Quality evaluation device, quality evaluation method, and quality evaluation program for rotary machine
The quality evaluation device for rotating machinery addresses the challenge of wear estimation by integrating AE signal data to determine the wear amount, facilitating targeted maintenance and reducing costs and time through accurate wear assessment.
Patent Information
- Application Number
- PCT/JP2024/006901
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-18
- Filing Date
- 2024-02-27
- Publication Date
- 2024-11-21
AI Technical Summary
Current methods for detecting rubbing in rotating machinery, such as steam turbines, can only determine instantaneous rubbing and not the degree of wear caused by accumulated rubbing, leading to costly and time-consuming maintenance as they require frequent inspection and replacement of components.
A quality evaluation device and method that uses AE signals to calculate a rubbing detection index and estimate the wear amount of rotating machinery by integrating this index over time, allowing for the determination of necessary maintenance and targeted part replacement.
Enables accurate estimation of wear on rotating machinery, reducing maintenance costs and time by identifying when and where maintenance is required, thereby improving maintenance efficiency.
Smart Images

Figure JP2024006901_21112024_PF_FP_ABST
Abstract
Description
Quality evaluation device, quality evaluation method, and quality evaluation program for rotating machines
[0001] This application claims priority to Japanese Patent Application No. 2023-082386, filed on May 18, 2023, with the Japan Patent Office, the contents of which are incorporated herein by reference.
[0002] In rotary machines with rotating parts that can rotate relative to stationary parts, such as steam turbines, thermal deformation of the casing during operation can cause the gap between the stationary and rotating parts to shrink, potentially resulting in rubbing, where the rotating body comes into contact with the stationary part. Rubbing causes wear on the components of the rotary machine, increasing shaft vibration and causing performance degradation due to the widening of the gap.
[0003] As a technique for determining whether rubbing has occurred, a method using an AE sensor capable of detecting an AE (Acoustic Emission) signal generated during rubbing is known. For example, Patent Document 1 discloses a method that makes it possible to determine whether rubbing has occurred based on information related to the phase of the AE signal. Patent Document 2 discloses a method that makes it possible to determine not only whether rubbing has occurred but also the location where the rubbing has occurred based on AE signals detected by multiple AE sensors arranged at different positions.
[0004] JP 2021-76533 A JP 2022-151125 A
[0005] The above-mentioned patent documents 1 and 2 are capable of determining the presence or absence of rubbing and the location of rubbing based on AE signals. However, these methods only assess instantaneous rubbing and are unable to evaluate the degree of damage caused by accumulated wear due to rubbing. As described above, rubbing causes wear to the components of a rotating machine, necessitating maintenance such as component replacement depending on the degree of wear. This type of maintenance requires opening the casing to access the interior of the rotating machine, which requires time and cost. Therefore, it is desirable to replace only the necessary parts at the necessary time. Understanding the amount of wear in a rotating machine caused by rubbing provides useful information for identifying the timing and target parts for such maintenance, and is expected to reduce the waiting time and costs associated with arranging parts by preparing replacement parts in advance.
[0006] At least one embodiment of the present disclosure has been made in consideration of the above-mentioned circumstances, and aims to provide a quality evaluation device, a quality evaluation method, and a quality evaluation program for a rotating machine that can appropriately evaluate the quality of a rotating machine by estimating the amount of wear of the rotating machine based on an AE signal.
[0007] In order to solve the above problem, a quality evaluation device for a rotating machine according to at least one embodiment of the present disclosure is a quality evaluation device for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, and includes: an AE signal acquisition unit for acquiring an AE signal from at least one AE sensor provided on the stationary part; a rubbing detection index calculation unit for calculating a rubbing detection index based on the AE signal; and a wear amount estimation unit for estimating the wear amount of the rotating machine by accumulating the rubbing detection index.
[0008] In order to solve the above-mentioned problems, a quality evaluation method for a rotating machine according to at least one embodiment of the present disclosure is a quality evaluation method for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, comprising: a step of acquiring an AE signal from at least one AE sensor provided on the stationary part; a step of calculating a rubbing detection index based on the AE signal; and a step of estimating the amount of wear of the rotating machine by integrating the rubbing detection index.
[0009] In order to solve the above-mentioned problems, a quality evaluation program for a rotating machine according to at least one embodiment of the present disclosure is a quality evaluation program for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, and is capable of executing, on a computer device, the steps of: acquiring an AE signal from at least one AE sensor provided on the stationary part; calculating a rubbing detection index based on the AE signal; and estimating the amount of wear of the rotating machine by accumulating the rubbing detection index.
[0010] According to at least one embodiment of the present disclosure, it is possible to provide a quality evaluation device, a quality evaluation method, and a quality evaluation program for a rotating machine that can appropriately evaluate the quality of the rotating machine by estimating the amount of wear of the rotating machine based on an AE signal.
[0011] 11 is a schematic diagram showing an enlarged peripheral configuration of the low-pressure unit of FIG. 6. FIG. 12 is a diagram showing a signal intensity distribution obtained when an impact is applied to each target portion of FIG. 10. FIG. 13 is a diagram showing an example of a weighting coefficient based on the signal intensity of FIG. 11. FIG. 14 is a diagram showing an example of an estimated value of the amount of wear for each target portion of the rotary machine 1. FIG. 15 is a diagram showing a manner in which a reference value for determining whether maintenance is necessary is determined from existing inspection results. FIG. 16 is a diagram showing an example of correction of the weighting coefficient based on the inspection results.
[0012] Hereinafter, several embodiments of the present disclosure will be described with reference to the accompanying drawings. However, the dimensions, materials, shapes, relative arrangements, etc. of the configurations described as embodiments or shown in the drawings are merely illustrative examples and are not intended to limit the scope of the present disclosure.
[0013] 1 is a cross-sectional structural view of a rotary machine 1 according to one embodiment. The rotary machine 1 includes a stationary part 2 and a rotating part 4 that is rotatable relative to the stationary part 2. In this embodiment, a steam turbine that can drive the rotating part 4 by a working fluid that is steam will be described as an example of the rotary machine 1, but the rotary machine 1 is not limited thereto.
[0014] The rotary machine 1 includes a high-pressure unit HP, an intermediate-pressure unit IP, and a low-pressure unit LP that can be driven using steam as a working fluid. High-pressure steam HS is supplied to the rotary machine 1 from the outside as a working fluid. The high-pressure steam HS is first supplied to the high-pressure unit HP. After performing work in the high-pressure unit HP, the high-pressure steam HS becomes medium-pressure steam IS as its pressure drops and is supplied to the medium-pressure unit IP. The medium-pressure steam IS becomes low-pressure steam LS as its pressure drops and is supplied to the low-pressure unit LP. The low-pressure steam LS becomes exhaust steam ES as its pressure drops and is discharged to the outside.
[0015] The high-pressure unit HP, the intermediate-pressure unit IP, and the low-pressure unit LP all share a rotating unit 4, which is a rotor, and each has a corresponding casing 6a-6c, which is a stationary unit 2. In each unit, a gap D (not shown) is provided between the rotating unit 4 and the casing 6a-6c. (More precisely, the gap D is primarily formed between the innermost peripheries of the stator vanes provided in the casings 6a-6c and the outermost peripheries (tips) of the rotor blades of the rotating unit 4, but it may also extend between the casings 6a-6c, to which the stator vanes are attached, and the rotating unit 4, to which the rotor blades are attached.) As described above, each working fluid is introduced into the gap D, thereby driving the rotating unit 4. During operation of the rotary machine 1, deformation of at least one of the stationary unit 2 and the rotor unit 4 due to heat or other factors may cause the gap D to shrink, resulting in rubbing. Such rubbing can be detected based on an AE signal detected by an AE sensor 10, which will be described later.
[0016] Each of the casings 6a to 6c has end portions 12a to 12f (so-called cat feet) on both axial sides thereof that are formed to be smaller in radial size than the central portions 11a to 11c. The end portions 12a to 12f integrally with the central portions 11a to 11c constitute the casings 6a to 6c, which are the stationary portions 2.
[0017] The rotary machine 1 also has bearings 8a to 8d (radial bearings) for supporting the rotating unit 4. The bearing 8a rotatably supports the end of the rotating unit 4 on the low-pressure unit LP side. The bearing 8b rotatably supports the portion of the rotating unit 4 between the low-pressure unit LP and the medium-pressure unit IP. The bearing 8c rotatably supports the portion of the rotating unit 4 between the medium-pressure unit IP and the high-pressure unit HP. The bearing 8d rotatably supports the end of the rotating unit 4 on the high-pressure unit HP side. These bearings 8a to 8d are housed in bearing housings 9a to 9d, respectively.
[0018] The AE sensor 10 is configured to detect an acoustic emission (AE) signal and is provided in the stationary portion 2 of the rotating machine 1. While FIG. 1 illustrates a typical example of the placement of the AE sensor 10 in the bearing housing 9a, as described below, the AE sensor 10 may be provided in the other bearing housings 9b to 9d or in the casings 6a to 6c. The rotating machine 1 generates AE waves when, for example, a seal or the like attached to the thermally deformed stationary portion 2 rubs against the rotating portion 4. For example, the AE waves generated at the location of the rubbing propagate as elastic waves through the stationary portion 2 and the rotating portion 4 and are detected as an AE signal by the AE sensor 10 installed in the rotating machine 1. AE waves generally have a frequency in the sound wave range of several tens of kilohertz to several megahertz.
[0019] Next, a quality evaluation device 100 for evaluating the quality of the rotating machine 1 having the above configuration will be described. The quality evaluation device 100 is composed of, for example, a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), and a computer-readable storage medium. A series of processes for realizing various functions is stored in a storage medium or the like in the form of a program, for example. The CPU reads this program into the RAM or the like and executes information processing and arithmetic operations to realize various functions. The program may be pre-installed in a ROM or other storage medium, provided in a state stored in a computer-readable storage medium, or distributed via wired or wireless communication means. Examples of computer-readable storage media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, and semiconductor memories.
[0020] Fig. 2 is a block diagram of the quality evaluation device 100 of Fig. 1. The quality evaluation device 100 includes an AE signal acquisition unit 102, a rubbing detection index calculation unit 104, a wear amount estimation unit 106, and a determination unit 108.
[0021] The AE signal acquisition unit 102 is configured to acquire an AE signal. As described above, the AE sensor 10 is attached to the stationary part 2 of the rotating machine 1, and the AE signal acquisition unit 102 acquires the AE signal detected by the AE sensor 10.
[0022] The rubbing detection index calculation unit 104 is configured to calculate the rubbing detection index a based on the AE signal acquired by the AE signal acquisition unit 102. For example, the rubbing detection index a is calculated based on information about the phase of the AE signal. Specifically, the rubbing detection index a is calculated using the following formula: Rubbing detection index=1 / (1+(AE signal phase variance)^0.5) Formula (1) The AE signal detected by the AE sensor 10 contains, for example, a considerable amount of noise signal from the rotating part 4 during operation of the rotating machine 1. Even when such noise signal is relatively large, the presence or absence of rubbing can be suitably determined by using the rubbing detection index a calculated based on information about the phase of the AE signal. Note that the detailed calculation method for the rubbing detection index a exemplified in the above formula (1) is described in Japanese Patent Application Laid-Open No. 2003-222244.
[0023] The wear amount estimation unit 106 is configured to estimate the wear amount A of the rotating machine 1 by accumulating the rubbing detection index a calculated by the rubbing detection index calculation unit 104. Here, several methods for estimating and calculating the wear amount A by the wear amount estimation unit 106 in Fig. 2 will be specifically described with reference to Fig. 3 and Fig. 4. Fig. 3 is an explanatory diagram showing one example of the method for estimating and calculating the wear amount A by the wear amount estimation unit 106 in Fig. 2, and Fig. 4 is an explanatory diagram showing another example of the method for estimating and calculating the wear amount A by the wear amount estimation unit 106 in Fig. 2.
[0024] 3 , the wear amount estimation unit 106 estimates the wear amount A by integrating the rubbing detection index a over a region where the rubbing detection index a exceeds a preset threshold ath. The threshold ath is set in advance as a reference value for determining whether or not instantaneous rubbing has occurred based on the rubbing detection index a. That is, the region where the rubbing detection index a exceeds the threshold ath is a time period where rubbing has occurred, and the region where the rubbing detection index a is equal to or less than the threshold ath is a time period where rubbing has not occurred. The wear amount estimation unit 106 can estimate the wear amount A using the following equation by integrating the rubbing detection index a(t) at a certain time t over the former region.
[0025] 4, the wear amount estimation unit 106 estimates the wear amount A by integrating the time during which the rubbing detection index a exceeds a preset threshold ath. Specifically, the wear amount A is estimated by the following equation.
[0026] The determination unit 108 is configured to compare the wear amount A estimated by the wear amount estimation unit 106 with a preset reference value Aref to determine whether maintenance of the rotating machine 1 is required. The determination unit 108 compares the wear amount A, which is quantitatively calculated by integrating the rubbing detection index a, with the preset reference value Aref. As a result, when the wear amount A exceeds the reference value Aref and it is determined that the wear amount A of the rotating machine 1 has become sufficiently large, it can be suitably determined that maintenance of the rotating machine 1 is required.
[0027] Next, a description will be given of a quality evaluation method implemented by the quality evaluation device 100 having the above configuration. Fig. 5 is a flowchart showing a quality evaluation method according to an embodiment.
[0028] When the quality evaluation method is started, first, the AE signal acquisition unit 102 acquires the AE signal detected by the AE sensor 10 (step S1). Next, the rubbing detection index calculation unit 104 calculates the rubbing detection index a based on the AE signal acquired in step S1 (step S2). The acquisition of the AE signals in step S1 is performed continuously in time according to a predetermined sampling period, and in step S2, the rubbing detection index a is calculated based on each of the AE signals.
[0029] Next, the wear amount estimating unit 106 estimates the wear amount A by integrating the rubbing detection index a calculated in step S2 over the evaluation period (step S3). As described above, the rubbing detection index a is calculated continuously over time, and in step S3, the rubbing detection index a is integrated over a predetermined evaluation period that is set in advance to determine an estimated value of the wear amount A. The method for calculating the estimated value of the wear amount A is as described above.
[0030] Next, the determination unit 108 acquires a reference value Aref corresponding to the wear amount A (step S4). The reference value Aref is a threshold value for determining whether maintenance is necessary, and is determined in advance by experimentally, theoretically, or by simulation-based evaluation of the wear amount A at which maintenance is required. The reference value Aref is readably stored in advance in, for example, a storage device (not shown).
[0031] Next, the determination unit 108 determines whether the wear amount A estimated in step S3 is greater than the reference value Aref acquired in step S4 (step S5). If the wear amount A is greater than the reference value Aref (step S5: YES), the determination unit 108 determines that the degree of wear is large and determines that "maintenance is required" (step S6). On the other hand, if the wear amount A is equal to or less than the reference value Aref (step S5: NO), the determination unit 108 determines that the degree of wear is small and determines that "maintenance is not required" (step S7). Note that in steps S6 and S7, a predetermined output (e.g., screen display, audio, etc.) may be performed according to the determination result.
[0032] Fig. 6 is a cross-sectional structural view of a rotary machine according to another embodiment. This embodiment differs from the embodiment shown in Fig. 1 in that multiple AE sensors 10 are provided on each of the bearing housings 9a to 9d and the end portions 12a to 12f. More specifically, of the multiple AE sensors 10, AE sensors 10a to 10d are provided on the bearing housings 9a to 9d, respectively, and AE sensors 10e to 10j are provided on the end portions 12a to 12f, respectively. Each of these AE sensors 10 has an identical configuration, and the AE signals detected by each AE sensor 10 can be acquired by the quality evaluation device 100.
[0033] 6, target portions for estimating the amount of wear are set by dividing each of the compartments 6a to 6c, which are the stationary portion 2, into three. Specifically, a left side portion LPL, a central portion LPC, and a right side portion LPR are set as target portions in the compartment 6a, a left side portion IPL, a central portion IPC, and a right side portion IPR are set as target portions in the compartment 6b, and a left side portion HPL, a central portion HPC, and a right side portion HPR are set as target portions in the compartment 6c.
[0034] The wear amount estimating unit 106 may estimate the wear amount A' for each target portion by using a predetermined weighting coefficient to add up the wear amounts A estimated using each AE sensor 10. Specifically, the wear amount A' for each target portion is expressed by the following equation using the wear amounts Aa to Aj estimated based on the AE signals from the AE sensors 10a to 10j and the corresponding weighting coefficients αa to αj. Wear amount A' of target part LPL = αa (= 0.5) × Aa + αe (= 1.0) × Ae Formula (4-1) Wear amount A' of target part LPC = αa (= 0.5) × Aa + αb (= 0.25) × Ab Formula (4-2) Wear amount A' of target part LPR = αb (= 0.25) × Ab + αf (= 1.0) × Af Formula (4-3) Wear amount A' of target part IPL = αb (= 0.25) × Ab + αg (= 1.0) × Ag Formula (4-4) Wear amount A' of target part IPC = αb (= 0.25) × Ab + αc (= 0.25) × Ac Formula (4-5) Wear amount A' of target part IPR = αc (= 0.25) × Ac + αh (= 1.0) × Ah Equation (4-6) Wear amount A' of target part HPL = αc (= 0.25) × Ac + αi (= 1.0) × Ai Equation (4-7) Wear amount A' of target part HPC = αc (= 0.25) × Ac + αd (= 0.5) × Ad Equation (4-8) Wear amount A' of target part HPR = αd (= 0.5) × Ad + αj (= 1.0) × Aj Equation (4-9) Note that each weighting coefficient αa to αj for each target part is standardized so that their respective sums become a predetermined value (for example, "1").
[0035] The wear amount A' of the target parts (LPL, LPR, IPL, IPR, HPL, HPR) that are biased to one side of the center (LPC, IPC, HPc) of each casing 6a to 6c is calculated by adding the wear amounts A calculated using a first AE sensor provided on a bearing housing that houses a bearing at a position closest to the target part and a second AE sensor provided at an end of the casing between the target part and the bearing housing, using a predetermined weighting coefficient α. In the above equations (4-1), (4-3), (4-4), (4-6), (4-7), and (4-9), the first term represents the wear amount corresponding to the first AE sensor, and the second term represents the wear amount corresponding to the second AE sensor. The first AE sensor is provided on a bearing housing that houses a bearing at a position closest to the target part, and therefore is the AE sensor that has the greatest influence on the target part. The second AE sensor is also provided at the end closest to the target part, and therefore has a large influence on the target part. In this way, the wear amount A' in the target portion biased from the center can be estimated with high accuracy by adding the wear amount A calculated based on the AE sensor, which is greatly affected by these factors.
[0036] Furthermore, when calculating the wear amount A' in each target portion, the wear amount estimating unit 106 may use a third AE sensor in addition to the first and second AE sensors. The third AE sensor is an AE sensor provided in a bearing housing that houses the bearing on the opposite side of the target portion from the first AE sensor. In this case, by adding a third term corresponding to the third AE sensor to each of the above equations, the wear amount A' in each target portion can be calculated more accurately.
[0037] The wear amount A' of the central portion (LPC, IPC, HPc) of each of the target portions 6a to 6c is calculated by adding together the wear amounts A calculated using a pair of AE sensors provided in a pair of bearing housings that house bearings on both sides of the target portion, using a predetermined weighting coefficient α. In the above equations (4-2), (4-5), and (4-8), the first term represents the wear amount corresponding to one of the pair of AE sensors, and the second term represents the wear amount corresponding to the other. In this way, the wear amount A' of the target portion in the central portion can be accurately estimated by adding together the wear amounts A calculated based on the pair of AE sensors that have the greatest influence.
[0038] In this embodiment, the wear amount A' for each target part is obtained by adding up the wear amount A estimated for each target part, but the wear amount A' for each target part may also be obtained by adding up the rubbing detection index a calculated for each target part using a predetermined weighting coefficient α and integrating the addition result over the evaluation period.
[0039] Fig. 7 is a cross-sectional structural view of a rotary machine 1 according to another embodiment. This embodiment differs from the embodiment shown in Fig. 1 in that a plurality of AE sensors 10 are provided in each of the bearing housings 9a to 9d. More specifically, of the plurality of AE sensors 10, AE sensors 10a to 10d are provided in each of the bearing housings 9a to 9d. Each of these AE sensors 10 has the same configuration, and the AE signals detected by each AE sensor 10 can be acquired by the quality evaluation device 100.
[0040] 7, the target portions for which the wear amount is to be estimated are set by dividing each of the compartments 6a to 6c, which are the stationary portion 2, into two parts. Specifically, a left side portion LPL and a right side portion LPR are set as target portions in the compartment 6a, a left side portion IPL and a right side portion IPR are set as target portions in the compartment 6b, and a left side portion HPL and a right side portion HPR are set as target portions in the compartment 6c.
[0041] The wear amount estimating unit 106 may estimate the wear amount A' for each target portion by using a predetermined weighting coefficient to add up the wear amounts A estimated using each AE sensor 10. Specifically, the wear amount A' for each target portion is expressed by the following equation using the wear amounts Aa to Ad estimated based on the AE signals from the AE sensors 10a to 10d and the corresponding weighting coefficients αa to αd. Wear amount A' of target part LPL = αa (= 0.75) × Aa + αb (= 0.167) × Ab Formula (5-1) Wear amount A' of target part LPR = αa (= 0.25) × Ab + αb (= 0.33) × Ab Formula (5-2) Wear amount A' of target part IPL = αb (= 0.33) × Ab + αc (= 0.167) × Ac Formula (5-3) Wear amount A' of target part IPR = αb (= 0.167) × Ab + αc (= 0.33) × Ac Formula (5-4) Wear amount A' of target part HPL = αc (= 0.33) × Ac + αd (= 0.25) × Ad Formula (5-5) Wear amount A' of target part HPR = αc (= 0.167) × Ac + αd (= 0.75) × Ad Equation (5-6) Note that each weighting coefficient αa to αd for each target part is normalized so that their sum becomes a predetermined value (for example, "1").
[0042] FIG. 8 is a cross-sectional structural view of a rotary machine 1 according to another embodiment. In this embodiment, compared to the embodiment shown in FIG. 1, a pair of bearings 8b1 and 8b2 arranged along the axial direction is provided instead of the bearing 8b between the casings 6a and 6b, and a pair of bearings 8c1 and 8c2 arranged along the axial direction is provided instead of the bearing 8c between the casings 6b and 6c. The bearings 8b1 and 8b2 are housed in independent bearing housings 9b1 and 9b2, respectively, and the bearings 8c1 and 8c2 are housed in independent bearing housings 9c1 and 9c2, respectively. Another difference is that a plurality of AE sensors 10 are provided in each of the bearing housings 9a to 9d and the end portions 12a to 12f. More specifically, of the multiple AE sensors 10, AE sensors 10a to 10d are provided on bearing housings 9a to 9d, respectively (sensors 10b1 and 10b2 are provided on bearing housings 9b1 and 9b2, respectively, and sensors 10c1 and 10c2 are provided on bearing housings 9c1 and 9c2, respectively), and AE sensors 10g to 10j are provided on end portions 12a to 12f, respectively. Each of these AE sensors 10 has the same configuration, and the AE signals detected by each AE sensor 10 can be acquired by the quality evaluation device 100.
[0043] 8, target portions for which wear amounts are to be estimated are set by dividing each of the compartments 6a to 6c, which are the stationary portion 2, into three parts. Specifically, a left side portion LPL, a central portion LPC, and a right side portion LPR are set as target portions in the compartment 6a, a left side portion IPL, a central portion IPC, and a right side portion IPR are set as target portions in the compartment 6b, and a left side portion HPL, a central portion HPC, and a right side portion HPR are set as target portions in the compartment 6c.
[0044] The wear amount estimating unit 106 may estimate the wear amount A' for each target portion by using a predetermined weighting coefficient to add up the wear amounts A estimated using each AE sensor 10. Specifically, the wear amount A' for each target portion is expressed by the following equation using the wear amounts Aa to A1 estimated based on the AE signals from the AE sensors 10a to 10l and the corresponding weighting coefficients αa to αl. Wear amount A' of target part LPL = αa (= 0.5) × Aa + αg (= 1.0) × Ag Formula (6-1) Wear amount A' of target part LPC = αa (= 0.5) × Aa + αb (= 0.5) × Ab Formula (6-2) Wear amount A' of target part LPR = αb (= 0.5) × Ab + αh (= 1.0) × Ah Formula (6-3) Wear amount A' of target part IPL = αc (= 0.5) × Ab + αi (= 1.0) × Ai Formula (6-4) Wear amount A' of target part IPC = αc (= 0.5) × Ac + αd (= 0.5) × Ad Formula (6-5) Wear amount A' of target part IPR = αd (= 0.5) × Ad + αj (= 1.0) × Aj Equation (6-6) Wear amount A' of target part HPL = αd (= 0.5) × Ad + αk (= 1.0) × Ak Equation (6-7) Wear amount A' of target part HPC = αe (= 0.5) × Ae + αf (= 0.5) × Af Equation (6-8) Wear amount A' of target part HPR = αf (= 0.5) × Af + αl (= 1.0) × Al Equation (6-9) Note that the weighting coefficients αa to αl for each target part are normalized so that their sum becomes a predetermined value (for example, "1").
[0045] 9 is a cross-sectional structural view of a rotary machine 1 according to another embodiment. This embodiment differs from the embodiment shown in FIG. 7 in that a plurality of AE sensors 10 are provided on each of the bearing housings 9a to 9d. More specifically, of the plurality of AE sensors 10, AE sensors 10a to 10d are provided on each of the bearing housings 9a to 9d (sensors 10b1 and 10b2 are provided on bearing housings 9b1 and 9b2, respectively, and sensors 10c1 and 10c2 are provided on bearing housings 9c1 and 9c2, respectively). Each of these AE sensors 10 has an identical configuration, and the AE signals detected by each AE sensor 10 can be acquired by the quality evaluation device 100.
[0046] 9, the target portions for which the wear amount is to be estimated are set by dividing each of the compartments 6a to 6c, which are the stationary portion 2, into two parts. Specifically, a left side portion LPL and a right side portion LPR are set as target portions in the compartment 6a, a left side portion IPL and a right side portion IPR are set as target portions in the compartment 6b, and a left side portion HPL and a right side portion HPR are set as target portions in the compartment 6c.
[0047] The wear amount estimating unit 106 may estimate the wear amount A' for each target portion by using a predetermined weighting coefficient to add up the wear amounts A estimated using each AE sensor 10. Specifically, the wear amount A' for each target portion is expressed by the following equation using the wear amounts Aa to Af estimated based on the AE signals from the AE sensors 10a to 10d and the corresponding weighting coefficients αa to αf. Wear amount A' of target part LPL = αa (= 0.75) × Aa + αb (= 0.25) × Ab Formula (7-1) Wear amount A' of target part LPR = αa (= 0.25) × Aa + αb (= 0.75) × Ab Formula (7-2) Wear amount A' of target part IPL = αc (= 0.75) × Ac + αd (= 0.25) × Ad Formula (7-3) Wear amount A' of target part IPR = αc (= 0.25) × Ac + αd (= 0.75) × Ad Formula (7-4) Wear amount A' of target part HPL = αe (= 0.75) × Ae + αf (= 0.25) × Af Formula (7-5) Wear amount A' of target part HPR = αe (= 0.25) × Ae + αf (= 0.75) × Af Equation (7-6) Note that each weighting coefficient αa to αf for each target part is normalized so that their sum becomes a predetermined value (for example, "1").
[0048] As described above, the wear amount estimation unit 106 can estimate the wear amount A' at the target portion by integrating the rubbing detection index a for each target portion provided at different positions along the axial direction of the rotating part 4. This makes it possible to quantitatively grasp the wear amount for each portion of the rotating machine, and to determine whether maintenance is required for each portion.
[0049] The weighting coefficient α used in the above embodiment may be determined by actual measurements based on the response when an impact is applied to the actual rotating machine 1 to be evaluated. Fig. 10 is an enlarged schematic diagram showing the peripheral configuration of the low-pressure unit LP in Fig. 6, Fig. 11 is a diagram showing the signal intensity distribution obtained when an impact is applied to each target portion in Fig. 10, and Fig. 12 is a calculation example of the weighting coefficient α based on the signal intensity in Fig. 11.
[0050] The impact on the target part may be applied from the inside of the casing, which is the stationary part 2. Rubbing in an actual rotating machine 1 occurs when the rotating part 4 comes into contact with the stationary part 2 from the inside. Therefore, by applying the impact from inside the casing, it is possible to suitably simulate the situation in which rubbing actually occurs.
[0051] In this case, as shown in Fig. 10, a strike is applied to a position corresponding to each target portion (LPL, LPC, LPR), and at that time, the weighting coefficient α is determined based on the signal intensity distribution of the AE signals obtained by the multiple AE sensors 10. Fig. 11 shows the signal intensity distribution obtained by each AE sensor 10 for each target portion to which a strike is applied in Fig. 10.
[0052] Specifically, when an impact is applied to the target location LPL, the AE sensor 10a (corresponding to the first AE sensor described above) provided on the bearing housing 9a closest to the target location LPL provides the strongest signal. The next strongest signal is provided by the AE sensor 10e (corresponding to the second AE sensor described above) provided at the end 12a of the casing 6a between the target location LPL and the bearing housing 9a. The next strongest signal is provided by the AE sensor 10b (corresponding to the third AE sensor described above) provided on the bearing housing 9b housing the bearing 6b on the opposite side of the AE sensor 10a from the target location LPL. On the other hand, almost no signal is provided by the AE sensor 10f.
[0053] When a strike is applied to the target part LPC, a pair of AE sensors 10a and 10b provided in a pair of bearing housings 9a and 9b that house the bearings 8a and 8b on both sides of the target part LC respectively obtain a strong signal intensity, while the AE sensors 10e and 10f obtain almost no signal intensity.
[0054] When a strike is applied to the target portion LPR, the AE sensor 10f located at the end b closest to the target portion LPR produces the strongest signal. The next strongest signal is produced by the AE sensor 10b (corresponding to the first AE sensor) located on the end bearing housing 9b closest to the target portion LPR. The next strongest signal is produced by the AE sensor 10a (corresponding to the third AE sensor) located on the bearing housing 9a housing the bearing 8a on the opposite side of the target portion LPR from the AE sensor 10b. On the other hand, almost no signal is produced by the AE sensor 10e.
[0055] FIG. 12 shows an example of calculating the weighting coefficient α based on the signal intensity distribution shown in FIG. 11 . In this example, the weighting coefficient α is calculated by non-dimensionalizing the ratio of the signal intensities of the AE sensors 10 for each target portion shown in the upper row. In this manner, in this embodiment, a situation in which rubbing occurs in the target portion due to contact between the stationary portion 2 and the rotating portion 4, for example, when an operator manually strikes the target portion, is simulated. The signal intensity distribution of the AE signals obtained by the multiple AE sensors 10 at this time reflects the type of AE signal obtained by each AE sensor 10 when rubbing occurs in the target portion. Therefore, by calculating the weighting coefficient α based on this signal intensity distribution, the wear amount A of the rotating machine 1, which is obtained by integrating the AE signals detected by the multiple AE sensors 10, can be more accurately estimated.
[0056] As described above, when the wear amount A is estimated for each target part by the wear amount estimation unit 106, the estimated values of the wear amount A for each target part can be compared to evaluate the degree of damage for each target part of the rotating machine 1. Fig. 13 is a diagram showing an example of the estimated values of the wear amount A for each target part of the rotating machine 1. In the example of Fig. 13 , the wear amount A of the target part LPL shown on the far left is particularly the largest, and therefore, when performing maintenance, it is possible to determine to open and inspect only the target part LPL corresponding to the wear amount A.
[0057] 13, by checking the amount of wear A for each target part, it is possible to arrange in advance for parts that are expected to need replacement work during the next maintenance. This reduces the waiting time while arranging for replacement parts, thereby improving the efficiency of the cost and time required for maintenance.
[0058] Furthermore, if there is an existing inspection record, the reference value Aref for determining whether or not maintenance is required may be determined based on the inspection record from the estimated value of the wear amount A. Fig. 14 is an explanatory diagram showing how the reference value Aref for determining whether or not maintenance is required is determined from the existing inspection results.
[0059] In FIG. 14 , an example of an existing inspection result is shown on the left side. This inspection result indicates that the target part LPC among multiple target parts has a large amount of wear A, and therefore requires maintenance. Based on this inspection result, the amount of wear (actual measurement value) of the target part LPC in the inspection result can be determined as the reference value Aref. The reference value Aref determined in this manner can be used to determine whether maintenance is necessary based on the estimated value of the amount of wear A from the next time onwards, as shown on the right side of FIG. 14 . In this example, the amount of wear A when the target part LPR is reevaluated exceeds the reference value Aref, and therefore it can be predicted that part replacement of the target part LPR will be required at the time of the next maintenance.
[0060] The weighting coefficient α used in each of the above embodiments may also be corrected based on the results of actual measurements taken during maintenance. For example, if there is a target portion where the estimated value of the wear amount A is small but the actually measured inspection result shows that the wear amount A is large, the sensitivity of the AE sensor 10 in that target portion can be determined to be low, and the weighting coefficient α corresponding to that AE sensor 10 can be corrected to be large. On the other hand, if there is a target portion where the estimated value of the wear amount A is large but the actually measured inspection result shows that the wear amount A is small, the sensitivity of the AE sensor 10 in that target portion can be determined to be high, and the weighting coefficient α corresponding to that AE sensor 10 can be corrected to be small.
[0061] FIG. 15 is an explanatory diagram showing an example of correcting the weighting coefficient α based on inspection results. In FIG. 15, the left side shows the wear amount A, which is the actual measurement value obtained from the inspection results for each target part, and the right side shows the estimated wear amount A for each target part. On the left side of this example, the average wear amount for each evaluation part is calculated from the inspection results and non-dimensionalized by the total value for each part, thereby calculating a correction coefficient for correcting the weighting coefficient α. On the other hand, on the right side, the estimated wear amount A is non-dimensionalized by the total value for each target part. When correcting the weighting coefficient α, the deviation between the non-dimensionalized inspection results and the non-dimensionalized estimated wear amount A can be determined by calculating the ratio. Therefore, by using this ratio as a correction coefficient for the weighting coefficient α, the accuracy of the estimated wear amount A can be improved by reflecting the inspection results.
[0062] In addition, within the scope of the present disclosure, the components in the above-described embodiments may be replaced with well-known components as appropriate, and the above-described embodiments may be combined as appropriate.
[0063] The contents described in each of the above embodiments can be understood, for example, as follows.
[0064] (1) A quality evaluation device for a rotating machine according to one embodiment is a quality evaluation device for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, and includes: an AE signal acquisition unit for acquiring an AE signal from at least one AE sensor provided on the stationary part; a rubbing detection index calculation unit for calculating a rubbing detection index based on the AE signal; and a wear amount estimation unit for estimating the wear amount of the rotating machine by integrating the rubbing detection index.
[0065] According to the above aspect (1), the rubbing detection index is calculated based on the AE signal acquired from the AE sensor provided on the stationary part, and the rubbing detection index is integrated, thereby making it possible to suitably estimate the amount of wear caused in the rotating machine due to the occurrence of rubbing. The rubbing detection index integrated to calculate the amount of wear can broadly include indices for determining whether or not instantaneous rubbing has occurred.
[0066] (2) In another aspect, in the aspect (1), a determination unit is further provided for determining whether maintenance of the rotating machine is required by comparing the amount of wear with a preset reference value.
[0067] According to the above aspect (2), the wear amount quantitatively calculated by integrating the rubbing detection index is compared with a preset reference value. As a result, when the wear amount exceeds the reference value and it is determined that the wear amount of the rotating machine has become sufficiently large, it can be suitably determined that maintenance of the rotating machine is necessary.
[0068] (3) In another aspect, in the aspect (1) or (2) above, the wear amount estimation unit estimates the wear amount in the target portion by accumulating the rubbing detection index for each target portion provided at a different position along the axial direction of the rotating portion.
[0069] According to the above aspect (3), the amount of wear is estimated by integrating the rubbing detection index for each target part provided at a different position along the axial direction of the rotating part, which makes it possible to quantitatively grasp the amount of wear for each part of the rotating machine and determine the need for maintenance for each part.
[0070] (4) In another aspect, in any one of the above aspects (1) to (3), the wear amount estimation unit estimates the wear amount by integrating the rubbing detection indicator over an area where the rubbing detection indicator exceeds a preset threshold value.
[0071] According to the above aspect (4), the rubbing detection index exceeds the threshold value, and thus the rubbing detection index is integrated over the area where it is determined that rubbing has occurred, thereby making it possible to suitably estimate the amount of wear.
[0072] (5) In another aspect, in any one of the above aspects (1) to (3), the wear amount estimation unit estimates the wear amount by accumulating the time during which the rubbing detection indicator exceeds a preset threshold value.
[0073] According to the above aspect (5), the amount of wear can be suitably estimated by integrating the time during which the rubbing detection index exceeds the threshold value.
[0074] (6) In another aspect, in any one of the above aspects (1) to (5), the at least one AE sensor includes a plurality of AE sensors, and the wear amount estimation unit estimates the wear amount by adding the rubbing detection index calculated based on the rubbing signals from the plurality of AE sensors using a weighting coefficient.
[0075] According to the above aspect (6), the rubbing detection index used for estimating the amount of wear by accumulating the AE signals detected by the plurality of AE sensors is calculated by adding the AE signals detected by the plurality of AE sensors using a weighting coefficient. By using the plurality of AE signals detected by the plurality of AE sensors in this way, the amount of wear can be estimated more accurately than when only the AE signal detected by a single AE sensor is used.
[0076] (7) In another aspect, in the aspect (6) above, when the target portion for which the wear amount is to be estimated is biased to one side of the center of the compartment that houses the rotating portion of the stationary portion, the plurality of AE sensors include a first AE sensor provided in a bearing box that houses the bearing at a position closest to the target portion, and a second AE sensor provided at an end of the compartment between the target portion and the bearing box.
[0077] According to the above aspect (7), when a target portion for which a wear amount is to be estimated is located to one side of the center of the passenger compartment, the wear amount is estimated based on the AE signals detected by the first AE sensor and the second AE sensor. The first AE sensor is provided on the bearing housing located closest to the target portion, and the second AE sensor is provided between the target portion and the bearing housing on which the first AE sensor is provided. These AE sensors are provided in positions that are sensitive to the target portion that is located to one side of the center of the passenger compartment, so the AE signals detected by these AE sensors can be used to accurately estimate the wear amount of the target portion.
[0078] (8) In another aspect, in the aspect (7), the plurality of AE sensors include a third AE sensor provided in a bearing housing that houses the bearing on the opposite side of the target portion from the first AE sensor.
[0079] According to the above aspect (8), when the target portion for which the amount of wear is to be estimated is located to one side of the center of the passenger compartment, the amount of wear is estimated based on an AE signal detected by a third AE sensor. The third AE sensor is provided on another bearing housing that is located on the opposite side of the bearing housing on which the first AE sensor is provided and to which the AE signal is likely to propagate via the rotating part. By further using the AE signal detected by the third AE sensor in this way, the amount of wear in the target portion can be estimated with greater accuracy.
[0080] (9) In another aspect, in any one of the above aspects (6) to (8), when the target portion for which the wear amount is estimated is a central portion of the vehicle interior that houses the rotating part, the plurality of AE sensors include a pair of AE sensors respectively provided in a pair of bearing housings that respectively house the bearings on both sides of the target portion.
[0081] According to the above aspect (9), when a target portion for which the amount of wear is to be estimated is located in the center of the passenger compartment, the amount of wear is estimated based on AE signals detected by a pair of AE sensors provided on bearing housings on both sides of the center. These AE sensors are provided in positions that are sensitive to the target portion in the center of the passenger compartment, so the AE signals detected by these AE sensors can be used to accurately estimate the amount of wear in the target portion.
[0082] (10) In another aspect, in any one of the above aspects (6) to (9), the weighting coefficient is calculated based on the signal intensity distribution of the AE signals obtained by the plurality of AE sensors when an impact is applied to the target portion for which the wear amount is to be estimated.
[0083] According to the above aspect (10), for example, an operator manually strikes the target part, thereby simulating a situation in which a rotating part comes into contact with a stationary part in the target part, causing rubbing. The signal intensity distribution of the AE signals obtained by the multiple AE sensors reflects the AE signals obtained by each AE sensor when rubbing occurs in the target part. Therefore, by calculating a weighting coefficient based on this signal intensity distribution, the amount of wear of the rotating machine, which is obtained by integrating the AE signals detected by the multiple AE sensors, can be more accurately estimated.
[0084] (11) In another aspect, in the aspect (10), the impact is applied to the stationary part from the inside.
[0085] According to the above aspect (11), by applying an impact from the inside of the stationary part, it is possible to suitably simulate a situation in which rubbing occurs due to the rotating part coming into contact with the stationary part from the inside. By calculating a weighting coefficient based on the signal intensity distribution of the AE signal detected by the AE sensor when such an impact is applied, it is possible to accurately estimate the amount of wear of the rotating machine.
[0086] (12) In another aspect, in the above aspects (6) to (11), the weighting coefficient is corrected based on the actual wear amount measured for each target portion provided at a different position along the axial direction of the rotating part.
[0087] According to the above aspect (12), the weighting coefficients used to estimate the amount of wear are corrected based on actual measurements obtained by, for example, opening the rotating machine during maintenance and measuring the amount of wear for each target part. By using the weighting coefficients corrected based on the actual measurements in this way, it is possible to effectively improve the accuracy of subsequent estimations of the amount of wear.
[0088] (13) In another aspect, in any one of the above aspects (1) to (12), the rubbing detection index is calculated based on information about the phase of the AE signal.
[0089] According to the above aspect (13), the wear amount of the rotating machine can be suitably estimated by integrating the rubbing detection index calculated based on the phase information of the AE signal. The AE signal detected by the AE sensor contains a considerable amount of noise signal from a rotating part such as a turbine during operation of the rotating machine. Even when such noise signal is relatively large, the occurrence of rubbing can be suitably determined by using the rubbing detection index calculated based on the phase information of the AE signal.
[0090] (14) In another aspect, in the aspect (13), the rubbing detection index is calculated by the following formula: rubbing detection index=1 / (1+(AE signal phase dispersion)^0.5)
[0091] According to the above aspect (14), by using the rubbing detection index defined by the above formula, it is possible to determine whether rubbing has occurred based on the AE signal containing a relatively large noise signal, and by integrating this, it is possible to suitably estimate the amount of wear of the rotating machine.
[0092] (15) A quality evaluation method for a rotating machine according to one embodiment is a quality evaluation method for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, comprising: a step of acquiring an AE signal from at least one AE sensor provided on the stationary part; a step of calculating a rubbing detection index based on the AE signal; and a step of estimating the amount of wear of the rotating machine by integrating the rubbing detection index.
[0093] According to the above aspect (15), the rubbing detection index is calculated based on the AE signal acquired from the AE sensor provided on the stationary part, and the rubbing detection index is integrated, thereby making it possible to suitably estimate the amount of wear caused in the rotating machine due to the occurrence of rubbing. The rubbing detection index integrated to calculate the amount of wear can broadly include indices for determining whether or not instantaneous rubbing has occurred.
[0094] (16) A quality evaluation program for a rotating machine according to one embodiment is a quality evaluation program for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, and is capable of executing the following steps on a computer device: acquiring an AE signal from at least one AE sensor provided on the stationary part; calculating a rubbing detection index based on the AE signal; and estimating the amount of wear of the rotating machine by accumulating the rubbing detection index.
[0095] According to the above aspect (16), the rubbing detection index is calculated based on the AE signal acquired from the AE sensor provided on the stationary part, and the rubbing detection index is integrated, thereby making it possible to suitably estimate the amount of wear caused in the rotating machine due to the occurrence of rubbing. The rubbing detection index integrated to calculate the amount of wear can broadly include indices for determining whether or not instantaneous rubbing has occurred.
[0096] REFERENCE SIGNS LIST 1 Rotating machine 2 Stationary part 4 Rotating part 6a Low pressure unit 6b Medium pressure unit 6c High pressure unit 8a to 8d Bearings 9a to 9d Bearing housing 10 AE sensors 11a to 11c Center part 12a to 12f End part 100 Rubbing determination device 102 AE signal acquisition part 104 Rubbing detection index calculation part 106 Wear amount estimation part 108 Determination part D Gap
Claims
1. A quality evaluation device for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, comprising: an AE signal acquisition unit for acquiring an AE signal from at least one AE sensor provided on the stationary part; a rubbing detection index calculation unit for calculating a rubbing detection index based on the AE signal; and a wear amount estimation unit for estimating the amount of wear of the rotating machine by integrating the rubbing detection index.
2. A quality evaluation device for a rotating machine as claimed in claim 1, further comprising a determination unit for determining whether or not maintenance of said rotating machine is required by comparing said wear amount with a preset reference value.
3. A quality evaluation device for a rotating machine as described in claim 1 or 2, wherein the wear amount estimation unit estimates the amount of wear in the target parts by accumulating the rubbing detection index for each target part provided at different positions along the axial direction of the rotating part.
4. A quality evaluation device for a rotating machine as described in claim 1 or 2, wherein the wear amount estimation unit estimates the wear amount by integrating the rubbing detection index over a region where the rubbing detection index exceeds a preset threshold value.
5. A quality evaluation device for a rotating machine as described in claim 1 or 2, wherein the wear amount estimation unit estimates the wear amount by integrating the time during which the rubbing detection index exceeds a preset threshold value.
6. A quality evaluation device for a rotating machine as described in claim 1 or 2, wherein the at least one AE sensor includes a plurality of AE sensors, and the wear amount estimation unit estimates the wear amount by adding up the rubbing detection index calculated based on the rubbing signals from the plurality of AE sensors using a weighting coefficient.
7. A quality evaluation device for a rotating machine as described in claim 6, wherein, when the target portion for which the amount of wear is to be estimated is biased to one side of the center of the casing that houses the rotating portion among the stationary portions, the multiple AE sensors include a first AE sensor provided in a bearing box that houses the bearing at a position closest to the target portion, and a second AE sensor provided at an end of the casing between the target portion and the bearing box.
8. A quality evaluation device for a rotating machine as described in claim 7, wherein the plurality of AE sensors includes a third AE sensor provided in a bearing box that houses the bearing on the opposite side of the target area from the first AE sensor.
9. A quality evaluation device for a rotating machine as described in claim 6, wherein when the target area for estimating the amount of wear is the center of a compartment accommodating the rotating part, the multiple AE sensors include a pair of AE sensors respectively provided in a pair of bearing housings that respectively accommodate the bearings on both sides of the target area.
10. A quality evaluation device for a rotating machine as described in claim 6, wherein the weighting coefficient is calculated based on the signal intensity distribution of the AE signals obtained by the multiple AE sensors when an impact is applied to the target part for which the wear amount is to be estimated.
11. The quality evaluation device for a rotating machine according to claim 10, wherein the impact is applied to the stationary part from inside.
12. A quality evaluation device for a rotating machine as described in claim 6, which corrects the weighting coefficient based on actual values of the wear amount measured for each target portion provided at different positions along the axial direction of the rotating part.
13. A quality evaluation device for a rotating machine as described in claim 1 or 2, wherein the rubbing detection index is calculated based on information on the phase of the AE signal.
14. The quality evaluation device for a rotating machine according to claim 13, wherein the rubbing detection index is calculated by the following formula: rubbing detection index=1 / (1+(variance of the phase of the AE signal)^0.5) 15. A quality evaluation method for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, comprising the steps of: acquiring an AE signal from at least one AE sensor provided in the stationary part; calculating a rubbing detection index based on the AE signal; and estimating an amount of wear of the rotating machine by integrating the rubbing detection index.
16. A quality evaluation program for a rotating machine for evaluating the quality of a rotating machine having a rotating part rotatably supported relative to a stationary part by a bearing, the program being capable of executing, on a computer device, the steps of: acquiring an AE signal from at least one AE sensor provided on the stationary part; calculating a rubbing detection index based on the AE signal; and estimating the amount of wear of the rotating machine by accumulating the rubbing detection index.
Citation Information
Patent Citations
Detection method of turbine collision and abrasion faults
CN102519579A
Abrasion detector
JP2007003299A
Vehicle control device
JP2021015063A
Device and method for detecting rubbing of rotary machine
JP2021076533A
