JP5083956B2 - Clogging detection system - Google Patents

Description

  The present invention relates to a clogging detection system for detecting clogging inside a pipe in a flow path of a fluid medium composed of a pump and a pipe connected to the pump.

  In many cases, a fluid medium such as water used in the production process of food products, industrial products, and medicines is supplied or discharged through a path composed of a pump and piping. And when the inside of the piping which comprises a path | route is clogged with the residue and the deposit | attachment, it will be necessary to wash | clean the inside of piping. However, since it is usually impossible to visually recognize the inside of the pipe, it is difficult to specify the clogged portion when clogging occurs, and the piping must be disassembled and inspected each time. There is.

  In order to make it easy to identify such a clogged portion, Patent Document 1 discloses a system in which a capsule provided with a photographing device is inserted into a pipe and foreign matter inside the pipe is detected from a photographed image. In Patent Document 2, an elastic wave detection sensor is installed outside the pipe, air compressed by an air compressor is blown from a nozzle inserted inside the pipe, and an elastic wave generated by a collision between the blown-off foreign object and the pipe. And a system for determining the presence or absence of a foreign substance based on the waveform level is disclosed. Further, Patent Documents 3 and 4 disclose a foreign matter detection apparatus using X-rays.

JP 2007-10513 A Japanese Patent Laid-Open No. 9-210973 JP 2004-317334 A JP 2006-10737 A

  However, in the techniques of Patent Documents 1 and 2, after an inspection instrument such as a capsule or a nozzle is inserted into the pipe, the captured image must be confirmed or air blown while the instrument is moved. There is a problem that work efficiency is poor. In addition, if there is no gap for inserting equipment at either end of the piping, it is necessary to remove the piping or provide an insertion port in the middle, and the system should be easily used for existing piping routes. There is a problem that can not be. On the other hand, in the case of foreign matter detection by X-rays as shown in Patent Documents 3 and 4, it is not necessary to insert an inspection instrument inside the pipe, but the internal clogging can be grasped only in the X-ray irradiation region. Therefore, in order to inspect the entire piping, a large-scale system for performing X-ray irradiation over the entire area is required, and there is a problem that the cost is high.

  The present invention has been made in view of such problems, and its purpose is that it is easy to identify a clogged portion, has high work efficiency, and can be easily used for existing piping routes. An object of the present invention is to provide a clogging detection system that is simple and low-cost as a whole system.

  The present invention adopts the following means in order to solve the above problems. The reference numerals used in the description of the best mode for carrying out the invention and the drawings to be described later are appended in parentheses for reference, but the constituent elements of the present invention are not limited to the appended description.

  The clogging detection system according to the present invention is in close contact with each of a plurality of predetermined locations along the shape of the outer circumference of the pipe in the path of the fluid medium composed of the pump (200) and the pipe (300) connected to the pump. And the film-like piezoelectric sensors (1 to 3) installed in the pipe, and reference data based on the output signal of the piezoelectric sensor during operation of the pump when there is no clogging in the pipe (output signal and power in the case of no clogging) Reference data storage means (storage device 53) for storing spectrum, transfer function), reference data stored in the reference data storage means, and target data (real time) based on an output signal of the piezoelectric sensor for a predetermined period Output means (monitor 60) that outputs the output signal, power spectrum, and transfer function) in a comparable manner.

  The reference data storage means (storage device 53) is measured in the vicinity of the frequency of pressure pulsation generated by the pump, which is measured based on an output signal of the piezoelectric sensor during the pump operation in a state where the pipe is not clogged. The power is stored, and the output means (monitor 60) measures the frequency of the pressure pulsation measured based on the power stored in the reference data storage means and the output signal of the piezoelectric sensor for a predetermined period. The measured power in the vicinity is output so as to be comparable.

  Further, the reference data storage means (storage device 53) is a frequency of pressure pulsation generated in the pump, which is measured based on an output signal of the piezoelectric sensor during the pump operation in a state where the pipe is not clogged. The transmission coefficient between adjacent piezoelectric sensors regarding power in the vicinity is stored, and the output means (monitor 60) outputs the transfer function stored in the reference data storage means and the output of the piezoelectric sensor for a predetermined period. A transfer function between adjacent piezoelectric sensors relating to the measured power in the vicinity of the frequency of the pressure pulsation, which is measured based on the signal, is output in a comparable manner.

  Here, the piezoelectric sensors (1 to 3) are respectively installed at the upstream and downstream portions of the curved portion of the pipe (300) (sensor A, sensor B), and the reference data value storage means (memory) The device 53) measures the upstream and downstream piezoelectric sensors related to power in the vicinity of the frequency of the pressure pulsation, which is measured based on the output signal of the piezoelectric sensor during the pump operation in a state where the piping is not clogged. The output coefficient (monitor 60) is measured based on the transfer function stored in the reference data storage means and the output signal of the piezoelectric sensor for a predetermined period. The transfer function between the upstream and downstream piezoelectric sensors relating to the measured power in the vicinity of the frequency of the pressure pulsation is output so as to be comparable.

  Furthermore, the clogging detection system according to the present invention detects a peak in the lowest frequency band excluding power in the vicinity of 0 Hz based on a frequency analysis result regarding the output signal of the piezoelectric sensor, and determines the frequency related to the peak as the pressure. It further comprises frequency specifying means (control unit 52) for specifying the frequency of pulsation.

  According to the clogging detection system of the present invention, since a film-shaped piezoelectric sensor is installed on the outer periphery of the pipe, the entire system can be simplified and reduced in cost and can be easily applied to an existing pipe path. Available to: Further, it is possible to easily identify the clogged portion based on the output signal of the piezoelectric sensor, and the work efficiency is good.

  According to the clogging detection system of the present invention, the power in a state without clogging can be compared with the actually measured power as the target data in the vicinity of the frequency of pressure pulsation, so that the power attenuation due to clogging can be grasped. As a result, compared with the case where the output signals of the piezoelectric sensors are simply compared, the clogged portion can be accurately identified.

  According to the clogging detection system of the present invention, it is possible to compare the transfer function without clogging with the transfer function as target data in the vicinity of the frequency of the pressure pulsation, so that it is possible to grasp the decrease in the transfer function due to clogging. . As a result, compared with the case where the output signals of the piezoelectric sensors are simply compared, the clogged portion can be accurately identified.

  According to the clogging detection system of the present invention, it is possible to compare the transfer functions in the pipe bending portion where clogging is likely to occur, so it is easy to grasp the occurrence of clogging.

  According to the clogging detection system of the present invention, the frequency of pressure pulsation can be automatically specified. That is, it is not necessary for the system operator to calculate the pressure pulsation frequency or to specify the frequency pulsation result from the frequency analysis result, thereby preventing erroneous determination due to human error. Further, when the number of rotations of the pump is changed, the pressure pulsation frequency is automatically changed even on the system side, so that erroneous determination due to the difference in the pressure pulsation frequency can be prevented.

[1. Configuration of clogging detection system]
Hereinafter, an embodiment of a clogging detection system of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a clogging detection system 100 of the present invention. As shown in FIG. 1, the clogging detection system 100 is used for a path of a fluid medium composed of a pump 200 and a pipe 300 connected to the pump. The clogging detection system 100 receives a piezoelectric sensor 1 installed at a plurality of locations in a path, a charge amplifier 40 connected to the output side of each piezoelectric sensor 1, and an output signal from each charge amplifier 40, and performs signal processing. The computer 50 is configured by a monitor 60 which is a display device.

  In this example, the pump 200 is a positive displacement rotary pump, and is a so-called screw type pump that discharges a fluid medium by causing a volume change by rotating a screw-shaped rotor. The pump 200 discharges a constant amount of fluid medium according to the number of rotations of the rotor. The rated discharge amount in the pump 200 of this example is 8.8 [l / min] based on 100 rpm. In general, positive displacement rotary pumps generate pressure pulsations with a frequency corresponding to the rotational speed of the rotor. As an example, a screw-type pump [rotational speed / 60] Hz, gear type The frequency of pressure pulsation can be calculated from the number of rotations of the pump, such as [number of rotations × number of teeth / 60] Hz. Since the pump 200 of this example is a screw type and the rotation speed is set to 26.4 rpm, the frequency of the pressure pulsation generated in the pump 200 is 26.4 / 60 = 0.44 Hz.

  As shown in FIG. 1, a hopper 201 is connected to the upper part of the main body casing of the pump 200. When the fluid medium is supplied from the upper part of the hopper 201, the fluid medium is discharged from the lower part of the hopper 201 and filled in the main casing of the pump 200. And it is discharged to the piping 300 by rotation of the rotor in the pump 200. Then, the discharged fluid medium passes through a route formed by the pipe 300 and is then supplied to the hopper 201 again. In this example, the fluid medium is water.

  The pipe 300 is a stainless steel pipe (JIS standard G3447), and the material is SUS304. As shown in FIG. 2, the pipe 300 is cylindrical, has an outer diameter of 38.1 mm, and a thickness of 1.2 mm.

  The film-like piezoelectric sensor 1 is a rectangular sheet of 15 mm × 30 mm, and is installed in close contact with the outer periphery of the pipe as shown in FIG. The piezoelectric sensor 1 and the pipe 300 are bonded by, for example, an adhesive using an epoxy resin. Hereinafter, the structure of the piezoelectric sensor 1 will be described in detail with reference to FIG. In place of the piezoelectric sensor 1, the piezoelectric sensor 2 or the piezoelectric sensor 3 shown in FIG. 3B or 3C can be used.

  FIG. 3A is a cross-sectional view showing a laminated structure of the film-like piezoelectric sensor 1. The substrate 10 is a film made of a flexible polymer material, and a piezoelectric layer 11 is formed on the upper surface side of the substrate 10. For example, a polyimide film can be used as the substrate 10. The piezoelectric layer 11 is a thin film of aluminum nitride (AlN). Aluminum nitride as a piezoelectric body shows c-axis orientation even on a polyimide film and has been confirmed to have sufficient piezoelectricity. Polyimide films are also excellent in heat resistance.

  The piezoelectric layer 11 is preferably an aluminum nitride thin film, but other piezoelectric metal compounds can also be used. As the piezoelectric metal compound, in addition to aluminum nitride, zinc oxide, gallium nitride, indium nitride, and lithium niobate can be used. That is, as the piezoelectric layer 11, aluminum nitride, zinc oxide, gallium nitride, indium nitride, lithium niobate and the like can be used, but aluminum nitride and zinc oxide are desirable from the viewpoint of heat resistance, temperature characteristics of detection output, and the like. Aluminum nitride is most suitable.

  The piezoelectric layer 11 can be formed by sputtering, but other thin film forming methods such as ion plating and CVD may be used. In particular, when the thin film formation of the piezoelectric layer 11 is performed using a sputtering method, it is desirable because the crystal orientation degree of the piezoelectric layer 11 can be increased and the piezoelectric characteristics can be improved. As the sputtering method, for example, a high-frequency magnetron sputtering method can be used. The film thickness of the piezoelectric layer 11 is preferably 0.5 μm to 10 μm, but if it is too thin, the film quality will not be constant, and if it is too thick, it takes time to form the film.

  When aluminum nitride formed by sputtering is used for the piezoelectric layer 11, aluminum nitride does not have a Curie point, and therefore does not lose piezoelectricity even in a high temperature environment of 600 degrees or more, and has excellent heat resistance. It can be. Moreover, the temperature dependence of the sensitivity characteristic is extremely small, and no extra compensation circuit is required. Therefore, measurement errors and failures can be prevented even when the temperature of the fluidized medium is high. As the substrate 10 on which the thin film piezoelectric body is laminated, a polymer film such as polyethylene naphthalate (PEN) or polyethylene terephthalate (PET) can be used in addition to polyimide (PI).

  The substrate 10 desirably has a flexible and sufficient mechanical strength and a heat resistance of 100 ° C. or more, and is most preferably a polyimide having excellent mechanical strength and heat resistance. If the thickness of the substrate 10 is too thin, the mechanical strength becomes insufficient, and if it is too thick, the electrical loss increases, so that the thickness is desirably in the range of 1 μm to 10 μm.

  Further, a first electrode layer 12 and a second electrode layer 13 are formed on the lower surface side of the substrate 10 and the upper surface side of the piezoelectric layer 11, respectively. As the first electrode layer 12 and the second electrode layer 13, a platinum (Pt) thin film having a thickness of about 0.1 μm (100 nm) can be used. For these electrode layers, various metals such as platinum, gold, silver, copper, and conductive materials can be used. In order to prevent corrosion of the electrode layer, it is preferable to use platinum. The electrode layer can be formed by a commonly used method such as sputtering, vapor deposition, or screen printing. In order to improve the adhesion to the substrate 10 and the piezoelectric layer 11, it is preferably formed by a sputtering method.

  FIG. 3B is a cross-sectional view illustrating a configuration of the piezoelectric sensor 2 according to another embodiment. The piezoelectric sensor 1 shown in FIG. 3A is formed by laminating two layers. The piezoelectric sensor 2 is produced as follows. First, as described with reference to FIG. 3A, the piezoelectric layer 1, the first electrode layer 12, and the second electrode layer 13 are formed on the substrate 10 to form the piezoelectric sensor 1. A connection hole 14 for connecting a detection terminal, a lead wire, or the like is formed in the electrode layer on the other piezoelectric sensor to form a piezoelectric sensor 1a.

  Then, the second electrode layer 13 side of the piezoelectric sensor 1 and the second electrode layer 13a side of the piezoelectric sensor 1a are overlapped and bonded so as to contact each other. The second electrode layer 13 and the second electrode layer 13a are electrically connected. A detection terminal, a lead wire, and the like are connected to the second electrode layer 13a through the connection hole. The first electrode layer 12 and the first electrode layer 12a are also electrically connected. The other detection terminals and lead wires are connected to the first electrode layer 12 and the first electrode layer 12a.

  Since the piezoelectric sensor 2 is configured by superposing two layers corresponding to the piezoelectric sensor 1 in this way, the area of the piezoelectric layer that receives pressure is doubled, and a highly sensitive pressure sensor can be obtained. Further, since the inside of the piezoelectric sensor 2 is shielded by the first electrode layer 12 and the first electrode layer 12a, electromagnetic induction noise from the outside can be eliminated and the S / N ratio can be improved. The second electrode layer 13 and the second electrode layer 13a can be laminated so that they are on the outside, but the piezoelectric layers 11 and 11a are protected when the second electrode layer 13 is on the inside as shown in the figure. Therefore, it is preferable.

  Further, in FIG. 3B, the two piezoelectric sensors 1 and the piezoelectric sensor 1a are stacked and laminated, but the single piezoelectric sensor 1 is folded so that the second electrode layer 13 is on the inside. It may be glued. Also in this case, the connection hole 14 is formed in advance at an appropriate position of the piezoelectric sensor 1. Note that the adhesive method of the piezoelectric sensor uses an adhesive such as silicon rubber or epoxy, but silicon rubber is preferable in order to maintain the flexibility of the sensor.

  FIG. 3C is a cross-sectional view showing a configuration of a piezoelectric sensor 3 of still another form. Protection films 31 and 32 are adhered to the upper and lower surfaces of the sensor corresponding to the piezoelectric sensor 2 shown in FIG. 3B so as to cover the entire sensor. Since the piezoelectric sensor main body is protected by the protective films 31 and 32, heat resistance and durability are further improved. If the metal shield layers 33 and 34 are formed on the respective protective films 31 and 32, the shielding property against electromagnetic induction noise and the like can be improved, and the S / N ratio can be further improved.

  Since the piezoelectric sensors 1 to 3 are extremely thin and rich in flexibility, they can be attached in close contact with the outer peripheral shape of the pipe 300. Then, the pressure pulsation generated by the pump 200 is propagated to the pipe 300 by a fluid medium or the like inside the pipe, and as a result, a minute distortion occurs in the pipe wall. Therefore, as shown in FIGS. 5 and 6, the piezoelectric sensors 1 to 3 that are in close contact with the outer periphery of the pipe detect minute distortion of the pipe wall, and as a result, pressure pulsation can be detected.

  On the other hand, with conventional vibration sensors and acceleration sensors, it is difficult to accurately detect pressure pulsations because output due to parallel movement of the pipe 300 is mixed. Also, strain gauges and laser displacement gauges have a very narrow measurement range and are measured almost at points, so that sensitivity is insufficient to detect strain in piping due to internal pressure pulsation. On the other hand, in the piezoelectric sensors 1 to 3, the piezoelectric layer is formed on the entire film substrate, and the entire sheet is installed in close contact with the pipe wall, so that the pipe wall is very small due to slight pressure pulsation inside the pipe. Distortion can be detected by integrating over the entire sheet area. That is, since the sensitivity for detecting the strain of the piping wall is proportional to the contact area between the piezoelectric sensors 1 to 3 and the piping wall, the entire sheet-like piezoelectric sensors 1 to 3 are brought into close contact with the piping wall. By covering a wide range, it is possible to integrate and detect minute distortions in the piping wall over the entire seat area, thereby obtaining sufficient sensitivity to detect pressure pulsations.

  Here, as a method for improving the detection sensitivity by the conventional vibration sensor or the like, an indirect countermeasure method such as increasing an amplification factor of an amplifier that electrically amplifies an output signal is mainly used. For this reason, there is a drawback that the S / N ratio is lowered due to noise of the amplifier or the like. On the other hand, in the case of the piezoelectric sensors 1 to 3, the sensor has a multilayer structure as shown in FIGS. 3B and 3C, or the sensor is wound around the pipe 300 or spirally wound. The detection sensitivity can be improved by a direct method without reducing the S / N ratio.

  In addition, the above-described conventional vibration sensor or the like may be limited in installation depending on the diameter or shape of the pipe 300, but the piezoelectric sensors 1 to 3 of the present invention may be bonded to the outer periphery of the pipe. There are no such restrictions on installation, the entire system can be simplified and reduced in cost, and can be easily used for existing piping paths.

  Note that the thickness of the piezoelectric sensor 1 shown in FIG. 3A can be 30 μm or less as a whole, and the thickness of the piezoelectric sensors 2 and 3 shown in FIGS. 3B and 3C is overall. It can be set to 60 μm or less. For this reason, since it becomes an extremely thin film-like piezoelectric sensor and the mass is also extremely small, the influence on the propagation of pressure pulsation can be greatly reduced. Furthermore, the piezoelectric sensors 1 to 3 do not require power supply for detecting pressure pulsations, and can realize a clogging detection system with low power consumption.

  Returning to FIG. 1, each piezoelectric sensor 1 is installed at each of a plurality of predetermined locations in a flow path of a fluid medium including a pump 200 and a pipe 300. In the example of this figure, it is installed in four places. At this time, the adjacent piezoelectric sensors 1 may be installed on the same pipe, and there may be a joint between them. Here, as shown in FIGS. 1 and 4, the piezoelectric sensor 1 is installed at each of an upstream portion and a downstream portion of a curved portion of a pipe 300 constituting a path. Hereinafter, the piezoelectric sensor 1 on the upstream side (pump side) of the bending portion is referred to as sensor A, and the piezoelectric sensor 1 on the downstream side of the bending portion is referred to as sensor B.

  The electric signal output from each piezoelectric sensor 1 is amplified by the charge amplifier 40 and input to the A / D converter 51 provided in the PC 50. In the A / D converter 51, an analog signal is sampled at a predetermined frequency and converted into a digital signal. The control unit 52 includes a CPU, a RAM, and the like, and is a unit that executes various programs and performs various processes related to the PC 50. The output signal converted into the digital signal is stored in the storage device 53 in association with the sensor ID that can identify the piezoelectric sensor 1, and is subjected to fast Fourier transform (FFT) by the frequency analysis program executed by the control unit 52. A power spectrum is obtained. That is, a power spectrum corresponding to the output signal of each piezoelectric sensor 1 is obtained. The power spectrum of each piezoelectric sensor 1 is stored in the storage device 53 in association with the sensor ID of the piezoelectric sensor 1.

  Further, the control unit 52 calculates a transfer function between adjacent piezoelectric sensors based on the power spectrum of each piezoelectric sensor 1. From the relational expression of So = H (f) Si (f) where the upstream side is the input signal Si (f) and the downstream side is the output signal So (f), the transfer function H ( f) is calculated. Here, f represents a frequency. The transfer function H (f) is calculated for each frequency related to the power spectrum, and is stored in the storage device 53 in association with the sensor ID of the upstream or downstream piezoelectric sensor 1.

  As a result, the storage device 53 stores an output signal, a power spectrum, and a transfer function in association with the sensor ID of each piezoelectric sensor 1. These can be displayed on a monitor 60 connected to the PC 50. Since the output signal from each piezoelectric sensor 1 is constantly changing, the output signal stored in the storage device 53 is updated to the latest data every predetermined period (for example, every 60 seconds). Along with the update of the output signal, the power spectrum and the transfer function are also updated every predetermined period. That is, the output signal, power spectrum, and transfer function of the storage device 53 are updated in real time, and the display on the monitor 60 is also updated accordingly. The operator can observe the real-time output signal, power spectrum, and transfer function. In this embodiment, the real-time output signal, power spectrum, and transfer function are the target data to be compared with reference data described later.

[2. Operation of clogging detection system]
Next, the operation of the clogging detection system 100 will be described. In the clogging detection system 100 of the present invention, first, reference data based on the output signal of the piezoelectric sensor 1 during pump operation is acquired in a state where there is no clogging in the pipe, and this is stored in the storage device 53. Cost. In this example, the output signal, power spectrum, and transfer function are stored as reference data.

  Specifically, the operator performs a predetermined operation on the PC 50 (for example, selection of the “reference data acquisition” button displayed on the monitor 60) in a state where there is no clogging of residue or deposits in the piping. To execute the reference data acquisition program. Thereby, for each piezoelectric sensor 1, an output signal for a predetermined period (for example, 60 seconds), a power spectrum corresponding to the output signal, and a transfer function between adjacent piezoelectric sensors for each frequency related to the power spectrum are set as a reference. Data is stored in the storage device 53 in association with the sensor ID of the piezoelectric sensor 1. These reference data are stored in a storage area different from the aforementioned real-time output signal, power spectrum, and transfer function, and are not updated.

  Next, the clogging detection system 100 acquires a real-time output signal, a power spectrum, and a transfer function that are target data to be compared with the reference data. In order to acquire the target data in the state where clogging has occurred, as shown in FIG. 4, clay was adhered to the inner wall of the curved portion of the pipe. This adhesion location corresponds to the middle between the sensor A and the sensor B, and the amount of clay is such that half of the inner diameter of the pipe is clogged. This clogging affects the propagation of pressure pulsations and consequently the real-time output signal, power spectrum, and transfer function.

  In this state, when the operator performs a first operation (for example, selection of the “real time mode (1)” button displayed on the monitor 60) on the PC 50, as shown in FIG. Output signal (FIG. 5A) and a real-time output signal as the target data (FIG. 5B) are output to the monitor 60 for comparison. Further, as shown in FIG. 6, for the sensor B, the output signal (FIG. 6A) as reference data and the real-time output signal (FIG. 6B) as target data can be compared with the monitor 60. Is output.

  In the output signal as the reference data shown in FIG. 5A, the operator can grasp that a signal component having a period of about 2 seconds is included. And since the rotation speed per second of a pump is 0.44, it can be estimated that the signal component of the period is a pressure pulsation. Further, in the real-time output signal shown in FIG. 5B, it can be understood that the signal component having a period of about 2 seconds is included as in FIG. It can be inferred that this is a pulsation. Moreover, it can be estimated that the amplitude of the pressure pulsation concerning both output signals is comparable. That is, comparing the reference data with the real-time output signal, the period and amplitude of the pressure pulsation are approximately the same, so that the factor that inhibits the propagation of the pressure pulsation and attenuates the pressure pulsation is upstream of the sensor A. Is estimated to be absent. As a result, the operator can estimate from the comparison in FIG. 5 that no clogging has occurred upstream of the sensor A.

  In the output signal as the reference data shown in FIG. 6A, the operator can grasp that a signal component having a period of about 2 seconds is included although it is weak. And since the rotation speed per second of a pump is 0.44, it can be estimated that the signal component of the period is a pressure pulsation. On the other hand, in the real-time output signal shown in FIG. 6B, it is impossible to visually recognize a signal component having a period of about 2 seconds. That is, since the pressure pulsation in the real-time output signal is attenuated as compared with the reference data, if there is a factor on the upstream side of the sensor B that inhibits the propagation of the pressure pulsation and attenuates the pressure pulsation. Presumed. As a result, the operator can deduce that clogging has occurred on the downstream side of sensor A and on the upstream side of sensor B from the comparison of FIGS. 5 and 6.

  As described above, it is possible to easily identify the clogged portion based on the output signal of the piezoelectric sensor 1, and it is not necessary to insert an inspection tool and perform manual operation when specifying the clogged portion. Work efficiency is also good. It is also possible to determine the degree of clogging based on the real-time attenuation of pressure pulsation compared to the reference data.

  Next, when the operator performs a second operation (for example, selection of the “real time mode (2)” button displayed on the monitor 60) on the PC 50, as shown in FIG. And a real-time power spectrum as target data are output to the monitor 60 so that they can be compared. Further, as shown in FIG. 8, for the sensor B, the power spectrum as the reference data and the real-time power spectrum as the target data are output to the monitor 60 so that they can be compared.

  In both the power spectrum as the reference data shown in FIG. 7 and the real-time power spectrum, the operator can visually recognize a plurality of peaks except for the DC component in the vicinity of 0 Hz. Of these, the peak associated with the lowest frequency indicates the power density of the pressure pulsation, and this frequency is the fundamental frequency of the pressure pulsation. That is, the first peak indicates the power density of the fundamental wave. The second and third peaks indicate the power density of the second and third harmonics of the pressure pulsation, respectively. Therefore, the operator can calculate the basic frequency of pressure pulsation also from the number of revolutions of the pump as described above, and can grasp the basic frequency of pressure pulsation from the peak of the power spectrum as shown in FIG. .

  As will be described later, when the reference data and the real-time power spectrum are compared in a state where clogging occurs, a clear difference occurs in the power density of the fundamental wave. Specifically, the real-time power density is characterized by being lower than the power density of the reference data. On the other hand, there is a feature that there is no clear difference in power density of harmonics such as second harmonic and third harmonic.

  Based on this feature, when comparing the power density of the fundamental wave in FIG. 7, both the reference data and the real-time pattern density are about −40 db, and it cannot be said that there is a clear difference between them. Therefore, it is estimated that there is no factor on the upstream side of the sensor A that inhibits the propagation of the pressure pulsation and attenuates the pressure pulsation. As a result, the operator can estimate from the comparison of FIG. 7 that no clogging has occurred on the upstream side of the sensor A.

  On the other hand, when the power density of the fundamental wave is compared in FIG. 8, the reference data is about −50 dB, whereas the real-time power density is about −60 dB, so the real-time power density is about 10 dB lower than the reference data. It can be said that they are doing. Therefore, it is estimated that a factor that inhibits the propagation of the pressure pulsation and attenuates the pressure pulsation exists on the upstream side of the sensor B. As a result, the operator can estimate that clogging has occurred on the downstream side of sensor A and on the upstream side of sensor B from the comparison of FIGS. 7 and 8. In addition, as shown in FIG. 8, even in the state where clogging occurs, the power density of harmonics such as second harmonic and third harmonic does not attenuate.

  In this way, at the fundamental frequency of pressure pulsation, the power density in the state without clogging can be compared with the real-time power density, so it is easy to identify the clogging location by grasping the decrease in power density due to clogging. Is possible. In addition, it is possible to easily determine the degree of clogging according to the degree of decrease in power density. Furthermore, since only the power density related to the fundamental frequency of the pressure pulsation can be compared, it is possible to more accurately identify the clogged portion and determine the clogging condition than when simply comparing the output signals of the piezoelectric sensor 1.

  Here, as described above, the transfer function as the reference data and the real-time transfer function as the target data are calculated for each frequency related to the power spectrum and stored in the storage device 53. And since the operator can grasp | ascertain the fundamental frequency of a pressure pulsation from the peak of the calculation formula mentioned above or a power spectrum, it becomes the said fundamental frequency from each of the reference data memorize | stored in the memory | storage device 53, and a real-time transfer function. Matching transfer functions can be selected and both transfer functions displayed on the monitor 60 for comparison. In the example of FIG. 9, in addition to the fundamental wave, transfer functions of second and third harmonics are also selected and displayed in a graph format.

  As shown in FIG. 9, when reference data and a real-time transfer function are compared in a state where clogging occurs, a clear difference occurs in the transfer function of the fundamental wave. Specifically, the real-time transfer function is characterized by being lower than the transfer function of the reference data. On the other hand, there is a characteristic that a clear difference does not occur in the transfer functions of harmonics such as second harmonic and third harmonic.

  When the transfer functions of the fundamental wave are compared in FIG. 9, the gain of the real-time transfer function is lower than that of the reference data. In this example, the real-time transfer function is 0.1 with respect to 0.4 of the reference data, and it can be said that there is a clear difference. Therefore, it is estimated that there is a factor between the sensor A and the sensor B that inhibits the propagation of the pressure pulsation and attenuates the pressure pulsation. As a result, the operator can estimate from FIG. 9 that clogging has occurred on the downstream side of sensor A and on the upstream side of sensor B.

  In this way, at the fundamental frequency of pressure pulsation, the transfer function in the state without clogging can be compared with the real-time transfer function, so it is easy to identify clogged parts by grasping the drop in transfer function due to clogging. Is possible. Moreover, it is possible to easily determine the degree of clogging according to the degree of decrease in the transfer function. Furthermore, since only the transfer functions related to the fundamental frequency can be compared, it is possible to more accurately identify the clogged portion and determine the clogging condition than when simply comparing the output signals of the piezoelectric sensor 1.

  Further, in the case where the clogged portion is specified by the comparison of the output signals or the comparison of the power spectrum, the reference data and the real-time data are not confirmed for the output signals or power spectra of the sensors A and B. It was not possible to specify that there was a blockage between sensor A and sensor B. That is, it is necessary to refer to four data (reference data and real-time data of sensor A, reference data and real-time data of sensor B). On the other hand, when the clogged portion is specified by contrasting the transfer functions, it is easier to refer to the sensor A and the sensor B only by referring to two data (transfer function as reference data and real-time transfer function). It can be identified that there is a blockage in between.

  Here, in the above example, the operator selects a transfer function that matches the fundamental frequency from the reference data in the storage device 53 and the real-time transfer function, but this operation can also be automated. Specifically, the frequency analysis program executed by the control unit 52 excludes power near 0 Hz from the power spectrum as reference data, and detects a peak in the lowest frequency band. And the frequency concerning the said peak is specified as a fundamental frequency of pressure pulsation. Furthermore, a transfer function matching the fundamental frequency is automatically extracted from the reference data and the real-time transfer function of the storage device 53, and both transfer functions are automatically displayed so as to be comparable as shown in FIG. Further, as in the example of FIG. 9, in addition to the fundamental wave, the transfer functions of the second harmonic and the third harmonic may be automatically extracted, and these may be collectively displayed in a graph format.

  Thus, the control unit 52 can automatically specify the frequency of the pressure pulsation from the power spectrum. That is, there is no need for the system operator to calculate the pressure pulsation frequency or to specify the frequency analysis result. In addition, since a transfer function that matches the frequency is automatically displayed, it is not necessary to select a transfer function that matches the frequency, and erroneous determination due to human error can be prevented. Further, when the number of rotations of the pump is changed, the pressure pulsation frequency is automatically changed even on the system side, so that erroneous determination due to the difference in the pressure pulsation frequency can be prevented.

  Further, there is no need to provide an observation mechanism and a processing path for separately observing the rotational speed of the pump, and the pressure pulsation frequency can be simplified more simply by using the piezoelectric sensor 1 originally installed for observing the pressure pulsation in the pipe. Can be specified. Furthermore, as described above, the calculation formula for calculating the pressure pulsation frequency from the rotational speed differs depending on the type and structure of the pump, so even if the rotational speed can be observed, the pressure pulsation frequency is calculated from the rotational speed. In this case, complicated processing is required. On the other hand, the present method for specifying the frequency of pressure pulsation directly from the output signal does not require such a complicated process and can flexibly deal with various pumps.

  In FIG. 9, when the reference data and the real-time transfer function are displayed in comparison, the transfer function itself is displayed. However, as shown in FIG. 10, the ratio of each transfer function may be displayed. The transfer function as the reference data is Hn (f), the real-time transfer function is Hb (f), and from the relational expression Γ (f) = Hb (f) / Hn (f), the transfer function ratio Γ ( f) is calculated. As a result, as shown in FIG. 10, it becomes easy to grasp the ratio of the real-time transfer function to the reference data, and as a result, it becomes easier to identify the clogged portion and determine the clogging condition. As a general rule, the ratio of the transfer function may be calculated only for the fundamental frequency. However, in the example of FIG. 10, the ratio of the transfer function is calculated for the second and third harmonics in addition to the fundamental frequency. Are displayed in graph format.

  Here, a predetermined threshold is set for the ratio of the transfer function at the fundamental frequency, and an alarm is issued from the speaker of the PC 50 or the like when the ratio falls below the threshold. As a result, when the clogging in the pipe reaches a certain level, the operator can recognize this and easily identify the clogged portion and deal with it.

  As described above, according to the clogging detection system 100 of the present invention, since the piezoelectric sensor 1 is installed on the outer periphery of the pipe, the entire system can be simplified and reduced in cost, and the existing pipe path can be reduced. It can be used easily. In addition, by attaching the plurality of piezoelectric sensors 1 to the outer periphery of the pipe 300, it is possible to easily identify a clogged portion based on an output signal, a power spectrum, or a transfer function. Further, since it is not necessary to insert an inspection instrument and perform manual operation, work efficiency is good, and contamination with foreign matter inside the piping can be prevented to ensure a sanitary condition.

  In addition, since the sensor A is installed upstream of the curved portion of the pipe where clogging is likely to occur, and the sensor B is installed downstream, it is possible to compare the transfer function of the reference data with the real-time transfer function.

  Here, there is also a method of facilitating the identification of the clogged portion by making the pipe transparent, but the transparent pipe itself is relatively expensive, and the already constructed route is disassembled. Since the piping must be replaced with a transparent one, there is a problem in terms of cost and labor. In addition, since there is a possibility that a transparent material cannot be used depending on the properties of the fluid medium, the clogging detection system 1 according to the present invention is advantageous.

[3. Modified example of clogging detection system]
Finally, a modification of the clogging detection system according to the present invention will be described.

  According to the present invention, a film-like piezoelectric sensor is disposed in close contact with the outer circumference of a pipe in a flow path of a fluid medium including a pump and a pipe, and target data based on reference data and an output signal of the piezoelectric sensor. Is output so as to be comparable, so that the clogged portion can be specified. Therefore, not only the pressure pulsation but also when any pressure fluctuation occurs in the piping, the target data based on the output signal of the piezoelectric sensor can be compared with the reference data to identify the clogged portion.

  For example, when a step-like pressure fluctuation associated with the opening and closing of a valve in the piping path occurs in the piping, the amplitude of the output signal is large when there is no clogging, but the amplitude of the output signal when there is clogging There is a feature that is small. Therefore, it is possible to estimate the clogging location and the clogging degree due to the difference in amplitude. That is, even if pressure pulsation in the pipe cannot be detected, if there is some pressure fluctuation in the pipe, it is possible to estimate the clogging location and clogging by detecting this with a piezoelectric sensor. It becomes possible.

  In the above-described embodiment, an example in which target data to be compared with reference data is a real-time output signal, a power spectrum, and a transfer function has been described. However, target data is not limited to these real-time data. For example, the output signal is stored in a storage device for a relatively long period (for example, 24 hours or more), the operator designates an arbitrary period in the period, the output signal of the designated period, the power spectrum of the output signal, In addition, a transfer function related to the power spectrum may be used as target data. That is, it is possible to set the data for any period of the stored data as the target data.

  In the above embodiment, the transfer function as the reference data and the real-time transfer function are calculated for each frequency related to the power spectrum and stored in the storage device 53, but as described above, the pressure pulsation is calculated from the power spectrum. Therefore, the transfer function may be calculated and stored only for the fundamental frequency. As a result, the calculation process can be largely omitted, and the storage area for the transfer function can be reduced.

FIG. 1 is a block diagram illustrating a configuration example of a clogging detection system according to the present invention. FIG. 2 is a sectional view of the piping. FIG. 3 is a cross-sectional view showing the laminated structure of the piezoelectric sensor. FIG. 4 is a diagram illustrating the clogging of the pipe bending portion. FIG. 5A is an example of an output signal of the sensor A when there is no clogging, and FIG. 5B is an example of an output signal of the sensor A when there is a clogging. 6A is an example of an output signal of the sensor B in a state where there is no clogging, and FIG. 6B is an example of an output signal of the sensor B in a state where there is a clogging. FIG. 7 is an example of a power spectrum corresponding to the output signal of the sensor A in each of the clogged state and the clogged state. FIG. 8 is an example of a power spectrum corresponding to the output signal of the sensor B in each of the clogged state and the clogged state. FIG. 9 is an example of a transfer function between the sensor A and the sensor B in a state where there is no clogging and a state where there is clogging. FIG. 10 is a diagram showing the transfer function of FIG. 9 as a ratio to the reference data.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Piezo sensor 2 ... Piezo sensor 3 ... Piezo sensor 50 ... PC
52 ... Control unit 53 ... Storage device 60 ... Monitor 100 ... Clogging detection system 200 ... Pump 300 ... Piping

Claims (4)

A film-like piezoelectric sensor installed in close contact with each other at a plurality of predetermined locations along the shape of the outer circumference of the pipe in the path of the fluid medium composed of the pump and the pipe connected to the pump;
Reference data storage means for storing reference data based on an output signal of the piezoelectric sensor during the pump operation in a state where there is no clogging in the pipe;
Output means for outputting the reference data stored in the reference data storage means and the target data based on the output signal of the piezoelectric sensor for a predetermined period in a comparable manner;
The reference data storage means stores power in the vicinity of the frequency of pressure pulsation generated by the pump, which is measured based on an output signal of the piezoelectric sensor during operation of the pump when the pipe is not clogged. Every
The output means outputs the power stored in the reference data storage means and the measured power near the frequency of the pressure pulsation, which is measured based on the output signal of the piezoelectric sensor for a predetermined period, in a comparable manner. A clogging detection system characterized by: A film-like piezoelectric sensor installed in close contact with each other at a plurality of predetermined locations along the shape of the outer circumference of the pipe in the path of the fluid medium composed of the pump and the pipe connected to the pump;
Reference data storage means for storing reference data based on an output signal of the piezoelectric sensor during the pump operation in a state where there is no clogging in the pipe;
Output means for outputting the reference data stored in the reference data storage means and the target data based on the output signal of the piezoelectric sensor for a predetermined period in a comparable manner;
The reference data storage means is an adjacent piezoelectric related to the power in the vicinity of the frequency of the pressure pulsation generated by the pump, which is measured based on the output signal of the piezoelectric sensor during the pump operation when the pipe is not clogged. Store the transfer coefficient between sensors,
The output means is measured between the transfer function stored in the reference data storage means and the output signal of the piezoelectric sensor for a predetermined period, and between adjacent piezoelectric sensors related to the measured power in the vicinity of the pressure pulsation frequency. The clogging detection system is characterized in that the transfer function is output in a comparable manner. A clogging detection system according to claim 2 ,
The piezoelectric sensors are respectively installed in an upstream portion and a downstream portion of a curved portion of the pipe,
The reference data storage means is measured based on an output signal of the piezoelectric sensor during the pump operation in a state where there is no clogging in the pipe, and the upstream portion and the downstream portion relating to power in the vicinity of the frequency of the pressure pulsation. Memorize the transmission coefficient between piezoelectric sensors,
The output means includes the upstream portion and the downstream portion related to the measured power near the frequency of the pressure pulsation, which is measured based on the transfer function stored in the reference data storage means and the output signal of the piezoelectric sensor for a predetermined period. A clogging detection system characterized in that a transfer function between the piezoelectric sensors of a part is output in a comparable manner. The clogging detection system according to claim 1, which is selected from claims 1 to 3 ,
A frequency specifying unit that detects a peak in the lowest frequency band excluding power in the vicinity of 0 Hz based on a frequency analysis result related to the output signal of the piezoelectric sensor, and specifies a frequency related to the peak as a frequency of the pressure pulsation. A clogging detection system further comprising: