JP7799530B2 - Fatigue diagnosis device - Google Patents

Description

This disclosure relates to a fatigue diagnosis device.

A conventional vehicle fatigue damage diagnosis system includes a sensor and a vehicle ECU. The sensor is attached to the object to be diagnosed. The object to be diagnosed is a component attached to the vehicle. The vehicle ECU calculates the fatigue damage level of the object to be diagnosed based on the detection values obtained from the sensor (see, for example, Patent Document 1).

JP 2013-79920 A

In the conventional fatigue and damage assessment system described above, fatigue assessment is performed based on the fatigue and damage level at the location where the sensor is attached. For this reason, the sensor must be attached to the appropriate location on the subject to be assessed, i.e., a location where fatigue assessment can be performed accurately. However, depending on the installation location and orientation of the subject to be assessed, it may not be possible to attach the sensor to the appropriate location, which could result in a decrease in the accuracy of the assessment.

The present disclosure has been made to solve the above-mentioned problems, and aims to provide a fatigue diagnosis device that can perform fatigue diagnosis of the diagnostic target component with stable accuracy.

The fatigue diagnosis device of the present disclosure includes an acceleration acquisition unit that acquires acceleration measurement values measured by an acceleration sensor attached to the component to be diagnosed, a stress estimation unit that calculates a stress estimate at at least one evaluation point that is part of the component to be diagnosed based on the acceleration measurement values and a finite element model that reproduces the component to be diagnosed using the finite element method, and a fatigue determination unit that determines the degree of fatigue of the component to be diagnosed based on the stress estimate value, wherein the stress estimation unit calculates stress estimates at two or more evaluation points, and the fatigue determination unit determines the degree of fatigue of the component to be diagnosed based on the stress estimate value at each evaluation point .

The fatigue diagnosis device disclosed herein can perform fatigue diagnosis of the target component with stable accuracy.

1 is a block diagram showing an overall configuration according to a first embodiment; 4 is a flowchart showing a process for creating an analysis program by the fatigue diagnosis device of FIG. 1 . FIG. 10 is a diagram illustrating a method for discretizing a degenerate model of a continuous system. 2 is a flowchart showing an acceleration measurement process performed by the acceleration sensor of FIG. 1 . 4 is a flowchart showing a stress estimation value calculation process performed by the fatigue diagnosis device of FIG. 1 . 4 is a flowchart showing a fatigue determination process performed by the fatigue diagnosis device of FIG. 1 . FIG. 7 is a diagram illustrating the process of tallying the number of stress occurrences in FIG. 6 . 7 is a diagram illustrating a process for calculating the cumulative damage level of the evaluation target point shown in FIG. 6. FIG. FIG. 2 is a configuration diagram showing a first example of a processing circuit that realizes the fatigue diagnosis device of the first embodiment. FIG. 10 is a configuration diagram showing a second example of a processing circuit that realizes the fatigue diagnosis device according to the first embodiment.

The following describes the embodiments with reference to the drawings.

Embodiment 1.
FIG. 1 is a block diagram showing the overall configuration according to the first embodiment.

In FIG. 1, the fatigue diagnosis system 1000 is a system that performs fatigue diagnosis on a diagnosis target component 200 provided in a vehicle body 100. Here, the vehicle body 100 is, for example, the main body of a railway vehicle. The diagnosis target component 200 is, for example, an inverter device. The diagnosis target component 200 is provided below the vehicle body 100.

The fatigue diagnosis system 1000 includes an acceleration sensor 300 and a fatigue diagnosis device 400.

The acceleration sensor 300 is attached to the lower end of the diagnostic target component 200. The acceleration sensor 300 includes a sensor body 310 and an amplifier 320.

The sensor body 310 detects vibrations at the location where it is attached and converts the detected vibrations into an electric charge signal.

The amplifier 320 amplifies the charge signal converted by the sensor body 310 and converts the amplified signal into acceleration.

The acceleration sensor 300 performs the above operation at regular intervals to measure acceleration values. The acceleration sensor 300 transmits data to the fatigue diagnosis device 400 for each detection cycle in which it measures acceleration values. The data transmitted here is referred to as acceleration measurement values 11. The acceleration sensor 300 sequentially transmits the acceleration measurement values 11 via communication means (not shown). Alternatively, the acceleration sensor 300 may accumulate acceleration measurement values 11 for a certain period of time in a storage device (not shown), and transmit the accumulated data set to the fatigue diagnosis device 400.

The fatigue diagnosis device 400 includes an acceleration acquisition unit 410, a stress estimation unit 420, a fatigue determination unit 430, an alarm issuing unit 440, and an analysis program creation unit 450.

The acceleration acquisition unit 410 acquires the acceleration measurement value 11 measured by the acceleration sensor 300.

The stress estimation unit 420 calculates a stress estimation value at at least one evaluation point that is part of the diagnosis target component 200. The stress estimation unit 420 acquires the acceleration measurement value 11 from the acceleration acquisition unit 410. The stress estimation unit 420 acquires the analysis program 13 that was created in advance. The stress estimation unit 420 then incorporates the acceleration measurement value 11 into the analysis program 13 and executes the analysis program 13. By executing this analysis program 13, a stress estimation value 12 at the evaluation point is calculated.

In embodiment 1, evaluation points P1, P2, and P3 shown in Figure 1 are the locations where stress estimates are calculated. Here, evaluation points P1 and P2 are located at the joint between the vehicle body 100 and the diagnosis target component 200. Evaluation point P3 is located at the center of the top surface of the diagnosis target component 200.

The fatigue determination unit 430 determines the degree of fatigue of the diagnosis target component 200 based on the stress estimation value 12 calculated by the stress estimation unit 420. If the degree of fatigue of the diagnosis target component 200 is equal to or greater than a threshold value, the diagnosis target component 200 instructs the alarm generation unit 440 to generate an alarm.

When the alarm issuing unit 440 receives an alarm instruction, it issues an alarm such as an audio notification via a speaker or a warning display on a monitor. The alarm issuing unit 440 may be provided external to the fatigue diagnosis device 400.

The analysis program creation unit 450 creates the analysis program 13. The analysis program creation unit 450 acquires sensor position information, evaluation target point position information, and other necessary parameters set by the operator. The analysis program creation unit 450 creates the analysis program 13 based on the various set data. Here, the sensor position information is position information of the acceleration sensor 300 in the diagnosis target component 200. The evaluation target point position information is position information of the evaluation target points P1 to P3 in the diagnosis target component 200.

Figure 2 is a flowchart showing the process of creating the analysis program 13 by the fatigue diagnosis device 400 of Figure 1.

In step S101, the analysis program creation unit 450 of the fatigue diagnosis device 400 constructs a finite element model that reproduces the diagnosis target component 200. The operator specifies the type of element, such as solid element, shell element, or beam element, and the analysis program creation unit 450 acquires the specified element type. The operator also specifies the number of dimensions, such as a two-dimensional model or a three-dimensional model, and the analysis program creation unit 450 acquires the specified number of dimensions.

In addition to the type and number of dimensions of the above elements, the analysis program creation unit 450 also acquires the material properties of the diagnosis target component 200 specified by the operator. Material properties include the Young's modulus, Poisson's modulus, density, etc. The material properties used may be either isotropic or anisotropic.

In step S102, the analysis program creation unit 450 performs eigenvalue analysis on the constructed finite element model. At this time, the analysis program creation unit 450 acquires the boundary conditions specified by the operator. The boundary conditions are conditions under which the displacement of the diagnostic target component 200 at a fixed position is set to zero.

The analysis program creation unit 450 performs eigenvalue analysis to extract natural frequencies, mode vectors, and modal stresses.

Natural frequency is numerical data that represents the resonant frequency. When a force is applied to an object repeatedly at a certain period, for example every 0.1 seconds, the object resonates and vibrates violently. The reciprocal of this period is the natural frequency.

A mode vector is the distribution of an object's movement when it resonates at its natural frequency. The mode vector of parts that vibrate strongly due to resonance will have a high value.

Modal stress is the distribution of stress that occurs in an object when it resonates at its natural frequency. Modal stress in parts that support the vibration of a component will have a high value.

In step S103, the analysis program creation unit 450 creates a degenerated continuous system model based on the extracted natural frequencies, mode vectors, and modal stresses. The analysis program creation unit 450 performs a simulation using the natural frequencies, mode vectors, and modal stresses for the finite element model prepared as described above.

The analysis program creation unit 450 performs simulations to obtain data for each of the first, second, ..., nth natural vibration modes. This natural vibration mode data includes the natural frequency, mode vector, and modal stress information extracted in step S102.

In addition, when performing the simulation, the analysis program creation unit 450 acquires the damping ratio, load input points, number of acceleration sensors 300, sensor position information, number of evaluation target points, and evaluation target point position information from the operator for each natural vibration mode. Here, the load input points are points where a load is applied. Specifically, the load input points are the connection points between the vehicle body 100 and the diagnosis target member 200, and are evaluation target points P1 and P2 in Figure 1.

The analysis program creation unit 450 creates a degenerated model of the continuous system based on this data. The following equation 1 expresses the degenerated model of the continuous system in state space representation.

The state space representation of Equation 1 is composed of the state equation shown in the upper part and the observation equation shown in the lower part. Furthermore, matrices A to D are matrices with fixed values calculated by the processing in step S103.

The determinants of matrix A are the eigenvalues and damping ratios. The determinants of matrix B are the modal displacements of the load input points. The determinants of matrix C are the eigenvalues, damping ratios, modal displacements of the response points, and modal stresses of the response points. The determinants of matrix D are the modal displacements of the load input points and modal displacements of the response points.

In Equation 1, the state quantities of displacement and stress in the real world are represented by {W, σ}. That is, the acceleration acquired by the acceleration sensor 300 is represented on the left side of the observation equation as the second derivative of the displacement W. Furthermore, the estimated value of the desired stress is represented on the left side of the observation equation as σ.

In Equation 1, the number of rows in matrices C and D increases as the number of evaluation points increases. Furthermore, the number of rows in matrices C and D increases as the number of acceleration sensors 300 increases.

Returning to the explanation of the flowchart in Figure 2, in step S104, the analysis program creation unit 450 discretizes the created continuous system degenerated model. The acceleration measurement value 11 is data that is discrete for each detection cycle of the acceleration sensor 300. Therefore, the analysis program creation unit 450 discretizes the continuous system degenerated model with respect to time.

As shown in Equation 1, a degenerated model of a continuous system consists of two equations: a state equation and an observation equation. Of these two equations, only the state equation is a continuous system equation. For this reason, the analysis program creation unit 450 performs discretization on the state equation.

Equation 2 below expresses the degenerated model of the discretized system in state space representation.

3 is a diagram illustrating a method for discretizing a degenerate model of a continuous system. To perform discretization, it is necessary to find matrices _A2 and _B2 in the above equation 2. Here, matrices _A2 and _B2 are found using a fourth-order Runge-Kutta method.

Equation F1 in Figure 3 is a discretized equation for the fourth-order Runge-Kutta method. If we assume that the load u in this equation F1 is constant, the discrete matrix shown in equation F2 in Figure 3 is calculated.

In Formula F2, by replacing AdT with k _1a , BdT with k _1b , ..., matrix _A2 and matrix _B2 become as shown in Formula F3, where I in Formula F3 represents a unit matrix.

Returning to the explanation of the flowchart in Figure 2, the analysis program creation unit 450 creates the analysis program 13 in step S105. The analysis program 13 is a program that combines a degenerated model after discretization, i.e., a state space representation after discretization, with a Kalman filter.

The analysis program 13 created in this manner is called by the stress estimation unit 420 during the stress estimation value calculation process described below.

Figure 4 is a flowchart showing the acceleration measurement process performed by the acceleration sensor 300 in Figure 1.

In step S201, the acceleration sensor 300 starts measuring acceleration. For example, the acceleration sensor 300 starts measuring acceleration when it receives a start operation from the worker or when a specified time arrives.

In step S202, the sensor body 310 of the acceleration sensor 300 converts vibrations occurring at the location where it is attached into an electric charge signal.

In step S203, the amplifier 320 amplifies the signal from the sensor main body 310 and converts it into a voltage signal. The amplifier 320 converts the amplified voltage signal into acceleration using a coefficient calculation.

In step S204, the acceleration sensor 300 transmits the measured acceleration to the fatigue diagnosis device 400 as an acceleration measurement value 11. This acceleration measurement value 11 is acquired by the acceleration acquisition unit 410 of the fatigue diagnosis device 400.

In step S205, the acceleration sensor 300 determines whether to end acceleration measurement. For example, the acceleration sensor 300 ends acceleration measurement when it receives an end operation from the worker or when a specified time arrives. If the acceleration sensor 300 does not end acceleration measurement, it returns the process to step S202.

Figure 5 is a flowchart showing the stress estimation value calculation process performed by the fatigue diagnosis device 400 of Figure 1. Here, load control on the diagnosis target component 200 is not taken into consideration. That is, in the flowchart of Figure 5, matrix D shown in Equation 2 is not used. Note that if the load applied to the diagnosis target component 200 is known, matrix D is also used.

The stress estimation unit 420 of the fatigue diagnosis device 400 calls the acceleration measurement value 11 and the analysis program 13 to perform the stress estimation value calculation process.

In step S301, the stress estimation unit 420 sets initial values for the state estimate X and the error covariance matrix P. Here, the state estimate X refers to the state variable z described above.

In the initial state, there is no vibration, so the state estimate X is 0. Also, the error covariance matrix P is set to the identity matrix as the initial value I.

In addition to the above initial values, constant parameters for system noise Q and observation noise R are also set. System noise Q and observation noise R are values appropriately determined by the operator according to the noise level.

In step S302, the stress estimator 420 uses the state estimate X and matrix _A2 to calculate a prior state estimate ^X- , where the prior state estimate ^X- is an estimate of the state variable z at time t based on data available up to time t-1.

In step S303, the stress estimation unit 420 calculates a priori error covariance matrix P ⁻ using the error covariance matrix P, matrix A ₂ , matrix B ₂ , system noise Q, matrix A ₂ ′, and matrix B ₂ ′, where matrix A ₂ ′ is the transposed matrix of matrix A ₂ , and matrix B ₂ ′ is the transposed matrix of matrix B ₂ .

In step S304, the stress estimation unit 420 calculates a Kalman gain matrix G using the a priori error covariance matrix P ⁻ , matrix C, matrix C′, and observation noise R. Here, matrix C′ is a matrix obtained by transposing matrix C.

In step S305, the stress estimation unit 420 calculates a state estimation value X using the Kalman gain matrix G, the prior state estimation value X ⁻ , the matrix C′, and the acceleration measurement value 11. The stress estimation unit 420 updates the state estimation value X that has been used so far to the calculated one.

In step S306, the stress estimation unit 420 calculates an estimated value σ of stress at the evaluation point based on the updated state estimation value X. The stress estimation unit 420 substitutes the state estimation value X as the state variable z(t+Δt) into the left side of the state equation shown in Equation 2. If the load u(t) is not taken into account or is constant, the stress estimation unit 420 can determine the state variable z(t) from one cycle ago from the state equation in Equation 2.

The stress estimation unit 420 then substitutes the state variable z(t) into the observation equation in Equation 2. If the load u(t) is not taken into account or is constant, the stress estimation unit 420 can determine the estimated value σ of stress at the evaluation point. Here, the estimated values σ of stress at evaluation points P1 to P3 are simultaneously determined.

The stress estimation unit 420 also writes the obtained stress estimate σ to the previously stored data. In this way, the stress estimate 12 is generated.

In step S307, the stress estimating unit 420 calculates the error covariance matrix P using the initial value I of the error covariance matrix, the Kalman gain matrix G, the matrix C', and the prior error covariance matrix ^P. The stress estimating unit 420 updates the error covariance matrix P that has been used so far to the calculated one.

In step S308, the stress estimation unit 420 determines whether a new acceleration measurement value 11 will be acquired within a specified time. If the stress estimation unit 420 acquires an acceleration measurement value 11 within the specified time, the processing returns to step S302. The stress estimation unit 420 then performs the processing from step S302 onwards on the newly acquired acceleration measurement value 11. If the stress estimation unit 420 does not acquire an acceleration measurement value 11 within the specified time, the processing ends.

Figure 6 is a flowchart showing the fatigue assessment process performed by the fatigue diagnosis device of Figure 1. The flowchart in Figure 6 is performed for each of the evaluation points P1 to P3, but as an example, the process for evaluation point P1 will be described here.

In step S401, the fatigue determination unit 430 tallies the number of times stress occurs at the evaluation point P1 for each stress estimation value.

FIG. 7 is a diagram illustrating the process of tallying the stress occurrence counts in FIG. 6. The graph shown in FIG. 6 shows the waveform of the stress estimation value 12 at the evaluation point P1, with the horizontal axis representing time and the vertical axis representing the stress estimation value. The fatigue determination unit 430 applies the Rainflow method to the waveform of this stress estimation value. As a result, the fatigue determination unit 430 tallys up the number of occurrences of stress occurring at the evaluation point P1 for each stress estimation value. Here, the number of occurrences _M1 , _M2 , _M3 , ..., _Mn are calculated for each of the stress estimation values _σ1 , _σ2 , _σ3 , ..., _σn .

In step S402, the fatigue determination unit 430 calculates the cumulative damage level d for the evaluation point P1.

Fig. 8 is a diagram illustrating the process of calculating the cumulative damage level d of the evaluation point in Fig. 6. When calculating the cumulative damage level d, the fatigue determination unit 430 uses the SN curve of the diagnosis target component 200. This allows the number of repetitions _N1 , _N2 , _N3 , ..., _Nn until component fracture to be determined, corresponding to each of the estimated stress values _σ1 , _σ2 , _σ3 , ..., _σn .
The fatigue determination unit 430 calculates the cumulative damage level d at the evaluation point P1 using the following equation 3.

In step S403, the fatigue determination unit 430 determines whether the cumulative damage level d is equal to or greater than a threshold value. The threshold value may be, for example, 0.3.

If the cumulative damage level d is equal to or greater than the threshold, the fatigue determination unit 430 instructs the alarm issuing unit 440 to issue an alarm in step S404. Upon receiving the alarm issuing instruction, the alarm issuing unit 440 issues an audio alert via a speaker, displays a warning on a monitor, or the like.

If the cumulative damage level d is not greater than or equal to the threshold, the fatigue determination unit 430 terminates processing.

The stress estimation unit 420 of this embodiment calculates stress estimates at evaluation points P1 to P3, which are part of the diagnosis target component 200. The stress estimates are calculated based on acceleration measurements and a finite element model that recreates the diagnosis target component using the finite element method.

By using a finite element model to perform fatigue diagnosis, the fatigue diagnosis device 400 can perform fatigue diagnosis on locations where the acceleration sensor 300 is not attached. Therefore, the fatigue diagnosis device 400 can perform fatigue diagnosis of the diagnosis target component 200 with stable accuracy even if the acceleration sensor 300 is not attached in an appropriate location.

The stress estimation unit 420 also calculates stress estimates at two or more evaluation points. The fatigue determination unit 430 determines the degree of fatigue of the diagnosis target component 200 based on the stress estimates at each evaluation point. Therefore, even if measurement values at multiple positions are required, there is no need to prepare multiple sensors.

The stress estimation unit 420 also calculates a stress estimate based on the state equation and the observation equation. This observation equation includes at least an acceleration element and a stress element at the evaluation point. Therefore, the stress estimation unit 420 can perform calculation processing that takes into account acceleration in the real world and stress at the evaluation point in the real world.

The stress estimation unit 420 also calculates stress estimates by executing a pre-created analysis program. This eliminates the need to create an analysis program each time a deterioration diagnosis is performed, allowing for smoother diagnostic work.

The analysis program creation unit 450 also creates an analysis program based on the sensor position information and the evaluation target point position information. This allows for the creation of an analysis program that takes into account the position of the acceleration sensor 300 and the positions of the evaluation target points P1 to P3.

The fatigue determination unit 430 calculates the cumulative damage level d, which is an index value indicating the degree of fatigue, based on the stress estimation value. If the cumulative damage level d is equal to or greater than the threshold value, the fatigue determination unit 430 instructs the alarm unit to issue an alarm. Therefore, by checking whether an alarm has been issued, the worker can determine whether the diagnosis target component 200 needs to be replaced.

Furthermore, strain gauges are generally used as a means of detecting data for fatigue diagnosis. In contrast, in embodiment 1, an acceleration sensor 300 is used. Therefore, when installing the acceleration sensor 300, the same level of precision as with a strain gauge is not required, making installation easier. Acceleration sensors also output a relatively large voltage signal. This makes them more resistant to external noise than strain gauges.

In the first embodiment, fatigue diagnosis is performed on a diagnosis target component 200 provided on the vehicle body 100 of a railway vehicle. However, this is not limited to this, and the diagnosis target component may also be a component provided on a moving body other than a railway vehicle. Furthermore, the diagnosis target component is not limited to components provided on a moving body, but may also be a component provided on equipment, facilities, and structures. For example, a communication antenna may be the diagnosis target component.

In addition, in embodiment 1, the degenerated model is constructed using the mode degeneration method. However, the degenerated model may be constructed using another method as long as degradation can be determined.

In addition, in embodiment 1, a discretization formula based on the Runge-Kutta method is used, but other discretization formulas such as the Euler method or the Hoyn method may also be used.

Furthermore, a linear Kalman filter is incorporated into the analysis program 13 of embodiment 1. However, from the perspective of versatility or improved accuracy, a non-linear Kalman filter may also be incorporated.

Furthermore, when constructing a finite element model, there are no restrictions on the elements used, such as solid elements, shell elements, beam elements, etc., as long as they can reproduce the vibration state of the target. Furthermore, there are no restrictions on the dimensions of the model, such as 3D models and 2D models.

In addition, there is no limit to the number of eigenvalue analyses that can be performed to construct a reduced model. Furthermore, there are no restrictions on the positions or numbers of load input points, acceleration sensors 300, and evaluation points.

Furthermore, there are no restrictions on the type of acceleration sensor 300, such as whether it is a single-axis sensor or a multi-axis sensor. As long as it can measure acceleration in the required direction, there are no restrictions on the type.

The acceleration sensor 300 may be attached to the diagnosis target component 200 at a position where the diagnosis target component 200 vibrates significantly.

In addition, in embodiment 1, the rainflow method is used when determining deterioration, but this is not a limitation. For example, cycle counting methods such as peak counting and range counting may also be used. Alternatively, fatigue limit diagrams such as the modified Goodman rule, Gerber rule, and Söderberg diagram may also be used. Alternatively, damage assessment formulas such as the modified Miner's rule, Miner's rule, and Heibach's rule may also be used. Alternatively, rather than using these theoretical approaches, assessment may be performed using AI.

Furthermore, the fatigue diagnosis device 400 sequentially acquires acceleration measurement values 11. However, as an alternative operation method, for example, acceleration measurement values 11 may be accumulated over a certain period of time and processed collectively. For example, the acceleration may be sent externally using IoT technology, and fatigue diagnosis processing may be performed externally.

Furthermore, each function of the fatigue diagnosis device 400 of embodiment 1 is realized by a processing circuit. Figure 9 is a configuration diagram showing a first example of a processing circuit that realizes each function of the fatigue diagnosis device 400 of embodiment 1. The processing circuit 50 of the first example is dedicated hardware.

The processing circuit 50 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination of these. Furthermore, each function of the fatigue diagnosis device 400 may be realized by a separate processing circuit 50. Alternatively, the functions of the fatigue diagnosis device 400 may be realized collectively by a processing circuit 50.

Furthermore, Figure 10 is a diagram showing a second example of a processing circuit that realizes each function of the fatigue diagnosis device 400 of embodiment 1. The processing circuit 60 of the second example includes a processor 61 and a memory 62.

In the processing circuit 60, each function of the fatigue diagnosis device 400 is realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs. The software and firmware are then stored in memory 62. The processor 61 realizes the functions of each unit by reading and executing the programs stored in memory 62.

The programs stored in memory 62 can be said to cause the computer to execute the procedures or methods of each of the above-mentioned sections. Here, memory 62 refers to non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), and EEPROM (Electrically Erasable and Programmable Read Only Memory). Magnetic disks, flexible disks, optical disks, compact disks, minidisks, DVDs, etc. also qualify as memory 62.

Some of the functions of the above-mentioned components may be realized by dedicated hardware, and some by software or firmware.

In this way, the processing circuit can realize the functions of each of the above-mentioned parts through hardware, software, firmware, or a combination of these.

11 Acceleration measurement value, 12 Stress estimation value, 13 Analysis program, 200 Diagnosis target component, 300 Acceleration sensor, 400 Fatigue diagnosis device, 410 Acceleration acquisition unit, 420 Stress estimation unit, 430 Fatigue determination unit, 440 Alarm generation unit, 450 Analysis program creation unit.

Claims (5)

an acceleration acquisition unit that acquires acceleration measurement values measured by an acceleration sensor attached to the diagnosis target component;
a stress estimating unit that calculates a stress estimated value at at least one evaluation target point that is a part of the diagnosis target member based on the acceleration measurement value and a finite element model that reproduces the diagnosis target member by a finite element method; and a fatigue determining unit that determines a degree of fatigue of the diagnosis target member based on the stress estimated value ,
the stress estimation unit calculates the stress estimation values at two or more of the evaluation target points;
The fatigue determination unit determines the degree of fatigue of the diagnosis target component based on the stress estimated value at each of the evaluation target points . the stress estimator calculates the stress estimated value based on a state equation and an observation equation, each of which is derived based on the finite element model;
2. The fatigue diagnostic device according to claim 1 , wherein the observation equation includes at least an acceleration element and a stress element at the evaluation point. the stress estimation unit calculates the stress estimated value by executing an analysis program;
3. The fatigue diagnosis device according to claim 2 , wherein the analysis program is a program created in advance based on a combination of a state space representation formed by the state equation and the observation equation and a Kalman filter. further comprising an analysis program creation unit that creates the analysis program;
The analysis program creation unit
acquiring sensor position information, which is position information of the acceleration sensor in the diagnosis target component, and evaluation target point position information, which is position information of the evaluation target point in the diagnosis target component;
4. The fatigue diagnostic device according to claim 3 , wherein the analysis program is created based on the sensor position information and the evaluation point position information. 5. A fatigue diagnosis device as claimed in claim 1, wherein the fatigue determination unit calculates an index value indicating the degree of fatigue of the diagnosed component based on the stress estimation value, and if the index value is greater than or equal to a threshold value, instructs the alarm unit to issue an alarm.