Pipe life prediction method

Abstract

To predict the life of a pipe earlier.SOLUTION: A life prediction method of a pipe formed of stainless steel through which high-temperature fluid flows, includes the steps of: polishing a surface of the pipe; collecting a replica sample from the surface of the pipe after polishing; observing the replica sample with a scanning electron microscope or a laser microscope to obtain a generation amount of minute voids in the replica sample; and predicting the service life of the pipe based on the generation amount of minute voids.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] This disclosure relates to a method for predicting the life of a pipe.

Background Art

[0002] Pipes through which a high-temperature fluid flows in a boiler or the like are affected by the pressure and heat of the high-temperature fluid. Therefore, in order to perform maintenance on the pipe at an appropriate timing, it has been carried out to predict the pipe life until creep damage or the like occurs in the pipe and the pipe becomes unusable. Various non-destructive inspection methods may be used for predicting the pipe life.

[0003] As one of such non-destructive inspection methods, for example, Patent Document 1 discloses a configuration in which a flaw detection inspection such as ultrasonic flaw detection is performed on a target part of a pipe.

[0004] Another non-destructive inspection method is a method of measuring the outer diameter of a pipe. In this method, paying attention to the fact that a pipe used for a long time deforms so as to bulge due to the influence of pressure and heat, the bulge on the surface of the pipe is grasped by measuring the outer diameter of the pipe, and the life of the pipe is estimated.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, the outer diameter of a pipe using a material with high strength at high temperature changes measurably from the latter stage to the final stage of the pipe life. Therefore, with the method of measuring the outer diameter of a pipe, it is difficult to grasp the pipe life earlier.

[0007] This disclosure was made to solve the above-mentioned problems and aims to provide a pipe life prediction method that can predict the lifespan of pipes at an earlier stage. [Means for solving the problem]

[0008] To solve the above problems, the pipe life prediction method according to the present disclosure is a pipe made of stainless steel through which a high-temperature fluid flows, and includes the steps of: polishing the surface of the pipe; taking a replica sample from the surface of the pipe after polishing; observing the replica sample with a scanning electron microscope or laser microscope and obtaining the amount of microvoids generated in the replica sample; and predicting the service life of the pipe based on the amount of microvoids generated. [Effects of the Invention]

[0009] According to the pipe life prediction method of this disclosure, the lifespan of the pipes can be predicted at an earlier stage. [Brief explanation of the drawing]

[0010] [Figure 1] This is a cross-sectional view showing an example of a pipe whose lifespan is to be predicted according to the pipe lifespan prediction method according to the embodiment of this disclosure. [Figure 2] This figure shows the state of occurrence of microvoids, voids, and microcracks in the above-mentioned piping. [Figure 3] This figure shows the correlation between the size of voids formed in steam pipes with different lifetime ratios and the void number density. [Figure 4] This flowchart shows the procedure for a method of predicting the lifespan of piping according to an embodiment of this disclosure. [Figure 5] This is an example of an image obtained by observing a replica sample with a scanning microscope. [Figure 6] This is an example of an observation image of the surface of an actual pipe, obtained using a scanning microscope. [Figure 7] This shows an example of specific conditions for the test specimen when creating master data using the above-mentioned pipe life prediction method. [Modes for carrying out the invention]

[0011] The following describes embodiments for implementing the piping life prediction method according to this disclosure, with reference to the attached drawings. However, this disclosure is not limited to these embodiments.

[0012] (Piping) The pipe life prediction method in this embodiment predicts the life of a pipe through which a high-temperature fluid flows. As shown in Figure 1, the pipe 1 whose life is to be predicted has an annular cross-section and extends continuously in a cylindrical shape in the direction of extension. The pipe 1 is used in equipment through which high-temperature steam, such as boilers, flows as the high-temperature fluid. The pipe 1 is used, for example, in a power plant. The pipe 1 is made of stainless steel that is resistant to corrosion even when exposed to high-temperature steam. The pipe 1 is, for example, a pipe material made of high-strength stainless steel with an outer diameter of about 30 mm to 40 mm and a wall thickness of about 3 mm to 4 mm. The pipe 1 is affected by the heat and pressure of the high-temperature fluid flowing inside it.

[0013] The inventors have found that in a stainless steel pipe 1 through which such a high-temperature fluid flows, minute voids 101 occur on the surface of the pipe 1 much earlier than the time when creep damage or the like occurs in the pipe 1. Here, a minute void is an extremely small defect of 1 nm to 1 μm that occurs from the middle of the pipe's lifespan onward. Hereinafter, defects of this size will be referred to as "minute voids".

[0014] Specifically, as shown in Figure 2, according to the findings obtained by the inventors, if the period from the start of use of pipe 1 until creep damage occurs and pipe 1 reaches the end of its lifespan is considered as 100%, then no voids occur in pipe 1 until the lifespan ratio reaches 50%. After that, as time passes and the lifespan ratio reaches about 60%, minute voids 101 begin to occur in pipe 1 near the surface 1f facing the outer Dr in the radial direction Dr. As more time passes, at a stage where the lifespan ratio is about 80%, voids 102 larger than minute voids 101 occur, and the area in which voids 102 occur expands from the surface 1f to the inner Dr in the radial direction Dr. As even more time passes, at a stage where the lifespan ratio is about 90%, multiple voids 102 connect to form microcracks 103. In this way, at stages where many voids 102 and microcracks 103 have occurred (for example, at a lifespan ratio of 80% or 90%), the outer diameter of pipe 1 begins to change to a measurable level. On the other hand, at the stage where only minute voids 101 have formed (for example, 60% of the lifetime), the outer diameter of pipe 1 does not change to a measurable extent.

[0015] Conventionally, the size of voids detectable by optical microscopes or ultrasonic testing is, for example, 0.2 μm or larger. As shown in Figure 3, the size of void 102 at the stage where the lifetime ratio is about 80% is about 1 μm to 20 μm, and the occurrence of void 102 can be detected by observation with an optical microscope. In contrast, the size of minute voids 101 that occur at the stage where the lifetime ratio is about 60% is between 1 nm and 1.0 μm, and it is difficult to detect minute voids 101 by observation with an optical microscope. Therefore, conventionally, it was thought that no detectable damage would occur inside pipe 1 until its outer diameter began to change to a measurable level. However, the inventors have found that minute voids 101 occur earlier than the time when voids 102 of a size detectable by observation with an optical microscope or ultrasonic testing occur.

[0016] (Method for predicting the lifespan of piping) In the pipe life prediction method S10 according to the present embodiment shown below, the life of the pipe 1 is predicted by detecting the generation of minute voids 101 that are difficult to detect with an optical microscope. As shown in FIG. 4, the pipe life prediction method S10 includes a step S11 of polishing the pipe, a step S12 of collecting a replica sample, a step S13 of obtaining the generation amount of minute voids, a step S14 of predicting the service life of the pipe, and a step S21 of creating master data.

[0017] In the step S11 of polishing the pipe, the surface 1f of the pipe 1 is polished. The polishing of the surface 1f of the pipe 1 is preferably performed by a method that can suppress the detachment of the deposits deposited on the surface 1f and enable the detection of minute voids 101. For such polishing of the surface 1f of the pipe 1, for example, CMP (Chemical Mechanical Polishing) using colloidal silica or electrolytic polishing using oxalic acid is used.

[0018] In the step S12 of collecting a replica sample, a replica sample is collected from the surface 1f of the polished pipe 1. The replica sample is collected without destroying the polished pipe 1. The replica sample is collected by a known replica method. For example, a liquid resin is applied to the surface 1f of the pipe 1, and by curing the resin, a replica sample on which the surface 1f of the pipe 1 is transferred can be obtained. Also, by heat-press molding a solid polymer on the surface 1f of the pipe 1, a replica sample on which the surface 1f of the pipe 1 is transferred can be obtained. When a replica sample is collected from the surface of the polished pipe 1, if minute voids 101 are generated on the surface 1f of the pipe 1, the minute voids 101 are transferred to the replica sample.

[0019] In step S13 of obtaining the generation amount of minute voids, the replica sample collected in step S12 is observed with a scanning electron microscope (SEM) or a laser microscope, and the generation amount of minute voids in the replica sample is obtained. With a scanning electron microscope or a laser microscope, minute voids 101 of the size as described above can be sufficiently observed. FIG. 5 is an example of an observation image of a replica sample observed with a scanning microscope. FIG. 6 is an example of an observation image of the surface 1f of the actual pipe 1 observed with a scanning microscope for comparison. As shown in FIGS. 5 and 6, by observing the replica sample obtained as described above with a scanning electron microscope or a laser microscope, the generation of minute voids 101 similar to those in the actual pipe 1 can be observed. In step S13, as the generation amount of minute voids 101 in the replica sample, for example, a void number density indicating the number of minute voids 101 per unit area is obtained.

[0020] In step S14 of predicting the service life of the pipe, the service life of the pipe 1 is predicted based on the generation amount of minute voids obtained in step S13. In step S14 in the present embodiment, the service life of the pipe 1 is predicted by comparing the generation amount of minute voids obtained in step S13 with data indicating the correlation between the generation amount of minute voids and the service life of the pipe 1. In step S14, the generation state (generation amount) of minute voids 101 in the actual pipe 1 is compared with master data described later to predict the service life of the actual pipe 1. Specifically, among a plurality of data included in the master data, data in which the generation amount of minute voids 101 obtained in step S13 is closest to the generation state (generation amount) of minute voids 101 in the actual pipe 1 is specified. Then, the service life associated with the specified data is obtained as the service life of the pipe 1.

[0021] When using master data in process S14, the master data is created by performing process S21 prior to process S14, which predicts the service life of pipe 1. In process S21, which creates the master data, the master data used in process S14, which predicts the service life of pipe 1, is created. The master data is created using test specimens that simulate pipe 1, rather than the actual pipe 1. In this process, multiple test specimens that simulate multiple pipes 1 with different usage conditions are used. The test specimens have the same material, diameter, and wall thickness as pipe 1. By performing creep tests on multiple test specimens and observing them, the data necessary for creating the master data is obtained.

[0022] For the test specimen, a creep test is performed by applying predetermined temperature and stress to simulate the flow of a high-temperature fluid. During this test, for example, the elapsed time from the start of the test is varied in multiple stages, and the generation of microvoids 101 on the surface of the specimen is observed at each stage. Furthermore, the temperature and applied stress during the test are changed, and the generation of microvoids 101 on the surface of the specimen is observed at each stage. The test continues until the specimen finally ruptures due to creep damage. In this way, the test is conducted on multiple specimens simulating multiple pipes 1 under different operating conditions.

[0023] Furthermore, it is preferable to conduct tests on the test specimen under conditions more severe than those caused by the high-temperature fluid flowing through the actual pipe 1. This allows for the creation of master data in a shorter time.

[0024] When observing the test specimen, first, the surface of the test specimen is polished using the same polishing method as in the pipe polishing process S11. Then, a replica sample of the surface of the test specimen is obtained in the same manner as in the replica sample collection process S12. Furthermore, in the same manner as in the process S13 for obtaining the amount of microvoids 101 generated, the obtained replica sample of the test specimen is observed with a scanning electron microscope or laser microscope, and the amount of microvoids 101 generated in the replica sample is obtained. In this way, multiple data points are obtained from the surfaces of multiple test specimens that simulate multiple pipes 1 under different usage conditions. In addition, for example, the lifetime ratio (T / Tb) for each test specimen is calculated based on the elapsed time T from the start of the test to the observation and the elapsed time Tb from the start of the test to fracture. The elapsed time T at the time of observation is the timing when the replica sample is obtained, and the timing when the amount of microvoids 101 generated is obtained. Then, a replica sample is obtained from the surface of the test specimen according to the lifetime ratio (T / Tb). This allows data on the time-dependent change in the amount of microvoids 101 generated in the test specimen to be obtained. In other words, the correlation between the time-dependent change in the amount of microvoids 101 generated and the elapsed time (lifetime) until fracture is obtained as master data.

[0025] Figure 7 shows an example of the specific conditions of the test specimen when creating the master data in step S21 described above. As shown in Figure 7, for example, a creep test was performed on the test specimen under conditions of 750°C and 61 MPa. The test was interrupted every 1000 hours from the start of the test, a replica sample was taken from the surface of the test specimen, and the condition of the replica sample (surface condition of the test specimen) was observed using a laser microscope. The test specimen fractured due to creep damage at 11911 hours after the start of the test. Therefore, master data was created by relating the amount of microvoids 101 generated and the lifetime ratio at 7000 hours, 8000 hours, and 9000 hours from the start of the test until fracture.

[0026] Note that Figure 7 is merely one example of specific test specimen conditions when creating master data. The test specimen conditions when creating master data should be appropriately selected according to the type of piping 1 that actually improves lifespan and the environment in which it is used.

[0027] In this way, the amount of microvoids 101 obtained is compared with the master data to predict the service life of pipe 1.

[0028] (Effects and Benefits) In the pipe life prediction method S10 with the above configuration, the surface 1f of pipe 1 is polished, and a replica sample is taken from the polished surface of pipe 1. If minute voids 101 are present on the surface 1f of pipe 1, the minute voids 101 are transferred to the replica sample. By observing the replica sample on which the minute voids 101 have been transferred using a scanning electron microscope or laser microscope, the amount of minute voids present in the replica sample can be obtained with high accuracy. Based on newly acquired knowledge that there are minute voids 101 that occur earlier than the time when voids grow to a size detectable by optical microscopy or ultrasonic flaw detection, the amount of minute voids 101 is obtained by observing the minute voids 101. As a result, the service life of pipe 1 can be predicted at an earlier stage compared to measuring the outer diameter of pipe 1, observing with an optical microscope, or performing flaw detection using ultrasonic flaw detection. Therefore, the life of pipe 1 can be predicted earlier. This makes it possible to predict the mid-term life of pipe 1, which was previously difficult to evaluate. Furthermore, the ability to predict the lifespan over the medium term will enable the efficient planning of various construction projects, such as replacement work for pipes during inspections.

[0029] Furthermore, in step S14, which predicts the service life of the piping, the service life of piping 1 is predicted based on master data showing the correlation between the amount of microvoids generated and the service life of piping 1, and the actual amount of microvoids generated. Because the master data is created in advance, the correlation between the amount of microvoids 101 generated and the service life of piping 1 can be determined in advance with high accuracy. As a result, the service life of piping 1 can be predicted more easily and accurately.

[0030] Furthermore, replica samples are obtained from the surface of multiple test specimens that simulate multiple pipes 1 under different usage conditions. Then, the amount of microvoids generated in the replica samples of multiple test specimens is obtained to create master data. By creating master data in this way, it is possible to obtain data that is close to the generation state of microvoids 101 in actual pipes 1 under different usage conditions. By predicting the service life of pipe 1 based on the master data obtained in this way, a more accurate prediction of the service life of pipe 1 can be made.

[0031] (Other embodiments) Although embodiments of this disclosure have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments and may include design changes and the like that do not depart from the gist of this disclosure.

[0032] In the above embodiment, the procedure for the life prediction method S10 of the piping 1 was described, but the details of the work content and procedures in each step can be changed as appropriate.

[0033] For example, the method is not limited to predicting the lifespan of pipe 1 using master data that shows the correlation between the amount of microvoids generated and the service life of the pipe. In that case, the process of predicting the service life of the pipe S14 may be performed without performing the process of creating the master data S21. In that case, in the process of predicting the service life of the pipe S14, data on the actual amount of microvoids 101 generated when pipe 1 was measured in the past is stored in storage or the like in advance, and the stored data is used as data that shows the correlation between the amount of microvoids generated and the service life of pipe 1. Therefore, the service life may be predicted by comparing the amount of microvoids 101 generated with the stored data.

[0034] Furthermore, even when performing the process S21 to create master data, if pipe 1 is inspected periodically multiple times, the process S21 to create master data may be performed only once or each time. In other words, the same master data created in advance may be used for each periodic inspection of pipe 1, or new master data may be created each time pipe 1 is inspected.

[0035] Furthermore, the process S21 for creating master data is not limited to conducting tests on multiple test specimens that simulate multiple pipes 1 under different usage conditions. For example, in the process S21 for creating master data, master data may be created by simulation rather than actual testing.

[0036] Furthermore, the high-temperature fluid flowing inside pipe 1 is not limited to steam. It may also be a high-temperature gas or liquid that causes damage such as corrosion to pipe 1, which is made of stainless steel.

[0037] Furthermore, the polishing method in step S11, which involves polishing the piping, is not limited to CMP using colloidal silica or electrolytic polishing using oxalic acid. It is preferable to apply the polishing method that is most suitable for the pipe 1 being predicted by first trying different types of polishing agents and methods on the pipe 1 or test specimen.

[0038] <Note> The life prediction method S10 for the piping 1 described in the embodiment can be understood, for example, as follows.

[0039] (1) The first embodiment of the method for predicting the lifespan of a pipe 1 is a method for predicting the lifespan of a pipe 1 made of stainless steel through which a high-temperature fluid flows, and includes the steps of: polishing the surface 1f of the pipe 1 in step S11; taking a replica sample from the surface 1f of the pipe 1 after polishing in step S12; observing the replica sample with a scanning electron microscope or laser microscope and obtaining the amount of microvoids generated in the replica sample in step S13; and predicting the service life of the pipe 1 based on the amount of microvoids generated in step S14.

[0040] The inventors have found that, when a high-temperature fluid flows through a stainless steel pipe 1, over time, due to the influence of the temperature and pressure of the high-temperature fluid, minute voids 101 are generated near the surface 1f in the radial direction Dr of the pipe 1. These minute voids 101 begin to form earlier than the time when the outer diameter of the pipe 1 undergoes a measurable change. Furthermore, the minute voids 101 are much smaller than voids that can be detected by observation with an optical microscope or by ultrasonic testing. In other words, the minute voids 101 form earlier than the time when voids grow to a size that can be detected by observation with an optical microscope or by ultrasonic testing.

[0041] In this pipe life prediction method S10, the amount of minute voids generated in the replica sample is obtained based on the newly acquired knowledge described above. This allows for an earlier prediction of the service life of pipe 1 compared to measuring the outer diameter of pipe 1, observing with an optical microscope, or performing flaw detection using ultrasonic testing. Therefore, the life of pipe 1 can be predicted at an earlier stage.

[0042] (2) The second embodiment of the pipe life prediction method S10 is the pipe life prediction method S10 of (1), further comprising a step S21 for creating master data in advance, which is a step of creating master data that shows a correlation between the amount of minute voids generated and the service life of the pipe 1, wherein in a step S14 for predicting the service life of the pipe 1, the service life of the pipe 1 is predicted based on the master data and the amount of minute voids generated.

[0043] Because master data is created in advance, the correlation between the amount of micro-voids 101 generated and the service life of pipe 1 can be determined with high accuracy beforehand. As a result, the service life of pipe 1 can be predicted more easily and accurately.

[0044] (3) The third embodiment of the pipe life prediction method S10 is the pipe life prediction method S10 of (2), wherein in the step S21 of creating the master data, replica samples are obtained from the surface of a plurality of test specimens that mimic a plurality of pipes 1 with different usage conditions, the amount of minute voids generated in the replica samples of the plurality of test specimens is obtained, and the master data is created.

[0045] By creating master data in this way, it is possible to obtain data that closely resembles the occurrence of minute voids 101 in actual pipes 1 under different usage conditions. By predicting the service life of pipe 1 based on the master data obtained in this way, a more accurate prediction of the service life of pipe 1 can be made. [Explanation of Symbols]

[0046] 1…Piping 1f…Surface 101... minute void 102...Void 103... Microcrack Dr…Radial direction Dri…inside Dro... outside S10... Method for predicting the lifespan of piping S11... Process of polishing pipes S12…Process of taking replica samples S13... Process for obtaining the amount of minute voids generated. S14…Process for predicting the service life of piping. S21... The process of creating master data. T...Time elapsed during observation Tb…Time elapsed from the start of the test until fracture

Claims

1. A method for predicting the life of a pipe made of stainless steel through which a high-temperature fluid flows, comprising the steps of: polishing a surface of the piping before an outer diameter of the piping begins to measurably change; taking a replica sample from the surface of the pipe after polishing; Observing the replica sample with a scanning electron microscope or a laser microscope to obtain the amount of microvoids, which are extremely small defects of 1 nm to 1 μm, generated in the replica sample; and predicting the service life of the pipe based on the amount of microvoids generated.

2. The method further includes a step of creating master data in advance, the master data indicating a correlation between the amount of microvoids generated and the service life of the pipe, 2. The method for predicting a service life of a pipe according to claim 1, wherein in the step of predicting a service life of the pipe, the service life of the pipe is predicted based on the master data and the amount of microvoids generated.

3. A method for predicting the life of a pipe as described in claim 2, wherein in the process of creating the master data, replica samples are obtained from a plurality of test bodies simulating a plurality of the pipes with different usage conditions, the amount of microvoids generated in the replica samples of the plurality of test bodies is obtained, and the master data is created.

Images