement sample is calculated as the fatigue life _Nf . The last measurement before breakage is the nth measurement shown in Fig. 7.

Next, proceed to step S104, and use the data of temperature fluctuations for each measurement from the number of repetitions (X ₁ to X _n ) at each starting point until a predetermined period (a predetermined number of repetitions) has elapsed to calculate the thermoelastic temperature amplitude ΔT _p and the average temperature T _p for each of the total n measurements, and obtain the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p for each measurement (where P is a numerical value (natural number) indicating any one of the measurements 1 to n). The calculation (computation) of the thermoelastic temperature amplitude ΔT _p and the average temperature T _p may be performed using the first calculation means and the second calculation means described above. That is, the data of temperature change over time for each measurement (such as temperature data for all frames of a predetermined number of frames) obtained in the measurement in step S102 described above is input to a first calculation means, and the data of temperature change over time for each measurement is frequency analyzed in the first calculation means to calculate (calculate) the thermoelastic temperature amplitude ΔT _p (P is a value indicating the number of measurements for which the thermoelastic temperature amplitude was obtained) of the preliminary measurement sample for each measurement (see FIG. 3), and the average temperature T _p (P is a value indicating the number of measurements for which the average temperature was obtained) of the preliminary measurement sample for each measurement is calculated (calculated) using the above formula (2) based on the temperature change data over time for each measurement. Next, the thermoelastic temperature amplitude ΔT _p for each measurement obtained by the first calculation means and the average temperature T _p for each measurement are input to a second calculation means, and the second calculation means calculates (calculates) the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p for each measurement.

Next, in step S105, the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p for each measurement of the preliminary measurement sample is normalized by the non-dimensional thermoelastic temperature amplitude (ΔT/T) ₁ at the early stage of fatigue of the preliminary measurement sample to obtain the normalized thermoelastic temperature amplitude _Yp for each measurement. In the flow shown in FIG. 6, the non-dimensional thermoelastic temperature amplitude (ΔT/T) ₁ obtained in the first measurement (first time: P=1) shown in FIG. 7 is used as the non-dimensional thermoelastic temperature amplitude at the early stage of fatigue. That is, in this embodiment, the time point when the number of repetitions (300 cycles) of the first (first) measurement shown in FIG. 7 is reached is regarded as the early stage of fatigue, and the non-dimensional thermoelastic temperature amplitude at the early stage of fatigue of the preliminary measurement sample is obtained (note that such number of repetitions at the early stage of fatigue (number of repetitions of the first measurement) is not limited to that shown in FIG. 7, and the setting may be changed as appropriate). Then, in step S105, the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p for each measurement of the preliminary measurement sample is divided by the non-dimensional thermoelastic temperature amplitude (ΔT/T) ₁ at the early stage of fatigue of the preliminary measurement sample (first measurement (p=1)) to obtain the normalized thermoelastic temperature amplitude _Yp (=(ΔT/T) _p /(ΔT/T) ₁ ) for the preliminary measurement sample. Such calculation may be performed by an external computer connected to the measurement unit 1. In addition, in such normalization, a state in which the measurement is started on an unused sample and the number of load repetitions reaches a specified number for the first time (for example, _X1 in FIG. 7) is assumed to be the early stage of fatigue, and the non-dimensional thermoelastic temperature amplitude of the first measurement is used as the non-dimensional thermoelastic temperature amplitude (ΔT/T) ₁ at the early stage of fatigue to perform the calculation.

Here, the reason for obtaining the normalized thermoelastic temperature amplitude _Yp and using it in the measurement in the present invention will be briefly described. Hereinafter, an example will be described in which measurements are performed on two pre-measurement samples with different stress amplitudes Δσ. When measurements are performed on two pre-measurement samples by using different stress amplitudes Δσ for each sample, the relationship between the number of load repetitions (prescribed number: X ₁ to X _n ) of each measurement and the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p of each measurement is shown in a graph as shown in FIG. 8. The graph shown in FIG. 8 is an example of a graph showing the relationship between the number of load repetitions _Xp ( _Xp is the number of repetitions at the start point of each measurement) and the non-dimensional thermoelastic temperature amplitude (ΔT/T) p of each measurement when different stress amplitudes _Δσ are set for two pre-measurement samples (samples A and B in FIG. 8) and measurements are performed. In this way, when two pre-measurement samples made of the same material are measured with different stress amplitudes Δσ, different non-dimensional thermoelastic temperature amplitudes (ΔT/T) _p are measured for each sample. This is because the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p depends on the stress amplitude Δσ, as is clear from the above formula (3). Therefore, when the stress amplitudes Δσ are set to different values for the two pre-measurement samples and measurements are actually performed, a graph with different values of (ΔT/T) _p is obtained as shown in FIG. 8. Considering the graph shown in FIG. 8, it can be understood that a more accurate measurement can be performed by normalizing the non-dimensional thermoelastic temperature amplitude. For example, when the load applying means 11 is set to a constant magnitude for measurement, even if the magnitude of the load fluctuates (varies) or fluctuates during the repeated application of loads by the load applying means 11 many times, it is considered that a more accurate measurement can be performed when the non-dimensional thermoelastic temperature amplitude is normalized and used. In view of this, in the present invention, the dimensionless thermoelastic temperature amplitude is normalized and used to identify (determine) the degree of fatigue damage.

The reason for calculating the normalized thermoelastic temperature amplitude Yp and using _it in the measurement has been briefly explained above, but in step S105, the normalized thermoelastic temperature amplitude _Yp is calculated and the life ratio _Dp is calculated for each measurement. That is, in step S105, the specified number of measurements (the number of repetitions at the start of each measurement ( _X1 to _Xn )) is normalized by the fatigue life _Nf (the number of repetitions at which the sample breaks) calculated in step S103, and the life ratio _Dp (= _Xp / _Nf : _Xp indicates the number of repetitions at the start of each measurement) is calculated for each measurement.

Next, in step S106, an approximation line and an approximation formula are obtained from the relationship between the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp of the preliminary measurement sample as the relationship between the normalized thermoelastic temperature amplitude _Yp and the fatigue damage degree of the preliminary measurement sample. For example, when the relationship between the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp of the preliminary measurement sample is a graph as shown in Fig. 9 (Fig. 9 is a graph showing the relationship between the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp when the sample B shown in Fig. 8 is the preliminary measurement sample), the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp can be approximated from the data by the least squares method to obtain an approximation line and an approximation formula. In this way, in step S106, an approximation formula of the relationship between the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp is obtained, and this is used as the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree. That is, by inputting the value of the normalized thermoelastic temperature amplitude Y of a measurement object whose fatigue damage degree is unknown into such an approximation formula, the life ratio D of the measurement object can be calculated (computed), and the fatigue damage degree of the measurement object can be specified based on the value of D. Therefore, in this embodiment, the above-mentioned approximation formula is used as the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree (here, since the life ratio _Dp is used to specify the fatigue damage degree, it is possible to estimate the life as well as specify the fatigue damage degree). Note that in step S106, the calculation to obtain the approximation formula by the least squares method may be performed using an automatic calculator, by inputting the values of the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp for each measurement and performing automatic calculation.

In the present invention, the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree is preferably a relationship obtained as described in the embodiment of the flowchart shown in Fig. 6. That is, the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree is obtained by first using a specimen (measurement object sample) made of the same material as the measurement object, applying a cyclic load to the specimen at a predetermined frequency until the specimen is destroyed, and each time at least a plurality of predetermined repetitions ( _X1 to _Xn ) are reached, data on the temperature change of the specimen from the number ( _X1 to _Xn ) to a predetermined number of times of the cyclic load application process (for example, a load repetition number corresponding to a specific number of frames) is used to obtain the thermoelastic temperature amplitude ΔT and average temperature T of the specimen (measurement object sample) for each of the predetermined repetitions ( _X1 to _Xn ), and then calculating the normalized thermoelastic temperature amplitude (ΔT/T) of the measurement object of the specimen for each of the predetermined repetitions ( _X1 to _Xn ), and using the number of repetitions at which the specimen (measurement object sample) is destroyed as the repetition number of the life, and calculating the temperature amplitude ΔT and average temperature T of the specimen (measurement object sample) for each of the predetermined repetitions ( _X1 to _Xn) . ) by the number of cycles of the life, and use the result as a value relating to the fatigue damage degree.

As described above, a preferred method for determining in advance the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree has been described with reference to the flow chart shown in Fig. 6, but the method for determining in advance the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree is not limited to the above method. For example, in the above-mentioned step S106, an approximation formula is determined using only the results of one pre-measurement sample (sample B in Fig. 8), but the method for determining the approximation formula is not limited to this. A plurality of pre-measurement samples may be prepared, and the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp for each measurement may be determined for each sample, and the approximation formula may be determined by the least squares method using the data of the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp for all samples. For example, two pre-measurement samples may be prepared, steps S101 to S105 may be performed on each sample, the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp may be obtained for each measurement of each sample, and then all the data on the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp for each measurement of the two pre-measurement samples may be used, plotted, and the normalized thermoelastic temperature amplitude _Yp and the life ratio _Dp may be approximated by the least squares method to obtain an approximation line and an approximation formula. Here, when measuring the two pre-measurement samples, it is preferable to set the load applying means 11 so that the magnitude of the stress amplitude Δσ differs for each sample, and to perform the measurements with different stress amplitudes Δσ. This is because the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p depends on the stress amplitude Δσ, and by setting the stress amplitude Δσ to different values and performing measurements, a graph with different values of (ΔT/T) _p is obtained, and by using this, it is possible to obtain the relationship between the fatigue damage degree and the normalized thermoelastic temperature amplitude, taking into account the case where the load magnitude fluctuates (varies) or fluctuates when the load is repeatedly applied multiple times by the load application means 11. An example of the case where an approximation line is obtained in this way is shown in Figure 10 (note that Figure 10 is a graph obtained using the results of samples A and B shown in Figure 8). When a plurality of such pre-measurement samples are prepared and an approximation formula is obtained, it is possible to take into account the fluctuation of the stress amplitude Δσ and obtain an approximation formula based on more data, and therefore it is considered that the fatigue damage degree can be measured with higher accuracy.

The method for determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage level" in advance has been explained above. Below, we will explain the third calculation means and fatigue damage level determination means provided in the fatigue damage level determination device for determining the fatigue damage level using the relationship between normalized thermoelastic temperature amplitude and fatigue damage level determined in this manner (the approximate formula as described above).

In the third calculation means, the non-dimensional thermoelastic temperature amplitude (ΔT/T) of the measurement object (test piece 10) is normalized by the non-dimensional thermoelastic temperature amplitude at the beginning of fatigue of the measurement object (test piece 10) to calculate the normalized thermoelastic temperature amplitude of the measurement object. Therefore, the third calculation means provided in the external computer described above may be a calculation unit configured to perform a calculation (configured so as to enable a desired calculation process (the calculation)) when the non-dimensional thermoelastic temperature amplitude (ΔT/T) of the measurement object (test piece 10) calculated (calculated) by the second calculation means described above is input, and the non-dimensional thermoelastic temperature amplitude (ΔT/T) is normalized by the non-dimensional thermoelastic temperature amplitude (ΔT/T) ₁ at the beginning of fatigue of the measurement object (test piece 10). Here, the "non-dimensional thermoelastic temperature amplitude at the beginning of fatigue" of the measurement object (test piece 10) used in the third calculation means is set as the same time as the initial fatigue period set when normalizing the non-dimensional thermoelastic temperature amplitude for the preliminary measurement sample when the above-mentioned "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" is obtained in advance, as the initial fatigue period for the measurement object (test piece 10) (for example, when the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" is obtained in advance in the flow shown in Figure 6 described above, the measurement result of the first measurement is considered to be the initial fatigue period, so the number of load repetitions of the first measurement corresponds to the initial fatigue period). In the present invention, when the measurement object (test piece 10) for which the fatigue damage degree is to be specified is still unused, a sample similar to the test piece 10 can be prepared in advance and the non-dimensional thermoelastic temperature amplitude at the initial fatigue period set using the sample can be measured in advance and used.

The fatigue damage degree identification means uses the normalized thermoelastic temperature amplitude of the measurement object calculated (computed) by the third calculation means to identify the fatigue damage degree of the measurement object from the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree previously determined using a sample (the pre-measurement sample) of the same material as the measurement object. Therefore, the fatigue damage degree identification means provided in the external computer described above may be a calculation unit configured to be able to identify (calculate) the fatigue damage degree of the measurement object (configured to enable desired calculation processing) by using the relationship (the above-mentioned approximation formula, etc.) between the normalized thermoelastic temperature amplitude and the fatigue damage degree previously determined using a sample (the pre-measurement sample) of the same material as the measurement object when the normalized thermoelastic temperature amplitude of the measurement object calculated (computed) by the third calculation means is input.

By using a fatigue damage level determination device consisting of a measuring unit such as that shown in Figure 1 and an external computer equipped with the first to third calculation means and fatigue damage level determination means described above, it is possible to determine the fatigue damage level of a measurement object whose fatigue damage level is unknown.

A preferred embodiment of the fatigue damage level identification device of the present invention has been described above with reference to FIG. 1. Below, a preferred embodiment of the fatigue damage level identification method of the present invention will also be described by describing a method (a preferred example of a procedure for identifying fatigue damage level) that can be preferably adopted when identifying fatigue damage level using the fatigue damage level identification device of this embodiment (a fatigue damage level identification device having a measuring unit as shown in FIG. 1).

Figure 11 is a flowchart showing an example of a procedure that can be suitably adopted when determining the degree of fatigue damage. In the following explanation of this flowchart, the object for which the degree of fatigue damage is to be determined will sometimes be referred to as a "measurement object with an unknown degree of fatigue damage."

In such a process of identifying the degree of fatigue damage, first, in step S201, the non-dimensional thermoelastic temperature amplitude (ΔT/T)1-est at the early stage of fatigue for the object to be measured whose degree of fatigue damage _{is unknown is} input in advance to the third calculation means of the external computer so that the non-dimensional thermoelastic temperature amplitude of the object to be measured whose degree of fatigue damage is unknown can be normalized in the third calculation means of the external computer. Note that for the sake of convenience, the "non-dimensional thermoelastic temperature amplitude at the early stage of fatigue" of the object to be measured whose degree of fatigue damage is unknown is expressed as "(ΔT/T)1-est" so as to be easily distinguished from the "non-dimensional thermoelastic temperature amplitude at the early stage of fatigue" of the pre-measurement sample used in determining the above-mentioned "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree. _"

Here, the non-dimensional thermoelastic temperature amplitude (ΔT/T) _1-est at the early stage of fatigue of the measurement object whose fatigue damage degree is unknown, which is input in step S201, may be a value (a value measured in advance) obtained by preparing a test piece having the same shape as the test piece 10 and measuring it in advance when the measurement object whose fatigue damage degree is to be specified is unused (a brand new stage where fatigue has not occurred), or, when the measurement object whose fatigue damage degree is to be specified is in the unused stage, a test piece having the same shape as the test piece 10 is prepared in advance and stored, and when it becomes necessary to measure the fatigue damage degree, the unused test piece of the measurement object is used to separately measure the non-dimensional thermoelastic temperature amplitude at the early stage of fatigue, and the measured value is used as the "non-dimensional thermoelastic temperature amplitude (ΔT/T) _1-est at the early stage of fatigue" of the measurement object whose fatigue damage degree is unknown. Note that the "early stage of fatigue" here has the same meaning as the "early stage of fatigue" employed when determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" in advance using a sample for advance measurement. In addition, the non-dimensional thermoelastic temperature amplitude (ΔT/T) _1-est at the initial stage of fatigue can be determined by employing a method similar to the method for determining the “non-dimensional thermoelastic temperature amplitude at the initial stage of fatigue” of the pre-measurement sample used in determining the above-mentioned “relationship between normalized thermoelastic temperature amplitude and fatigue damage degree,” except that an unused test piece of the measurement object is used instead of the pre-measurement sample.

Next, in step S202, the application of a repeated load to the measurement object whose fatigue damage degree is unknown is started, and the temperature change of the measurement object from the start to a predetermined number of load repetitions is measured. That is, in step S202, a process of applying a repeated load to the measurement object at a predetermined frequency and a process of measuring the temperature change of the measurement object are executed. The application of a repeated load to the measurement object whose fatigue damage degree is unknown may be performed by starting the operation of the load application means 11 (e.g., a hydraulic servo type fatigue testing machine, etc.) of the measurement unit 1 shown in FIG. 1. Note that the conditions such as the form (size and shape) of the test piece 10 of the measurement object whose fatigue damage degree is unknown used in such step S202, the magnitude of the repeated load F applied to the test piece 10, and the frequency of the repeated load must be the same as the conditions used when determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" in advance using a sample for advance measurement. In addition, when determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree," if multiple pre-measurement samples are used and different stress amplitude Δσ conditions are adopted to determine the normalized thermoelastic temperature amplitude, etc., it is desirable to perform the measurement using conditions (stress amplitude Δσ) similar to the measurement conditions (stress amplitude Δσ) of at least one of the multiple pre-measurement samples. In addition, even when multiple pre-measurement samples are used to determine the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree," each pre-measurement sample and the test piece 10 of the measurement object with unknown fatigue damage degree must be the same size and shape.

In addition, the measurement of the temperature fluctuation (temperature change) of the measurement object in step S202 is performed by starting the application of a repeated load to the measurement object whose fatigue damage degree is unknown, and measuring the change in the temperature state of the measurement object over a time series from the start of the application of the repeated load (from the start of the application of the repeated load to a predetermined number of load repetitions) (for example, temperature change data as shown in FIG. 2). Such measurements are performed by the temperature measurement means 12, and the operation of the temperature measurement means 12 can be controlled by an external computer connected to the measurement unit 1, and the measurement is performed using the same method as that used when the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" is obtained in advance using a sample for advance measurement. Note that the "predetermined period (the period from the start to a predetermined number of load repetitions)" is set to the same period as the period set when measuring the surface temperature of the sample for each measurement when obtaining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" (for example, if the measurement is performed for a period of 190 load repetitions when obtaining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree", the measurement is also performed for a period of 190 cycles in step S202). In step S202, the object to be measured, whose degree of fatigue damage is unknown, is measured for a predetermined period from the start of the application of the repeated load, in order to measure the degree of fatigue damage of the object to be measured at the start of the measurement. Note that an image relating to the temperature measured by the temperature measuring means 12 (a thermographic image if the temperature measuring means 12 is an infrared camera) may be appropriately analyzed in an external computer to determine the relationship as shown in Fig. 2 (the relationship between the number of repeated loads and the temperature), which may be used in the subsequent step S203. In addition, the analysis of the image related to temperature may be performed, for example, by configuring the first calculation means so that such analysis can also be performed, and the first calculation means may analyze the time-series image data input from the temperature measurement means 12, or a separate analysis may be performed using another image processing means, and the analysis data may be used (when a separate image processing means is used, for example, image processing (image analysis) may be performed using a computer other than the computer equipped with the first calculation means, or analysis software may be loaded into the computer equipped with the first calculation means, and the computer may be equipped with a calculation unit for image processing configured to be able to perform separate calculations, separate from the first calculation means, and image processing (image analysis) may be performed within the same computer). In addition, the "temperature of the measurement object" to be measured is the same temperature as the temperature defined as the temperature of the pre-measurement sample when determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree". That is, when determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage," image analysis is performed for each frame, and the image of each frame is used to determine the average surface temperature of the sample for all pixels in a specified measurement area for each image, and this is used as the temperature of the pre-measurement sample at each time. In this case, for measurement objects with unknown fatigue damage, a similar measurement area is set on the test piece 10, and the average surface temperature of the test piece for all pixels in each measurement area is similarly determined from the image of each frame, and this average is used as the temperature of the measurement object with unknown fatigue damage at each time. In this way, data on temperature fluctuations (temperature changes) of the measurement object can be obtained.

Next, in step S203, the temperature amplitude ΔT _est and the average temperature T _est of the measurement object whose fatigue damage degree is unknown are calculated (computed). In order to easily distinguish the "temperature amplitude ΔT" and the "average temperature T" of the pre-measurement sample used in obtaining the above-mentioned "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree", in step S203, for convenience, the temperature amplitude of the measurement object whose fatigue damage degree is unknown is expressed as "ΔT _est ", and the average temperature of the measurement object whose fatigue damage degree is unknown is expressed as "T _est ". Such step S203 is a step of calculating (computing) the thermoelastic temperature amplitude ΔT est and the average temperature T est of the measurement object whose fatigue damage degree is unknown at the current stage (time when the fatigue damage degree is to be obtained) from the data of the temperature change of the measurement object measured in step _S202 (data of the temperature change of the measurement object for a predetermined number of repetitions: measurement _data during a predetermined period from the start of the application of the repetitive load). Such calculations may be performed by employing the same method as that employed in determining the thermoelastic temperature amplitude ΔT and average temperature T of the preliminary measurement sample when determining the "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree" except that data on temperature fluctuations (temperature changes) of the measurement object with unknown fatigue damage degree is used, and may be executed by the first calculation means. For example, when determining the thermoelastic temperature amplitude ΔT and average temperature T of the preliminary measurement sample by the method described in step S104 above, data on temperature changes may be input to the first calculation means, and the data on the time series of temperature changes may be frequency analyzed to calculate (calculate) the thermoelastic temperature amplitude ΔT _est of the preliminary measurement sample for each measurement (see FIG. 3), and the average temperature T _est of the measurement object (test piece 10) may be calculated (calculated) based on the time series of temperature change data of each measurement using the above formula (2).

Next, in step S204, for a measurement object whose degree of fatigue damage is unknown, the ΔT _est and T _est calculated in step S203 (the thermoelastic temperature amplitude ΔT _est and average temperature T _est of the measurement object whose degree of fatigue damage is unknown, calculated by the first calculation means) are used to calculate (calculate) the non-dimensional thermoelastic temperature amplitude (ΔT/T) _est of the measurement object at the time of measurement (current stage). Note that, for convenience, the non-dimensional thermoelastic temperature amplitude of a measurement object whose degree of fatigue damage is unknown is denoted as "(ΔT/T) est" so as to be easily distinguished from the "non-dimensional thermoelastic temperature amplitude (ΔT/T)" of the pre-measurement sample used when determining the above-mentioned "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree _" . Such step S204 is a step of non-dimensionalizing the thermoelastic temperature amplitude ΔT _est of the measurement object having an unknown degree of fatigue damage with the average temperature T _est (ΔT _est is divided by T _est ) to calculate the non-dimensional thermoelastic temperature amplitude (ΔT/T) _est of the measurement object having an unknown degree of fatigue damage. This calculation can be performed by the second calculation means by inputting ΔT _est and T _est calculated in step S203.

Next, in step S205, a normalized thermoelastic temperature amplitude Y _est is calculated (computed) for the measurement object with an unknown degree of fatigue damage. For convenience, the non-dimensional thermoelastic temperature amplitude of the measurement object with an unknown degree of fatigue damage is denoted as "Y est " so as to be easily distinguished from the "non-normalized thermoelastic temperature amplitude _Y " of the pre-measurement sample used in determining the above-mentioned "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree." Step S205 is a process of normalizing the non-dimensional thermoelastic temperature amplitude (ΔT/T) _est of the measurement object with an unknown degree of fatigue damage by the non-dimensional thermoelastic temperature amplitude (ΔT/T) _1-est at the beginning of fatigue to calculate the normalized thermoelastic temperature amplitude Y _est of the measurement object with an unknown degree of fatigue damage. The value of the non-dimensional thermoelastic temperature amplitude (ΔT/T) _1-est at the initial stage of fatigue is the value input in step S201, and the value of the non-dimensional thermoelastic temperature amplitude (ΔT/T) _est of the measurement object with an unknown degree of fatigue damage is the value found in step S204. Such calculation (computation) of the normalized thermoelastic temperature amplitude Y _est can be performed in the third calculation means by using the value input in step S201 and inputting the value of the non-dimensional thermoelastic temperature amplitude (ΔT/T) _est of the measurement object with an unknown degree of fatigue damage found in step S204.

Next, in step S206, based on the normalized thermoelastic temperature amplitude Y _est of the measurement object with unknown fatigue damage, the fatigue damage degree D _est of the measurement object with unknown fatigue damage is specified from the relationship between the normalized thermoelastic temperature amplitude Y and the fatigue damage degree D previously obtained using a sample of the same material as the measurement object. Note that the fatigue damage degree of the measurement object with unknown fatigue damage specified in this step is denoted as "D _est " so as to be easily distinguished from the "fatigue damage degree D" of the pre-measurement sample used when obtaining the above-mentioned "relationship between normalized thermoelastic temperature amplitude and fatigue damage degree". Such step S206 is a process of specifying (determining) the fatigue damage degree D _est from the relationship between the normalized thermoelastic temperature amplitude Y _est obtained in step S205 and the fatigue damage degree previously obtained for a sample of the same material as the measurement object. Such a fatigue damage degree D _est may be specified by utilizing the relationship between the normalized thermoelastic temperature amplitude and the fatigue damage degree (for example, the above-mentioned approximation formula, etc.) previously obtained using a pre-measurement sample, and may be specified in a fatigue damage degree specifying means (a calculation unit in an external computer) that is a calculation unit (a calculation unit programmed to enable the above-mentioned calculation processing) configured to specify (calculate) the fatigue damage degree of a measurement object whose fatigue damage degree is unknown based on such a relationship. Note that, as a method for specifying the fatigue damage degree, for example, when the relationship between the normalized thermoelastic temperature amplitude Y and the fatigue damage degree D previously obtained using a pre-measurement sample is an approximation formula showing the relationship between Y and D, a method can be adopted in which the fatigue damage degree D is calculated by introducing the value of Y _est into the approximation formula instead of Y, and the calculated value of D is assumed to be the fatigue damage degree D _est of the unknown measurement object.

A preferred embodiment of the fatigue damage identification device and fatigue damage identification method of the present invention has been described above with reference to the drawings, but the fatigue damage identification device and fatigue damage identification method of the present invention are not limited to the above embodiment. For example, in the embodiment shown in FIG. 1, an infrared camera is used as the temperature measurement means 12, but the temperature measurement means 12 is not limited to an infrared camera and may be a thermocouple. Note that even when a thermocouple is used as the temperature measurement means 12, it is possible to easily measure the fatigue damage degree.

The present invention will be described in more detail below with reference to examples and comparative examples, but the present invention is not limited to the following examples.

[Test Example 1]
<Preparation of test specimen>
Three sheets (plate-like bodies) made of the same carbon fiber reinforced plastic (CFRP) and having a thickness of 3 mm were prepared, and test pieces 10 of the measurement object (CFRP) as shown in FIG. 12 were prepared from each sheet, and were designated as test pieces A, B, and C. As shown in FIG. 12, each test piece was a strip-like test piece having a surface with a length ( _Ls ): 200 mm and a width ( _Ws ): 25 mm and a thickness of 3 mm. A black body paint was applied to a part of the surface of each test piece, an area A1 (an area with a length (L _A1 ): 80 mm and a width: 25 mm), to form a paint application area A1. As shown in FIG. 12, the position where the area A1 was formed was set so that the distance (L _B ) from each end in the longitudinal direction of the test piece was 60 mm. In addition, when measuring the temperature data, an area A2 (length (L _A2 ): 70 mm, width (W _A2 ): 23 mm) indicated by a dashed line within the area A1 of the test piece 10 shown in FIG. 12 was used as the area for measuring the temperature.

Four samples for tensile testing were prepared using the CFRP constituting the sheets used to prepare test pieces A, B, and C, and the ultimate tensile strength of the CFRP was determined separately and the average value was calculated (note that such tensile strength (UTS) was measured in advance to determine the magnitude of the load applied to the test pieces by the load applying means 12. Hereinafter, the average value of the ultimate tensile strength thus determined is referred to simply as "tensile strength (UTS)" in Table 1).

<Equipment used for measurement>
For the measurements, an apparatus having the same configuration as the measurement unit 1 shown in Fig. 1 (an apparatus simulating the apparatus in Fig. 1) was used. Here, a hydraulic servo type fatigue testing machine was used as the load applying means 11, and an infrared thermography camera was used as the temperature measuring means 12. Each test piece 10 was used such that the area other than the paint application area A1 was fixed by the gripping tool of the hydraulic servo type fatigue testing machine. In this way, the measurements were performed using the measurement unit 1 configured so that the surface temperature distribution of the test piece 10 could be measured using the infrared camera 12 installed in front of the fatigue testing machine 11.

<Measurement of thermoelastic temperature amplitude and average temperature of test pieces A to C>
Using each of the test pieces A to C as the test pieces 10, a load was repeatedly applied to _the test pieces 10 by the load applying means 11 under the load conditions shown in Table 1, and the temperature change in the area corresponding to the area A2 on the surface of the test piece 10 was measured by the temperature measuring means 12 under the conditions shown in Table 2 every time the specified number of repetitions (X 1 to X ₂₅ ) shown in FIG. 13 was reached. In addition, in such measurements, the magnitude of the stress amplitude was changed between the test piece A and the test pieces B and C as shown in Table 1. Furthermore, the operation of the load applying means 11 (the repeated load application process) was performed until each test piece was broken. As a result, the test piece A broke at 13,160 cycles, the test piece B broke at 23,860 cycles, and the test piece C broke at 75,720 cycles. Therefore, the test piece A was measured up to the 12th time (X ₁ to X ₁₂ ), the test piece B was measured up to the 14th time (X ₁ to X ₁₄ ), and the test piece C was measured up to the 25th time (X ₁ to X ₁₅ ).

By such measurements, for each test piece, a time series distribution of surface temperature over a period of 190 cycles from the specified number of repetitions shown in Fig. 13 was obtained (surface temperature data of the test piece for the specified number of repetitions (190 cycles) was obtained for each measurement). Here, when analyzing the temperature of the test piece, for the image of the time series temperature distribution, the temperature of each pixel in the part of the region A2 was obtained for each image of each frame, and the average value was calculated and adopted as the temperature of the test piece in that frame (Note that in this analysis, 4009 frames of images are used as data for each measurement each time the specified number of repetitions ( _X1 to _X25 ) is reached (see Table 2), so that time series temperature fluctuation data of 4009 points is obtained for all pixels in the part of the region A2 for each measurement).

In this way, the temperature of each test piece was analyzed for each measurement, and data on the number of load repetitions and temperature (data as shown in Figure 2: which can also be said to be the relationship between elapsed time and temperature) was obtained. Then, the temperature data of each test piece was subjected to frequency analysis (Fourier transform was used here) to obtain the relationship between frequency and temperature amplitude (relationship as shown in Figure 3), and the temperature amplitude ΔT (absolute value) of the test piece at the same frequency (10 Hz) as the load frequency (frequency of repeated load) was obtained. In addition, the above formula (2) was calculated based on the temperature data of the test piece for each frame to obtain the average temperature T of each test piece. In this way, the temperature amplitude at the same frequency (10 Hz) as the repeated load was obtained as the thermoelastic temperature amplitude ΔT from the relationship between the temperature amplitude obtained by frequency analysis of the time-series temperature fluctuation data of 4009 points and the frequency, and the temperature amplitude at the same frequency (10 Hz) as the repeated load was obtained as the thermoelastic temperature amplitude ΔT, and the average value of the time-series temperature fluctuation data of the test piece for 4009 points was used as the average temperature T.

The number of repeated load cycles until the test specimen was broken was defined as the fatigue life _Nf of each test specimen. The fatigue lives _Nf were 13160 cycles for test specimen A, 23860 cycles for test specimen B, and 75720 cycles for test specimen C.

<Consideration of the relationship between dimensionless thermoelastic temperature amplitude and number of repeated loads>
For test pieces A and B, the thermoelastic temperature amplitude ΔT (absolute value) and average temperature T of each measurement were obtained by the above-mentioned measurement of the thermoelastic temperature amplitude and average temperature, and using the measurement results, a graph was obtained for test pieces A and B showing the relationship between the dimensionless thermoelastic temperature amplitude (ΔT/T) _P and the number of specified load repetitions for each measurement (here, P is a natural number indicating the number of measurements). The obtained results are shown in FIG.

As shown in Fig. 14, the value (absolute value) of the non-dimensional thermoelastic temperature amplitude (ΔT/T) decreased with an increase in the number of load cycles X for both test pieces A and B. The reason why the value (absolute value) of the non-dimensional thermoelastic temperature amplitude (ΔT/T) differs for each test piece is obviously due to the difference in the magnitude of the stress amplitude during temperature measurement between test pieces A and B, as shown in Table 1 (as is clear from the above formula (3), this is because the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p depends on the stress amplitude Δσ). In addition, since the value (absolute value) of the non-dimensional thermoelastic temperature amplitude (ΔT/T) gradually decreases with an increase in the number of load cycles X, it can also be seen that the value of the non-dimensional thermoelastic temperature amplitude (ΔT/T) is a value related to the degree of fatigue damage.

<Consideration of the relationship between normalized thermoelastic temperature amplitude and life ratio>
As described above, since the non-dimensional thermoelastic temperature amplitude (ΔT/T) _p depends on the stress amplitude Δσ, the use of a normalized value (normalized thermoelastic temperature amplitude) was considered so that measurements could be made taking into account the fluctuations and fluctuations of the stress amplitude Δσ during measurement, and normalization was performed as shown below. That is, the state in which a repeated load is applied for 300 cycles, which is the specified number of times for the first measurement (X ₁ ), is defined as the early stage of fatigue, and the non-dimensional thermoelastic temperature amplitude for the first measurement is defined as the non-dimensional thermoelastic temperature amplitude (ΔT/T) ₁ at the early stage of fatigue, and the non-dimensional thermoelastic temperature amplitude for each measurement is normalized by the non-dimensional thermoelastic temperature amplitude at the early stage of fatigue. That is, such normalization is performed using the following formula:
Y _p = (ΔT/T) _p / (ΔT/T) ₁
(P is a natural number indicating the number of measurements), and the value obtained for each measurement was taken as the normalized thermoelastic temperature amplitude Y for each measurement. In addition, since it is considered that the fatigue life differs depending on conditions such as the magnitude of the stress amplitude Δσ, the number of repetitions was also calculated using the following formula:
_Dp = _Xp / _Nf
(In the formula, _Xp represents the specified number of repetitions for each measurement, _Nf represents the fatigue life of the test piece, and _Dp represents the life ratio.)
Normalization was performed by

The relationship between the dimensionless thermoelastic temperature amplitude (ΔT/T) and the number of repetitions for test pieces A and B was rearranged using normalized values, and the relationship between the normalized thermoelastic temperature amplitude Y and the life ratio D for each measurement was obtained. The results are shown in Figure 15. The solid line in Figure 15 is an approximation line obtained by approximating all data points using the least squares method. When approximating with a quartic function, the normalized thermoelastic temperature amplitude Y changed monotonically with respect to the life ratio D, so the approximation line was a quartic function curve.

As is clear from the results shown in FIG. 15, when the dimensionless thermoelastic temperature amplitude and the number of repetitions are normalized, it was confirmed that both test pieces A and B can be plotted on approximately the same line. In addition, as shown in FIG. 15, both test pieces A and B can be plotted on approximately the same line, so that even for a test piece of a measurement object with an unknown degree of fatigue damage, if the measurement conditions for the thermoelastic temperature amplitude and average temperature are the same as those for the measurement conditions for either test piece A or B, and the thermoelastic temperature amplitude and average temperature of the test piece are obtained, and the normalized thermoelastic temperature amplitude Y of the test piece is obtained, it is possible to obtain a life ratio indicating the degree of fatigue damage from the value of Y and the relationship shown in FIG. 15. In this way, when the "relationship between the normalized thermoelastic temperature amplitude Y and the life ratio D related to the degree of fatigue damage" is obtained in advance, the life ratio D (value related to the degree of fatigue damage) can be obtained by obtaining the normalized thermoelastic temperature amplitude Y of a test piece of a measurement object with an unknown degree of fatigue damage based on the relationship (a quartic function in the example shown in FIG. 15), and it can be seen that the degree of fatigue damage can be specified.

<Considerations on determining fatigue life>
First, the degree of fatigue damage was specified (estimated) using the measurement results of the test piece C as follows. That is, first, based on the measurement results of the thermoelastic temperature amplitude and the average temperature of the test piece C, the non-dimensional thermoelastic temperature amplitude for each measurement of the test piece C was calculated in the same manner as the test pieces A and B, and the relationship between the non-dimensional thermoelastic temperature amplitude and the number of load repetitions was obtained. Note that a graph showing the relationship between the non-dimensional thermoelastic temperature amplitude and the number of load repetitions of the test piece C is shown in FIG. 16. Next, in the same manner as the test pieces A and B, the non-dimensional thermoelastic temperature amplitude for each measurement of the test piece C was normalized by the non-dimensional thermoelastic temperature amplitude (ΔT/T) ₁ at the initial stage of fatigue (first measurement) of the test piece C to obtain the normalized thermoelastic temperature amplitude _Yp for each measurement. Next, the normalized thermoelastic temperature amplitude _Yp for each measurement of the test piece C obtained was used to specify (estimate) the life ratio of the test piece C at the time of each measurement from the relationship between the normalized thermoelastic temperature amplitude and the life ratio (approximation equation of a quartic function) shown in FIG. 15. Hereinafter, for the sake of convenience, the "estimated value of the life ratio" of the test piece C for each measurement obtained from the relationship between the normalized thermoelastic temperature amplitude and the life ratio shown in FIG. 15 (approximation of a fourth-order function) is denoted as "D _est ".

Furthermore, the number of repetitions (prescribed number of times) of each measurement was normalized using the value of the fatigue life _Nf (actual measured value) for the test specimen C to determine the life ratio _Dp for each measurement. Note that, for the sake of convenience, the "actual life ratio" of each measurement of the test specimen C determined by normalization using the value of the fatigue life _Nf is hereinafter referred to as " _Dact ."

Next, the actual life ratio D _act of the test piece C thus obtained was compared with the estimated value D _est of the life ratio. That is, the relationship between the "estimated value D est of the life ratio" obtained by estimating the life ratio (degree of fatigue damage) from the relationship between the normalized thermoelastic temperature amplitude and the life ratio obtained in advance (FIG. 15: approximation of a quartic function) and the "actual life ratio D _{act "} obtained from the above-mentioned measurement results was examined. Here, a graph showing the relationship between the actual life ratio D _act and the estimated value D _est of the life ratio is shown in FIG. 17. Note that the solid line shown in FIG. 17 is a line drawn for convenience of evaluation when D _act and D _est are equal (D _est =D _act ), and the broken line is a line drawn for convenience of evaluation when the estimation error of the life ratio is ±10% (when the life ratio differs by 0.1).

17, it was found that the error of the estimated value D _est of the life ratio with respect to the actual life ratio (value obtained by experiment) D _act was basically within a range of ±10% up to the region where the life ratio was 0.8. From these results, it was confirmed that it is possible to specify (estimate) the life ratio (fatigue damage degree) with extremely high accuracy by specifying (estimating) the life ratio (fatigue damage degree) from the relationship between the normalized thermoelastic temperature amplitude and the life ratio obtained in advance (approximation equation of a quartic function).

In addition, if the life ratio can be estimated, it is possible to estimate the fatigue life and remaining fatigue life from the life ratio D _est , and it is also possible to evaluate the remaining fatigue life at the same time. For example, by using the measured value D _est at 10,000 cycles of the load repetition number (X ₁₂ ) of the 12th measurement of the test piece C, and by modifying the formula for the life ratio D (D _p =X _p /N _f ), the estimated value N _est of the fatigue life can be calculated using the following formula:
D _est = X ₁₂ / N _est
From the relationship, it was found that the estimated value N _est of the fatigue life of the test specimen C at the time when a cyclic load of 10,000 cycles was applied was 68,427 cycles, and the remaining fatigue life at the time when a cyclic load of 10,000 cycles was applied was estimated to be 58,427 cycles. On the other hand, the actual fatigue life N _f of the test specimen C was 75,720 cycles, and the actual remaining fatigue life of the test specimen C at the time when a cyclic load of 10,000 cycles was applied was 65,720 cycles. From such a comparison of the remaining fatigue lives, it was found that the fatigue life and the remaining fatigue life can be measured with high accuracy from the life ratio D _est .

As explained above, since the non-dimensional thermoelastic temperature amplitude (ΔT/T) is a physical quantity related to the degree of fatigue damage of the material (CFRP in Test Example 1), by determining in advance the change in the normalized value of the non-dimensional thermoelastic temperature amplitude (ΔT/T) and determining in advance the relationship between the normalized thermoelastic temperature amplitude and the degree of fatigue damage (life ratio), it is possible to determine the degree of fatigue damage of the measurement object by measuring the thermoelastic temperature amplitude ΔT and the average temperature T using a test piece of the measurement object with an unknown degree of fatigue damage, and by using the measured value to determine the normalized thermoelastic temperature amplitude of the test piece, it is possible to determine the degree of fatigue damage of the measurement object and also to determine the remaining fatigue life from the specified value of the degree of fatigue damage.

As described above, the present invention makes it possible to provide a fatigue damage determination device and a method for determining fatigue damage that can measure the fatigue damage level of a measurement object with high accuracy.

The fatigue damage identification device of the present invention is therefore useful as a technique for estimating the fatigue damage level and remaining life of a measurement object whose fatigue damage level is unknown.

1... measuring part, 10... test piece, 11... load application means, 12... temperature measurement means.

Claims (4)

A load applying means for applying a repeated load to the object to be measured at a predetermined frequency;
A temperature measuring means for measuring a temperature change of the measurement object;
A first calculation means for calculating a thermoelastic temperature amplitude and an average temperature of the object to be measured from data of a temperature change of the object to be measured;
A second calculation means for non-dimensionalizing the thermoelastic temperature amplitude by the average temperature to calculate the non-dimensional thermoelastic temperature amplitude of the measurement object;
A third calculation means for normalizing the non-dimensional thermoelastic temperature amplitude of the measurement object by the non-dimensional thermoelastic temperature amplitude at the initial stage of fatigue to calculate a normalized thermoelastic temperature amplitude of the measurement object;
a fatigue damage degree specifying means for specifying a fatigue damage degree of the measurement object based on a relationship between a normalized thermoelastic temperature amplitude and a fatigue damage degree obtained in advance using a sample made of the same material as the measurement object;
A fatigue damage identification device comprising: The fatigue damage identification device according to claim 1, characterized in that the object to be measured is made of fiber-reinforced plastic. applying a cyclic load to a measurement object at a predetermined frequency;
Measuring a temperature change of the measurement object;
Calculating a thermoelastic temperature amplitude and an average temperature of the object from the data of the temperature change of the object;
A step of non-dimensionalizing the thermoelastic temperature amplitude with the average temperature to calculate the non-dimensional thermoelastic temperature amplitude of the measurement object;
A step of normalizing the non-dimensional thermoelastic temperature amplitude of the measurement object by the non-dimensional thermoelastic temperature amplitude at the initial stage of fatigue to calculate a normalized thermoelastic temperature amplitude of the measurement object;
determining the degree of fatigue damage from a relationship between a normalized thermoelastic temperature amplitude and a degree of fatigue damage previously obtained for a sample made of the same material as the measurement target;
A method for identifying fatigue damage, comprising: The method for determining the degree of fatigue damage according to claim 3, characterized in that the object to be measured is made of fiber-reinforced plastic.