Vibration monitoring system, method, and program

Abstract

To provide a technique of monitoring vibration of a fluid machine from the start of rotation of an impeller to a maximum rotational frequency with one vibration sensor.SOLUTION: A vibration monitoring system includes: a storage unit that stores time-series waveform signal data measured by a vibration sensor that measures vibration of a fluid machine; a frequency analysis unit that performs frequency analysis of the waveform signal data and outputs the intensity of the signal at each frequency; a weighting unit that performs weighting processing on the intensity based on the corresponding frequency and outputs a value obtained from the weighting; and an output unit that outputs the frequency and the value obtained from the weighting.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] An embodiment of the present invention relates to a technique for monitoring vibrations of a fluid machine. [Background technology]

[0002] A non-positive displacement pump, which is a type of fluid machinery, uses the power of a motor to rotate an impeller inside a casing, giving energy to the fluid. A centrifugal pump is an example of a non-positive displacement pump. A volute pump, which is a type of centrifugal pump, transports the liquid by passing it through a rotating impeller inside a casing and applying pressure using centrifugal force.

[0003] Such pumps need to monitor the vibration caused by the rotation of the impeller, and with the rise of the Internet of Things (IoT), there is a demand for remote monitoring of pump vibration.

[0004] Patent Document 1 discloses a motor anomaly detection system that uses measurement data of a motor under normal conditions measured by multiple vibration sensors as a reference, compares the correlation between each measurement data over a period in which the motor's rotation speed is gradually changed, and detects abnormalities such as scratches on the motor's bearings. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 2016-003875 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in Patent Document 1, the measurement data of the motor during normal operation contains noise, and there is a possibility that false detection occurs due to noise when the surrounding environment changes. In addition, there is a problem that multiple sensors are required to determine whether or not there is noise.

[0007] The present invention was completed through extensive research focusing on these issues, and its purpose is to provide a technology that can monitor the vibrations of a fluid machinery from the start of rotation of the impeller to the maximum rotational frequency using a single vibration sensor. [Means for solving the problem]

[0008] In order to solve the above problems, the present invention provides a vibration monitoring system comprising a memory unit that stores time-series waveform signal data measured by a vibration sensor that measures vibrations of a fluid machinery, a frequency analysis unit that performs frequency analysis on the waveform signal data and outputs the signal intensity for each frequency, a weighting unit that weights the intensity with the corresponding frequency and outputs the weighted value, and an output unit that outputs the frequency and the weighted value.

[0009] Another invention is a vibration monitoring method comprising a frequency analysis step of frequency-analyzing time-series waveform signal data measured by a vibration sensor that measures vibrations of a fluid machine and outputting the signal intensity for each frequency, a weighting step of weighting the intensity with the corresponding frequency and outputting the weighted value, and an output step of outputting the frequency and the weighted value.

[0010] Another invention is a vibration monitoring program executable by a computer to perform a frequency analysis step of frequency-analyzing time-series waveform signal data measured by a vibration sensor that measures vibrations of a fluid machine and outputting the signal strength for each frequency, a weighting step of weighting the strength with the corresponding frequency and outputting the weighted value, and an output step of outputting the frequency and the weighted value. Effect of the Invention

[0011] According to the present invention, it is possible to provide a technique for monitoring the vibration of a fluid machine from the start of rotation of an impeller to the maximum rotation frequency using a single vibration sensor. [Brief description of the drawings]

[0012] [Figure 1] 1 is a functional block diagram of a vibration monitoring system according to an embodiment of the present invention. [Diagram 2] 4 is a flowchart of signal processing according to an embodiment of the present invention. [Diagram 3] 5 is a graph showing the results of signal processing according to an embodiment of the present invention. [Figure 4] FIG. 4 is a functional block diagram of a vibration monitoring system according to a first modified example of the present invention. [Diagram 5] FIG. 11 is a functional block diagram of a vibration monitoring system according to a second modified example of the present invention. [Figure 6] FIG. 11 is a functional block diagram of a vibration monitoring system according to a third modified example of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] An embodiment of the present invention will be described with reference to the drawings.

[0014] (Vibration monitoring system) 1 is a functional block diagram of a vibration monitoring system according to an embodiment of the present invention. The vibration monitoring system 300 monitors the vibration of a fluid machine 100, and outputs the monitoring results to a monitoring terminal 400 via the Internet 500.

[0015] The fluid machine 100 includes a pump 130 that pumps out a fluid, a motor 110 that operates the pump, and a rotating shaft 120 that transmits the power of the motor 110 to the pump. The pump 130 includes an impeller 132 that rotates in accordance with the rotation speed of the rotating shaft 120, and a casing 131 that houses the impeller 132. The casing 131 is provided with a suction port 133 that draws in the fluid, and a discharge port 134 that discharges the fluid.

[0016] The pumping capacity of pump 130 is the capacity to draw up liquid from suction port 133 and discharge liquid from discharge port 134 by applying pressure to the liquid in casing 131 by centrifugal force as impeller 132 rotates. The amount of liquid that pump 130 can discharge in a certain period of time is called the flow rate. The flow rate is determined by the outer diameter of the impeller, the number of blades, the blade outlet angle, the rotation speed, etc. Impeller 132 has five blades and a rotation frequency of 50 Hz.

[0017] The pump 130 of this embodiment is a chemical pump used in building water supply and chemical plants. The pump 130 is made of corrosion-resistant resin, which is suitable for transporting and circulating highly corrosive liquids such as plating solutions used in plating factories. The impeller 132 is a semi-open impeller for transporting cleaning solutions and liquids containing particles that are necessary for manufacturing semiconductors and printed circuit boards. In addition, the gap between the impeller 132 and the casing 131 is wide, allowing high particle passage.

[0018] In this embodiment, the type of impeller 132 is not limited to a semi-open type in which one side of the blade is covered with a disk, but can also be an open type in which the blade is exposed, or a closed type in which both sides of the blade are covered with a disk. Also, the material of the pump 130 is not limited to resin, but metal can also be used.

[0019] Vibration sensor 200 is installed outside casing 131 and in the vicinity of discharge port 134. The reason for installing it in such a location is that the vicinity of discharge port 134 is a location that is greatly affected by the rotation of the impeller of pump 130 and vibrations caused by air bubbles generated by interaction with the fluid transported by the rotation of the impeller, and therefore vibrations that are undesirable to the pump other than malfunctions, such as cavitation, occur around discharge port 134. However, if it is only desired to detect abnormal vibrations caused by malfunctions, vibration sensor 120 can be installed anywhere on the outside of casing 131 for detection.

[0020] The vibration sensor 200 of this embodiment is a piezoelectric acceleration sensor. A piezoelectric element inside the sensor generates an electric signal by the piezoelectric effect to measure vibration. The piezoelectric acceleration sensor is small, lightweight, inexpensive, and easy to install, and can be installed around the discharge port 134. The vibration sensor 200 is also often used in IoT (Internet of Things) in the manufacturing industry. In this embodiment, the vibration of the pump 130 is about several Hz to 1000 Hz. The vibration sensor 200 transmits the measurement value to the vibration monitoring system 300 via a signal cable or the like. Then, the processing result of the vibration monitoring system 300 is transmitted to the monitoring terminal 400 via the Internet 500.

[0021] The vibration monitoring system 300 is configured as vibration monitoring software operable on a computer capable of transmitting and receiving data to and from the Internet 500. The vibration monitoring system 300 includes a storage unit 310 that stores vibration data of the pump 130 sent from the vibration sensor 200, a frequency analysis unit 320 that performs frequency analysis on the vibration data, a weighting unit 330 that performs a weighting process (described later) on the frequency-analyzed data, and an output unit 340 that outputs the weighted data. The vibration monitoring system 300 can be implemented by implementing each functional block of such vibration monitoring software in an executable form on a computer of a virtual server provided by a cloud service provider.

[0022] The monitoring terminal 400 can operate the vibration monitoring system 300 via the Internet 500. The monitoring terminal 400 can graph the data output from the output unit 340 and display it on a display. A user of the monitoring terminal 400 can monitor the vibration of the pump 130 by viewing the graph displayed on the display.

[0023] (Signal Processing) 2 is a flowchart of signal processing according to an embodiment of the present invention. The vibration sensor 200 measures the vibration of the pump 130 and outputs electrical signal data. The electrical signal data is time-series amplitude waveform signal data, and is stored in the memory unit 310 of the vibration monitoring system 300.

[0024] The frequency analysis unit 320 performs Fourier frequency analysis to analyze the frequency components of the time-series waveform signal data. This frequency analysis outputs the signal strength for each frequency. This frequency analysis result data can be represented as a graph with the horizontal axis representing frequency and the vertical axis representing power spectrum density. Power spectrum density is also called energy density.

[0025] The frequency analysis unit 320 may have a low-pass filter unit that performs low-pass filtering when performing frequency analysis on the waveform signal data. S110 is a step of performing low-pass filtering on the frequency analysis result data. The initial value of the center frequency of the low-pass filter is set to 300 Hz. The reason for setting the initial value to 300 Hz is that the number of blades of the impeller 132 is five, and the rotation frequency is 50 Hz. The initial value of the center frequency is not limited to 300 Hz, and can be changed as appropriate.

[0026] Furthermore, the frequency analysis unit 320 may have a tracking filter unit that performs tracking-type filtering when performing frequency analysis on the waveform signal data. In step S120, low-pass filtering is performed at 1.3 times the center frequency f (e.g., 300 Hz), in step S130, frequency analysis is performed, and in step S140, it is determined whether a peak frequency can be identified, and if it can be identified (Yes in S140), the center frequency f is set to the peak frequency F (S150), and the processing from S120 onwards is executed again. On the other hand, if the peak frequency cannot be identified (No in S140), the tracking-type low-pass filtering ends.

[0027] A band-pass filter may be applied as the filtering process of the tracking method. The advantages of applying a low-pass filter to filtering process of the tracking method as in this embodiment will be described below. For example, assuming that there is a failure in which one blade of the impeller 132 is broken, the number of blades of the impeller 132 is reduced from five to four. The vibration frequency of the impeller 132 is reduced from 250 Hz (= rotation frequency 50 Hz × 5 blades) to 200 Hz (= 50 Hz × 4 blades). When it is necessary to widely guarantee a situation in which monitoring can be performed while taking into account the above-mentioned failure, it is preferable to apply a low-pass filter to filtering process of the tracking method.

[0028] In the present embodiment, the frequency analysis unit 320 performs low-pass filtering and / or tracking filtering. However, it should be noted that these filtering processes are not essential.

[0029] The weighting unit 330 receives the signal strength (energy density) for each frequency, which is the output of the frequency analysis unit 320, and performs weighting processing on the signal strength (energy density) for each frequency with the corresponding frequency. S160 is a specific weighting process, which is a step of multiplying the frequency by the strength (energy density). A weighted value is output by this weighting process. The data resulting from this weighting process can be represented in a graph with the horizontal axis being frequency and the vertical axis being energy.

[0030] The output unit 340 outputs a set of the frequency and the weighted value (energy). The monitoring terminal 400 receives the set of the frequency and the energy via the Internet 500, and displays a graph on the display with the horizontal axis representing frequency and the vertical axis representing energy. The user of the monitoring terminal 400 can monitor the vibration of the pump 130 by viewing the graph displayed on the display.

[0031] 3 is a graph showing the results of signal processing according to an embodiment of the present invention. This is the result of processing the signal output by the vibration sensor 200 when the impeller 132 has five blades and a rotation speed of 50 Hz. Note that no filtering such as low-pass filtering is performed.

[0032] 1A is a graph showing the results of a Fast Fourier Transform (FFT), which is one type of Fourier frequency analysis. The horizontal axis is a logarithmic scale of frequency (Hz), and the vertical axis is signal strength (energy density). The numerical values ​​on the vertical axis are the numerical values ​​of the raw data output by the vibration sensor 200, and correspond to the energy density.

[0033] In FIG. 2(a), two large vibration peaks are displayed, namely, a vibration peak near 48 Hz (see the solid circle) and a vibration peak near 250 Hz (see the dashed circle).

[0034] On the other hand, FIG. 1B is a graph showing the results of the weighted FFT of this embodiment. The horizontal axis represents the logarithmic scale of frequency (Hz), and the vertical axis represents energy. In FIG. 1B, a large vibration peak (see dashed circle) is displayed near 250 Hz. However, it can be seen that the vibration peak (see solid circle) near 48 Hz is suppressed by the weighting process.

[0035] (Action and effect) By performing the frequency weighting process of this embodiment, it is found that the vibration peak around 48 Hz is some kind of noise when monitoring the vibration of the pump 130. The frequency-weighted energy density corresponds to energy. Furthermore, noise has no or small energy in principle. Therefore, by comparing the output signal of the vibration sensor 200 with the vertical axis representing energy, it is possible to further suppress low-frequency noise.

[0036] According to the vibration monitoring system 300 of this embodiment, it is possible to monitor the vibration of the impeller 132 from the beginning of rotation of the impeller (low frequency band of about 1 Hz) to the maximum rotation frequency (50 Hz) with one vibration sensor 200.

[0037] In addition, in the weighted FFT of Fig. 3(b), filtering such as a low-pass filter is not performed. For this reason, a vibration peak is also drawn near 487 Hz. However, since the possible maximum frequency can be estimated from the maximum rotation frequency of the impeller 132, the vibration peak near 487 Hz can be eliminated by performing filtering such as a low-pass filter described in Fig. 2. In order to explain that filtering such as a low-pass filter is not an essential configuration for this embodiment, Fig. 3(b) shows a graph of the results of weighted FFT without filtering.

[0038] (Variation 1: Also use a motor current monitoring system) Figure 4 is a functional block diagram of a vibration monitoring system 301 according to the first modified example of the present invention. The same components as those in the vibration monitoring system 300 in Figure 1 are given the same reference numerals, and their description will be omitted. The difference from Figure 1 is that a current sensor 210 is installed in the motor 110 to measure the current value of the motor 110 in order to monitor the current of the motor 110 that operates the pump 130.

[0039] The time-series current signal output from current sensor 210 is stored in memory unit 310 of vibration monitoring system 301 via a signal cable or the like. Estimation unit 350 estimates the rotation speed of impeller 132 from the current signal. Then, comparison unit 360 detects the possibility of damage to the impeller or the occurrence of other malfunctions by comparing the impeller rotation speed estimated by estimation unit 350 with the vibration frequency weighted by weighting unit 330 (corresponding to FIG. 3(b)).

[0040] The output unit 340 outputs the vibration monitoring results of the pump 130 and the above-mentioned detection results to the monitoring terminal 400 via the Internet 500. Since the above-mentioned detection results are displayed on the display of the monitoring terminal 400 in addition to the graph of Fig. 3, the user of the monitoring terminal 400 can perform so-called double monitoring of the fluid machinery 100. For example, when the impeller rotation speed estimated from the current value and the impeller rotation speed estimated from the vibration sensor deviate from each other, it becomes possible to diagnose an abnormality such as a low amount of liquid in the pump.

[0041] (Variation 2: Use the motor tachometer as well) Fig. 5 is a functional block diagram of a vibration monitoring system 302 according to the second modification of the present invention. The same components as those in the vibration monitoring system 301 in Fig. 4 (i.e., the first modification) are given the same reference numerals, and their description will be omitted. The difference from the first modification is that in order to directly monitor the rotation speed of the motor 110 that operates the pump 130, a tachometer 220 is provided on the motor 110, and the rotation speed of the motor 110 is directly measured.

[0042] The time-series rotation speed signal output from the tachometer 220 is stored in a memory unit 310 of the vibration monitoring system 302 via a signal cable or the like. A comparison unit 360 detects the possibility of impeller damage or other malfunctions by comparing the rotation speed signal (i.e., impeller rotation speed) with the vibration frequency weighted by the weighting unit 330 (corresponding to FIG. 3(b)). Thus, the vibration monitoring system 302 is also capable of abnormality diagnosis similar to that of the first modification.

[0043] (Variation 3: Adding an amplifier to the vibration monitoring system) FIG. 6 is a functional block diagram of a vibration monitoring system 303 according to a third modification of the present invention. The same components as those in the vibration monitoring system 300 in FIG. 1 are assigned the same reference numerals, and their description will be omitted. The difference from FIG. 1 is that the system includes an amplifier unit 370 that amplifies an analog voltage signal from the vibration sensor 200, and an AD converter unit 380 that converts the amplified analog voltage signal into a digital signal and outputs the converted digital signal to the storage unit 310 to facilitate signal processing such as FFT. The amplifier unit 370 and the AD converter unit 380 may use the functions of a computer in which vibration monitoring software is installed, or may use an external dedicated circuit. If the amplifier unit 370 is added and the output signal of the vibration sensor 200 is amplified, when it is desired to monitor a vibration that is very small compared to the vibration of the pump, the pump vibration and the very small vibration can be simultaneously monitored by one system, even if the amplifier magnification is such that the signal of the pump vibration is saturated.

[0044] Since the input voltage range of the AD converter unit 380 is limited (usually 0 to 5 V), even if the input voltage of the vibration sensor 200 is, for example, 7 V, it is converted to 5 V by the AD converter unit 380 and input. In this state, it is assumed that an extremely large vibration occurs due to a malfunction such as damage to the impeller, resulting in an input voltage that exceeds the set input voltage range (for example, 6 V).

[0045] In such a case, the waveform signal data sent from vibration sensor 200 to storage unit 310 is saturated and no longer a waveform. In this manner, a constant signal (for example, 5 V) is sent to vibration monitoring system 303 for several seconds.

[0046] If conventional FFT processing is performed on such signal data, the graph of the processing results will show up as low-frequency noise (for example, as shown in FIG. 3(a)). In this case, the low-frequency noise can be said to be a signal with no energy. In contrast, the weighted FFT of this embodiment has a large effect of suppressing such low-frequency noise, making it possible to monitor vibrations that are much smaller than those of the pump itself.

[0047] Although the embodiments (including modified examples) of the present invention have been described above, two or more of these embodiments may be combined and implemented. Alternatively, one of these embodiments may be partially implemented. Furthermore, two or more of these embodiments may be partially combined and implemented.

[0048] Furthermore, the present invention is not limited to the above description of the embodiments of the invention. Various modified embodiments are also included in the present invention within the scope of the claims and within the scope that can be easily conceived by a person skilled in the art. For example, the vibration sensor 200 may be an acceleration sensor other than a piezoelectric type. Furthermore, depending on the sampling frequency and the data transfer speed, the signal data from the vibration sensor 200 and the like may be sent to the vibration monitoring system 300 not only via a signal cable but also via a wired or wireless local area network or the Internet. Furthermore, the computer on which the vibration monitoring software is installed may be a cloud-type server. [Explanation of symbols]

[0049] 100 Fluid machinery 110 Motor 120 Rotational Axis 130 Pump 131 Casing 132 Impeller 133 Intake port 134 Discharge port 200 Vibration Sensor 210 Current Sensor 220 Tachometer 300, 301, 302, 303 Vibration monitoring system 310 Storage section 320 Frequency Analysis Section 330 Weighting section 340 Output section 350 Estimation Department 360 Comparison Section 370 Amplifier section 380 AD converter section 400 Monitoring terminal 500 Internet

Claims

1. a storage unit that stores time-series waveform signal data measured by a vibration sensor that measures vibrations of the fluid machine; a frequency analysis unit that performs frequency analysis on the waveform signal data and outputs a signal intensity for each frequency; a weighting unit that weights the intensity by the corresponding frequency and outputs the weighted value; an output unit that outputs the frequency and the weighted value; A vibration monitoring system comprising:

2. 2. The vibration monitoring system according to claim 1, wherein the vibration sensor measures vibration around a discharge port of a pump of the fluid machinery.

3. 2. The vibration monitoring system according to claim 1, wherein the frequency analysis section includes a low-pass filter section that performs low-pass filtering when analyzing the frequency of the waveform signal data.

4. 2. The vibration monitoring system according to claim 1, wherein the frequency analysis section includes a tracking filter section that performs a filtering process using a tracking method when analyzing the frequency of the waveform signal data.

5. the storage unit also stores a current signal measured by a current sensor that measures a current value of a motor that operates a pump of the fluid machine; an estimation unit that estimates a rotation speed of an impeller of the pump from the current signal; a comparison unit that compares the rotation speed estimated by the estimation unit with the weighted value by the weighting unit; The vibration monitoring system of claim 1 further comprising:

6. The storage unit also stores a rotation speed signal measured by a rotation meter that measures the rotation speed of a motor that operates a pump of the fluid machine, a comparison unit that compares the rotation speed signal with the value weighted by the weighting unit; The vibration monitoring system of claim 1 further comprising:

7. an amplifier unit that amplifies an analog voltage signal from the vibration sensor; an AD converter unit that converts the amplified analog voltage signal into a digital signal and outputs the digital signal to the storage unit; The vibration monitoring system of claim 1 further comprising:

8. a frequency analysis step of performing frequency analysis on time-series waveform signal data measured by a vibration sensor that measures vibrations of the fluid machine, and outputting a signal intensity for each frequency; a weighting step of weighting the intensities with the corresponding frequencies and outputting the weighted values; an output step of outputting the frequency and the weighted value; A vibration monitoring method comprising:

9. a frequency analysis step of performing frequency analysis on time-series waveform signal data measured by a vibration sensor that measures vibrations of the fluid machine, and outputting a signal intensity for each frequency; a weighting step of weighting the intensities with the corresponding frequencies and outputting the weighted values; an output step of outputting the frequency and the weighted value; the computer to execute the vibration monitoring program.

Images