Acoustic signal device, acoustic signal processing system, acoustic signal processing method, and acoustic signal processing program

Abstract

Automatically sets the lower voltage limit for the sliding noise signal. [Solution] The system includes an input unit 41 that receives a sliding sound signal output by an acoustic sensor 2 attached to the valve 1a that detects the sliding sound of the valve 1a; an envelope generation unit 11 that generates an envelope of the sliding sound signal; a threshold voltage determination unit 12 that divides the space between the maximum and minimum absolute voltages of the first envelope generated by the envelope generation unit 11 to determine a plurality of threshold voltages; a counting unit 13 that counts the number of data points that take a value greater than or equal to the threshold voltage for each threshold voltage; a maximum gradient point determination unit 14 that determines the maximum gradient point of the curve of the number of data points with respect to the threshold voltage; and a lower limit voltage setting unit 15 that sets the lower limit voltage of the sliding sound signal to be analyzed based on the threshold voltage at the maximum gradient point.

Description

Technical Field

[0001] The present invention relates to an acoustic signal device, an acoustic signal processing system, an acoustic signal processing method, and an acoustic signal processing program, and for example, relates to a preprocessing device that is executed before analyzing the sliding sound of a valve.

Background Art

[0002] A valve having a function of closing the flow of a fluid stops the flow of the fluid by closing the flow path between the valve body and the valve seat by a drive unit, and also prevents the fluid from leaking from the valve. For example, in an electric valve, elements including a seat portion, which is a sliding portion between the valve body and the valve seat, interlock to stop the flow of the fluid, but sliding occurs in these elements. In these sliding portions, deterioration of the sliding portions due to friction occurs, and the risk of leakage from the seat portion and the loss of driving force increase.

[0003] Since the above sliding portion deteriorates due to sliding according to the number of operating times, valves with advanced sliding deterioration are extracted by disassembly inspection, and component replacement and maintenance are performed to prevent problems in the valves. However, since a lot of labor and cost are required for disassembly inspection, a device for determining sliding deterioration is useful.

[0004] Patent Document 1 discloses a technique for detecting the valve seat sliding sound of the sliding portion between the valve body and the valve seat, obtaining the number of operating times or the operating time of the sliding portion from the spectral intensity of the detected valve seat sliding sound, and estimating the leakage amount, surface roughness, or friction coefficient of the sliding portion. Patent Document 2 also discloses a technique for calculating AE energy corresponding to an elastic wave signal, calculating the fractal dimension of the AE energy, and outputting a particle size distribution constant of raw materials for steelmaking.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

[0006] The sliding sound signal contains not only the sound component of the sliding between the valve body and valve seat (sliding component), but also a noise component. The peak voltage of the sliding component and the voltage of the noise component differ depending on the valve and the installation environment. Therefore, it is not possible to set a fixed value for the lower limit voltage required to cut the noise component from the sliding sound signal. In other words, it is preferable to be able to automatically set an appropriate value for the lower limit voltage required to cut the noise from the sliding sound signal.

[0007] This invention has been made in view of these circumstances, and aims to provide an acoustic signal device, an acoustic signal processing system, an acoustic signal processing method, and an acoustic signal processing program that can automatically set the lower limit voltage of a sliding sound signal. [Means for solving the problem]

[0008] To achieve the above objective, the acoustic signal processing device of the present invention is characterized by comprising: an input unit that receives a sliding sound signal output by an acoustic sensor attached to a valve and used to detect the sliding sound of the valve; an envelope generation unit (envelope calculation unit, envelope detection unit) that generates an envelope of the sliding sound signal; a threshold voltage determination unit that divides the space between the maximum and minimum absolute voltages of the first envelope generated by the envelope generation unit and determines one or more first threshold voltages; a counting unit that counts the number of first data points that take a value equal to or greater than the first threshold voltage for each of the first threshold voltages; a maximum gradient point determination unit that determines the first maximum gradient point of the curve of the number of first data points with respect to the first threshold voltage; and a lower limit voltage setting unit that sets the lower limit voltage of the sliding sound signal to be analyzed based on the first threshold voltage at the first maximum gradient point. Note that the symbols and characters in parentheses are symbols and the like that used in embodiments and do not limit the present invention. [Effects of the Invention]

[0009] According to the present invention, the lower limit voltage of the sliding sound signal can be automatically set. [Brief explanation of the drawing]

[0010] [Figure 1] This is a diagram showing the configuration of an acoustic signal system, which is a first embodiment of the present invention. [Figure 2] This diagram shows the cross-section of the valve and the installation location of the acoustic sensor. [Figure 3] This diagram shows the cross-section of a manual globe valve and the installation location of the acoustic sensor. [Figure 4] This figure shows an example of a sliding sound signal used in the first embodiment of the present invention. [Figure 5] This waveform shows an example of a sliding sound signal and its envelope used in the first embodiment of the present invention. [Figure 6] This diagram shows the relationship between threshold voltage and the number of data points. [Figure 7] This diagram shows the relationship between threshold voltage and the number of data points, illustrating the characteristics before (before degradation) and after degradation. [Figure 8] This waveform shows an example of a sliding sound signal used when performing fractal analysis. [Figure 9] Figure (1) shows an example of fractal analysis. [Figure 10] Figure (1) shows an example of fractal analysis. [Figure 11] This is a flowchart illustrating the acoustic signal processing method performed by an acoustic signal processing device, which is a first embodiment of the present invention. [Figure 12] This waveform shows an example of a sliding sound signal and its envelope used in a second embodiment of the present invention. [Modes for carrying out the invention]

[0011] The following description is an explanation of an embodiment of the present invention (hereinafter referred to as "this embodiment") with reference to the drawings. Note that each drawing only schematically shows the embodiment to an extent that enables sufficient understanding. Also, in each drawing, common components and similar components are denoted by the same reference numerals, and redundant explanations thereof are omitted.

[0012] (First Embodiment) FIG. 1 is a configuration diagram of an acoustic signal system according to the first embodiment of the present invention. The acoustic signal system 200 includes a valve 1a with an acoustic sensor 2 attached thereto and an acoustic signal device 100. The valve 1a is an electric shut-off valve including a valve body 29 (FIG. 2), a valve seat 30 (FIG. 2), an acoustic sensor 2, a valve stem 4, and a drive unit 5. The acoustic sensor 2 detects acoustic emission (elastic wave) generated by the opening / closing operation of the valve as sliding sound at a sliding part such as the contact surface between the valve body 29 and the valve seat 30. Note that the drive unit 5 is driven by a motor 22 (FIG. 2).

[0013] The acoustic signal device 100 performs spectral analysis on the sliding sound component from the sliding sound signal output by the acoustic sensor 2, and performs fractal analysis after removing noise. The acoustic signal device 100 estimates the deterioration state of the valve 1, performs abnormality determination, and estimates the maintenance timing based on the analysis result. The acoustic sensor 2 is an acoustic emission sensor, a piezoelectric sensor, an ultrasonic sensor, or an acceleration sensor.

[0014] FIG. 2 is a diagram showing the cross-section of the valve and the installation position of the acoustic sensor. At the sliding part between the valve body 29 and the valve seat 30, sliding occurs during the opening / closing operation of the valve 1a. Therefore, the acoustic sensor 2 is installed at a position on the outer surface of the valve 1a that serves as the propagation path of the direct wave of the sliding sound in the vicinity of these sliding parts.

[0015] Note that in addition to between the valve body 29 and the valve seat 30, the sliding parts include the contact surface of the screw threads between the worm gear 24 and the valve stem screw part 25, and the joint surface between the valve stem 4 and the gland packing 28. The acoustic sensor 2 is also attached to these parts.

[0016] When the acoustic sensor 2 is installed directly near the sliding part, a couplant 21 is interposed between the acoustic sensor 2 and the detection surface as an acoustic transmission medium. The material of the couplant 21 may be grease, wax, adhesive, lubricating oil, water, gel, etc.

[0017] Figure 2 illustrates an electrically operated gate valve (valve 1a), but the acoustic sensor 2 can also be installed on a manually operated globe valve.

[0018] Figure 3 shows a cross-section of the manual globe valve and the installation position of the acoustic sensor 2. When valve 1b, which is a manual globe valve, is closed from the open position, rotating the manual handle 23 in the closing direction causes the valve stem 4 to descend, resulting in sliding at the sliding portion between the valve body 29 and the valve seat 30. The acoustic sensor 2 is installed on the handle and the side of the valve body and detects the sound of the manual handle 23 being operated (handle operation sound) and the sliding sound of the sliding parts of the valve body 29 and the valve seat 30 (valve seat sliding sound).

[0019] Furthermore, the acoustic sensor on the manual handle 23 detects not only the sound of the handle being operated, but also the acoustic waves transmitted by the sliding sound generated at the valve seat. In addition, the acoustic sensor 2 on the side of the valve body detects not only the sliding sound at the valve seat, but also the acoustic waves transmitted by the sound of the handle being operated by the manual handle 23.

[0020] Returning to the explanation of Figure 1, the acoustic signal device 100 is configured to include a signal processing unit 20, an input unit 41, a display operation unit 42, and a control unit 10.

[0021] The signal processing unit 20 is a hardware processing unit that processes the sliding sound signals output by the acoustic sensors 2 attached to the valves 1a (Figure 2) and 1b (Figure 3). The signal processing unit 20 is composed of an amplification unit 6 that amplifies the sliding sound signals to a predetermined amplitude level, a filter 7, and an A / D converter 8 that converts the output signal of the acoustic sensors 2 from analog to digital. The filter 7 allows ultrasonic frequencies of 10 kHz or higher to pass through, for example. The filter 7 may also be a bandpass filter that cuts out elastic waves, vibration components, noise, etc. in a specific frequency range. The input unit 41 is an interface that inputs the digital signal output by the A / D converter 8 to the control unit 10. The input unit 41 may also temporarily store the digital signal output by the A / D converter 8 in a non-volatile memory unit.

[0022] Figure 4 shows an example of a sliding sound signal. The horizontal axis represents time (s), and the vertical axis represents amplitude voltage (mV). The sliding sound signal input via the input unit 41 is a signal having a peak (sliding component) with an amplitude of approximately ±15mV, accompanied by noise components (background noise signals) with an amplitude of approximately ±3mV before and after it. The acoustic signal device 100 of this embodiment aims to remove the noise component from the sliding sound signal. However, since the relationship between the peak voltage of the sliding component of the sliding sound signal and the voltage of the noise component differs depending on the valves 1a, 1b and the installation environment, it is not possible to cut the noise component by setting a predetermined fixed value as the lower limit voltage.

[0023] Returning to the explanation of Figure 1, the control unit 10 is a CPU (Central Processing Unit) and implements the following functions by executing an acoustic signal processing program. Specifically, the control unit 10 implements the functions of an envelope generation unit 11, a threshold voltage determination unit 12, a counting unit 13, a maximum gradient point determination unit 14, a lower limit voltage setting unit 15, a fractal analysis unit 16, a spectral analysis unit 17, a deterioration state estimation unit 18, an abnormality determination unit 19, and a maintenance timing estimation unit 43. The sliding sound signal extraction unit 44, shown by a dashed line, will be explained in the second embodiment. The fractal analysis unit 16 and the spectral analysis unit 17 function as sliding sound analysis units.

[0024] The envelope generation unit 11 is an envelope calculation unit that calculates the root mean square (rms) of the envelope of the sliding sound signal detected by the acoustic sensor 2 within a predetermined range. This predetermined range is a predetermined time T1 (Figures 4 and 5, for example, T1 = 5 mSec) and is a range that includes the sliding component (peak). In other words, the calculated envelope is obtained by dividing the detected sliding sound signal into a predetermined range. Here, an envelope is a curve that is tangent to all of a group of curves that follow certain conditions. The equation of the envelope of a group of curves represented by the equation f(x,y,c)=0 including one parameter c can generally be obtained by eliminating c from f=0 and ∂f / ∂c=0. Note that the envelope generation unit 11 may be replaced by a hardware detection unit (envelope detection unit (not shown)) inserted between the filter 7 and the A / D converter 8 inside the signal processing unit 20.

[0025] Figure 5 shows an example waveform of a sliding sound signal and its envelope. The vertical axis represents amplitude voltage (V), and the horizontal axis represents time (s). The sliding sound signal shown by the solid line is, for example, an AC waveform with positive and negative peaks, accompanied by noise before and after the sliding component (peak). The dashed line is the root mean square of the envelope of the sliding sound signal. As shown in Figure 5, the envelope removes very small noise signals that are not sliding sound signals but background noise signals caused by the surrounding environment, etc., whose signal intensity is less than or equal to the root mean square of the envelope. In other words, by using the envelope, low-intensity noise signals can be removed. Therefore, analyzing the sliding sound signal with the envelope allows for more accurate analysis compared to directly analyzing the acquired signal.

[0026] The threshold voltage determination unit 12 determines the maximum value (maximum voltage Vmax) and minimum value (minimum voltage Vmin) of the calculation result (first envelope within a predetermined range) calculated by the envelope generation unit 11. The threshold voltage determination unit 12 may also subtract the root mean square of the first envelope from the magnitude of the sliding sound signal and determine the maximum voltage Vmax and minimum voltage Vmin of the subtracted value. In this case, the fractal analysis unit 16 can perform fractal analysis on the signal from which low-intensity noise signals have been removed. The threshold voltage determination unit 12 further divides the space between the maximum voltage Vmax and the minimum voltage Vmin into equal intervals. Each of the divided boundary voltages is called a threshold voltage. For each threshold voltage determined by the threshold voltage determination unit 12, the counting unit 13 counts the number of data points that take a value equal to or greater than the threshold voltage at a predetermined time T1. The maximum gradient point determination unit 14 plots the multiple threshold voltages determined by the threshold voltage determination unit 12 on the horizontal axis (threshold voltage axis) and the number of data points counted by the counting unit 13 on the vertical axis (number of data points axis).

[0027] Figure 6 shows the relationship between threshold voltage and the number of data points. The horizontal axis represents the threshold voltage (mV), and the vertical axis represents the number of data points (×10). 6 ) Furthermore, the number of division points divided in the voltage direction by the threshold voltage determination unit 12 is 1000 divisions. Also, the threshold voltage (horizontal axis) uses the value obtained by removing very small noise signals with a signal strength less than or equal to the root mean square of the envelope. The maximum gradient point at this time is with a threshold voltage V1 = 3.6mV and 8 × 10 data points.6 This is the point. Note that below a threshold voltage of 3mV, it is the noise region 91. Also, the number of data points saturates, and the saturation point is approximately 16 × 10⁻¹⁶. 6 In addition, Figure 6 shows parallel lines (referred to as "saturation parallel lines") that are parallel to the threshold voltage axis and pass through the saturation point of the number of data points.

[0028] The maximum gradient point determination unit 14 determines the maximum gradient point where the slope of the plotted data points is greatest. This determines the threshold voltage V1 of the maximum gradient point. At this time, the threshold voltage at which the tangent to the maximum gradient point intersects the threshold voltage axis is defined as V2. Also, the threshold voltage at which the tangent to the maximum gradient point intersects the saturation parallel line is defined as V3. At this time, signals with voltage values ​​of V2 or higher are generally the sliding sound signals to be acquired, and signals with voltage values ​​of V3 or lower are generally the noise signals to be removed. The lower limit voltage setting unit 15 sets a lower limit voltage VL that limits the lower limit of the sliding sound signal, using the threshold voltage V1 as the reference value (threshold voltage reference value) (setting the lower limit voltage of the sliding sound signal to be analyzed based on the threshold voltage at the maximum gradient point).

[0029] In Figure 6, which shows the relationship between threshold voltage and the number of data points, the threshold voltage value used is the value obtained by removing very small noise signals by taking the envelope. Therefore, in Figure 6, the threshold voltage V1 at the maximum gradient point is shifted to a higher voltage value than in the figure (not shown) which shows the relationship between the voltage value of the sliding sound signal itself and the number of data points. Consequently, it is necessary to correct the shifted voltage to the equivalent of the maximum gradient voltage value of the original acquired signal, and various methods can be considered for this. For example, as a result of repeated experimental trials, it was found that in many cases, setting the lower limit voltage VL in the lower limit voltage setting unit 15 to 0.4 to 0.6 times the threshold voltage reference value (threshold voltage V1) results in appropriate correction. Also, in the relationship between the threshold voltage and the number of data points, including the shift, based on the envelope in Figure 6, signals with a voltage value higher than the threshold voltage V2 at which the tangent of the maximum gradient point passes through the threshold voltage axis are sliding component signals and not noise, so the lower limit voltage setting unit 15 may set the threshold voltage V2 to the lower limit voltage VL of the sliding sound signal. Furthermore, signals with voltage values ​​greater than the threshold voltage V3 will have a smaller proportion of noise components, even if they are present. In other words, the lower limit voltage setting unit 15 can set the lower limit voltage VL to any voltage value including the maximum gradient point threshold voltage V1, which lies between the threshold voltage V2 where the tangent to the maximum gradient point passes through the threshold voltage axis and the threshold voltage V3 where it passes through the saturation parallel line.

[0030] Figure 7 shows the relationship between threshold voltage and the number of data points, illustrating the characteristics before (before degradation) and after degradation. Similar to Figure 6, the horizontal axis represents the threshold voltage (V), and the vertical axis represents the number of data points. The initial state of the valve (before degradation) is plotted with "○", and the state after valve degradation is plotted with "△". As the valve degrades, the threshold voltage increases, and the slope becomes slightly gentler. In other words, as the valve degrades, the threshold voltage at the point of maximum slope increases. Note that in Figure 7, data in the noise region 91 has been manually removed.

[0031] The fractal analysis unit 16 (Figure 1) performs fractal analysis on sliding sound signals that are above the lower limit voltage VL (VL=0.4~0.6V1 or VL=V3~V2) set by the lower limit voltage setting unit 15. As a result, fractal analysis is performed on the sliding sound signal from which noise has been removed.

[0032] Figure 8 shows an example waveform of a sliding sound signal used when performing fractal analysis. Similar to Figure 4, the horizontal axis represents time (s), and the vertical axis represents amplitude voltage (mV). Also, unlike Figure 4, the sliding region includes multiple peaks, with the maximum peak voltage being approximately ±2mV.

[0033] Figures 9 and 10 show examples of fractal analysis. The vertical axis represents the ring-down count, which is the number of peaks above the threshold voltage. The horizontal axis represents the threshold voltage (V). Figure 9 shows the embodiment in which noise is cut at the lower voltage limit VL, while Figure 10 shows the embodiment in which noise is not cut. In other words, the noise region 91 plotted in Figure 10 is not plotted in the same region in Figure 9.

[0034] Returning to the explanation of Figure 1, the spectral analysis unit 17 is a functional unit that analyzes the frequency spectrum of the sliding sound signal. The deterioration state estimation unit 18 estimates the deterioration state (e.g., surface roughness) of the valve body 29 (Figure 2) and valve seat 30 (Figure 2) based on the analysis results of the fractal analysis unit 16 and the spectral analysis unit 17. The abnormality determination unit 19 determines whether the valve body 29 and valve seat 30 are abnormal based on the analysis results of the fractal analysis unit 16 and the spectral analysis unit 17. The maintenance timing estimation unit 43 estimates the maintenance timing of the valve 1 based on the deterioration state estimated by the deterioration state estimation unit 18.

[0035] The display operation unit 42 is a touch-enabled LCD (Liquid Crystal Display) panel. The display operation unit 42 displays the analysis results from the fractal analysis unit 16 and the spectral analysis unit 17, the abnormality determination results from the abnormality determination unit 19, and the estimated results from the maintenance timing estimation unit 43.

[0036] Figure 11 is a flowchart illustrating the acoustic signal processing method performed by an acoustic signal processing device, which is a first embodiment of the present invention. When a sliding sound signal is input, the control unit 10 (Figure 1) calculates the envelope of the sliding sound signal (step S1). After processing in step S1, the control unit 10 divides the space between the maximum voltage Vmax (Figure 5) and minimum voltage Vmin (Figure 5) of the root mean square of the envelope into equal intervals and determines the divided voltages (multiple threshold voltages) (step S2). Alternatively, the control unit 10 may subtract the root mean square of the envelope from the sliding sound signal, divide the space between the maximum voltage Vmax (Figure 5) and minimum voltage Vmin (Figure 5) of the subtracted value into equal intervals and determine the divided voltages (multiple threshold voltages). After processing in step S2, the control unit 10 counts the number of data points above each threshold voltage determined in step S2 (step S3). After processing in step S3, the control unit 10 determines the point of the maximum slope of the curve plotted with the horizontal axis representing the threshold voltage and the vertical axis representing the number of data points (S4). After processing in S4, the control unit 10 determines the lower limit voltage VL of the sliding sound signal based on the threshold voltage V1 of the maximum gradient point (S5). The lower limit voltage VL is set, for example, to 0.4 to 0.6 times the threshold voltage V1. The control unit 10 can set the lower limit voltage VL between the threshold voltage V1 of the maximum gradient point and the threshold voltage V2 at which the tangent to the maximum gradient point passes through the threshold voltage axis. After processing in step S5, the control unit 10 performs fractal analysis (S6).

[0037] As explained above, the acoustic signal device 100 can automatically set the lower limit voltage VL for removing noise contained in the sliding sound signal to an appropriate value. In other words, even when the peak voltage and noise level of the sliding sound signal are unknown and the lower limit voltage VL of the sliding sound signal cannot be fixed, the acoustic signal processing device 100 can automatically set an appropriate lower limit voltage VL by using the point of maximum slope of the curve showing the relationship between the threshold voltage and the number of data points as a reference. Furthermore, the acoustic signal device 100 can perform fractal analysis and spectral analysis using the sliding sound signal from which noise below the lower limit voltage VL has been cut.

[0038] (Second Embodiment) In the first embodiment described above, noise below the lower limit voltage VL was cut from the sliding sound signal. However, it is also possible to cut out the sliding sound signal so that it includes the envelope crossover time T3 (Figure 12), thereby cutting out the noise before and after the crossover.

[0039] The configuration of the second embodiment is the same as that of the first embodiment (Figure 1), but differs in that a sliding sound signal extraction unit 44, shown by a dashed line, is added. In this embodiment as well, similar to the first embodiment, the envelope generation unit 11 (Figure 1) divides the sliding sound signal at a predetermined time T1 (for example, T1 = 5 mSec) to generate an envelope (first envelope) shown by a dashed line in Figure 5. The threshold voltage determination unit 12 determines a plurality of threshold voltages. The counting unit 13 counts the number of data points that are greater than or equal to each threshold voltage. The maximum gradient point determination unit 14 determines the maximum gradient point (Figure 6) where the slope of the number of data points with respect to the threshold voltage is maximum.

[0040] Figure 12 shows a waveform of a sliding sound signal used in a second embodiment of the present invention, and an example of its envelope. The solid line (effectively blacked out) represents the sliding sound signal, and the dashed line is its envelope. Similar to the first embodiment (Figure 4), the sliding sound signal has noise components before and after the sliding component, which has positive and negative peaks.

[0041] In this embodiment, after the maximum gradient point determination unit 14 determines the maximum gradient point (Figure 6), the envelope generation unit 11 (Figure 1) divides the sliding sound signal over a predetermined time T2 >> T1 (for example, when T1 (Figures 4, 5) = 5 msec, T2 = 50 msec) to generate an envelope (second envelope) shown by the dashed line in Figure 12 (a second envelope is generated that divides the signal for a longer period than the first envelope). The intersection points (two intersection points at times t2 and t3) of the second envelope and the lower limit voltage VL (for example, threshold voltage V1) set based on the threshold voltage V1 of the maximum gradient point (Figure 6) calculated by the maximum gradient point determination unit 14 are determined. This (t3-t1)=T3 is called the envelope crossing time. The sliding sound signal extraction unit 44 extracts a time period from the sliding sound signal that includes the envelope crossing time T3 (for example, the time T4 from 0.5 seconds before and after the envelope crossing time T3 = (t2-t1) = (t4-t3)) (the sliding sound signal is extracted so as to include the time of the two intersection points where the lower limit voltage VL and the second envelope intersect).

[0042] As described above, according to this embodiment, the acoustic signal device 100 can extract the sliding sound signal (sliding region) at the envelope crossing time T3 without setting a lower limit voltage VL of the sliding sound signal. Furthermore, even when there are many peaks in the first envelope within the envelope crossing time T3, it can be correctly extracted.

[0043] (modified version) The present invention is not limited to the embodiments described above, and various modifications are possible, for example, as follows. (1) In the above embodiment, the envelope of the root mean square (rms) of the sliding sound signal was calculated. Alternatively, the absolute value of the envelope of the sliding sound signal could be used instead of calculating the root mean square. In other words, the threshold voltage determination unit 12 would determine the maximum voltage Vmax and minimum voltage Vmin of the absolute value of the first envelope. The threshold voltage determination unit 12 may also subtract the absolute value of the first envelope from the magnitude of the sliding sound signal and determine the maximum voltage Vmax and minimum voltage Vmin of the subtracted value. [Explanation of Symbols]

[0044] 1 valve 1a Valve (electric gate valve) 1b Valve (Manual Globe Valve) 2 Acoustic sensors 4 valve stems 5 Drive Unit 6. Amplification section (signal processing section) 7. Filters (signal processing unit, high-pass filter) 8. A / D converter (signal processing unit) 10 Control Unit 11 Envelope generation section 12. Threshold voltage determination unit 13 Counting section 14. Maximum gradient point determination section 15 Lower limit voltage setting section 16. Fractal Analysis Section (Sliding Sound Analysis Section) 17. Spectrum Analysis Unit (Sliding Sound Analysis Unit) 18. Deterioration state estimation unit 19 Abnormality determination section 20 Signal Processing Unit 21 Couprants 22 motors 23 Manual handle 24 Worm Gear 25 Valve stem threaded section 28 Gland Packing 29 Valve body 30 valve seats 41 Input section 42 Display operation section 43 Maintenance timing estimation unit 44 Sliding sound signal extraction part 91 Noise Region 100 Acoustic signaling device 200 Acoustic Signal Systems

Claims

1. An input section that receives the sliding sound signal output by an acoustic sensor attached to the valve that detects the sliding sound of the valve, An envelope generation unit that generates the envelope of the sliding sound signal, A threshold voltage determination unit divides the range between the maximum and minimum absolute voltages of the first envelope generated by the envelope generation unit and determines a plurality of threshold voltages, A counting unit that counts the number of data points that take a value equal to or greater than the threshold voltage for each of the threshold voltages, A maximum gradient point determination unit that determines the maximum gradient point of the curve of the number of data points with respect to the threshold voltage, A lower limit voltage setting unit sets the lower limit voltage of the sliding sound signal to be analyzed based on the threshold voltage at the point of maximum gradient. An acoustic signal processing device characterized by comprising:

2. The lower limit voltage setting unit sets the lower limit voltage to an intermediate voltage between the threshold voltage at which the tangent to the maximum gradient point intersects a saturation parallel line, which is parallel to the axis of the threshold voltage and passes through the saturation point of the number of data points, and the threshold voltage at which the tangent to the maximum gradient point intersects the axis of the threshold voltage. The acoustic signal processing device according to feature 1.

3. The threshold voltage determination unit determines multiple threshold voltages by dividing the range between the maximum and minimum voltages of the subtraction value obtained by subtracting the absolute value of the first envelope generated by the envelope generation unit from the magnitude of the sliding sound signal. The lower limit voltage setting unit sets the lower limit voltage to a value between 0.4 and 0.6 times the threshold voltage of the maximum gradient point. The acoustic signal processing device according to feature 1.

4. The envelope generation unit divides the sliding sound signal in time and outputs an envelope for each divided sliding sound signal. The threshold voltage determination unit divides the range between the maximum and minimum absolute voltages of the first envelope into equal intervals and determines a plurality of threshold voltages. The acoustic signal processing device according to feature 1.

5. The envelope generation unit generates a second envelope which is divided to be longer than the first envelope. The sliding sound signal extraction unit further comprises a sliding sound signal extraction unit that extracts the sliding sound signal such that the sliding sound signal includes the time of two intersection points where the sliding sound signal is extracted by the sliding sound signal extraction unit, which extracts the sliding sound signal such that extraction unit includes the time of two intersection points where the sliding sound signal extraction unit, which has been set by the sliding sound signal extraction unit based on the threshold The acoustic signal processing device according to feature 4.

6. The system further includes a fractal analysis unit that performs fractal analysis on sliding sound signals whose absolute value is equal to or greater than the lower limit voltage. The acoustic signal processing apparatus according to any one of claims 1 to 4.

7. The envelope generation unit calculates the root mean square of the sliding sound signal and uses the calculation result as the envelope. The acoustic signal processing device according to feature 1.

8. The envelope generation unit is an envelope detection unit that performs envelope detection on the sliding sound signal. The acoustic signal processing device according to feature 1.

9. An acoustic sensor attached to the valve to detect the sound of the valve sliding, An envelope generation unit that generates the envelope of the sliding sound signal output by the acoustic sensor as the first envelope, A threshold voltage determination unit divides the range between the maximum and minimum absolute voltages of the first envelope generated by the envelope generation unit and determines a plurality of threshold voltages, A counting unit that counts the number of data points that take a value equal to or greater than the threshold voltage for each of the threshold voltages, A maximum gradient point determination unit that determines the maximum gradient point of the curve of the number of data points with respect to the threshold voltage, A lower limit voltage setting unit sets the lower limit voltage of the sliding sound signal to be analyzed based on the threshold voltage at the point of maximum gradient. An acoustic signal processing system characterized by comprising the following features.

10. An acoustic signal processing method performed by the control unit of an acoustic signal processing device, An envelope generation step in which an envelope of the sliding sound signal output by an acoustic sensor attached to the valve and used to detect the sliding sound of the valve is generated as the first envelope, A threshold voltage determination step involves dividing the range between the maximum and minimum absolute voltages of the first envelope generated in the envelope generation step and determining a plurality of threshold voltages, A counting step in which the number of data points that take a value equal to or greater than the threshold voltage is counted for each threshold voltage, A maximum gradient point determination step, which determines the maximum gradient point of the curve of the number of data points with respect to the threshold voltage, A lower limit voltage setting step, which sets the lower limit voltage of the sliding sound signal to be analyzed based on the threshold voltage at the point of maximum gradient, and A method for processing acoustic signals, characterized by performing the following:

11. An acoustic signal processing program to be executed by the control unit of an acoustic signal processing device, An envelope generation step in which an envelope of the sliding sound signal output by an acoustic sensor attached to the valve and used to detect the sliding sound of the valve is generated as the first envelope, A threshold voltage determination step involves dividing the range between the maximum and minimum absolute voltages of the first envelope generated in the envelope generation step and determining a plurality of threshold voltages, A counting step in which the number of data points that take a value equal to or greater than the threshold voltage is counted for each threshold voltage, A maximum gradient point determination step, which determines the maximum gradient point of the curve of the number of data points with respect to the threshold voltage, A lower limit voltage setting step, which sets the lower limit voltage of the sliding sound signal to be analyzed based on the threshold voltage at the point of maximum gradient, and A sound signal processing program characterized by causing the following to be executed.

Images