JP2022079251A - Predictive maintenance assessment device, predictive maintenance assessment method, and program - Google Patents

Abstract

To provide a predictive maintenance assessment device, a predictive maintenance assessment method, and a program, with can know the occurrence of abnormality that affects apparatus operation before the abnormality actually occurs.SOLUTION: An average value calculation unit (first calculation unit) of the predictive maintenance assessment device acquires an AE output (signal) outputted in time series by at least one AE sensor installed on the surface of a metal housing of a gear box (apparatus), and calculates a first quantity (e.g., average value) pertaining to the magnitude of the AE output for a prescribed duration of time. A standard deviation calculation unit (second calculation unit) calculates a second quantity (e.g., standard deviation) pertaining to variation in the magnitude of the AE output. Meanwhile, a state display unit displays the first quantity pertaining to the magnitude of the AE output and the second quantity pertaining to variation in the magnitude of the AE output by plotting them on a two-dimensional map with each axis representing each of the first quantity and the second quantity. Since the positions of plotted points indicate the operation state of the gear box, it is possible to find a sign of abnormality occurring to the gear box in advance.SELECTED DRAWING: Figure 6

Description

The present invention relates to a predictive maintenance determination device, a predictive maintenance determination method, and a program for predicting the occurrence of an abnormality in a device.

It is known that when a solid material is deformed, the strain energy accumulated up to that point is emitted as a sound wave (AE wave). Conventionally, a damage detection device for detecting damage to a gear by detecting an AE wave with an AE sensor and analyzing the waveform is known.

For example, the gear damage detection device described in Patent Document 1 analyzes the output of the AE sensor and detects the occurrence of gear damage by detecting the signal strength in a specific frequency region.

Japanese Unexamined Patent Publication No. 2009-42151

However, the damage detection device of Patent Document 1 has a problem that the abnormality cannot be detected unless an abnormality such as damage actually occurs in the device. Therefore, when an abnormality is detected, it is necessary to immediately stop the equipment to inspect and maintain the abnormal part, replace consumable parts (bearings, seal parts, etc.), and perform cleaning. Therefore, it may be necessary to stop the equipment at an unexpected timing, and it may be necessary to take measures such as stopping not only the equipment but also the production line. This could have a significant impact on the production process.

The present invention has been made in view of the above, and is a predictive maintenance determination device, a predictive maintenance determination method, and a program that can know the occurrence of an abnormality that affects the operation of the device before the abnormality actually occurs. The purpose is to provide.

In order to solve the above-mentioned problems and achieve the object, the predictive maintenance determination device according to the present invention acquires a signal output in time series by at least one AE sensor installed on the surface of the housing of the device. Calculated as a first calculation unit that calculates a first amount related to the magnitude of the signal for a predetermined time, and a second calculation unit that calculates a second amount related to the variation in the magnitude of the signal. It is characterized by including a state display unit for plotting and displaying the first quantity and the second quantity on a two-dimensional map having the first quantity and the second quantity on each axis. And.

According to the predictive maintenance determination device according to the present invention, the occurrence of an abnormality that affects the operation of the device can be known before the abnormality actually occurs, that is, when the sign of the abnormality is detected. Therefore, it is possible to set in advance the timing for inspecting and servicing the equipment, replacing consumable parts, cleaning, and the like. Therefore, it is possible to maintain the operating state of the production line by operating another device while the device is stopped.

FIG. 1 is a diagram illustrating acoustic emission. FIG. 2 is a diagram illustrating an AE sensor. FIG. 3 is a block diagram showing an example of the configuration of the predictive maintenance determination system using the predictive maintenance determination device according to the first embodiment. FIG. 4 is a structural diagram showing an example of the structure of the extruder according to the first embodiment. FIG. 5 is a hardware block diagram showing an example of the hardware configuration of the predictive maintenance determination device according to the first embodiment. FIG. 6 is a functional block diagram showing an example of the functional configuration of the predictive maintenance determination device according to the first embodiment. FIG. 7 is a first diagram showing an example of an evaluation result of the operating state of the gear box displayed by the state display unit. FIG. 8 is a second diagram showing an example of the evaluation result of the operating state of the gear box displayed by the state display unit. FIG. 9 is a flowchart showing an example of the flow of processing performed by the predictive maintenance determination device according to the first embodiment. FIG. 10 is a functional block diagram showing an example of the functional configuration of the predictive maintenance determination device according to the first modification of the first embodiment. FIG. 11 is a first diagram showing an example of an evaluation result of an operating state of a gear box according to a first modification of the first embodiment. FIG. 12 is a second diagram showing an example of an evaluation result of the operating state of the gear box according to the first modification of the first embodiment. FIG. 13 is a functional block diagram showing an example of the functional configuration of the predictive maintenance determination device according to the second modification of the first embodiment. FIG. 14 is a first diagram showing an example of an evaluation result of an operating state of a gear box according to a second modification of the first embodiment. FIG. 15 is a second diagram showing an example of an evaluation result of the operating state of the gear box according to the second modification of the first embodiment. FIG. 16 is a system block diagram showing an example of the system configuration of the predictive maintenance determination system of the second embodiment.

[Explanation of Acoustic Emission (AE)]
Prior to the description of the embodiment, the acoustic emission (hereinafter referred to as AE) used for determining the predictive maintenance of the device will be described. AE is a phenomenon in which when a solid material is deformed, the strain energy accumulated up to that point is emitted as a sound wave (elastic wave, AE wave). By detecting the AE wave, anomalies in solid materials can be predicted. The frequency band of the AE wave is said to be about several tens of kHz to several MHz, and has a frequency band that cannot be detected by a general vibration sensor or an acceleration sensor. Therefore, in order to detect the AE wave, a dedicated AE sensor is used. The AE sensor will be described in detail later.

FIG. 1 is a diagram illustrating acoustic emission. As shown in FIG. 1, when deformation, contact, friction, or the like occurs in the AE generation source P inside the solid material Q, an AE wave W is generated. The AE wave W spreads radially from the AE source P and propagates inside the solid material Q at a speed corresponding to the solid material Q.

The AE wave W propagating inside the solid material Q is detected by the AE sensor 20 installed on the surface of the solid material Q. Then, the AE sensor 20 outputs the detection signal D. Since the detection signal D is a signal representing vibration, it is an AC signal having positive and negative values. However, since it is difficult to handle the detection signal D (AE wave W) as it is when performing various operations, it is common to treat the negative portion of the detection signal D as a rectified waveform obtained by half-wave rectification. Further, when analyzing the AE wave W, it is generally treated as a value obtained by averaging the root mean square value of the rectified waveform at a predetermined time interval and taking a square root, that is, an effective value (RMS (Root Mean Square) value).

The propagation speed of the AE wave W differs between the longitudinal wave and the transverse wave (the longitudinal wave is faster than the transverse wave), but the difference can be ignored in consideration of the size (propagation distance) of the solid material Q. So, we do not distinguish between longitudinal waves and transverse waves. That is, the AE wave W detected within a predetermined time is used as a measurement signal for analysis regardless of whether it is a longitudinal wave or a transverse wave.

FIG. 2 is a diagram illustrating an AE sensor. As shown in FIG. 2, the AE sensor 20 is included in the shield case 23a. Then, a receiving surface 23b that receives the AE wave W is formed on the bottom surface of the AE sensor 20. The wavefront surface 23b is made of an insulating material. Further, a magnet 23c is installed near the bottom surface of the shield case 23a, and the AE sensor 20 is fixed to the metal housing 30a of the device 30 subject to predictive maintenance by the magnet 23c. At that time, the wave receiving surface 23b is installed in close contact with the surface of the metal housing 30a of the device 30.

A thin-film film 23d made of copper or the like is formed on the wave receiving surface 23b. A piezoelectric element 23e such as lead zirconate titanate (PZT) is installed on the upper portion of the vapor-film vapor film 23d. The piezoelectric element 23e receives the AE wave W via the wave receiving surface 23b and outputs an electric signal corresponding to the amplitude of the AE wave W. The electric signal output by the piezoelectric element 23e is output as a detection signal D via the vapor deposition film 23f and the connector 23g. Since the detection signal D is weak, a preamplifier (not shown in FIG. 2) is installed inside the AE sensor 20 to suppress noise mixing, and the detection signal D is amplified and output in advance. You may.

Since the AE wave W is also generated by minute scratches and friction, it is possible to detect signs of equipment abnormality at an early stage. Further, since the AE wave W spreads radially from the AE source P, if it is a metal housing, by installing the AE sensor 20, the AE wave W can be observed at any position of the housing and the detection signal D can be observed. It is possible to get. A specific analysis method for the detection signal D will be described later. Further, since the frequency band of the signal that can be detected differs depending on the type of the AE sensor 20, it is desirable to consider the material and the like of the device to be measured when selecting the AE sensor 20 to be used.

Hereinafter, the predictive maintenance determination device, the predictive maintenance determination method, and the embodiment of the program according to the present disclosure will be described in detail with reference to the drawings. The present invention is not limited to these embodiments. Further, the components in the following embodiments include those that can be easily conceived by those skilled in the art, or those that are substantially the same.

[First Embodiment]
The first embodiment of the present disclosure is an example of the predictive maintenance determination device 12a that detects and notifies a sign that an abnormality occurs in the device.

[Explanation of the outline configuration of the predictive maintenance judgment device]
First, with reference to FIG. 3, the overall configuration of the predictive maintenance determination system 10a using the predictive maintenance determination device 12a in the present embodiment will be described. FIG. 3 is a block diagram showing an example of the configuration of a predictive maintenance determination system using the predictive maintenance determination device according to the embodiment. The predictive maintenance determination system 10a uses the predictive maintenance determination device 12a of the present disclosure to reduce the rotational driving force of the motor 22 input from the input shaft 31 to drive the extruder 40 via the output shaft 32. It is applied to the judgment of predictive maintenance of the box 30. The gear box 30 is an example of the device 30 in the present disclosure. The gear box 30 is configured by meshing a plurality of gears to reduce the rotational driving force of the motor 22 connected to the input shaft 31 and transmit it to the output shaft 32. The predictive maintenance determination system 10a detects and notifies the signs of abnormality such as cracks and wear generated in the gear and wear of the shaft supporting the gear. The device configuration described below is an example, and the device subject to predictive maintenance is not limited to the gear box 30. Further, the drive target of the gear box 30 is not limited to the extruder 40. The outline of the extruder 40 will be described later (see FIG. 4).

The predictive maintenance determination device 12a acquires the AE output M (t) output in time series by at least one AE sensor 20 installed on the surface of the metal housing 30a of the gear box 30 connected to the extruder 40. Then, the predictive maintenance determination device 12a performs predictive maintenance of the gear box 30 by analyzing the AE output M (t). The AE output M (t) is an example of the signal in the present disclosure.

As the AE sensor 20, a sensor having a frequency band capable of detecting the AE wave W propagating inside the metal housing 30a is used. In particular, when the frequency band of the AE wave W to be detected is known, it is desirable to use the AE sensor 20 having high sensitivity in the frequency band. For example, in the present embodiment, the AE sensor 20 having high sensitivity in the frequency band including 150 kHz is used.

The mounting position of the AE sensor 20 with respect to the metal housing 30a of the gear box 30 does not matter, but it is desirable to mount the AE sensor 20 in the vicinity of a place where an abnormality is likely to occur in the gear box 30. For example, it is desirable to mount the AE sensor 20 in the vicinity of the output shaft 32 of the gearbox 30. Further, although the mounting direction of the AE sensor 20 does not matter, it is desirable to detect the AE output M (t) from a plurality of directions as much as possible in order to accurately detect the AE wave W. Therefore, it is desirable to install a plurality of AE sensors 20 in different directions on the surface of the metal housing 30a of the gear box 30.

For example, as shown in FIG. 3, a plurality of AE sensors 20 (20a, 20b, 20c, 20d) are installed on the surfaces of the gearbox 30 corresponding to each XYZ axial direction, and the output of each AE sensor 20 is output. It is desirable to get it. In FIG. 3, for example, the AE sensor 20a is installed on the XY plane of the gear box 30 and accurately detects the AE wave W along the Z axis. Further, the AE sensor 20b is installed on the YZ plane of the gear box 30 and accurately detects the AE wave W along the X axis. Then, the AE sensor 20c is installed on the XZ plane on the input shaft 31 side of the gear box 30 and accurately detects the AE wave W along the Y axis. The AE sensor 20d is installed on the XZ plane on the output shaft 32 side of the gearbox 30 and accurately detects the AE wave W along the Y axis. By simultaneously measuring a plurality of channels in this way, it is possible to more accurately evaluate the operating state of the gearbox 30 in each axial direction. The operating state of the gear box 30 means the operating state of the internal parts (gears, bearings, etc.) of the gear box 30.

As a result of making a determination related to predictive maintenance, if it is determined that there is a sign that an abnormality has occurred in the operating state of the gear box 30, or that an abnormality has occurred, the predictive maintenance determination device 12a is not shown in FIG. The monitor, indicator, speaker, etc. of the system notify that there is a sign that an abnormality has occurred or that an abnormality has occurred.

[Explanation of extruder structure]
FIG. 4 is a structural diagram showing an example of the structure of the extruder according to the embodiment. The extruder 40 rotates, for example, a resin raw material by rotating a screw 42 installed at a position where the output shaft 32 is extended with the rotation of the output shaft 32 which is rotationally driven in response to the output of the gear box 30. And the powdery filler are kneaded. In particular, the extruder 40 shown in FIG. 4 is a twin-screw extruder having two output shafts 32 installed at a distance C between the shafts.

The two output shafts 32 are arranged in parallel inside the barrel portion 44 while maintaining a constant inter-axis distance C. The bases of the two screws 42 that rotate in the same direction while meshing with each other are connected to each output shaft 32. The output shaft 32 transmits the rotation of the motor 22 decelerated by the gear box 30 to the two screws 42. The two screws 42 rotate at a speed of, for example, 300 rpm.

Inside the barrel portion 44, two cylindrical insertion holes 46 into which each screw 42 is inserted are provided. The insertion hole 46 is a hole provided along the longitudinal direction of the barrel portion 44, and is formed in a state where a part of the cylinder is overlapped so that two screws 42 that mesh with each other can be inserted. A material supply port 47 for supplying the pellet-shaped resin raw material to be kneaded and the powder-like filler material to the insertion hole 46 is provided on one end side of the barrel portion 44 in the longitudinal direction. On the other end side of the barrel portion 44 in the longitudinal direction, a discharge port 48 for discharging the kneaded material while passing through the insertion hole 46 is provided. A heater 49 is provided on the outer periphery of the barrel portion 44 to heat the material supplied to the insertion hole 46 by heating the barrel portion 44.

The screw 42 has the first screw portion 42a, the second screw portion 42b, and the like from one end side of the barrel portion 44 provided with the material supply port 47 toward the other end side of the barrel portion 44 provided with the discharge port 48. It has a third screw portion 42c. Although detailed description is omitted, the first screw portion 42a, the second screw portion 42b, and the third screw portion 42c have different shapes in order to uniformly knead the materials.

Similarly, the barrel portion 44 also has the first screw portion 42a, the second screw portion 42b, and the third screw portion 42 of the screw 42 from one end side where the material supply port 47 is provided toward the other end side where the discharge port 48 is provided. It has a first barrel portion 44a, a second barrel portion 44b, and a third barrel portion 44c corresponding to the screw portion 42c, respectively. The gap between the screw 42 and the barrel portion 44 is formed so as to gradually decrease from the gear box 30 side toward the discharge port 48 side. As a result, the material supplied from the material supply port 47 is kneaded more uniformly.

The total length L in the longitudinal direction of the barrel portion 44, the length L1 of the first barrel portion 44a and the first screw portion 42a, the length L2 of the second barrel portion 44b and the second screw portion 42b, the third barrel portion 44c and the third barrel portion 44c. The length L3 of the screw portion 42c is appropriately determined according to the material to be kneaded.

In the vicinity of the tip of the screw 42, the molten resin is uniformly kneaded. Then, the molten resin that has passed through the screw 42 is discharged from the discharge port 48 in a uniformly kneaded state.

[Explanation of hardware configuration of predictive maintenance judgment device]
Next, the hardware configuration of the predictive maintenance determination device 12a will be described with reference to FIG. FIG. 5 is a hardware block diagram showing an example of the hardware configuration of the predictive maintenance determination device according to the embodiment.

The predictive maintenance determination device 12a includes a control unit 13, a storage unit 14, and a peripheral device controller 16.

The control unit 13 includes a CPU (Central Processing Unit) 13a, a ROM (Read Only Memory) 13b, and a RAM (Random Access Memory) 13c. The CPU 13a connects the ROM 13b and the RAM 13c via the internal bus 15. The CPU 13a reads out the control program P1 stored in the storage unit 14 and expands it in the RAM 13c. The CPU 13a controls the operation of the control unit 13 by operating according to the control program P1 expanded in the RAM 13c. That is, the control unit 13 has a general computer configuration that operates based on the control program P1.

The control unit 13 further connects to the storage unit 14 and the peripheral device controller 16 via the internal bus 15.

The storage unit 14 is a non-volatile memory such as a flash memory, an HDD (Hard Disk Drive), or the like in which the storage information is retained even when the power is turned off. The storage unit 14 stores the control program P1 and the AE output M (t) output by the AE sensor 20. The control program P1 is a program for exerting the functions of the control unit 13. The AE output M (t) is a time-series signal obtained by converting the effective value of the detection signal D output by the AE sensor 20 into a digital signal by the A / D converter 17.

The control program P1 described above may be provided by being incorporated in the ROM 13b in advance. Further, the control program P1 is a file in a format that can be installed in the control unit 13 or in an executable format, and can be read by a computer such as a CD-ROM, a flexible disc (FD), a CD-R, or a DVD (Digital Versatile Disc). It may be configured to be recorded and provided on various recording media. Further, the control program P1 may be stored on a computer connected to a network such as the Internet and provided by downloading via the network. Further, the control program P1 may be configured to be provided or distributed via a network such as the Internet.

The peripheral device controller 16 connects to the A / D converter 17, the display device 18, and the operation device 19. The peripheral device controller 16 controls the operation of various connected hardware based on the command from the control unit 13.

The A / D converter 17 converts the detection signal D output by the AE sensor 20 into a digital signal, and outputs the AE output M (t).

The display device 18 is, for example, a liquid crystal display. The display device 18 displays information related to the operating state of the gear box 30 based on the instruction from the control unit 13. The display device 18 may notify, for example, when a sign of abnormality is detected in the operating state of the gear box 30 (equipment) based on the instruction from the control unit 13. In that case, an indicator composed of LEDs or the like may be used as the display device 18. Further, when performing notification, a buzzer, a speaker, or the like that outputs sound or voice may be used as the display device 18.

The operation device 19 is, for example, a touch panel superimposed on the display device 18. The operation device 19 acquires the setting of the predictive maintenance determination device 12a and the operation information related to the operation of the operator, and transmits the operation information to the control unit 13.

[Explanation of the functional configuration of the predictive maintenance judgment device]
Next, the functional configuration of the predictive maintenance determination device 12a will be described with reference to FIG. FIG. 6 is a functional block diagram showing an example of the functional configuration of the predictive maintenance determination device according to the embodiment. The control unit 13 of the predictive maintenance determination device 12a expands the control program P1 into the RAM 13c and operates the signal acquisition unit 51, the signal analysis unit 52, the determination unit 53, and the state display unit 54 shown in FIG. Is realized as a functional part.

The signal acquisition unit 51 acquires the detection signal D output by the AE sensor 20 for at least a predetermined time, for example, at least 20 seconds. The signal acquisition unit 51 includes an amplifier to amplify the detection signal D, and the A / D converter 17 converts the effective value of the detection signal D, which is an analog signal, into the AE output M (t), which is a digital signal. do. Even if the signal acquisition unit 51 acquires the detection signal D for a longer time, converts it into an analog signal by the A / D converter 17, and then cuts out the AE output M (t) for a predetermined time. good. When a plurality of AE sensors 20 are installed in the gear box 30, the signal acquisition unit 51 has an AE output M (t) corresponding to each detection signal D from the detection signal D output by each AE sensor 20. To get.

The signal analysis unit 52 analyzes the AE output M (t) and calculates an evaluation value for determining the operating state of the gear box 30.

The signal analysis unit 52 further includes an average value calculation unit 52a and a standard deviation calculation unit 52b.

The average value calculation unit 52a calculates the average value Save of the AE output M (t) for a predetermined time. The average value Save of AE output M (t) is an example of the first quantity related to the magnitude of AE output M (t) in the present disclosure. Further, the average value calculation unit 52a is an example of the first calculation unit in the present disclosure.

The standard deviation calculation unit 52b calculates the standard deviation SD of the AE output M (t) for a predetermined time. The standard deviation SD of the AE output M (t) is an example of the second quantity in the present disclosure relating to the variation of the AE output M (t).

The determination unit 53 has a two-dimensional map 60a in which the mean value Save calculated by the mean value calculation unit 52a and the standard deviation SD calculated by the standard deviation calculation unit 52b are taken on each axis of the mean value Save and the standard deviation SD. Plot (see FIG. 7). Then, the determination unit 53 determines that the operating state of the gear box 30 is "good", "possible", "caution", or "danger" based on the plot position (Save, SD) in the two-dimensional map 60a. Is determined. The details of each operating state and the specific determination method will be described later.

The state display unit 54 uses the average value Save (first amount) of the AE output M (t) and the standard deviation SD (second amount) of the AE output M (t) as the average value Save and the standard deviation SD. Is plotted on a two-dimensional map for each axis to display the operating state of the gearbox 30 (equipment). Details will be described later (see FIGS. 7 and 8).

[Explanation of predictive maintenance judgment method]
Next, with reference to FIGS. 7 and 8, a method for determining the operating state of the gear box 30 performed by the determination unit 53 will be described. FIG. 7 is a first diagram showing an example of an evaluation result of the operating state of the gear box displayed by the state display unit. FIG. 8 is a second diagram showing an example of the evaluation result of the operating state of the gear box displayed by the state display unit.

The inventors have installed AE sensors 20 at a plurality of locations of a plurality of gear boxes 30 of the same type in various operating states, and have acquired a plurality of AE outputs M (t) for 20 seconds each. (The number of data points is about 2000 (sampling frequency is about 100 Hz)) was analyzed.

First, an example of the evaluation result of the operating state of the gear box will be described with reference to FIG. 7. FIG. 7 shows a two-dimensional map 60a in which the mean value Save calculated by the mean value calculation unit 52a is taken on the horizontal axis and the standard deviation SD calculated by the standard deviation calculation unit 52b is taken on the vertical axis. Note that FIG. 7 is an example of the evaluation result when the gear box 30 is operating in a state where the vibration acceleration is small. The magnitude of the vibration acceleration generated in the gear box 30 varies depending on, for example, the operating state of the extruder 40.

The diamond marks plotted in FIG. 7 indicate the mean value Save and standard deviation SD of the AE output M (t) acquired on the surface of the housing of the gearbox 30 with the extruder 40 in operation. The square marks plotted in FIG. 7 indicate the average value Save and standard deviation SD of the AE output M (t) acquired when the same gearbox 30 is maintained and then operated again in the same state. show. The maintenance of the extruder 40 means that the gear box 30 is inspected, consumable parts are replaced, and cleaning is performed while the extruder 40 is stopped. According to the experience of the inventors, the average value Save is generally highly correlated with the damage level. Specifically, the larger the average value Save, the larger the damage level. In addition, the standard deviation SD has a high correlation with the degree of damage progression. Specifically, the larger the standard deviation SD, the greater the degree of damage progression.

Regarding the parts of the gear box 30 corresponding to the plot positions (Save, SD) of the diamond mark acquired in the state before maintenance, the gear box 30 is disassembled and the internal state is confirmed, and "the possibility of progressing to failure". Is in a very high state, "is in a state where signs of failure begin to appear," is in a state that requires follow-up, or is in a state where there is no problem at present. gone. Then, information indicating the determination result was added to each plot position (Save, SD). As a result of performing the same evaluation on a plurality of gear boxes 30, a danger range 62a, a caution range 62b, a usable range 62c, and a good range 62d shown in FIG. 7 are formed on the two-dimensional map 60a. It turned out that.

The danger range 62a is a range indicating that the gearbox 30 is in a state in which the possibility of developing a failure is very high. If the evaluation result belongs to the danger range 62a, it may be necessary to replace the repair parts, so it is recommended to formulate a repair plan as soon as possible. It should be noted that just because the determination result belongs to the danger range 62a does not mean that a failure has occurred, so it is not necessary to stop the extruder 40 immediately.

The caution range 62b is a range indicating that the gear box 30 has begun to show signs of failure. If the evaluation result belongs to the caution range 62b, regular re-diagnosis (for example, after half a year) is recommended.

The usable range 62c is a range indicating that follow-up observation of the operating state of the gear box 30 is necessary. If the evaluation results fall within the usable range 62c, regular re-diagnosis (eg, one year later) is recommended.

The good range 62d is a range indicating that there is no problem in the operating state of the gear box 30. If the evaluation result belongs to the good range 62d, it is recommended to continue the regular diagnosis.

The danger range 62a, the caution range 62b, the usable range 62c, and the good range 62d are set based on the mean value Save of the AE output M (t) and the size of the standard deviation SD. Specifically, the danger range 62a is a range in which the mean value Save is the threshold value Ta3 or more, or the standard deviation SD is the threshold value Ts3 or more. The attention range 62b is a range in which the mean value Save is the threshold value Ta2 or more and less than the threshold value Ta3, or the standard deviation SD is the threshold value Ts2 or more and less than the threshold value Ts3. The usable range 62c is a range in which the mean value Save is the threshold value Ta1 or more and less than the threshold value Ta2, or the standard deviation SD is the threshold value Ts1 or more and less than the threshold value Ts2. The good range 62d is a range in which the mean value Save is less than the threshold value Ta1 and the standard deviation SD is less than the threshold value Ts1. In FIG. 7, when the average value Save is equal to or higher than the threshold value Ta4, clipping processing is performed to fix the average value Save to the threshold value Ta4. When the standard deviation SD is equal to or higher than the threshold value Ts4, clipping processing is performed to fix the standard deviation SD to the threshold value Ts4.

The specific threshold value differs depending on the type of the gear box 30 and the like. Therefore, it is necessary to conduct an evaluation experiment or the like in advance to set an appropriate threshold value.

The authors stopped and disassembled the gear box 30 from which the diamond-shaped mark plotted in FIG. 7 was acquired, and performed necessary maintenance. Then, after the maintenance was completed, the gearbox 30 was reassembled, the AE sensor 20 was attached again at the same position, the average value Save and the standard deviation SD of the AE output M (t) were acquired, and plotted on the two-dimensional map 60a. .. The result is the square mark shown in FIG. It was found that both the mean value Save and the standard deviation SD can be reduced by arranging the gear box 30. Then, in the two-dimensional map 60a, the plot positions (Save, SD) (square marks in FIG. 7) of the mean value Save and the standard deviation SD calculated in the state after maintenance are the above-mentioned usable range 62c and the good range 62d. It belongs to and. That is, it was confirmed that the gear box 30 can be returned to the normal state by performing maintenance.

FIG. 8 is an example of the two-dimensional map 60b when the gear box 30 is operating in a state where the vibration acceleration is larger than that in the state where the evaluation of FIG. 7 is performed. The diamond-shaped marks and square marks plotted in FIG. 8 are the same as those described in FIG. 7. That is, the diamond-shaped mark is the evaluation result before the maintenance of the gear box 30, and the square mark is the evaluation result after the maintenance of the gear box 30.

Comparing FIG. 7 and FIG. 8, the evaluation result before maintenance shows that the average value Save in FIG. 8 is larger than the average value Save in FIG. 7. On the other hand, the standard deviation SD in FIG. 8 has a smaller value than the standard deviation SD in FIG. 7. It is presumed that the reason why the average value Save in FIG. 8 is larger than the average value Save in FIG. 7 is that the amplitude of the AE output M (t) is increased due to the influence of the vibration acceleration. .. Further, the standard deviation SD in FIG. 8 has a smaller value than the standard deviation SD in FIG. 7 because the amplitude of the AE output M (t) becomes large due to the influence of the vibration acceleration, and the AE wave. It is presumed that this is because the variation in amplitude is small because W is buried.

Then, as shown in FIG. 8, even when the vibration acceleration of the gear box 30 is large, the two-dimensional maps 60a and 60b with the mean value Save of the AE output M (t) and the standard deviation SD as each axis are taken. It was found that the operating state of the gearbox 30 can be visualized by the same threshold as shown in FIG.

The timing at which the state display unit 54 plots the average value Save and the standard deviation SD of the AE output M (t) on the two-dimensional maps 60a and 60b may be appropriately determined. For example, the display based on the determination result of the AE output M (t) over the past predetermined time (for example, 20 seconds) may be performed at a predetermined time interval, for example, once per second.

Further, the display form of the two-dimensional maps 60a and 60b displayed by the state display unit 54 may be appropriately determined. For example, only the latest evaluation result may be plotted, or as shown in FIGS. 7 and 8, the latest evaluation result may be superimposed and displayed while the past evaluation result is displayed. Further, the previous evaluation result and the latest evaluation result may be connected by a line segment and displayed so that the time transition can be understood. Further, the latest evaluation results (good, acceptable, caution, dangerous) may be displayed in the margins of the two-dimensional maps 60a and 60b so that the area to which the latest evaluation results belong can be immediately known.

The two-dimensional maps 60a and 60b shown in FIGS. 7 and 8 may have the vertical axis and the horizontal axis interchanged. That is, even if the horizontal axis is the standard deviation SD and the vertical axis is the average value Save, the operating state of the gearbox 30 can be visualized in the same manner as described above.

[Explanation of the flow of processing performed by the predictive maintenance judgment device]
Next, with reference to FIG. 9, the flow of processing performed by the predictive maintenance determination device 12a according to the embodiment will be described. FIG. 9 is a flowchart showing an example of the flow of processing performed by the predictive maintenance determination device according to the embodiment.

The signal acquisition unit 51 acquires the AE output M (t) for a predetermined time from the AE output M (t) temporarily stored in the storage unit 14 (step S11). The signal acquisition unit 51 may directly acquire the result of the A / D conversion by the A / D converter 17 and perform the process of step S11 without going through the storage unit 14.

The average value calculation unit 52a calculates the average value Save of the AE output M (t) for a predetermined time (step S12).

The standard deviation calculation unit 52b calculates the standard deviation SD of the AE output M (t) for a predetermined time (step S13).

The determination unit 53 determines the operating state of the gearbox 30 based on the mean value Save calculated from the AE output M (t) and the standard deviation SD (step S14).

The state display unit 54 plots the mean value Save and the standard deviation SD calculated from the AE output M (t) on the two-dimensional map 60a (60b) and displays them on the display device 18 (step S15).

The status display unit 54 determines whether or not an instruction to end the determination has been input from the operation device 19 (step S16). When it is determined that the determination end instruction has been input (step S16: Yes), the predictive maintenance determination device 12a ends the process of FIG. On the other hand, if it is not determined that the instruction to end the determination has been input (step S16: No), the process returns to step S11 and the above-mentioned process is repeated.

As described above, the average value calculation unit 52a (first calculation unit) of the predictive maintenance determination device 12a of the embodiment is at least one AE sensor installed on the surface of the metal housing 30a of the gear box 30 (equipment). The AE output M (t) (signal) output by 20 in time series is acquired, and the first quantity (for example, average value Save) related to the magnitude of the AE output M (t) for a predetermined time is calculated. Further, the standard deviation calculation unit 52b (second calculation unit) calculates a second quantity (for example, standard deviation SD) related to the variation in the size of the AE output M (t). Then, the state display unit 54 sets the first amount related to the size of the AE output M (t) and the second amount related to the variation in the size of the AE output M (t) into the first amount and the first amount. The quantity of 2 is plotted and displayed on a two-dimensional map 60a (60b) taken for each axis. Therefore, since the operating state of the gear box 30 (equipment) can be visualized, it is possible to know the occurrence of an abnormality that affects the operation of the gear box 30 before the abnormality actually occurs.

Further, in the predictive maintenance determination device 12a of the embodiment, the first quantity related to the size of the AE output M (t) is the average value Save of the AE output M (t) of the AE sensor 20 for a predetermined time. Therefore, the first quantity related to the magnitude of the AE output M (t) can be calculated by a simple calculation.

Further, in the predictive maintenance determination device 12a of the embodiment, the second amount related to the variation in the size of the AE output M (t) is the standard deviation SD for a predetermined time of the AE output M (t) of the AE sensor 20. be. Therefore, the second amount can be easily calculated from the AE output M (t).

Further, in the predictive maintenance determination device 12a of the embodiment, the state display unit 54 has an average value Save (first amount) of the AE output M (t) in the two-dimensional map 60a or the two-dimensional map 60b and a standard deviation SD (first quantity). Based on the plot position (Save, SD) with the amount of 2), the gearbox 30 (equipment) is normal, signs of failure are beginning to appear, or there is a very high possibility that it will progress to failure. Which state is displayed. Therefore, the operating state of the gearbox 30 can be visualized based on the plot positions (Save, SD) of the two-dimensional map 60a or the two-dimensional map 60b. In particular, the operating state of the gearbox 30 can be visualized using the same two-dimensional map regardless of the magnitude of vibration generated when the gearbox 30 is in operation.

Further, the predictive maintenance determination device 12a of the embodiment determines the predictive maintenance of the gear box 30 (equipment) that drives the extruder 40. Therefore, the operating state of the gearbox 30 can be visualized while the extruder 40 is in operation. Further, based on the plot position (Save, SD) on the two-dimensional map 60a or the two-dimensional map 60b, the timing for stopping the extruder 40 to inspect and maintain the gearbox 30, replace consumable parts, clean, etc. (immediately). Whether the extruder 40 should be stopped and maintained, follow-up observation, etc.) can be planned in advance. This makes it possible to prevent the production line from being stopped at an unexpected timing.

[First Modification Example of First Embodiment]
Next, a first modification of the first embodiment will be described. In the first embodiment, the signal analysis unit 52 calculates the average value Save and the standard deviation SD of the AE output M (t), and the determination unit 53 uses the average value Save and the standard deviation SD as each axis. The obtained two-dimensional map 60a or two-dimensional map 60b was created to visualize the operating state of the gearbox 30.

At that time, the first amount related to the magnitude of the signal calculated by the signal analysis unit 52 is not limited to the average value Save. The predictive maintenance determination device 12b according to the first modification of the first embodiment uses the mode Smоd (mode: mode) instead of the average value Save of the AE output M (t), and the gear box 30 Evaluate the operating status of. The mode value Smоd is a value that appears most frequently among the outputs of the AE output M (t) for a predetermined time.

FIG. 10 is a functional block diagram showing an example of the functional configuration of the predictive maintenance determination device according to the first modification of the first embodiment.

The predictive maintenance determination device 12b has the same hardware configuration (see FIG. 5) as the predictive maintenance determination device 12a. Then, the control unit 13 realizes the signal acquisition unit 51, the signal analysis unit 52, the determination unit 53, and the state display unit 54 shown in FIG. 10 as functional units. Since the functions of the signal acquisition unit 51, the determination unit 53, and the status display unit 54 are the same as those of the predictive maintenance determination device 12a, the description thereof will be omitted.

The signal analysis unit 52 analyzes the AE output M (t) and calculates an evaluation value for determining the operating state of the gear box 30. The signal analysis unit 52 further includes a mode calculation unit 52c and a standard deviation calculation unit 52b.

The mode calculation unit 52c calculates the mode Smоd of the AE output M (t) for a predetermined time. The mode Smоd of the AE output M (t) is an example of the first quantity related to the magnitude of the AE output M (t) in the present disclosure. Further, the mode calculation unit 52c is an example of the first calculation unit in the present disclosure.

[Explanation of predictive maintenance judgment method]
Next, with reference to FIGS. 11 and 12, a method for determining the operating state of the gear box 30 performed by the determination unit 53 will be described. FIG. 11 is a first diagram showing an example of an evaluation result of an operating state of a gear box according to a first modification of the first embodiment. FIG. 12 is a second diagram showing an example of an evaluation result of the operating state of the gear box according to the first modification of the first embodiment.

The inventors calculated the mode Smоd and the standard deviation SD for the AE output M (t) plotted in FIGS. 7 and 8 described above. Then, two-dimensional maps 60c and 60d were created with the mode Smоd as the horizontal axis and the standard deviation SD as the vertical axis. The two-dimensional map 60c of FIG. 11 shows an example of the plot position (Smоd, SD) of the evaluation result when the gear box 30 is operating in a state where the vibration acceleration is small. The two-dimensional map 60c is an evaluation result based on the same AE output M (t) as the above-mentioned two-dimensional map 60a. Further, the two-dimensional map 60d of FIG. 12 shows an example of the plot position (Smоd, SD) of the evaluation result when the gear box 30 is operating in a state where the vibration acceleration is large. The two-dimensional map 60d is an evaluation result based on the same AE output M (t) as the above-mentioned two-dimensional map 60b.

Comparing FIGS. 11 and 12 with FIGS. 7 and 8, the plot positions of the two-dimensional maps 60c and 60d determined by the mode Smоd and the standard deviation SD regardless of the magnitude of the vibration acceleration of the gearbox 30. It can be seen that the operating state of the gearbox 30 can be visualized based on the above.

As described above, in the predictive maintenance determination device 12b of the embodiment, the first quantity related to the magnitude of the AE output M (t) is the mode of the AE output M (t) of the AE sensor 20 for a predetermined time. The value is Smоd. Therefore, the first quantity related to the magnitude of the AE output M (t) can be easily calculated by a simple calculation.

Although the standard deviation SD is plotted on the vertical axis of FIGS. 11 and 12, a numerical value representing the variation around the mode Smоd, for example, kurtosis may be plotted instead of the standard deviation SD. Kurtosis is a value indicating the sharpness of the frequency distribution, and is a quantity widely used as a statistic indicating the degree of variation in measured values.

[Second variant of the first embodiment]
Next, a second modification of the first embodiment will be described. The predictive maintenance determination device 12c according to the second modification of the first embodiment uses the median Smed (median) instead of the average value Save of the AE output M (t), and uses the median of the gear box 30. Evaluate the operating status. The median Smed is a value located at the center when the outputs of the AE output M (t) for a predetermined time are arranged in ascending or descending order.

FIG. 13 is a functional block diagram showing an example of the functional configuration of the predictive maintenance determination device according to the second modification of the first embodiment.

The predictive maintenance determination device 12c has the same hardware configuration (see FIG. 5) as the predictive maintenance determination device 12a. Then, the control unit 13 realizes the signal acquisition unit 51, the signal analysis unit 52, the determination unit 53, and the state display unit 54 shown in FIG. 13 as functional units. Since the functions of the signal acquisition unit 51, the determination unit 53, and the status display unit 54 are the same as those of the predictive maintenance determination device 12a, the description thereof will be omitted.

The signal analysis unit 52 analyzes the AE output M (t) and calculates an evaluation value for determining the operating state of the gear box 30. The signal analysis unit 52 further includes a median value calculation unit 52d and a standard deviation calculation unit 52b.

The median calculation unit 52d calculates the median Smed of the AE output M (t) for a predetermined time. The median Smed of AE output M (t) is an example of the first quantity relating to the magnitude of AE output M (t) in the present disclosure. Further, the median value calculation unit 52d is an example of the first calculation unit in the present disclosure.

[Explanation of predictive maintenance judgment method]
Next, with reference to FIGS. 14 and 15, a method for determining the operating state of the gear box 30 performed by the determination unit 53 will be described. FIG. 14 is a first diagram showing an example of an evaluation result of an operating state of a gear box according to a second modification of the first embodiment. FIG. 15 is a second diagram showing an example of an evaluation result of the operating state of the gear box according to the first modification of the first embodiment.

The inventors calculated the median Smed and the standard deviation SD for the AE output M (t) plotted in FIGS. 7 and 8 described above. Then, two-dimensional maps 60e and 60f were created with the median Smed as the horizontal axis and the standard deviation SD as the vertical axis. The two-dimensional map 60e of FIG. 14 shows an example of the plot position (Smed, SD) of the evaluation result when the gear box 30 is operating in a state where the vibration acceleration is small. The two-dimensional map 60e is an evaluation result based on the same AE output M (t) as the above-mentioned two-dimensional map 60a. Further, the two-dimensional map 60f of FIG. 15 shows an example of the plot position (Smed, SD) of the evaluation result when the gear box 30 is operating in a state where the vibration acceleration is large. The two-dimensional map 60f is an evaluation result based on the same AE output M (t) as the above-mentioned two-dimensional map 60b.

Comparing FIGS. 14 and 15 with FIGS. 7 and 8, the plot positions of the two-dimensional maps 60e and 60f determined by the median Smed and the standard deviation SD regardless of the magnitude of the vibration acceleration of the gearbox 30. Based on this, it can be seen that the operating state of the gearbox 30 can be visualized.

As described above, in the predictive maintenance determination device 12c of the embodiment, the first quantity related to the magnitude of the AE output M (t) is the median value of the AE output M (t) of the AE sensor 20 for a predetermined time. It is Smed. Therefore, the first quantity related to the magnitude of the AE output M (t) can be easily calculated by a simple calculation.

The standard deviation SD is plotted on the vertical axis of FIGS. 14 and 15, but instead of the standard deviation SD, a numerical value representing the variation around the median Smed, for example, when the data are arranged in ascending order, is shown below. You may plot the quadrant range by subtracting the number of data contained in the range of 1/4 from the number of data contained in the range of 3/4 from the bottom.

[Second Embodiment]
Next, a second embodiment of the present disclosure will be described with reference to FIG. FIG. 16 is a system block diagram showing an example of the system configuration of the predictive maintenance determination system of the second embodiment. The predictive maintenance determination system 10b is a system in which the predictive maintenance determination device 12d determines and displays the operating states of a plurality of gear boxes 70a, 70b, ... Connected via the Internet.

The predictive maintenance determination system 10b determines the predictive maintenance of the outputs (outputs amplified by the preamplifier) of the AE sensors 21a, 21b, ... Installed in the plurality of gear boxes 70a, 70b, ... Via the Internet 100, respectively. It is transmitted to the device 12d, and the predictive maintenance determination device 12d determines the operating state of each gear box 70a, 70b, .... The gear boxes 70a, 70b, ... Are rotationally driven by the motors 22a, 22b, ..., Respectively, to drive the extruders 40a, 41b, .... Further, it is assumed that the outputs of the AE sensors 21a, 21b, ... Are given identification information for identifying the gear box in which each AE sensor is installed.

Since the gear boxes 70a, 70b, ... And the predictive maintenance determination device 12d are connected via the Internet 100, the installation location of the predictive maintenance determination device 12d does not have to be in the vicinity of the gear boxes 70a, 70b, ... It may be located far away from the gear boxes 70a, 70b, .... Further, the gear boxes 70a, 70b, ... Connected to the predictive maintenance determination device 12d are not limited to the gear boxes installed in the same factory, and may be gear boxes installed in a plurality of factories.

The predictive maintenance determination device 12d has the same configuration as any of the predictive maintenance determination devices 12a, 12b, and 12c described in the first embodiment. The predictive maintenance determination device 12d determines the operating state of the gear boxes 70a, 70b, ... Using the determination method performed by the determination unit 53 described above for the outputs of the AE sensors 21a, 21b, ...

Then, the predictive maintenance determination device 12d displays the determined operation state in the form of the above-mentioned two-dimensional map.

Since the output of the AE sensors 21a, 21b, ... Is given identification information for identifying the gear box in which each AE sensor is installed, the gear boxes 70a, 70b, ... do not have to be of the same model. That is, the predictive maintenance determination device 12d includes a plurality of determination logics for determining different AE output M (t) obtained from different types of gear boxes, and the predictive maintenance determination device 12d receives the AE output M (t). ), The operating state of the gearbox may be determined by using the determination logic corresponding to the gearbox that has detected the AE output M (t).

Further, the predictive maintenance determination device 12d may return the determination result to each gear box (70a, 70b, ...) Via the Internet 100. Then, the operating state of each gear box (70a, 70b, ...) may be displayed as a two-dimensional map on a display device (not shown in FIG. 16) installed in each gear box (70a, 70b, ...). ..

As described above, the predictive maintenance determination device 12d of the second embodiment is via the AE sensors 21a, 21b, ... Installed on the surface of one or more gear boxes 70a, 70b, ... (Equipment) and the Internet 100. When connected, the outputs of the AE sensors 21a, 21b, ... Are acquired, and the operating states of the gearboxes 70a, 70b, ... Are determined. As a result, it is possible to determine the operating state of the gear box (equipment) at a place away from the gear box (equipment).

Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are also included in the scope of the invention described in the claims and the equivalent scope thereof.

10a, 10b ... Predictive maintenance determination system, 12a, 12b, 12c, 12d ... Predictive maintenance determination device, 20, 20a, 20b, 20c, 20d, 21a, 21b ... AE sensor, 22 ... motor, 30, 70a, 70b ... gear Box (equipment), 30a ... metal housing, 31 ... input shaft, 32 ... output shaft, 40 ... extruder, 51 ... signal acquisition unit, 52 ... signal analysis unit, 52a ... average value calculation unit (first calculation unit) ), 52b ... Standard deviation calculation unit (second calculation unit), 52c ... Mode calculation unit (first calculation unit), 52d ... Median value calculation unit (first calculation unit), 53 ... Judgment unit, 54 ... Status display unit, 60a, 60b, 60c, 60d, 60e, 60f ... Two-dimensional map, 62a ... Danger range, 62b ... Caution range, 62c ... Usable range, 62d ... Good range, D ... Detection signal, Save ... Mean (first quantity), SD ... standard deviation (second quantity), Med ... median (first quantity), Smоd ... mode (first quantity), Ta1, Ta2, Ta3, Ta4 , Ts1, Ts2, Ts3, Ts4 ... threshold, M (t) ... AE output (signal), W ... AE wave, (Save, SD), (Smоd, SD), (Smed, SD) ... Plot position

Claims (10)

With a first calculation unit that acquires a signal output in time series by at least one AE sensor installed on the surface of the housing of the device and calculates a first quantity related to the magnitude of the signal for a predetermined time. ,
A second calculation unit that calculates a second quantity related to the variation in signal magnitude, and a second calculation unit.
A state display unit that plots and displays the calculated first quantity and the second quantity on a two-dimensional map having the first quantity and the second quantity on each axis.
Predictive maintenance judgment device equipped with. The first quantity is an average value of the outputs of the AE sensor for the predetermined time.
The predictive maintenance determination device according to claim 1. The first quantity is the mode of the output of the AE sensor for the predetermined time.
The predictive maintenance determination device according to claim 1. The first quantity is the median output of the AE sensor for the predetermined time.
The predictive maintenance determination device according to claim 1. The second quantity is the standard deviation of the output of the AE sensor for the predetermined time.
The predictive maintenance determination device according to any one of claims 1 to 4. Based on the plot positions of the first quantity and the second quantity in the two-dimensional map, the status display unit indicates whether the device is normal, a sign of failure is beginning to appear, or the failure progresses. Shows whether it is in a very likely state or in which state,
The predictive maintenance determination device according to any one of claims 1 to 5. The device is a gearbox that drives the extruder.
The predictive maintenance determination device according to any one of claims 1 to 6. An AE sensor installed on the surface of one or more devices is connected via the Internet to acquire the output of the AE sensor.
The predictive maintenance determination device according to any one of claims 1 to 7. A first calculation process in which at least one AE sensor installed on the surface of the housing of a device acquires a signal output in time series and calculates a first quantity related to the magnitude of the signal for a predetermined time. ,
A second calculation process for calculating a second quantity related to the variation in signal magnitude, and a second calculation process.
A state display process in which the calculated first quantity and the second quantity are plotted and displayed on a two-dimensional map with the first quantity and the second quantity on each axis.
Predictive maintenance judgment method. A computer that controls a predictive maintenance judgment device that acquires signals output in chronological order by at least one AE sensor installed on the surface of the device housing.
A first calculation unit that acquires the signal and calculates a first quantity related to the magnitude of the signal for a predetermined time.
A second calculation unit that calculates a second quantity related to the variation in signal magnitude, and a second calculation unit.
A state display unit that plots and displays the calculated first quantity and the second quantity on a two-dimensional map having the first quantity and the second quantity on each axis.
A program that makes it work.

Images