JP2009174859A - Remaining lifetime evaluation method of machine part - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem, wherein clear remaining lifetime of a machine part, for which metal fatigue cracks are discovered, cannot be evaluated by conventional future life evaluation techniques and increase in the inspection number of times (increase in inspection cost), or the like, of the machine part is brought about. <P>SOLUTION: A stress concentration region of the machine part is selected by the analysis of stress under restriction and loading condition which is close to actual phenomenon for keeping a master curve, which is a relational curve of the maximum crack dimension and a remaining lifetime consumption ratio, preliminarily calculated by the analysis of the crack propagation from the stress concentration region which is a crack occurrence start point. The maximum dimension measured values of the cracks detected by the stereoscopic inspection of the machine part are adapted to the master curve, and the remaining lifetime of the machine part is calculated. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a remaining life evaluation method for a machine part, and more particularly, to a remaining life evaluation method for a machine part for evaluating the remaining life due to metal fatigue of the machine part.

In recent years, many destruction accidents due to metal fatigue due to repeated stress of machine parts have occurred. Under such circumstances, there is a need for a technique for diagnosing the remaining life of machine parts and taking appropriate measures. As a remaining life evaluation technique, a technique proposed in Patent Document 1 or the like has been known. However, these are methods for low-cycle thermal stress for gas turbines with large temperature changes, and are not methods for repeated stress due to load, and materials for creating a master curve for calculating fatigue life. Since a method of complementing the one obtained by the test is used, it cannot be applied to mechanical parts that are strongly influenced by geometric stress concentration or mechanical parts that are in a complicated stress state.
JP-A-10-160646

  On the other hand, in factories that have mechanical parts that are strongly influenced by geometric stress concentration or that have complex stress conditions, if cracks are found by inspection, life-prolonging measures are taken to remove the cracks by, for example, grooming. Often taken. In this method, it may be difficult to completely remove the crack, and it may become a new defect depending on the method of removing the crack by grinder care or the like. There is a possibility of shortening the fatigue life by decreasing the thickness and increasing the working stress. In addition, since the remaining life after removing the crack by grinder care or the like is not clear, it is necessary to frequently inspect, and the inspection cost may increase.

  In other words, with the conventional remaining life evaluation technology, there is a problem that a clear remaining life evaluation cannot be performed on a machine part in which a metal fatigue crack is found, resulting in an increase in the number of inspections of the machine part (increase in inspection cost). there were.

The inventors studied a method that can clearly evaluate the remaining life against metal fatigue due to repeated stress generated in mechanical parts. As a result, the stress concentration part of the target metal mechanical part was selected by the stress analysis under the constraint and loading conditions close to the actual phenomenon, and the crack propagation analysis with this selected part as the starting point of crack initiation By calculating a master curve that is a relationship curve between crack dimensions and life consumption rate, and calculating the remaining life by applying the maximum dimension measurement value of the crack detected by inspecting the mechanical part to this master curve, The inventors have conceived that the above problems can be solved, and have made the present invention. That is, the present invention is as follows.
[Claim 1] A stress concentration part of a target metal mechanical part is selected by a stress analysis under constraints and loading conditions close to actual phenomena, and a crack propagation analysis using the selected part as a starting point of crack generation. The master curve, which is the relationship curve between the maximum crack size and the life consumption rate, is calculated, and the remaining life is calculated by applying the maximum dimension measurement value of the crack detected in the physical inspection of the machine part to this master curve. A remaining life evaluation method for mechanical parts, characterized in that:
[Claim 2] The method for evaluating the remaining life of a machine part according to claim 1, wherein the substance inspection is performed by nondestructive inspection.
[Claim 3] The non-destructive inspection includes VT (visual observation), replica method, MT (magnetic particle inspection test), UT (ultrasonic inspection), PT (penetration inspection), and RT (radiation inspection). 3. The method for evaluating the remaining life of a machine part according to claim 2, wherein any one or more of them are used.

  According to the present invention, it is possible to clearly evaluate the remaining life against mechanical fatigue of mechanical parts that are strongly influenced by geometric stress concentration and mechanical parts that are in a complicated stress state, and therefore, the found cracks can be removed by grooming or the like. This makes it possible to accurately determine whether or not to perform the process, avoid unnecessary crack removal measures that may lead to shortening of the service life, and enable stable operation of the entire machine. Even when cracks are removed, the remaining life can be clearly calculated by the same stress analysis and crack propagation analysis using the machine part shape after crack removal as the initial condition, and the number of inspections can be minimized based on this. Can reduce the inspection cost.

The best mode for carrying out the present invention will be described below with reference to the drawings.
FIG. 1 is a flowchart showing an outline of the present invention.
First, step 1 (selection of stress concentration part) is executed. In this step, FEM (Finite Element Method) analysis is performed on the target mechanical parts using the constraints and loading conditions that can be considered to be close to real phenomena as the boundary conditions, and the maximum generated stress and its position are determined. calculate. At this time, attention should be paid so that the mesh size in the vicinity of the stress concentration portion has a size and a shape with sufficient accuracy.

Next, step 2 (maximum crack dimension measurement) and step 3 (master curve determination) are executed. Either of these steps 2 and 3 may be preceded, or these may be performed in parallel.
In step 2, in accordance with the periodic repair of the machine part, in particular, the mechanical part is inspected mainly about the stress concentration part calculated by the FEM analysis to detect a crack, and the maximum crack size is measured. The physical inspection for detecting cracks is preferably performed by non-destructive inspection because machine parts after remaining life evaluation may be used as they are. However, for mechanical parts that are determined to be replaced with new ones, a sample may be cut out and cracks may be measured.

  Nondestructive inspection methods include VT (visual observation), replica method, MT (magnetic particle inspection test), UT (ultrasonic inspection), PT (penetration inspection), RT (radiation inspection), etc. Any of these can be preferably used. Any one of these methods may be used, or two or more methods may be used in combination. The replica method is a method of observing the surface state by attaching a replica film to a test surface. As an example of a specific replica method, the test surface is mirror-polished, then acetone is dropped, a polymer replica film is attached, the acetone is volatilized, the replica film is peeled off, gold deposition is performed, and a laser microscope is used. It can be used to observe and measure microcracks.

  In step 3, a master curve is determined (calculated). The master curve is defined as a graph showing the relationship between the maximum crack size (taken on the vertical axis) and the life consumption rate (taken on the horizontal axis) generated in the target member (machine part). In order to calculate this graph, the relationship between the crack size and the lifetime consumption rate is calculated using the Paris law and crack growth analysis of the steel used for the member. The life consumption rate is calculated from the ratio between the number of breaks (see FIG. 6) and the current number. The number of fractures is determined from crack growth analysis using the shape of the structure and the physical constants of the material.

  In step 4, the remaining life is calculated from the value of the maximum crack size measured in step 2 and the master curve calculated in step 3.

Hereinafter, the present invention will be described in more detail with reference to examples. In the embodiment, the remaining life is evaluated for an end yoke that is a member of a steel mill.
(Step 1) FEM analysis is performed under constraints and loading conditions that are considered to be close to the actual use state of the end yoke, the stress distribution is obtained, and a stress concentration site is selected. FIG. 2 shows the stress distribution of the end yoke 5 obtained by this FEM analysis with iso-stress lines. In the case of this example, the multi-nested iso-stress line has a higher stress in the inner part of the nest. It can be seen that the generation position 6 of the maximum generated stress is located on the end side of the R corner.

  (Step 2) Substantive inspection (crack detection and crack dimension measurement) is performed centering on the site 6 where the maximum stress is generated. In this embodiment, since the stress concentration position is the surface, the crack is detected by MT with high crack detection accuracy. The maximum value of the detected crack lengths of the detected cracks is adopted as the maximum crack size. In this example, the maximum crack size measured was 3.2 mm in crack length.

  In order to conduct a more accurate investigation, it is preferable to use a replica method. Further, when the crack depth is known by the TOFD (Time of Flight Diffraction) method or the like, it is preferable to employ the deepest value among the measured crack depths as the maximum crack size. The TOFD method is a kind of UT method, and a pair of transmission and reception probes are arranged opposite to each other, and ultrasonic waves are transmitted and received so as to transmit the cross section to be inspected. Using the difference in propagation time between the surface transmitted wave (lateral wave) and the bottom reflected wave and the diffracted wave generated at the upper and lower ends of the flaw, geometrically the position of the flaw (flaw height, flaw indication length) This is a simple and highly accurate flaw detection, and in particular, can measure the dimension in the thickness direction with high precision (URL of Ishikawajima Inspection and Measurement Co., Ltd. [http: / /www.iic-hq.co.jp/work_contents/01inspection_measurement/01nondestructive_testing/KH-01.html]).

(Step 3) For example, a crack growth analysis is performed along a flow as shown in FIG. This flow will be described.
First, in step 31, an analysis model for crack growth analysis is created. This analysis model can be created based on an analysis program for crack growth analysis. Various general-purpose programs are sold as crack growth analysis programs. For example, there is an analysis method based on a boundary element method called BEASY of Computational Mechanics BEASY, and a crack growth analysis based on a finite element method called ZENCRACK of Zentech. Furthermore, in recent years, the X-FEM (extended Finite Element Method, reference material “Belytschko T., Black T.,“ Elastic Crack Growth in Finite Elements with Minimal Remeshing ”Int. J. Number. Meth. Engng. 45 (5), 601-620 (1999) ") and other analytical methods are also in practical use, and it has become possible to significantly reduce labor. Any model can be used as long as it can be accurately modeled and has equivalent or better performance.

  Next, in step 32, the initial crack is determined. Here, the assumed initial crack is given to the stress concentration site (R corner) obtained previously. The initial crack length is a value that takes into account the calculation accuracy such as the minimum crack length and mesh size in the experimental results of the Paris rule. Note that the Paris rule is a CT test piece as shown in FIG. 4 (from “Material Collection of Metallic Fatigue Crack Propagation Resistance” edited by Japan Society for Materials Science (1983, published by Shinchon Nishimura Co., Ltd.)). This is a curve relation law found by log-log plotting data of crack propagation rate da / dN and stress intensity factor ΔK obtained by the fatigue test used.

Next, in step 33, structural analysis calculation is performed. Here, the analysis is executed under the given analysis model, boundary condition, and crack condition, and the entire structural deformation state is calculated. At this time, the general crack portion (the portion other than the tip of the crack) has a structure that does not resist the load, or when the influence of contact, friction, etc. is large, the conditions may be taken into consideration.
Next, in step 34, the K value of the crack tip is determined. Since the crack tip is a singular field, calculation based on fracture mechanics is required to calculate deformation, stress, strain, and the like. Specifically, J integration is performed around the crack tip, and the K value is calculated from the obtained J integration value.

Next, in step 35, the crack position of the next step is calculated. The maximum circumferential stress criterion was used to determine the propagation direction angle when the crack propagates. This is based on the hypothesis that the crack of the next step progresses in the direction in which the stress is maximum in the circumferential direction surrounding the crack tip. For 2-dimensional analysis, K _I values obtained at each calculation step, the following equation using the K _II value (Equation 1)

Thus, a progress direction angle (a deviation angle from the progress direction in the previous calculation step) θ is obtained. The resulting crack progress of the crack is calculated from the stress intensity factor K _I value in a direction perpendicular to the extending direction (using Paris law here), to determine a new crack tip position.
In step 36, it is determined whether or not the new crack tip position exists in the mesh. If it exists, a crack including the new crack tip position is given to the analysis model, and step 33 and subsequent steps are repeated. If it does not exist, the analysis is terminated.

According to the analysis result in the present embodiment, as shown in FIG. 5, the crack 7 of the end yoke 5 propagates from the starting point C _{0 at} the R corner to the end point C ₁ on the outer surface side by propagating along the path as shown in the figure. The end yoke 5 is broken.
The relationship between the number of times of loading and the crack length as shown in FIG. 6 is obtained by the crack growth analysis described above. Here, the term “crack length” is used synonymously with “crack dimension”, which is a generic term for “crack length” in the member surface and “crack depth” in the thickness direction of the member.

The life consumption rate (specifically, fatigue life consumption rate) is calculated from the following break length (or the number of breaks) and the number of times of loading.
a) If data on the fatigue crack fracture surface of an actual rupture accident is obtained, the crack length or crack depth at the time of rupture shall be the rupture length (corresponding to the rupture length a2 in FIG. 6) The number of times the load is received during the service period is defined as the number of breaks (corresponding to the number of breaks N2 in FIG. 6).
b) In the crack growth analysis, the crack length that leads to a brittle fracture due to abrupt crack growth is the fracture length (corresponding to the fracture length a1 in FIG. 6), and the number of times is the number of fractures (corresponding to the number of fractures N1 in FIG. 6). . The lifetime consumption rate is calculated from (current number of times / number of breaks) as described above. “Current count” is the current value of the number of repetitions of step 33, and the crack length is calculated for each current value.

  As for the calculated number of breaks, it is desirable to use a value that takes into account a safety factor in consideration of an unexpectedly excessive load or material variation. Here, the crack length is generally considered as the crack length, but the master curve is usually calculated by the crack depth in many cases. However, the depth and length of the crack can be related to each other depending on the applied load and the crack propagation state. Therefore, by obtaining the relational expression of the depth and length at the assumed working load in advance, if one of the values is available, the other can be calculated. Either depth or length may be used.

From these, plot points of the relationship between the life consumption rate and the crack length can be obtained, and a master curve is assumed assuming an approximate function (approximate curve) for these values. The approximate curve is created by an appropriate method according to the distribution of plot points and the concept of safety factor (design concept).
FIG. 7 shows plot points obtained by the analysis in this example and some approximate curves assumed for the plot points. The parameter on the vertical axis is the crack length. Curves M1, M3, and M4 pass through approximately the center, approximately the upper limit, and approximately the lower limit of the distribution range estimated from the dispersion of the plot points of the crack length against the life consumption rate, respectively, and the curve M2 represents the plot points. Are connected smoothly. Which one is selected depends on the design concept and is not particularly limited. In this embodiment, the curve M1 is selected.

(Step 4) The relationship between the load of the end yoke 5 and the number of times of loading (and time) is calculated from the actual measurement value or the usage plan of the apparatus. From these, the life (N _f ) corresponding to the number of breaks is known, and the remaining life can be calculated from the life consumption rate. That is, as shown in FIG. 8, in the case of the present embodiment, the maximum crack length of 3.2 mm was detected in step 2 as described above. The value on the horizontal axis is calculated as the current life consumption rate (Φ _c ). This value is 0.4 as shown. From these values, the remaining life (N _fr ) can be determined by the following formula (Formula 2).

In the present embodiment, the end yoke 5 is used as it is without removing cracks only during a period in which the remaining life N _fr1 determined by substituting Φ _c = 0.4 into the formula 2 is multiplied by a predetermined safety factor. Subsequently, it could be used without breaking. After this service period, as well was to evaluate the remaining life, because Φ _c had become 0.9, was replaced with a new one. A sample was taken from the end yoke 5 that had been used up to the time of replacement, the crack depth was measured, and the master curve was corrected based on the data.

  As described above, according to the present invention, it is possible to avoid unnecessary crack removal measures that may lead to shortening of the life, and stable operation of the entire machine becomes possible.

It is a flowchart which shows the outline | summary of this invention. It is a stress distribution figure which shows the result calculated about the end yoke as an example of a stress analysis. It is a flowchart which shows the outline | summary of a crack growth analysis. It is a graph which shows the specific example of a Paris rule. It is a crack propagation path figure which shows the result calculated about the end yoke as an example of a crack propagation analysis. It is explanatory drawing which shows the example of the determination method of the frequency | count of fracture | rupture. It is explanatory drawing which shows the example of the determination method of a master curve. It is explanatory drawing which shows the example of the determination method of the lifetime consumption rate by a master curve.

Explanation of symbols

1, 2, 3, 4 Step 5 Machine parts (eg end yoke)
6 Maximum stress generation position 7 Crack
31,32,33,34,35,36 steps

Claims (3)

  The stress concentration part of the target metal mechanical part is selected by restraint that is close to the actual phenomenon and stress analysis under loading conditions, and the crack propagation analysis with this selected part as the starting point of crack generation is used to determine the maximum crack size. A master curve that is a relationship curve of the life consumption rate is calculated, and the remaining life is calculated by applying the maximum dimension measurement value of the crack detected by the physical inspection of the machine part to the master curve. A method for evaluating the remaining life of machine parts.   2. The remaining life evaluation method for machine parts according to claim 1, wherein the substance inspection is performed by non-destructive inspection.   The nondestructive inspection is any one of VT (visual observation), replica method, MT (magnetic particle inspection test), UT (ultrasonic inspection), PT (penetration inspection), RT (radiation inspection). The method for evaluating the remaining life of a machine part according to claim 2, wherein two or more are used.

Images