JPH075086A - Method for estimating superposed damage of creep and fatigue of high-temperature structure material - Google Patents

Abstract

PURPOSE:To estimate the life of a part which is subjected to superposed damage non-destructively, quickly, and easily by comparing a standardization parameter which is calculated by combining a plurality of material state quantities for reflecting damage state with that obtained experimentally in advance. CONSTITUTION:An interruption test of a superposed damage is performed in advance in a laboratory to obtain a quantitative index A parameter (number of damage grain boundaries/total number of grain boundaries) of a void for reflecting the damage accumulation state of a creep alone from the material organization of a test piece and the maximum crack length of a fine crack and Vickers hardness reduction quantity ratio, {(amount of softening due to damage)-(amount of softening due to heating)}/(amount of softening due to heating) for reflecting the damage accumulation state of fatigue alone. Then, the standardization parameter for compound evaluation is calculated from the three indexes and a superposed damage master curve is created. Then, the standardization parameter is calculated by measuring the A parameter, the maximum crack length, and hardness of the metal organization of an actual machine material and it is compared with the superposed damage curve, thus estimating the life of the actual machine material.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for diagnosing the remaining life of high temperature parts, which is typified by steam turbine casings, in which creep damage and fatigue damage work in combination.

[0002]

2. Description of the Related Art Steam turbine parts and the like used in a high temperature atmosphere are aged and damaged during operation. Moreover, in many cases, both fatigue damage mainly accumulated at the time of starting and stopping and creep damage accumulated due to continuation of steady operation are accumulated and accumulated. It is known that the lifespan during superposition is generally shorter than that when each damage acts alone, and when the sum of damage accumulation reaches a certain limit value, macroscopic crack initiation is caused and it becomes extremely large. Will cause damage. These damages do not accumulate uniformly on each part of the member, but the damage form and damage ratio differ for each member and site,
Accurately detecting the damage rate for each member is
It is a very important issue in terms of reliability and economy.

In the conventional method, the relationship between the physical quantity such as temperature and stress applied to the material and the amount of damage is experimentally obtained in advance, and the consumption life of the actual material is estimated and calculated from the usage conditions of the actual material. ing. However, with such an estimation method, the error due to the measurement and estimation of temperature and stress is large, and reliable estimation cannot be performed. Further, the stress acting on the actual machine member in use changes due to the stress redistribution and the redistribution of the stress due to the deformation, but it is difficult to make an estimation in consideration of these by the conventional method.

Therefore, in recent years, the amount of damage received by a material is
A simple method for estimating from the non-destructively measured change of material state quantity has come to be applied. Among them, the entity evaluation method for measuring changes in the metallographic structure and its state quantity improved the replica method technology for nondestructively observing the metallographic structure, and improved the accuracy and simplification of image processing technology. As a result, reliability and simplicity have improved dramatically. For example, a method using voids or hardness to estimate the amount of creep damage and a microcrack length of several microns to several millimeters to estimate the amount of fatigue damage have been considered. It has been applied to the evaluation. In addition, a method of evaluating the degree of damage using information obtained nondestructively from members using ultrasonic waves, electromagnetic waves, X-rays, etc. has been considered.

[0005]

However, in these methods, the change in the material state quantity that originally depends on both creep damage and fatigue damage is related to only one damage amount, and the influence of the other damage is small. Only limited to cases. In other words, although it is possible to estimate the consumption life and the remaining life of a member that receives only one of the damages, it is insufficient to evaluate the damages in which creep and fatigue are superimposed, which are accumulated in the actual machine member. An object of the present invention is to solve these various problems that have not been solved in the past and to provide a method for nondestructively estimating the consumption life of a portion that is damaged by superposition of creep and fatigue.

[0006]

In order to solve the above problems, the present invention reflects the damage accumulation state of fatigue alone in a method of estimating the consumption life of an actual machine member in which creep and fatigue superimposed damages are accumulated. Measure and measure non-destructive weight values, non-destructive weight values that reflect the damage accumulation state of creep alone, and non-destructive weight values that reflect the metallurgical state caused by damage accumulation of fatigue and creep. The value of the standardization parameter is calculated from the value, and the consumption life of the actual machine member is obtained from the value.

In high temperature structural materials, fatigue damage causes the generation of microscopic cracks, and creep damage promotes the generation of creep voids, but sometimes both are observed when superposed damage accumulates, and either one of them is observed. In some cases, it is only observed. Therefore, although the evaluation of a single parameter is not sufficient, it is possible to estimate the consumption life by performing a composite evaluation of a plurality of indexes using the standardized parameter of the present invention. Further, in the present invention, the consumption life of superposed damage can be obtained by a non-destructive method centered on the replica method, and the consumption life estimation that had to rely on conventional calculation can be performed only by the non-destructive method by actual measurement. It became possible to easily and accurately estimate in a short time.

[0008]

EXAMPLES The present invention will be specifically described below based on examples. In the example, a case will be described in which the consumption life of a steam turbine high-pressure casing (CrMoV cast steel) material, in which many forms of superimposed damage of creep and fatigue have been recognized in the past, is estimated.

FIG. 1 is a conceptual diagram showing the content of the present invention. One of the non-destructive measurement values that reflects the damage accumulation state of creep alone from the material structure of the interrupted test piece (2a) by conducting the interrupted test of superposed damage in advance in the laboratory (1). The quantification index maximum parameter of the microscopic crack, which is one of the non-destructive measurement values that reflect the damage accumulation state of fatigue alone and the quantification index A parameter of voids (number of damaged grain boundaries / total number of grain boundaries) Vickers hardness reduction ratio, which is one of the nondestructive measurement values of the metallographic state, is calculated in parallel with the crack length {(softening amount due to damage-softening amount due to heating) /
The amount of softening by heating} is calculated (3a). A standardization parameter, which is a composite evaluation parameter, is calculated from these three indexes (4a). The calculated standardized parameter value is plotted on the vertical axis, and the consumption life of superposed damage given in the interruption test is plotted on the horizontal axis, which is plotted in a graph to show the damage master curve (5).
Is created.

After obtaining the superposed damage master curve in the laboratory by the above-mentioned procedure, the metal structure of the actual material was observed (2
b), the A parameter and the maximum crack length are obtained, the hardness is simultaneously measured (3b), and the standardized parameter value is calculated from these measured values (4b). Note that the metal structure is sampled using a replica film in the same manner as the data measurement procedure for creating the superposed damage master curve, and the hardness is measured by using an echo chip hardness meter or the like to measure from the actual material. Is completely non-destructive. By comparing the normalized parameter value data calculated in the above procedure with the superposition damage master curve, the consumption life of the actual material is estimated.

FIG. 2 shows a procedure for creating a superposed damage master curve. In this example, as a superimposed damage imparting test for creating a master curve, a uniaxial holding cycle fatigue test is performed under axial strain control. Here, the holding cycle fatigue means a tensile side holding fatigue test, and when the tensile strain in each cycle of the fatigue test reaches a peak, the cycle of holding the state for a certain period of time is repeated. Further, the holding time is changed in several types in order to simulate all the operation patterns of the actual machine.

First, the test conditions are determined, the holding cycle fatigue at each set holding time is carried out, and the number of repeated breaks is obtained. Next, under the same conditions as those given above, the test material was damaged, and 20%, 40%, 60% of the number of repeated ruptures under each condition.
And make a material that has been given an 80% repeat cycle (1). Here, the ratio of the number of repetitions given to the test piece and the number of repetitions of fracture at the same holding time is defined as the consumption life.

After the surface of each damage imparting material is mirror-polished and the finished surface is etched, the metal structure is transferred to a replica film to collect the metal structure (2). The replica film is photographed with a microscope, the total number of crystal grain boundaries and the number of crystal grain boundaries in which voids are generated are measured, and the A parameter, which is the ratio, is determined (3). Similarly, microscopic cracks are selected from the photographed images, and the length of each microscopic crack is measured, and the maximum length of the microscopic crack is set as the maximum crack length (3). At this time, it is ideal to measure the microscopic crack length of the entire effective part of the test piece, but if it is not possible, statistically processing based on the observed area is performed to obtain the predicted maximum crack length data. And Furthermore, the Vickers hardness of the effective part of the test piece that was damaged by creep and the chuck part of the test piece that was only damaged by heat were measured, and the difference in hardness between the two was divided by the hardness of the heat damaged part Calculate (3). The above-mentioned A parameter and maximum crack length are measured by
Although the test piece may be directly observed and implemented, in the present example, the replica method is adopted in order to obtain consistency with the actual material data, considering that the data collection from the actual material is a replica film.

Based on the data obtained by the above procedure, the value of the standardization parameter is calculated (4), and the data of the unused material and the fractured material are also added, and the horizontal axis represents the consumption life and the vertical axis represents the standardization. Plotting the parameters gives the damage master curve (5). It should be noted that, from the damage imparting material measurement results at various holding times, there is no significant difference between the consumption life and the standardized parameters, and one overlapping damage master curve is sufficient regardless of the length of the holding time. It turns out to be.

FIG. 3 shows the procedure for calculating the standardized parameters in this embodiment. First, three types of individual damage master curves are created (4-1). Here, the single damage means both creep single damage and fatigue single damage. First, after performing a creep independent test to determine the rupture life, various damage ratios of creep damage were applied to the test pieces under the same conditions, and the A parameter value was measured from each of the test pieces to determine the single damage. The life ratio is plotted on the horizontal axis and the A parameter value is plotted on the vertical axis, and a single damage life ratio vs. A parameter master curve is created. Secondly, fatigue alone (without holding) fatigue test is performed, and the maximum crack length value is measured from each of the test pieces,
The individual damage life ratio is plotted on the horizontal axis, and the maximum crack length of microscopic cracks is plotted on the vertical axis to create a single damage life ratio vs. maximum crack length master curve. Thirdly, measure the Vickers hardness reduction ratio from the test piece with creep single damage and fatigue single damage, and take the single damage life ratio on the horizontal axis and the Vickers hardness reduction ratio on the vertical axis. A master curve is created for the damage life ratio vs. Vickers hardness reduction ratio.

Next, the A parameter, the maximum crack length, and the Vickers hardness reduction ratio are measured from the test material (4).
-2). Using these measured values, the individual damage life ratio is calculated back from the previously determined three types of individual damage master curves. In this example, if the single damage life ratio obtained from the A parameter is α, the single damage life ratio obtained from the maximum crack length is β, and the single damage life ratio obtained from the Vickers hardness decrease ratio is γ, The conversion parameter D is calculated by the following equation (a) (4-4). D = Lα + Mβ + Nγ ……………… (a) (L, M, N: constant)

In the unused material, the value of the standardization parameter is
While the value is almost 0, in the superposed damage imparting material, the D value increases as the damage accumulates. The constants L, M, and N in the equation (a) are the intergranular damage, the intragranular damage, and the decrease in hardness when the total damage of the test material is 100, rather than the fractured material of the superimposed damage The degree of dominance of the three indicators of damage due to is calculated and determined from the degree of dominance (4-3).

FIG. 4 is a graph showing the proportion of grain boundary damage, intragranular damage, and damage due to hardness reduction in the total damage, which is 100. This ratio is obtained by measuring 3 indices from the fractured material with superposed damage and comparing each with 3 indices at the time of fracture with independent damage. For example, for a material in which intergranular damage is predominant among intergranular damage, intragranular damage, and damage due to hardness reduction, L is M,
It is a higher value than N.

Further, in the present embodiment, the linear expression with α, β and γ as variables is used as the estimation expression of the standardized parameter D, but by using the multi-dimensional approximation or the like from the superposed damage observation example. , The function f shown in the equation (b) can also be adopted. D = f (α, β, γ) (b) By using this function f, it is possible to more accurately estimate the consumption life of the superposed damage.

FIG. 5 is an explanatory diagram based on a microstructure observation photograph of an unused material of the present test material. The steam turbine high pressure casing (CrMoV cast steel) material used in this example has a structure in which crystal grain boundaries are clearly observed. When creep-only damage is applied to this material, the creep damage appears at the grain boundaries, so that the structure has microvoids at the crystal grain boundaries as shown in Fig. 6, and the voids are connected and grown. It becomes a crack and eventually breaks. Therefore, in the case of only creep damage, A is the evaluation index of grain boundary damage.
Can be organized by parameters.

On the other hand, when fatigue alone (fatigue without holding) damage is applied, the damage mainly appears in the crystal grains, so as shown in FIG. The structure becomes cracked, and the crack propagates and coalesces to fracture. Therefore, in the case of fatigue-only damage, the maximum crack length, which is an evaluation index for intragranular damage, can be arranged.

In the superimposed damage imparting of creep and fatigue, there are cases where both voids and microscopic cracks occur and cases where only one of them occurs due to the length of the holding time.
The consumption life of superposed damage cannot be evaluated well with one parameter. Also, in the case of either single damage, the hardness tends to decrease uniformly as the damage progresses, but since the decreasing tendency is slightly different depending on the damage form,
The hardness reduction ratio cannot be used to unequivocally estimate the consumption life of superposed damaged materials.

However, the above-mentioned standardized parameter D value is
As it is an index that combines and evaluates three indicators, it increases both when either one of voids and microscopic cracks occurs, or when both occur, and damage mechanism mode changes due to the length of holding time of superimposed damage. Irrespective of the above, it increases monotonically as the life consumption progresses. In addition, if constants L, M, and N are used and weights peculiar to the respective materials are applied to the three indices, the indices with large changes can be effectively evaluated.

When the standardized parameter data collected by the above procedure are arranged with the horizontal axis representing the consumption life and the vertical axis representing the standardized parameters, the superposed damage master curve (5) shown in FIGS. 1 and 2 is obtained. To be

FIG. 8 is a plot of the results of estimating the consumption life of the same material cut from an actual machine member from the superposed damage master curve (5) and the actual life obtained by the material test. In this example, it was found that the grain boundary damage, the intragranular damage, and the decrease in hardness all appeared, and the control rates thereof were almost the same. Therefore, the calculation formula for the normalized parameter D value was as follows. The formula (c) shown is adopted. D = α + β + γ (c) As is apparent from FIG. 8, according to the method of the present invention, it is possible to accurately estimate the consumption life of the superimposed damage of creep and fatigue.

[0026]

According to the present invention, it is possible to accurately and non-destructively estimate the consumption life of damage in which creep and fatigue are superposed, which conventionally had to rely on calculation.

[Brief description of drawings]

FIG. 1 is a conceptual diagram showing the content of an embodiment of the invention.

FIG. 2 is a procedure for creating a superposed damage master curve in an example.

FIG. 3 is a procedure for calculating a standardized parameter according to an embodiment.

FIG. 4 shows the proportion of the intergranular damage, the intragranular damage, and the damage due to the decrease in hardness that occupy the total damage.

FIG. 5 is an observation photograph of a metal structure of an unused material in an example.

FIG. 6 is a photograph of the metallographic structure of the creep-only damage imparting material in the example.

FIG. 7 is a photograph of the metallographic structure observation of the material with single fatigue damage imparted in the example.

FIG. 8 is a diagram showing the relationship between the service life of the actual machine member and the service life estimated from the standardized parameters in the example.

Claims (3)

[Claims] 1. A non-destructive method for measuring a plurality of material state quantities reflecting a damage state in a portion where both creep and fatigue of a structural part made of a metal material used at high temperature are accumulated and accumulated, A standardized parameter representative of superposed damage is calculated by combining them, and the superposed damage is estimated in the light of the relationship between the standardized parameter and the superposed damage that have been experimentally obtained in advance. Method for estimating superimposed damage of creep and fatigue. 2. A plurality of material state quantities are a non-destructive metric value reflecting a damage accumulation state of fatigue alone, a non-destructive metric value reflecting a damage accumulation state of creep alone, and damages of both fatigue and creep. The method for estimating the superimposed damage of creep and fatigue of a high temperature structural material according to claim 1, which is a non-destructive metric value that reflects a metallic structure state caused by accumulation. 3. The superposition of creep and fatigue of the high temperature structural material according to claim 1, wherein the standardized parameter is the sum of the respective standard values of the respective nondestructive measured values according to claim 2. Damage estimation method.