Rubbing determination device for rotary machine, and rubbing determination method
The proposed rubbing determination device and method for rotating machinery utilize signal strength ratios and indices from AE signals to accurately detect rubbing, overcoming noise interference and phase convergence issues, enabling reliable rubbing detection in rotating machinery.
Patent Information
- Application Number
- PCT/JP2024/006902
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-19
- Filing Date
- 2024-02-27
- Publication Date
- 2024-11-28
AI Technical Summary
In rotating machinery, accurately determining rubbing occurrences is challenging due to noise interference in AE signals, especially when the phase of the AE signal may converge regardless of rubbing presence or absence, making it difficult to determine rubbing using existing methods.
A rubbing determination device and method that involves acquiring AE signals, performing envelope processing to identify local maxima and minima, calculating first and second frequency average spectra, and determining the presence of rubbing based on signal strength ratios and indices, which are independent of phase convergence.
This approach allows for accurate determination of rubbing in rotating machinery by focusing on signal strength ratios and indices, effectively filtering out noise and ensuring reliable detection regardless of phase convergence states.
Smart Images

Figure JP2024006902_28112024_PF_FP_ABST
Abstract
Description
Device and method for determining rubbing of a rotating machine
[0001] This application claims priority to Japanese Patent Application No. 2023-083194, filed on May 19, 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 reduce the gap between the stationary and rotating parts, resulting in rubbing, where the rotating body comes into contact with the stationary part. Rubbing can increase shaft vibration in rotary machines and lead to performance degradation due to the widening of the gap. Therefore, there is a demand for technology that can detect the occurrence of rubbing early and provide feedback to the operation of rotary machines. In particular, in recent years, there has been a trend toward narrowing the gap in rotary machines to accommodate high-speed load fluctuations and improve performance, so the risk of rubbing is expected to increase even further.
[0003] A known technique for determining the occurrence of rubbing is to use an AE (Acoustic Emission) sensor capable of detecting AE signals. AE sensors can be easily installed and are capable of detecting rubbing over a wide range. Therefore, they are expected to meet the need for a quick and easy evaluation of rubbing occurrence by detecting the occurrence and location of rubbing in rotating machinery during operation.
[0004] In rotating machines such as turbines, fluid flowing inside the machine can introduce significant noise into the AE signals detected by an AE sensor installed in the rotating machine, potentially reducing the accuracy of rubbing detection. Accurately detecting rubbing in such a noisy environment is a challenge. To address this challenge, Patent Document 1 proposes the following method. In this document, filtering is first performed on the AE signals acquired from the AE sensor. Based on the finding that the ratio of contact signals to noise signals (signal-to-noise ratio), which indicates rubbing, is high in high-frequency bands (e.g., around 175 kHz), a band-pass filter is used to extract only AE signals in high-frequency bands. Next, the phase of the rotation-synchronous component is calculated from the extracted AE signals, and the occurrence of rubbing is determined based on the convergence of the phase. More specifically, an index for quantifying the phase convergence is calculated, and the occurrence of rubbing is automatically determined when the index exceeds a threshold.
[0005] Japanese Patent Application Laid-Open No. 2021-076533
[0006] In the above-mentioned Patent Document 1, it is determined that rubbing has occurred when the phase calculated based on the AE signal acquired from an AE sensor attached to a stationary part of the rotating machine converges. However, depending on the installation position of the AE sensor on the rotating machine, the phase of the AE signal may always converge regardless of the presence or absence of rubbing. In such cases, it is difficult to determine the occurrence of rubbing in the rotating machine using the above-mentioned method.
[0007] At least one embodiment of the present disclosure has been made in consideration of the above circumstances, and aims to provide a rubbing determination device and a rubbing determination method for a rotating machine that can accurately determine the presence or absence of rubbing based on an AE signal.
[0008] To solve the above problem, at least one embodiment of the present disclosure provides a rubbing determination device for a rotary machine for determining rubbing in a rotary machine having a rotating part rotatably supported relative to a stationary part by a bearing, the device including: an AE signal acquisition unit for acquiring an AE signal from an AE sensor attached to the stationary part; a signal analysis unit for identifying maximum and minimum values of an envelope obtained by performing envelope processing on a temporal change in the AE signal, and calculating a first frequency average spectrum corresponding to a first time period including the maximum value of the temporal change, and a second frequency average spectrum corresponding to a second time period including the minimum value of the temporal change; a first signal intensity ratio calculation unit for calculating a first signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the first frequency average spectrum; and a second signal intensity ratio calculation unit for calculating a second signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the second frequency average spectrum. The optical disc drive includes a first index calculation unit for calculating a first index as a ratio of the first signal intensity ratio to the second signal intensity ratio, and a rubbing determination unit for determining whether or not rubbing is performed based on the first index.
[0009] To solve the above problem, a method for determining rubbing of a rotating machine according to at least one embodiment of the present disclosure includes the steps of: acquiring an AE signal from an AE sensor attached to the stationary part; identifying maximum and minimum values of an envelope obtained by performing envelope processing on a temporal change in the AE signal; calculating a first frequency average spectrum corresponding to a first time period including the maximum value of the temporal change, and a second frequency average spectrum corresponding to a second time period including the minimum value of the temporal change; calculating a first signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the first frequency average spectrum; calculating a second signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the second frequency average spectrum; and calculating a first index as a ratio of the first signal intensity ratio to the second signal intensity ratio. and determining whether or not the rubbing is performed based on the first index.
[0010] According to at least one embodiment of the present disclosure, it is possible to provide a rotating machine rubbing determination device and a rubbing determination method that can accurately determine the presence or absence of rubbing based on an AE signal.
[0011] 9 is a cross-sectional structural view of a rotary machine according to an embodiment. FIG. 1 is a configuration block diagram of the rubbing determination device of FIG. 1. FIG. 2 is a flowchart showing a procedure for calculating a first index. FIG. 3 is a diagram showing an example of a temporal change in an AE signal. FIG. 4 is a diagram showing a first frequency spectrum and a second frequency spectrum corresponding to the temporal change in the AE signal of FIG. 4. FIG. 5 is an example showing temporal changes in the amplitude, phase, first index (30-second average), and first index (2-minute average) of an AE signal. FIG. 6 is a flowchart showing a method for calculating a second index by the second index calculation unit of FIG. 2. FIG. 7 is a graph showing the amplitude of the AE signal for each rotation order. FIG. 8 is a flowchart showing a determination flow in the rubbing determination unit of FIG. 2. FIG. 9 is a flowchart showing a modified example of FIG.
[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] First, a rotary machine that is a target of a rubbing determination device according to at least one embodiment of the present disclosure will be described. Fig. 1 is a cross-sectional structural view of a rotary machine 1 according to one embodiment.
[0014] The rotary machine 1 includes a stationary part 2 and a rotating part 4 that is rotatable relative to the stationary part 2. The stationary part 2 is a casing of the rotary machine 1 and is stationary relative to the outside. The rotating part 4 is rotatably supported via bearings 5 housed in a bearing housing 6 and can be driven by any power. Like the stationary part 2, the bearing housing 6 is stationary relative to the outside.
[0015] A gap D is provided between the stationary part 2 and the rotating part 4 (more precisely, the gap D is mainly formed between the innermost peripheral parts of the stator vanes of the stationary part 2 and the outermost peripheral parts (tips) of the rotor blades of the rotating part 4, but may also extend between the stationary part 2 to which the stator vanes are attached and the rotating part 4 to which the rotor blades are attached). A working fluid W is introduced into the gap D from the outside, thereby driving the rotating part 4. The working fluid W that has driven the rotating part 4 is discharged to the outside via a discharge path (not shown). During operation of the rotary machine 1, at least one of the stationary part 2 and the rotor part 4 may deform due to the influence of heat or the like, causing the gap D to shrink and rubbing to occur. Such rubbing can be detected based on an AE signal detected by an AE sensor 10 (described later).
[0016] The rotary machine 1 includes a plurality of units U arranged along the axial direction (extension direction) of the rotating part 4. In the present embodiment, the rotary machine 1 includes a first unit Ua and a second unit Ub as the plurality of units U. The rotary machine 1 may include any number of units.
[0017] The first unit Ua and the second unit Ub share the rotating part 4 and each have an independent stationary part 2 (first stationary part 2a and second stationary part 2b). That is, the first unit Ua is composed of the first stationary part 2a and the rotating part 4, and the second unit Ub is composed of the second stationary part 2b and the rotating part 4.
[0018] The rotating part 4 is a rotor that can be rotated by any power source, and in this embodiment, power is imparted by the aforementioned working fluid W. The rotating part 4 may be configured as an integral unit, or may be configured by dividing the rotating part 4 into multiple axially separated members that are connected to each other at joints. In the latter case, the multiple members may be made of different materials.
[0019] The rotary machine 1 may use steam as the working fluid W, for example, and each unit U may be configured as a steam turbine. In this case, the flow path of the working fluid W may be independent in each unit U, or the units U may be configured to be connected in series or in parallel. For example, when the flow paths of the working fluid W of each unit U are connected in series, the first unit Ua may be a high-pressure turbine, and the second unit Ub may be a high-pressure turbine that can be driven by steam from the high-pressure turbine.
[0020] Each unit U is provided with an inlet 8 for introducing the working fluid W from the outside into the gap D, and an outlet (not shown) for discharging the working fluid W to the outside after it has completed its work in the gap D. Specifically, the first unit Ua is provided with a first inlet 8a for introducing the working fluid W into the first gap Da, and the second unit Ub is provided with a second inlet 8b for introducing the working fluid W into the second gap Db.
[0021] The rotary machine 1 also has at least one bearing 5 (radial bearing) for supporting the rotating part 4. In this embodiment, the bearings 5 include bearings 5a, 5b, and 5c. The bearing 5a is provided at one end of the rotating part 4, the bearing 5b is provided at an intermediate position of the rotating part 4 (specifically, between the first stationary part 2a constituting the first unit Ua and the second stationary part 2b constituting the second unit Ub), and the bearing 5c is provided at the other end of the rotating part 4 (the opposite side to the bearing 5a). The rotating part 4 is rotatably supported by these multiple bearings 5. As described above, the bearings 5a, 5b, and 5c are each housed in a bearing housing 6 that is stationary relative to the outside.
[0022] The AE sensor 10 is configured to detect an acoustic emission (AE) signal and is attached to a stationary component of the rotating machine 1. In this embodiment, the AE sensor 10 is attached to the bearing housing 6 that houses the bearing 5b. The rotating machine 1 generates AE waves, for example, when a seal or the like attached to the thermally deformed stationary component 2 rubs against the rotating component 4. For example, the AE waves generated at the location where the rubbing occurs propagate as elastic waves through the stationary component 2 and the rotating component 4 and are detected as AE signals by each AE sensor 10 installed in the rotating machine 1. AE waves generally have a frequency in the sound wave range of several tens of kHz to several MHz.
[0023] The AE sensor 10 may be attached to the bearing housing 6 that houses the bearing 5a or 5c, or may be attached to the first stationary portion 2a or the second stationary portion 2b.
[0024] Next, a rubbing determination device 100 for determining rubbing in the rotating machine 1 having the above configuration will be described. The rubbing determination device 100 is composed of, for example, a central processing unit (CPU), a random access memory (RAM), a read-only memory (ROM), 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 the 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.
[0025] Fig. 2 is a block diagram of the rubbing determination device 100 of Fig. 1. The rubbing determination device 100 includes an AE signal acquisition unit 102, a signal analysis unit 103, a first signal intensity ratio calculation unit 104, a second signal intensity ratio calculation unit 106, a first index calculation unit 108, a second index calculation unit 110, and a rubbing determination unit 112.
[0026] 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. In particular, the AE signal acquisition unit 102 is configured to be able to identify a change in the AE signal over time by continuously acquiring the AE signal over time.
[0027] The signal analysis unit 103 is configured to perform spectrum analysis of temporal changes in the AE signal. The first signal strength ratio calculation unit 104 and the second signal strength ratio calculation unit 106 are configured to calculate a first signal strength ratio POA / OA1 and a second signal strength ratio POA / OA2 based on the spectrum analysis results. The first index calculation unit 108 is configured to calculate a first index I1 based on the first signal strength ratio POA / OA1 and the second signal strength ratio POA / OA2.
[0028] A method for calculating the first signal strength L1, the second signal strength L2, and the first index I1 from the temporal change in the AE signal will now be described in detail with reference to Figures 3 to 5. Figure 3 is a flowchart showing the procedure for calculating the first index I1, Figure 4 is a diagram showing an example of the temporal change in the AE signal, and Figure 5 is a diagram showing the first frequency spectrum S1 and the second frequency spectrum S2 corresponding to the temporal change in the AE signal of Figure 4.
[0029] First, the AE signal acquisition unit 102 acquires an AE signal over a predetermined period of time, thereby obtaining a temporal change in the AE signal (step S100). The temporal change in the AE signal is subjected to envelope processing to obtain a corresponding envelope (step S101), and maximum and minimum values are identified (step S102). Fig. 4 shows an example of the temporal change in the AE signal, and shows maximum values Vm (peaks) and minimum values Vn (valleys) identified from the envelope obtained by envelope processing. The maximum values Vm and minimum values Vn appear alternately in time series.
[0030] As the spectral analysis, the signal analysis unit 103 performs a frequency transform (e.g., a Fourier transform) on a first time period T1 that includes the maximum value Vm among the temporal changes in the AE signal, thereby obtaining a corresponding first frequency average spectrum S1 (step S103). The first frequency average spectrum S1 is obtained by averaging the frequency spectra calculated for each of the maximum values Vm identified in step S102. The first time period T1 is defined as a predetermined period that includes the maximum value Vm. For example, the start time t1 and end time t2 of the first time period T1, and the time tm1 at which the maximum value Vm is obtained, can be set so as to satisfy the following equation: tm1 = (t1 + t2) / 2
[0031] Next, the signal analysis unit 103 performs a spectral analysis on a second time period T2 that includes a minimum value Vn in the temporal change of the AE signal, by performing a frequency transform (e.g., a Fourier transform) to obtain a corresponding second frequency average spectrum S2 (step S104). The second frequency average spectrum S2 is obtained by averaging the frequency spectra calculated for each minimum value Vn identified in step S102. The second time period T2 is defined as a predetermined period that includes the minimum value, and can be set, for example, so that the start time t3 and end time t4 of the second time period T2 and the time tm2 at which the maximum value is obtained satisfy the following equation: tm2 = (t3 + t4) / 2
[0032] Next, the first signal strength ratio calculation unit 104 calculates a first signal strength ratio POA / OA1 based on the first frequency average spectrum S1 calculated in step S103 (step S105). In step S105, the first signal strength ratio POA / OA1 is calculated as the ratio between a partial overall value POA, which corresponds to the signal strength in a frequency band higher than the reference frequency fref in the first frequency average spectrum S1, and an overall value OA, which corresponds to the signal strength in all frequency bands of the first frequency average spectrum S1.
[0033] The second signal strength ratio calculation unit 106 then calculates a second signal strength ratio POA / OA2 based on the second frequency average spectrum S2 calculated in step S104 (step S106). In step S106, the second signal strength ratio POA / OA2 is calculated as the ratio between a partial overall value POA, which corresponds to the signal strength in a frequency band higher than the reference frequency fref in the second frequency average spectrum S2, and an overall value OA, which corresponds to the signal strength in the entire frequency band of the second frequency average spectrum S2.
[0034] Next, the first index calculation unit 108 calculates a first index I1 as the ratio of the first signal intensity ratio POA / OA1 to the second signal intensity ratio POA / OA2 ((POA / OA1) / (POA / OA2)) (step S107). The reason why the first index I1 calculated in this way is an effective index for rubbing judgment will be explained below.
[0035] When the phase of the AE signal is converged, the amplitude of the AE signal fluctuates in synchronization with the rotation speed. Specifically, the relatively large amplitude peaks (near the maximum value Vm) of the temporal change in the AE signal correspond to the contact signal, while the relatively small amplitude valleys (near the minimum value Vn) correspond to the non-contact signal. As shown in FIG. 5 , when the first frequency average spectrum S1 and the second frequency average spectrum S2 are calculated by applying a Fourier transform to the AE signals corresponding to the peaks and valleys, the first frequency average spectrum S1 corresponding to the contact (rubbing) state has a higher signal intensity in the high frequency band (e.g., near 175 kHz) than the second frequency average spectrum S2 corresponding to the non-contact state. The first index I1 is an index that quantifies such characteristics.
[0036] FIG. 6 shows an example of the temporal transitions of the amplitude, phase, first index I1 (30-second average), and first index I1 (2-minute average) of the AE signal. In this example, the amplitude of the AE signal increases and the phase converges during periods ta, tb, and tc, suggesting the occurrence of rubbing. Focusing on the behavior of the first index I1, however, the first index I1 also temporarily increases during periods ta, tb, and tc. This indicates that the first index I1 is effective as an index for rubbing assessment. In particular, the behavior of the first index I1 during periods ta, tb, and tc is more pronounced for the 2-minute average than for the 30-second average, improving its usefulness as an index for rubbing assessment. Note that the phase always tends to converge during periods other than periods ta, tb, and tc, indicating that it is not sufficient as an index for rubbing assessment.
[0037] 6, since the amplitude of the AE signal increases when rubbing occurs, some believe that rubbing can be determined without using the first index I1. However, the amplitude of the AE signal varies depending on the output (load) of the rotating machine 1. Therefore, for rubbing determination in the rotating machine 1, including in a transient state, more accurate rubbing determination is possible by using the first index I1 instead of the amplitude of the AE signal.
[0038] 2, the second index calculation unit 110 is configured to calculate a second index I2 that indicates phase convergence in the temporal change of the AE signal. Specifically, the second index I2 is calculated by the following equation using the variance σ of the phase of the AE signal: second index I2=1 / (1+σ^0.5)
[0039] A specific method for calculating the second index I2 will now be described with reference to Fig. 7. Fig. 7 is a flowchart showing a method for calculating the second index I2 by the second index calculation unit 110 of Fig. 2.
[0040] First, the second index calculation unit 110 performs filtering using a predetermined filter on the AE signal acquired by the AE signal acquisition unit 102, and outputs the filtered AE signal so as to have frequency components in the pass band of the filter (step S200). The filtering is performed using a band-pass filter whose pass band is the predetermined frequency band of the AE signal.
[0041] Next, the second index calculation unit 110 performs envelope processing, resampling, and mean-zeroing processing on the AE signal or the filtered AE signal, thereby processing the data (step S201). Specifically, in the envelope processing, envelope processing is performed on the filtered AE signal or the AE signal, and the envelope-processed AE signal is output. In the resampling processing, resampling is performed on the envelope-processed AE signal, and the resampled AE signal is output. In the mean-zeroing processing, processing is performed on the resampled AE signal to set the average value of the amplitude in each period to zero, and the mean-zeroed AE signal is output.
[0042] Next, the second index calculation unit 110 performs frequency analysis on the AE signal that has been subjected to the mean value zeroing process, and outputs a rotation order analysis result F in which the frequency is represented by the rotation order, as shown in Fig. 8 (step S202). Fig. 8 is a graph showing the amplitude of the AE signal for each rotation order.
[0043] Next, the second index calculation unit 110 acquires the 1x rotation speed component C from the rotation order analysis result F, and calculates the second index I2 according to the above formula using the variance σ of the 1x rotation speed component extracted phase P acquired by sampling the 1x rotation speed component C (step S203). This variance σ is found for the 1x rotation speed component extracted phase P, which is found by performing a predetermined sampling on the phase of the 1x rotation speed component C. The 1x rotation speed component extracted phase P is obtained as a phase shift of the period of the 1x rotation speed component C with respect to the period of the rotation speed of the rotating part. The 1x rotation speed component C extracted phase P is acquired by sampling 5 to 10 times at intervals of several seconds, for example.
[0044] The rubbing determination unit 112 is configured to determine whether or not rubbing is present based on at least one of the first index I1 and the second index I2. This determination of rubbing may be made based on either the first index I1 or the second index I2, or may be made based on both the first index I1 and the second index I2.
[0045] Here, a specific description will be given of the rubbing determination method performed by the rubbing determination unit 112. Fig. 9 is a flowchart showing the determination flow in the rubbing determination unit 112 of Fig. 2. In the mode shown in Fig. 9, the rubbing determination unit 112 determines the presence or absence of rubbing based only on the first index I1.
[0046] First, the rubbing determination unit 112 acquires the first index I1 calculated by the first index calculation unit 108 (step S300). The first index calculation unit 108 sequentially calculates the first index I1, and the rubbing determination unit 112 acquires the first index I1 from the first index calculation unit 108 each time.
[0047] Next, the rubbing determination unit 112 acquires a first threshold value corresponding to the first index I1 (step S301). The first threshold value may be a calculated value calculated based on the operating state of the rotary machine 1, or may be a default value stored in advance in a storage device.
[0048] Next, the rubbing determination unit 112 determines whether the first index I1 obtained in step S300 is greater than the threshold value obtained in step S301 (step S302). If the first index I1 is greater than the threshold value (step S302: YES), the rubbing determination unit 112 determines that rubbing is present (step S303). On the other hand, if the first index I1 is equal to or less than the threshold value (step S302: NO), the rubbing determination unit 112 determines that rubbing is not present (step S304). Since such rubbing determination using the first index I1 is effective regardless of phase convergence as described above, it is possible to perform rubbing determination with high accuracy.
[0049] Fig. 10 is a flowchart showing a modification of Fig. 9. In this modification, the rubbing determination unit 112 determines the presence or absence of rubbing based on both the first index I1 and the second index I2.
[0050] First, the rubbing determination unit 112 obtains the first index I1 calculated by the first index calculation unit 108 and the second index I2 calculated by the second index calculation unit 110 (step S400). The first index calculation unit 108 and the second index calculation unit 110 sequentially calculate the first index I1 and the second index I2, and the rubbing determination unit 112 obtains the first index I1 from the first index calculation unit 108 and the second index I2 from the second index calculation unit 110 each time.
[0051] Next, the rubbing determination unit 112 acquires a first threshold value and a second threshold value corresponding to the first index I1 and the second index I2, respectively (step S401). The first threshold value and the second threshold value may be calculated values calculated based on the operating state of the rotary machine 1, or may be default values stored in advance in a storage device.
[0052] Next, the rubbing determination unit 112 determines whether the first index I1 or the second index I2 obtained in step S400 is greater than the first threshold value or the second threshold value obtained in step S401 (step S402). If at least one of the first index I1 or the second index I2 is greater than the first threshold value or the second threshold value (step S402: YES), the rubbing determination unit 112 determines that rubbing is present (step S403). On the other hand, if both the first index I1 and the second index I2 are equal to or less than the first threshold value or the second threshold value (step S402: NO), the rubbing determination unit 112 determines that rubbing is not present (step S404). By using both the first index I1 and the second index I2 in this way, rubbing determination can be performed with higher accuracy than the example described above with reference to FIG. 9.
[0053] 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.
[0054] The contents described in each of the above embodiments can be understood, for example, as follows.
[0055] (1) A rubbing determination device for a rotary machine according to one aspect is a rubbing determination device for determining rubbing in a rotary machine having a rotary part rotatably supported relative to a stationary part by a bearing, the device including: an AE signal acquisition unit for acquiring an AE signal from an AE sensor attached to the stationary part; a signal analysis unit for identifying maximum and minimum values of an envelope obtained by performing envelope processing on a temporal change in the AE signal, and calculating a first frequency average spectrum corresponding to a first time period including the maximum value of the temporal change, and a second frequency average spectrum corresponding to a second time period including the minimum value of the temporal change; a first signal intensity ratio calculation unit for calculating a first signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the first frequency average spectrum; and a second signal intensity ratio calculation unit for calculating a second signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the second frequency average spectrum. The optical disc drive includes a first index calculation unit for calculating a first index as a ratio of the first signal intensity ratio to the second signal intensity ratio, and a rubbing determination unit for determining whether or not rubbing is performed based on the first index.
[0056] According to the above aspect (1), the presence or absence of rubbing is determined based on a temporal change in an AE signal detected by an AE sensor attached to a stationary part of a rotating machine. The temporal change in the AE signal is subjected to envelope processing to create an envelope of the temporal change and identify its maximum and minimum values. A first frequency average spectrum corresponding to a first time period including a maximum value in the temporal change and a second frequency average spectrum corresponding to a second time period including a minimum value in the temporal change are calculated. A first signal intensity ratio is calculated as the ratio of the signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to the ratio of the signal intensity in the entire frequency band of the first frequency average spectrum. A second signal intensity ratio is calculated as the ratio of the signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to the ratio of the signal intensity in the entire frequency band of the second frequency average spectrum. In this aspect, a first index for determining the presence or absence of rubbing is calculated using the first signal intensity ratio and the second signal intensity ratio. The first index calculated in this manner reflects information about the presence or absence of rubbing, regardless of the convergence state of the phase determined from the temporal change. Therefore, by determining the presence or absence of rubbing based on the first index, it is possible to make a reliable determination of rubbing.
[0057] (2) In another aspect, in the aspect (1), the rubbing determination unit determines the presence or absence of rubbing based on a result of averaging the first index over a predetermined period of time.
[0058] According to the aspect (2) above, the first index calculated as described above is averaged over a predetermined period, thereby canceling out random components such as noise contained in the first index. By using the first index from which the random components have been removed in this manner, it becomes possible to make a more accurate rubbing determination.
[0059] (3) In another aspect, in the aspect (1) or (2) above, a second index calculation unit is further provided that calculates a second index that indicates phase convergence in the temporal change, and the rubbing determination unit determines the presence or absence of rubbing based on the first index and the second index.
[0060] According to the above aspect (3), by using the first index based on the signal strength and the second index based on the phase convergence as described above, it is possible to make a more accurate rubbing determination.
[0061] (4) In another aspect, in any one of the above aspects (1) to (3), the stationary portion is a bearing box that houses the bearing.
[0062] According to the above aspect (4), rubbing can be judged with high accuracy based on the AE signal detected by the AE sensor installed in the bearing housing.
[0063] (5) A method for determining rubbing of a rotary machine according to one aspect is a method for determining rubbing in a rotary machine having a rotary part rotatably supported relative to a stationary part by a bearing, the method comprising the steps of: acquiring an AE signal from an AE sensor attached to the stationary part; identifying a maximum value and a minimum value of an envelope obtained by performing envelope processing on a temporal change in the AE signal; calculating a first frequency average spectrum corresponding to a first time period including the maximum value of the temporal change, and a second frequency average spectrum corresponding to a second time period including the minimum value of the temporal change; calculating a first signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the first frequency average spectrum; calculating a second signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the second frequency average spectrum; and calculating a first index as a ratio of the first signal intensity ratio to the second signal intensity ratio. and determining whether or not the rubbing is performed based on the first index.
[0064] According to the above aspect (5), the presence or absence of rubbing is determined based on a temporal change in an AE signal detected by an AE sensor attached to a stationary part of a rotating machine. The temporal change in the AE signal is subjected to envelope processing to create an envelope of the temporal change and identify its maximum and minimum values. A first frequency average spectrum corresponding to a first time period including a maximum value in the temporal change and a second frequency average spectrum corresponding to a second time period including a minimum value in the temporal change are calculated. A first signal intensity ratio is calculated as the ratio of the signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to the ratio of the signal intensity in the entire frequency band of the first frequency average spectrum. A second signal intensity ratio is calculated as the ratio of the signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to the ratio of the signal intensity in the entire frequency band of the second frequency average spectrum. In this aspect, a first index for determining the presence or absence of rubbing is calculated using the first signal intensity ratio and the second signal intensity ratio. The first index calculated in this manner reflects information about the presence or absence of rubbing, regardless of the convergence state of the phase determined from the temporal change. Therefore, by determining the presence or absence of rubbing based on the first index, it is possible to make a reliable determination of rubbing.
[0065] REFERENCE SIGNS LIST 1 Rotating machine 2 Stationary portion 2a First stationary portion 2b Second stationary portion 4 Rotating portion 5 Bearing 6 Bearing housing 8 Inlet portion 8a First inlet portion 8b Second inlet portion 10 AE sensor 100 Rubbing determination device 102 AE signal acquisition portion 104 First signal strength calculation portion 106 Second signal strength calculation portion 108 First index calculation portion 110 Second index calculation portion 112 Rubbing determination portion Ua First unit Ub Second unit D Gap
Claims
1. A rotating machine rubbing determination device for determining rubbing in 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 an AE sensor attached to the stationary part; a signal analysis unit for identifying maximum and minimum values of an envelope obtained by performing envelope processing on a temporal change in the AE signal, and calculating a first frequency average spectrum corresponding to a first time period including the maximum value of the temporal change, and a second frequency average spectrum corresponding to a second time period including the minimum value of the temporal change; a first signal intensity ratio calculation unit for calculating a first signal intensity ratio as a ratio of signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to a ratio of signal intensity in an entire frequency band of the first frequency average spectrum; and a second signal intensity ratio calculation unit for calculating a second signal intensity ratio as a ratio of signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to a ratio of signal intensity in an entire frequency band of the second frequency average spectrum. a first index calculation unit for calculating a first index as a ratio of the first signal intensity ratio to a second signal intensity ratio; and a rubbing determination unit for determining the presence or absence of rubbing based on the first index.
2. The rubbing determination device for a rotating machine according to claim 1, wherein the rubbing determination unit determines the presence or absence of rubbing based on a result of averaging the first index over a predetermined period of time.
3. A rubbing determination device for a rotating machine as described in claim 1 or 2, further comprising a second index calculation unit that calculates a second index that indicates phase convergence in the time change, and the rubbing determination unit determines the presence or absence of rubbing based on the first index and the second index.
4. A rubbing determination device for a rotating machine as claimed in claim 1 or 2, wherein the stationary part is a bearing box that houses the bearing.
5. A method for determining rubbing in 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 an AE sensor attached to the stationary part; identifying a maximum value and a minimum value of an envelope obtained by performing envelope processing on a time change of the AE signal; calculating a first frequency average spectrum corresponding to a first time period including the maximum value of the time change, and a second frequency average spectrum corresponding to a second time period including the minimum value of the time change; calculating a first signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the first frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the first frequency average spectrum; calculating a second signal intensity ratio as a ratio of a signal intensity in a frequency band higher than a reference frequency of the second frequency average spectrum to a ratio of a signal intensity in an entire frequency band of the second frequency average spectrum; calculating a first index as a ratio of the first signal intensity ratio to the second signal intensity ratio; and determining the presence or absence of rubbing based on the first index. A method for determining rubbing in a rotating machine, comprising:
Citation Information
Patent Citations
Device and method for diagnosing audio and vibration for rotary machine
JP1995182035A
Rubbing judging method and apparatus for rotary machine
JP1996261817A
Diagnostic method and device of sliding bearing
JP2011180082A
Device and method for detecting rubbing of rotary machine
JP2021076533A
