TWI808433B - Omen judging device, omen judging system, omen judging method, and storage medium - Google Patents

Abstract

The sign judging device includes: a measurement result acquisition unit that acquires a measurement result of a current flowing through the motor; an analysis unit that performs frequency analysis and decomposes the measurement result into frequency components; and a prediction unit that determines whether there is a sign of abnormality in at least one of the motor and the load of the motor based on time-series data of the frequency components.

Description

Omen judging device, omen judging system, omen judging method, and storage medium

This disclosure relates to an omen judging device, an omen judging system, an omen judging method, and a storage medium.

Patent Document 1 discloses a technique of a motor abnormality diagnosis method for determining whether the motor is abnormal based on a reference amplitude probability density function obtained from the reference sine wave signal waveform of the rated current of the motor and an inspection-time amplitude probability density function obtained from the current waveform of the motor during operation. [Prior Art Literature] [Patent Document]

[Patent Document 1] JP-A-2011-257362

[Problem to be solved by the invention]

Patent Document 1 indicates a technique for determining whether a bearing of a load driven by a motor is abnormal. However, not only abnormality detection of bearings but also technology for detecting abnormalities other than bearings is required. The object of the present disclosure is to provide an omen judging device, an omen judging system, an omen judging method, and a storage medium capable of appropriately judging an omen judging object. [means to solve the problem]

The sign judging device of the present disclosure includes: a measurement result acquisition unit that acquires a measurement result of a current flowing through the motor; an analysis unit that performs frequency analysis on the measurement result and decomposes it into frequency components; and a prediction unit that determines whether there is a sign of abnormality in at least one of the motor and the load of the motor based on the time-series data of the frequency component.

The omen judging system of the present disclosure is an abnormal omen judging system formed by an omen judging device capable of communicating with a terminal device, wherein the terminal device has: a request unit for omen judgment regarding an abnormality in the load demand of the motor and the motor; wherein the omen judging device includes: a measurement result acquisition unit that acquires a measurement result of the current flowing through the motor according to a request from the terminal device; an analysis unit that performs frequency analysis on the measurement result and decomposes it into frequency components; And a predictor whether there is a sign of abnormality in at least one of the loads of the electric motor.

The omen judgment method disclosed in the present disclosure includes: obtaining the measurement result of the current flowing through the motor; based on the measurement result, performing frequency analysis and decomposing the aforesaid current into frequency components; based on the time series data of the aforesaid frequency components, judging whether there is an omen of abnormality in at least one of the aforesaid motor and the load of the aforesaid motor.

The memory medium disclosed in the present invention memorizes a program for causing a computer to execute the following content: obtain the measurement result of the current flowing through the motor; based on the measurement result, perform frequency analysis and decompose the aforementioned current into frequency components; based on the timing data of the aforementioned frequency component, determine whether there is a sign of abnormality in at least one of the aforementioned motor and the load of the aforementioned motor.

In the omen judging method of the present disclosure, the terminal device has: a request step for judging an omen of an abnormal load demand on the motor and the motor; wherein, the omen judging device capable of communicating with the terminal device has a measurement result obtaining step of obtaining a measurement result of a current flowing through the motor according to a request from the terminal device; an analysis step of frequency analyzing and decomposing the measurement result into frequency components; and judging whether there is a prediction of abnormality in at least one of the motor and the load of the motor based on the time series data of the frequency components. steps. [Effect of the invention]

According to at least one of the above aspects, it can be determined whether or not there is a sign of abnormality in at least one of the motor and the load of the motor.

<First Embodiment> "The Structure of the Omen Judgment System"

Hereinafter, the structure of the omen judgment system 1 according to the embodiment will be described in detail with reference to the drawings. The sign judging system 1 judges a sign of an abnormality of a load 13 driven by a motor 11 .

FIG. 1 is a diagram showing the structure of a sign judgment system 1 according to the first embodiment. The omen determination system 1 includes a power source 10 , a motor 11 , an electric wire 12 , a load 13 , a measuring device 16 , an inverter 17 , and an omen determination device 100 .

The electric power source 10 supplies electric current to the electric motor 11 via the electric wire 12 . The electric motor 11 receives current from a power source 10 via an electrical line 12 . The electric motor 11 receiving the electric current rotates the shaft 14A included in the electric motor 11 and rotates the shaft 14B included in the load 13 .

The shaft 14B is rotated by the motor 11 . That is, the load 13 is driven by the motor 11 . Moreover, the load 13 is provided with the bearing 15A and the bearing 15B which support the shaft 14B. Lubricating oil is supplied to the bearing 15A and the bearing 15B. The lubricating oil reduces the friction between the bearing 15A and the bearing 15B and the shaft 14B.

The measuring device 16 measures the current flowing through the wire 12 . That is, the measuring device 16 measures the current flowing through the motor 11 via the wire 12 . An example of the measuring device 16 is a current detector (Current Transducer, CT). The measuring device 16 measures the current to obtain an analog current waveform.

The converter 17 converts the analog current waveform acquired by the measuring device 16 into digital current data. The converter 17 sends the converted digital current data to the sign judging device 100 . That is, the inverter 17 transmits the measurement result of the current flowing through the motor 11 to the omen determination device 100 .

"Structure of the Omen Judgment Device" Hereinafter, the structure of the omen determination device 100 will be described. The sign determination device 100 determines whether there is a sign of abnormality in the bearing 15A and the bearing 15B of the load 13 driven by the motor 11 . As an example of an abnormality of the bearing 15A and the bearing 15B, there is a shape change of the bearing 15A and the bearing 15B due to friction between the shaft 14B and the bearing 15A and the bearing 15B.

FIG. 2 is a schematic block diagram showing the structure of the omen determination device 100 . The sign determination device 100 includes a measurement result acquisition unit 101, an analysis unit 102, an image generation unit 103, a detection unit 104, a first determination unit 105 (an example of a prediction unit), a calculation unit 106, a second determination unit 107, a identification unit 108, an update unit 109, an output unit 110, and a storage unit 111.

The measurement result acquisition unit 101 acquires digital current data sent from the inverter 17 . That is, the measurement result acquisition unit 101 acquires the measurement result of the current flowing through the motor 11 . The measurement results are digital current data. The analysis unit 102 decomposes the measurement result acquired by the measurement result acquisition unit 101 into complex frequency components by fast Fourier transform (FFT (Fast Fourier Transform)).

The video generation unit 103 generates a video of the time-series data representing the time-series data of each of the plurality of frequency components and the energy values associated with the frequency components. Also, the video generated by the video generation unit 103 expresses the frequency components as a color graph with time and frequency axes, and the energy value is displayed on the color graph. A colored graph is represented by a collection of points. FIG. 3 shows an example of video generated by the video generating unit 103 .

For example, the analysis unit 102 decomposes the measurement result into two frequency components. Figure 3 shows the color graph of the decomposed 2 frequency components. It is assumed that an abnormality of the bearing 15 of the load 13 occurs at time T2. The value of the frequency with respect to the two energy values up to time T1 is a value of a certain frequency band. When the load 13 is normal, the current flowing through the motor 11 does not change, and the values of the two frequency components resulting from the decomposition of the current measurement result are values in a constant frequency band until time T1. The value of the frequency component after time T2 changes by Z amount from the value of the fixed frequency band shown before time T1. Since an abnormality occurs in the bearing 15 at time T2, when the measurement result of the current flowing through the motor 11 is decomposed, it becomes a frequency component having a value different from that at time T1. 3 shows frequency on the horizontal axis and time on the vertical axis, but time may be shown on the horizontal axis and frequency on the vertical axis.

On the color graph of two frequency components, the value of the energy of each frequency component is represented by color. Examples of colors include blue, green, yellow, and red. When the energy value is low, it is represented in blue, and when the energy value is high, it is represented in the order of green, yellow, and red. The energy value of the frequency component representing the state of the bearing maintains a value above a certain value, and is also displayed in the same color (for example, red) before time T1, T1 to T2, and after T2. The video generation unit 103 may display colors not only on the color graphics but also at positions other than the color graphics. For example, the video generation unit 103 may display the positions other than the color graphics in dark blue in the generated video, and may indicate that the energy value is 0 or a value close to 0 at this frequency.

The detection unit 104 is provided with the video generated by the video generation unit 103 and detects the change value of the frequency component. For example, in the case of an image such as that shown in FIG. 3 , the detection unit 104 detects Z as a change value.

The first determination unit 105 compares the change value detected by the detection unit 104 with a preset first threshold value, and determines whether there is a sign of abnormality of the bearing 15 . For example, the first determination unit 105 determines that there is a sign of abnormality in the bearing 15 when the change value detected by the detection unit 104 is equal to or greater than the first threshold value. Furthermore, the first determination unit 105 determines that there is no sign of abnormality in the bearing 15 when the change value detected by the detection unit 104 is not greater than the first threshold value. When an abnormality of the bearing 15 occurs, since the value of the frequency component changes, it can be determined whether there is a sign of an abnormality of the bearing 15 by comparing the changed value with a predetermined threshold.

The calculation unit 106 calculates the degree of similarity based on the feature amount of the image associated with the cause of the abnormality of the bearing 15 and the feature amount of the image generated by the image generation unit 103 . Examples of factors include poor lubrication, poor installation, foreign matter intrusion, rust, and insufficient clearance. The user of the sign determination system 1 pre-records factor information, which is information to associate a factor with an image, in the memory unit 111 . The above image is an image representing the time-series data of frequency components and the value of energy. FIG. 4 is an example of an image in factor information. For example, the user of the sign determination system 1 records factor information associating the image of A in FIG. 4 with the factor A in the memory unit 111 . Moreover, the user of the sign determination system 1 records factor information associating the image of B in FIG. 4 with the factor B in the storage unit 111 .

For example, the calculation unit 106 extracts feature quantities associated with factors such as the image in FIG. 4 by means of a convolutional neural network (CNN). In addition, the calculation unit 106 extracts the feature amount of the video generated by the video generation unit 103 by means of a convolutional neural network. The calculation unit 106 calculates the degree of similarity between the extracted feature quantities. The calculation unit 106 is based on the similarity between the feature quantities of the plurality of images associated with the factors in the factor information and the feature quantities of the images generated by the video generation unit 103 . That is, the calculation unit 106 calculates the plural similarity between the respective images in the plural factor information. In addition, as an example of the operation of the calculation unit 106, before the user of the omen determination system 1 uses the omen determination system 1, it is extracted based on known factors and known images, and when the user of the omen determination system 1 uses the omen determination system 1, it is extracted based on a newly generated factor and a newly generated image.

The second determination unit 107 determines whether or not the degree of similarity calculated by the calculation unit 106 is greater than or equal to a preset second threshold value when the first determination unit 105 determines that there is a sign. When the calculation unit 106 calculates the plural similarity degrees, the second judgment unit 107 judges whether each of the plural similarity degrees is equal to or greater than the second threshold.

The identifying unit 108 identifies factors related to the degree of similarity calculated by the calculating unit 106 when it is determined that the second determining unit 107 is equal to or greater than the second threshold. For example, when the degree of similarity to the video of A in FIG. 4 is equal to or greater than the second threshold, the second determination unit 107 determines that the similarity is equal to or greater than the second threshold. The specifying unit 108 specifies a factor associated with the video of A in FIG. 4 . In addition, the site may be specified when the second determination unit 107 determines that the second threshold value is greater than or equal to the second threshold value, and the site may be specified when at least one of the first determination unit 105 or the second determination unit 107 determines that there is a sign or that the second threshold value is greater than or equal to the second threshold value.

The specifying unit 108 compares the frequency components with the site information associating the frequency components with the site where the abnormality occurs when the first determining unit 105 determines that there is a sign, and specifies the site. Examples of the aforementioned parts include the inner ring of the bearing 15 , the outer ring of the bearing 15 , and the ball between the inner ring and the outer ring of the bearing 15 . The user of the omen judgment system 1 records part information in the memory unit 111 in advance. In addition, the user of the omen determination system 1 may record part information newly generated when the omen determination system 1 is used in the memory unit 111 .

For example, assuming that the value of the frequency component in FIG. 3 changes from F1 by the amount of Z, the first determination unit 105 determines that there is a sign of abnormality in the bearing 15 . At this time, the specifying unit 108 compares the value of the frequency component, namely F1, with the part information. When the location information associates the value F1 of the frequency component with the inner ring of the bearing 15 at the location, the specifying unit 108 specifies the location where the inner ring of the bearing 15 is abnormal.

Furthermore, the identifying unit 108 collates the identified factor with time information associating the factor with the time until the abnormality occurs, and then specifies the time. The user of the sign determination system 1 records time information in the memory unit 111 in advance. In addition, the user of the sign judgment system 1 may record time information newly generated when using the sign judgment system 1 in the memory unit 111 .

Fig. 5 is a diagram showing an example of time information in the first embodiment. For example, it is assumed that the identifying unit 108 identifies the factor A. As shown in FIG. 5 , when the time information associates the cause A with the time T3 until the abnormality occurs, the specifying unit 108 specifies the time T3. The specifying unit 108 may specify the time when the abnormality occurs instead of the time T3 until the abnormality occurs.

The updating unit 109 receives an input from the outside when the second judging unit 107 judges that the degree of similarity is not equal to or greater than the second threshold, and updates the factor information. For example, it is assumed that the degree of similarity between the video generated by the video generation unit 103 and the video included in the factor information is determined not to be equal to or greater than the second threshold. The output unit 110 outputs the video generated by the video generation unit 103 to a display device (not shown) included in the sign determination system 1 . The user of the sign determination system 1 confirms the output video through the display device. The user of the omen determination system 1 identifies the cause of the abnormality of the bearing 15 using a device other than the omen determination device 100 or the like. The user of the omen judgment system 1 inputs the newly specified factor into the omen judgment system 1 by associating it with the output image as factor information. The update unit 109 receives input update factor information. Thereby, even if it is a factor that cannot be determined by the second determination unit 107 , the factor information can be updated, and factors that can be determined by the second determination unit 107 can be added.

The output unit 110 outputs the content specified by the specifying unit 108 to the notification device included in the sign determination system 1 . Examples of the notification device include a display device, a speaker, and the like. Signals output by the output unit 110 include signals representing video and signals related to audio. For example, the output unit 110 outputs the cause, location, and time specified by the specifying unit 108 to the display device. The user of the sign determination system 1 can confirm the cause of the abnormality of the bearing 15 , the location of the abnormality of the bearing 15 , and the time until the abnormality of the bearing 15 occurs based on the display of the display device. In this way, the output unit 110 outputs a signal for displaying an image on the display device, so that the user can easily grasp the content specified by the specifying unit 108 .

The memory unit 111 stores factor information, site information, and time information recorded by the user of the omen judgment system 1 . As an example of the storage unit 111, there is a hard disk.

"Operation of the Omen Judgment System" Hereinafter, the operation of the sign determination system 1 will be described. FIG. 6 is a flowchart showing the operation of the sign determination system 1 .

The measuring device 16 measures the current to obtain an analog current waveform (step S1). The converter 17 converts the analog current waveform acquired in step S1 into digital data (step S2).

The measurement result acquisition unit 101 acquires a digital data measurement result from the converter 17 (step S3). The analysis unit 102 decomposes the measurement result obtained in step S3 into frequency components by FFT (step S4 ).

The image generating unit 103 generates an image of a color graph representing time-series data of frequency components and energy values (step S5). The detection unit 104 detects the change value of the frequency component based on the image generated in step S5 (step S6).

The first determination unit 105 compares the change value detected in step S6 with the first threshold value, and determines whether there is a sign of abnormality in the bearing 15 (step S7). When it is judged that there is no sign of abnormality (step S7: NO), the sign judging system 1 returns to step S1, and performs operations from step S1. On the other hand, when it is judged that there is a sign of abnormality (step S7: YES), the calculation unit 106 calculates the similarity between the feature amount of the image in the factor information and the feature amount of the image generated in step S5 (step S8).

The second determination unit 107 determines whether or not the degree of similarity calculated in step S8 is equal to or greater than a second threshold (step S9). When it is determined that the degree of similarity is equal to or greater than the second threshold (step S9: YES), the specifying unit 108 specifies a factor related to the degree of similarity (step S10). The identification part 108 identifies the location where the abnormality of the bearing 15 occurs (step S11). Furthermore, the specifying unit 108 specifies the time until the abnormality of the bearing 15 occurs (step S12).

The output unit 110 outputs the content specified by the specifying unit 108 to the reporting device (step S13). The reporting device displays the content specified by the specifying unit 108 to the user of the omen determination system 1 (step S14).

On the other hand, when it is determined that the degree of similarity is less than the second threshold (step S9: NO), the output unit 110 outputs the video generated in step S5 to the display device (step S15). After that, the specifying unit 108 specifies the location where the abnormality of the bearing 15 occurs (step S11).

Based on the display output in step S15, the user of the omen judgment system 1 judges the cause of abnormal occurrence using a device different from the omen judgment device 100. The user of the sign judgment system 1 inputs the factor information on the factor specified in step S16 to the sign judgment system 1 . The updating unit 109 updates the factor information recorded in the storage unit 111 with the factor information input in step S17.

"Effect" The sign judging device 100 of the present disclosure includes: a measurement result acquisition unit 101 that acquires the measurement result of the current flowing through the motor 11; an analysis unit 102 that performs frequency analysis and decomposes the measurement result into frequency components; an image generation unit 103 that generates an image of time-series data representing the frequency component; and a first determination unit 105 that determines whether there is a sign of abnormality in the bearing 15 of the load 13 driven by the motor 11 based on the image.

The sign determination device 100 can determine whether there is a sign of abnormality in the bearing 15 of the load 13 driven by the motor 11 based on the image representing the frequency component of the current of the motor 11 . Furthermore, even if the omen judgment device 100 is not installed around the bearing 15 where vibrations are generated and affected by temperature changes, it is possible to remotely determine whether or not there is the omen.

Furthermore, the analysis unit 102 of the sign determination device 100 decomposes into complex frequency components by FFT, and displays the time-series data of each of the complex frequency components as an image.

The sign determination device 100 can determine whether there is a sign of abnormality in the bearing 15 of the load 13 driven by the motor 11 based on the image representing the complex frequency components decomposed by FFT.

In addition, the image of the sign judging device 100 further displays time-series data of energy values associated with the frequency components; a detection unit 104 is provided to detect the change value of the frequency component based on the image; the first determination unit 105 determines whether there is a sign corresponding to the first threshold value of the change value set in advance.

The sign determination device 100 can determine whether there is a sign of abnormality in the bearing 15 of the load 13 driven by the motor 11 based on the image representing the value of the frequency component and energy.

In addition, the image of the sign judging device 100 expresses the frequency of the energy value as a color graph with the time and frequency as axes, and the energy value is represented by color on the color graph of the color graph.

The sign judging device 100 can judge whether there is a sign of abnormality in the bearing 15 of the load 13 driven by the motor 11 based on the image of the color graph representing the value of the frequency component and energy.

In addition, the sign judging device 100 includes a specifying unit 108 for specifying a site by comparing the frequency component with site information associating the frequency component with the site where the abnormality occurs when it is determined that there is a sign.

The sign determination device 100 specifies the site where the abnormality occurs based on the site information. Thereby, the user of the sign judging device 100 can specify the location where the abnormality of the bearing 15 occurs.

Furthermore, the specifying unit 108 of the sign judging device 100, when it is judged that there is a sign, further specifies the cause based on the cause information associating the image with the cause of the abnormality.

The sign determination device 100 specifies the cause of the abnormality based on the cause information. Thereby, the user of the sign judging device 100 can identify the cause of the abnormality of the bearing 15 .

In addition, the sign determination device 100 is provided with feature quantities of images associated with factors; a calculation unit 106 that calculates a similarity based on feature quantities of images generated by the image generation unit 103; a second determination unit 107 that determines whether the similarity is greater than or equal to a preset second threshold when it is determined that there is a sign; and a specifying unit 108 that determines factors related to the calculated similarity when it is determined to be greater than the second threshold.

The sign determination device 100 calculates the similarity based on the feature value of the video to identify the cause of the abnormality. Thereby, the user of the sign judging device 100 can identify the cause of the abnormality of the bearing 15 .

Furthermore, the sign judging device 100 includes an updating unit 109 that receives an input from the outside and updates the factor information when it is judged that the degree of similarity is not equal to or greater than the second threshold.

The omen judging device 100 receives an input from the outside when the omen judging device 100 cannot identify a factor, and updates the factor information. Thereby, the sign determination device 100 can specify more factors.

Furthermore, the specifying unit 108 of the sign judging device 100 collates the specified factor with time information associating the factor with the time until the occurrence of the abnormality, and then specifies the time.

The sign determination device 100 specifies the time until abnormality occurs based on the time information. Thereby, the user of the sign judging device 100 can specify the time until the abnormality of the bearing 15 occurs.

Furthermore, the sign determination device 100 includes an output unit 110 for outputting the content identified by the identification unit 108 .

The sign determination device 100 outputs the cause, location, time, etc. specified by the specifying unit 108 . Thereby, the user of the sign judging device 100 can confirm the cause, location, time, etc. specified by the specifying unit 108 .

The method for judging the sign disclosed in the present disclosure includes: obtaining the measurement result of the current flowing through the motor 11; performing frequency analysis on the measurement result and decomposing it into frequency components; generating an image representing time-series data of the frequency component; based on the image, judging whether the bearing 15 of the load 13 driven by the motor 11 has a sign of abnormality.

The user of the omen determination method can determine whether there is a sign of abnormality in the bearing 15 of the load 13 driven by the motor 11 based on the image representing the frequency component of the current of the motor 11 by using the omen determination method.

The memory medium disclosed in the present disclosure stores a program that causes the computer to execute the following: obtain the measurement result of the current flowing through the motor 11; perform frequency analysis on the measurement result and decompose it into frequency components; generate an image representing time-series data of the frequency component; based on the image, determine whether the bearing 15 of the load 13 driven by the motor 11 has signs of abnormality.

The user of the program can determine whether the bearing 15 of the load 13 driven by the motor 11 has signs of abnormality based on the image representing the frequency component of the current of the motor 11 by executing the program.

<Second Embodiment> "The Structure of the Omen Judgment System" Hereinafter, the structure of the omen judgment system 1 according to the embodiment will be described in detail with reference to the drawings. The omen judging system 1 of the first embodiment is for judging omens of abnormality of the bearing 15A and the bearing 15B of the load 13 . The omen judging system 1 of the second embodiment judges omens of abnormalities including at least one of bearing 15, breakage of the rotor bar in the motor 11, tension change of the belt connecting the motor 11 and the load 13, and voids.

The structure of the sign judgment system 1 of the second embodiment is the same as that of the sign judgment system 1 of the first embodiment shown in FIG.

"Structure of the Omen Judgment Device" Hereinafter, the structure of the omen determination device 100 will be described. The sign judging device 100 judges whether there is a sign of abnormality of at least one of the bearing 15 and the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and a void.

Since the structure of the omen judging device 100 is the same as that of the omen judging device 100 according to the first embodiment shown in FIG. 2 , the description of each structure will be omitted, and only the different structures will be described. In addition, description of this embodiment will be given below using the same reference numerals for the structure.

When the video generation unit 103 has a sign of abnormality in the generated video, the frequency components vary greatly as in the video shown in FIG.

For example, the analysis unit 102 decomposes the measurement result into two frequency components. The image generated by the image generating unit 103 is represented as a color graph of two frequency components decomposed as shown in FIG. At this time, as in the first embodiment, the value of the frequency component after time T2 is changed by the amount of Z from the value of the constant frequency band shown before time T1. Since time T2 includes at least one abnormal occurrence of the bearing 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity, when the measurement result of the current flowing through the motor 11 is decomposed, it becomes a frequency component with a value different from that of time T1.

The energy value of the frequency component indicating the state of the bearing 15, the breakage of the rotor bar in the motor 11, and the tension change of the belt connecting the motor 11 and the load 13 is maintained at a constant value or more. Before time T1, from T1 to T2, and after time T2, when there is at least one abnormality including breakage of the bearing 15, the rotor bar in the motor 11, and a change in the tension of the belt connecting the motor 11 and the load 13, the energy value of the frequency component may change. Holes are identified based on an increase in the energy value of complex frequency components. When a hole occurs at time T1, it can be discriminated because the energy value of the complex frequency components after time T1 increases (for example, changes from blue to yellow).

The detection unit 104 is provided with the video generated by the video generation unit 103 and detects the change value of the frequency component. For example, in the case of an image such as that shown in FIG. 3 , the detection unit 104 detects Z as a change value.

The first determination unit 105 compares the change value detected by the detection unit 104 with a preset first threshold value, and determines whether there is a sign of at least one abnormality including breakage of the bearing 15 and the rotor bar in the motor 11, a change in tension of the belt connecting the motor 11 and the load 13, and a cavity. For example, when the change value detected by the detection unit 104 is equal to or greater than the first threshold value, the first determination unit 105 determines that there is an abnormal sign including at least one of the bearing 15, the breakage of the rotor bar in the motor 11, the change in the tension of the belt connecting the motor 11 and the load 13, and a void. In addition, the first determination unit 105 determines that there is no sign of abnormality including at least one of the breakage of the bearing 15 and the rotor bar in the motor 11, a change in the tension of the belt connecting the motor 11 and the load 13, and a void when the change value detected by the detection unit 104 is not greater than the first threshold value. When at least one abnormality including breakage of the bearing 15 and the rotor bar in the motor 11, a change in the tension of the belt connecting the motor 11 and the load 13, and a void occurs, the value of the frequency component changes. By comparing the change value with a certain threshold value, it can be determined whether there is a sign of at least one of the abnormality including the breakage of the bearing 15, the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the void.

The calculation unit 106 calculates the similarity based on the feature value of the image associated with at least one of abnormality factors including the breakage of the bearing 15 and the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity, and the feature value of the image generated by the image generation unit 103. As an example of the cause, there are poor lubrication, poor installation, foreign matter intrusion, rust, and too little space in the bearing 15, damage to the rotor guide bar in the motor 11, and deterioration over time in the tension change of the belt connecting the motor 11 and the load 13, and abnormality of the pump (not shown) operated by the motor 11 in the cavity.

As in the first embodiment, when the first determination unit 105 judges that there is a sign, the specifying unit 108 compares the frequency components with the site information that associates the frequency components with the site where the abnormality occurs, and specifies the site. As examples of parts, the inner ring of the bearing 15, the outer ring of the bearing 15, the ball between the inner ring and the outer ring of the bearing 15, the rotor bar for the breakage of the rotor bar in the motor 11, the belt for the tension change of the belt connecting the motor 11 and the load 13, and the pump operated by the motor 11 for the cavity.

For example, the value of the frequency component in FIG. 3 changes from F1 by the amount of Z, and the first judging unit 105 judges that there is a sign of abnormality regarding the breakage of the rotor bar in the motor 11 . At this time, the specifying unit 108 compares the value of the frequency component, namely F1, with the part information. When the location information associates F1, which is the value of the frequency component, with the rotor bar, the specifying unit 108 specifies the abnormal location of the rotor bar.

As in the first embodiment, when the second judging unit 107 judges that the degree of similarity is not equal to or greater than the second threshold value, the updating unit 109 receives an input from the outside and updates the factor information. For example, it is assumed that the degree of similarity between the video generated by the video generation unit 103 and the video included in the factor information is determined not to be equal to or greater than the second threshold. In such a judgment, the user of the omen judging system 1 can use a device other than the omen judging device 100 to identify the cause of at least one abnormality including the breakage of the bearing 15, the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

As in the first embodiment, the output unit 110 outputs the content identified by the identification unit 108 to the reporting device included in the omen determination system 1 . For example, the output unit 110 outputs the cause, location, and time specified by the specifying unit 108 to the display device. Based on the display of the display device, the user of the omen judgment system 1 can confirm the cause of occurrence of at least one abnormality, the location where the abnormality occurred, and the time until the abnormality occurred. In this way, the output unit 110 outputs a signal for displaying an image on the display device, so that the user can easily grasp the content specified by the specifying unit 108 .

"Operation of the Omen Judgment System" Hereinafter, the operation of the sign determination system 1 will be described. The flow chart showing the operation of the sign judgment system 1 at this time is the same as that shown in FIG. 6 , which is the flow chart showing the operation of the sign judgment system 1 according to the first embodiment. The description of each processing content will be omitted, and only the different processing content will be described.

In step S7, the first determination unit 105 compares the change value detected in step S6 with the first threshold value, and determines whether there is a sign of abnormality including at least one of the breakage of the bearing 15 and the rotor bar in the motor 11, a change in tension of the belt connecting the motor 11 and the load 13, and a void. In step S11 via step S10, the identification unit 108 identifies the location where at least one of the bearings 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity occurs abnormally. In addition, in step S12, the specifying unit 108 specifies the time until at least one abnormality occurs including the breakage of the bearing 15 and the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

On the other hand, in step S11 of step S15, the identification unit 108 identifies the location where at least one abnormality occurs including the breakage of the bearing 15, the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

In addition, in the present embodiment, although the analysis unit 102 decomposes into complex frequency components and uses the form of the frequency components, it is possible to acquire the frequency component that becomes the peak among the frequency components decomposed by the analysis unit 102 and use the frequency component that becomes the peak. In this case, a frequency component that becomes a peak value is set with a threshold value at that frequency, and when the threshold value is exceeded, a frequency component that becomes a peak value is acquired. Furthermore, by identifying the energy value of the frequency component that becomes the peak by color, it is possible to determine whether or not there is a sign of abnormality even when a peak occurs in an unexpected frequency component.

"Effect" The sign judging device 100 of the present disclosure includes: a measurement result acquisition unit 101 that acquires the measurement result of the current flowing through the motor 11; an analysis unit 102 that frequency-analyzes and decomposes the measurement result into frequency components; an image generation unit 103 that generates an image of time-series data representing the frequency component; based on the image, determines whether there is abnormality of at least one of the bearing 15, the breakage of the rotor guide bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and a void. The first judgment unit 105 of the sign.

The sign judging device 100 can judge whether there is an abnormal sign including at least one of the bearing 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity based on the image representing the frequency component of the electric current of the motor 11. Moreover, even if the omen determination device 100 is not installed around the motor 11 or the load 13, it can determine whether or not the omen is present at a long distance.

Furthermore, the analysis unit 102 of the sign determination device 100 decomposes into complex frequency components by FFT, and displays the time-series data of each of the complex frequency components as an image.

The sign determination device 100 can determine whether there is an abnormal sign including at least one of the bearing 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity based on the complex frequency components decomposed by FFT.

In addition, the image of the sign judging device 100 further displays time-series data of energy values associated with the frequency components; a detection unit 104 is provided to detect the change value of the frequency component based on the image; the first determination unit 105 compares the change value with a preset first threshold value to determine whether there is a sign.

The sign determination device 100 can determine whether there is an abnormal sign including at least one of the bearing 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity based on the image representing the frequency component and energy value.

In addition, the image of the sign judging device 100 expresses the frequency of the energy value as a color graph with time and frequency as axes, and the energy value is represented by color on the color graph.

The sign judging device 100 can judge whether there is a sign of abnormality of at least one of the bearing 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity based on the image of the color graph representing the value of the frequency component and energy.

In addition, the sign judging device 100 includes a specifying unit 108 for specifying a site by comparing the frequency component with site information associating the frequency component with the site where the abnormality occurs when it is determined that there is a sign.

The sign determination device 100 specifies the site where the abnormality occurs based on the site information. Thereby, the user of the sign judging device 100 can judge the location where at least one abnormality occurs including the breakage of the bearing 15 and the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

Furthermore, the specifying unit 108 of the sign judging device 100, when it is judged that there is a sign, further specifies the cause based on the cause information associating the image with the cause of the abnormality.

The sign determination device 100 specifies the cause of the abnormality based on the cause information. Thereby, the user of the omen judging device 100 can specify the cause of at least one abnormal occurrence including the breakage of the bearing 15 and the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

Furthermore, the sign determination device 100 includes a calculation unit 106 that calculates a degree of similarity based on the feature value of the image associated with the factor and the feature value of the image generated by the image generation unit 103; a second determination unit 107 that determines whether the degree of similarity is greater than or equal to a preset second threshold when determining that there is a sign;

The sign determination device 100 calculates the similarity based on the feature value of the video to identify the cause of the abnormality. Thereby, the user of the omen judging device 100 can specify the cause of at least one abnormal occurrence including the breakage of the bearing 15 and the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

Furthermore, the sign judging device 100 includes an updating unit 109 that receives an input from the outside and updates the factor information when it is judged that the degree of similarity is not equal to or greater than the second threshold.

The omen judging device 100 receives an input from the outside when the omen judging device 100 cannot identify a factor, and updates the factor information. Thereby, the sign determination device 100 can specify more factors.

Furthermore, the specifying unit 108 of the sign judging device 100 collates the specified factor with time information associating the factor with the time until the occurrence of the abnormality, and then specifies the time.

The sign determination device 100 specifies the time until abnormality occurs based on the time information. Thereby, the user of the omen judging device 100 can specify the time until at least one abnormality including the breakage of the bearing 15 and the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity occurs.

Furthermore, the sign determination device 100 includes an output unit 110 for outputting the content identified by the identification unit 108 .

The sign determination device 100 outputs the cause, location, time, etc. specified by the specifying unit 108 . Thereby, the user of the sign judging device 100 can confirm the cause, location, time, etc. specified by the specifying unit 108 .

The omen judgment method disclosed in the present disclosure includes: obtaining the measurement result of the current flowing through the motor 11; performing frequency analysis on the measurement result and decomposing it into frequency components; generating an image representing the time series data of the frequency component; based on the image, determining whether there is an abnormal omen including at least one of the bearing 15, the breakage of the rotor guide bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

By using the omen judgment method, the user of the omen judgment method can judge whether or not there is an omen of abnormality including at least one of the bearing 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity, based on the image representing the frequency component of the electric current of the motor 11.

The memory medium of the present disclosure memorizes a program for causing the computer to execute the following: obtain the measurement result of the current flowing through the motor 11; decompose the measurement result into frequency components through frequency analysis; generate an image representing time-series data of the frequency component; based on the image, determine whether there is any sign of abnormality including at least one of the bearing 15, the breakage of the rotor guide bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity.

By executing the program, the user of the program can determine whether or not there is a symptom of at least one abnormality including the breakage of the bearing 15, the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and the cavity based on the image representing the frequency component of the electric current of the motor 11.

<Third Embodiment> Hereinafter, the omen determination device 100 of the third embodiment will be described in more detail. In the sign judging device 100 according to the first embodiment and the second embodiment, the detection unit 104 detects the change value of the frequency component based on the generated image, and the first judgment unit 105 judges that there is a sign of abnormality when the change value detected by the detection unit 104 is equal to or greater than a first threshold value. In addition, it will be described that the calculation unit 106 extracts feature quantities of the video generated by the video generation unit 103 by using a convolutional neural network technique, and calculates the similarity between the extracted feature quantities. Also, the second determination unit 107 will be described. When the first determination unit 105 determines that there is a sign, it determines whether the degree of similarity calculated by the calculation unit 106 is greater than or equal to a preset second threshold value. However, in the sign judging device 100 of the third embodiment, the processing unit 112 described later specifies whether there is a sign of abnormality, and if there is a sign of abnormality, specifies the time until the abnormality occurs, the place where the abnormality occurs, and the cause of the abnormality.

As shown in FIG. 7 , the sign determination device 100 includes a measurement result acquisition unit 101 , an analysis unit 102 , a processing unit 112 (an example of a processing unit), an update unit 109 , an output unit 110 , and a memory unit 111 .

The processing unit 112 predicts the presence or absence of signs of abnormality, the time until the abnormality actually occurs, the place where the abnormality occurs, and the cause of the abnormality, based on each time-series data decomposed into complex frequency components by the analyzing unit 102 by FFT. The processing unit 112 predicts the presence or absence of signs of abnormality, the time until the abnormality actually occurs, the place where the abnormality occurs, and the cause of the abnormality, for example, by using a learned model (such as a convolutional neural network) that uses training data to determine parameters, which is one type of machine learning. Here, the learned model used by the processing unit 112 for each prediction will be described.

In addition, let the learned model used by the processing part 112 in order to predict the sign of the abnormality of the bearing 15 be the 1st learned model. In addition, the learned model used by the processing unit 112 to predict at least one abnormality including the breakage of the bearing 15, the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and a cavity is set to a second learned model. In addition, in the following description, in order to make it easy to understand how the parameters are determined, a first learned model that predicts signs of abnormality in the bearing 15 is given as a specific example of the learned model.

(The first learning model) First, the first learned model will be described. Furthermore, the processing unit 112 predicts signs of abnormality of the bearing 15 based on each time-series data decomposed into complex frequency components by the analyzing unit 102 by FFT. Herein, a learned model for predicting a sign of an abnormality of the bearing 15 based on each time-series data decomposed into complex frequency components by the analysis unit 102 by the processing unit 112 will be described.

At this time, each time-series data decomposed into complex frequency components by the analysis unit 102 by FFT becomes one input data. In addition, the time until an abnormality occurs with respect to the input data, the place where the abnormality occurred, and the cause of the abnormality become one of the output data. Next, a combination of input data and output data corresponding to the input data becomes one training data. For example, before the omen judging device 100 predicts the omen of abnormality, the analysis unit 102 used in other devices when predicting the omen of abnormality uses FFT to decompose each time-series data into complex frequency components, that is, input data, and output data (that is, data indicating the time until the abnormality actually occurs, the place where the abnormality occurs, and the cause of the abnormality) are specified. Or, for example, by conducting experiments and simulations, output data (that is, data indicating the time until the abnormality actually occurs in the time-series data, the place where the abnormality occurs, and the cause of the abnormality) are specified for each time-series data that is decomposed into complex frequency components by the analysis unit 102, that is, input data. Likewise, it is possible to prepare training data obtained by combining plural data of input data and output data. In addition, the training data are data used to determine the value of the parameter in the learning model in which the value of the parameter has not been determined.

FIG. 8 is a diagram showing an example of training data. Each time-series data decomposed into complex frequency components by the analysis unit 102, that is, input data, and output data corresponding to the input data (that is, data indicating the time until the abnormality actually occurred in the time-series data, the place where the abnormality occurred, and the cause of the abnormality) constitute a set of data. In the example shown in FIG. 8, the training data includes 10000 sets of data.

For example, consider a case where parameters in a learning model are determined using training data consisting of 10,000 sets of data shown in FIG. 8 . At this time, the training data is, for example, divided into training data, evaluation data, and test data. Examples of ratios to training data, evaluation data, and test data include 70%, 15%, 15%, 95%, 2.5%, and 2.5%. For example, the training data of data #1 to #10000 is divided into data #1 to #7000 as training data, data #7001 to #8500 as evaluation data, and 15% as test data to data #8501 to #10000. At this point, the training data, data #1, is input into the learning model, namely the convolutional neural network. The convolutional neural network outputs any one of occurrence of no abnormality, time until the abnormality actually occurred, the place where the abnormality occurred, and the cause of the abnormality. Whenever the input data of the training data is input to the convolutional neural network, no abnormality occurs, or the time until the abnormality actually occurs, the place where the abnormality occurs, and the cause of the abnormality are output from the convolutional neural network (at this time, whenever the data of each of the data #1 to #7000 is input to the convolutional neural network, it should be output by performing, for example, backpropagation, changing the additional weight parameter indicating the combination of data between nodes (that is, changing the model of the convolutional neural network). The parameters of the product neural network are tuned.

Next, the input data (data #7001-#8500) of the evaluation data are sequentially input to the convolutional neural network whose parameters are changed by the training data. The convolutional neural network, in response to the input evaluation data, outputs any of whether there is no abnormality, the time until the abnormality actually occurs, the place where the abnormality occurred, and the cause of the abnormality. Here, when the output data of the convolutional neural network is different from the output data associated with the input data in FIG. 8 , the parameters are changed so that the output of the convolutional neural network becomes the output data associated with the input data in FIG. 8 . In this way, the convolutional neural network (that is, the learning model) whose parameters are determined is the first learned model.

Next, as the final confirmation, the input data of the test data (data #8501 to #10000) are sequentially input to the convolutional neural network of the first learned model. The convolutional neural network that has learned the model outputs any of whether there is no abnormality, the time until the abnormality actually occurs, the place where the abnormality occurs, and the cause of the abnormality in response to the input test data. For all test data, when the output data of the convolutional neural network that has learned the model is consistent with the output data associated with the input data in Figure 8, the convolutional neural network that has learned the first model is the desired model. Also, even for one of the test data, if the data output by the convolutional neural network that has first learned the model does not match the output data associated with the input data in Fig. 8, new training data are used to determine the parameters of the learning model. The determination of the parameters of the learning model described above is repeated until the first learned model having desired parameters is obtained. When the first learned model having desired parameters is obtained, the first learned model is recorded in the memory unit 111 .

(The 2nd learning model) Next, the second learned model will be described. FIG. 9 is a diagram showing an example of training data. The parameters of the second learned model can be determined in the same manner as in the above-mentioned first learned model. However, as shown in FIG. 9 , besides the bearings, abnormalities may occur due to breakage of the rotor bar in the motor 11, changes in the tension of the belt connecting the motor 11 and the load 13, voids, etc., and there are multiple abnormalities.

"Operation of the Omen Judgment System" Hereinafter, the operation of the sign determination system 1 will be described. FIG. 10 is a flowchart showing the operation of the sign determination system 1 .

The measuring device 16 measures the current to obtain an analog current waveform (step S1). The converter 17 converts the analog current waveform acquired in step S1 into digital data (step S2).

The measurement result acquisition part 101 acquires the measurement result which is a digital data from the converter 17 (step S3). The analysis unit 102 decomposes the measurement result obtained in step S3 into frequency components by FFT (step S4 ).

The processing unit 112 inputs each time-series data decomposed into complex frequency components by the analyzing unit 102 into the learned model (step S21). The learned model of the processing unit 112 outputs the time until the abnormality actually occurred, the place where the abnormality occurred, and the cause of the abnormality (step S22).

The output unit 110 outputs the content specified by the processing unit 112 (that is, the time until the abnormality actually occurs, the place where the abnormality occurred, and the cause of the abnormality) to the reporting device (step S23). The notification means notifies the user of the sign judgment system 1 of the content specified by the processing unit 112 (step S24).

"Effect" In the sign judging device 100 of the present disclosure, the processing unit 112 predicts the presence or absence of a sign of abnormality, the time until the abnormality actually occurs, the place where the abnormality occurs, and the cause of the abnormality based on each time-series data decomposed into complex frequency components by the analyzing unit 102.

Based on the prediction result by the processing unit 112, the omen judging device 100 can judge whether there is an omen of abnormality including at least one of the bearing 15, the breakage of the rotor bar in the motor 11, the tension change of the belt connecting the motor 11 and the load 13, and a cavity.

<Fourth Embodiment> Hereinafter, the omen determination device 100 according to the fourth embodiment will be described. The sign judging device 100 may be information associating the cause of abnormal occurrence of the bearing 15 , images, and countermeasures against the cause as the cause information. Thereby, the user of the sign determination device 100 can confirm an appropriate countermeasure for the factor in addition to the factor.

In addition, the update unit 109 of the sign determination device 100 may receive an input from the outside to update time information or part information.

<Fifth Embodiment> Hereinafter, the omen determination device 100 according to the fifth embodiment will be described. Fig. 11 is a diagram showing an example of a sign judgment system 1 according to the fifth embodiment. In the omen determination system 1 , the omen determination device 100 may be configured as a system capable of communicating with the terminal device 200 having the request unit 2001 . In this case, the request unit 2001 of the terminal device 200 may be configured to make a request for the omen judgment of the abnormality of the motor 11 and the load 13 of the motor 11 to the omen judging device 100 . Furthermore, the terminal device 200 has the function of the output unit 110, and the terminal device 200 may be configured to output the result of the sign determination of the abnormality in the sign determination device 100 as the aforementioned notification device.

〈Computer structure〉 Fig. 12 is a schematic block diagram showing the structure of a computer in at least one embodiment. The computer 1100 includes a processor 1110 , a main memory 1120 , a storage 1130 , and an interface 1140 . The above-mentioned sign judging device 100 is implemented in a computer 1100 . Next, the operations of the above processing units are stored in the memory 1130 in the form of programs. The processor 1110 reads the program from the storage 1130 and expands it in the main memory 1120, and executes the above processing according to the program. Furthermore, the processor 1110 reserves memory areas corresponding to the above-mentioned memory units in the main memory 1120 according to the program.

The program may be used to realize a part of the functions for the computer 1100 to function. For example, the program can be used by the function performer by combining it with other programs stored in the memory 1130 or by combining it with other programs installed in other devices. In addition, in other embodiments, the computer 1100 may include a customized LSI (Large Scale Integrated Circuit) such as a PLD (Programmable Logic Device) in addition to or instead of the above structure. Examples of the PLD include PAL (Programmable Array Logic), GAL (Generic Array Logic), CPLD (Complex Programmable Logic Device), and FPGA (Field Programmable Gate Array). In this case, part or all of the functions realized by the processor 1110 may be realized by the integrated circuit.

Examples of the storage 1130 include magnetic disks, magneto-optical disks, semiconductor memories, and the like. The storage 1130 may be an internal medium directly connected to the bus of the computer 1100, or an external medium connected to the computer through the interface 1140 or a communication line. Also, when the program is distributed to the computer 1100 via the communication line, the computer 1100 that has received the distribution may expand the program in the main memory 1120 to execute the above-mentioned processing. In at least one embodiment, the storage 1130 is not a temporary tangible storage medium.

Also, the program may be used to realize a part of the aforementioned functions. Furthermore, this program may be realized by combining the aforementioned functions with other programs already stored in the memory 1130, which is a so-called differential file (difference program).

〈Appendix〉 The sign determination device 100 described in each embodiment is understood as follows, for example.

(1) The sign judging device 100 of the present disclosure includes: a measurement result acquisition unit 101 that acquires the measurement result of the current flowing through the motor 11; an analysis unit 102 that performs frequency analysis and decomposes the measurement result into frequency components; and a prediction unit 105 that determines whether there is a sign of abnormality in at least one of the motor 11 and the load 13 of the motor 11 based on the time series data of the frequency components.

The sign determination device 100 can determine whether at least one of the motor 11 and the load 13 of the motor 11 has a sign of abnormality based on the frequency component of the electric current of the motor 11 . Furthermore, the omen judging device 100 can remotely determine whether or not there is the omen even if it is not installed in the vicinity of the motor 11 and the load 13 that are affected by vibration, temperature change, or the like.

(2) Also, the analysis unit 102 of the sign determination device 100 decomposes into complex frequency components by FFT, and the prediction unit (105, 112) determines whether there is a sign of abnormality in at least one of the motor 11 and the load 13 based on the time series data of each of the complex frequency components.

The sign determination device 100 can determine whether or not at least one of the motor 11 and the load 13 has a sign of abnormality based on the complex frequency components decomposed by FFT.

(3) Also, the omen of abnormality of the motor 11 determined by the prediction unit (105, 112) of the omen determination device 100 includes at least one of breakage of the rotor bar in the motor 11, a change in tension of the belt connecting the motor 11 and the load 13, and a void.

The sign judging device 100 can judge whether or not there is a sign of at least one abnormality including breakage of the rotor bar in the motor 11 , a change in tension of the belt connecting the motor 11 and the load 13 , and a void.

(4) Furthermore, the predicting unit (112) of the sign judging device 100 includes a learned model for determining parameters using time-series data of frequency components as input data, and using training data as output data to determine whether there is a sign of abnormality in the input data.

The sign judging device 100 can judge whether or not at least one of the motor 11 and the load 13 has a sign of abnormality using a learned model whose parameters are determined using training data.

(5) Also, the sign judging device 100 includes an image generating unit 103 that generates time-series data representing frequency components; and the predicting unit 105 determines whether there is a sign of abnormality in at least one of the motor 11 and the load 13 based on the image.

The sign determination device 100 can determine whether or not at least one of the motor 11 and the load 13 has a sign of abnormality based on an image showing the frequency component of the current of the motor 11 . The omen judging device 100 can remotely determine whether or not there is the omen even if it is not installed in the vicinity of the motor 11 and the load 13 that are affected by vibration, temperature change, or the like.

(6) Furthermore, the analysis unit 102 of the sign determination device 100 decomposes into complex frequency components by FFT, and displays the time-series data of each of the complex frequency components as an image.

The sign determination device 100 can determine whether or not at least one of the motor 11 and the load 13 has a sign of abnormality based on an image representing the complex frequency components decomposed by FFT.

(7) Also, the images of the sign judging device 100 further display time-series data of energy values associated with the frequency components; a detection unit 104 is provided to detect the change values of the frequency components based on the images; and the prediction unit 105 compares the change values with a preset first threshold to determine whether there is a sign.

The sign determination device 100 can determine whether there is a sign of abnormality in at least one of the motor 11 and the load 13 based on the image representing the value of the frequency component and the energy.

(8) Also, the image of the sign judging device 100 expresses the frequency components as a color graph with the time and frequency axes as axes, and the energy value is represented by colors on the color graph.

The sign judging device 100 can judge whether or not at least one of the motor 11 and the load 13 has a sign of abnormality based on an image of a color graph representing a frequency component and an energy value.

(9) Furthermore, the sign judging device 100 is provided with a specifying unit 108 for specifying a site by comparing the frequency component with site information associating the frequency component with the site where the abnormality occurs when the sign is determined to be present.

The sign determination device 100 specifies the site where the abnormality occurs based on the site information. Thereby, the user of the omen determination device 100 can specify the location where at least one of the motor 11 and the load 13 is abnormal.

(10) Furthermore, when the identification unit 108 of the sign judging device 100 determines that there is a sign, it further specifies the cause based on the cause information linking the image and the cause of the abnormality.

The sign determination device 100 specifies the cause of the abnormality based on the cause information. Thereby, the user of the sign judging device 100 can identify the cause of the abnormality of the bearing 15 .

(11) Also, the sign determination device 100 is provided with a calculation unit 106 that calculates a degree of similarity based on the feature value of the image associated with the factor and the feature value of the image generated by the image generation unit 103; a second determination unit 107 that determines whether the degree of similarity is greater than or equal to a preset second threshold when it is determined that there is a sign;

The sign determination device 100 calculates the similarity based on the feature value of the video to identify the cause of the abnormality. Thereby, the user of the omen determination device 100 can identify the cause of abnormality in at least one of the motor 11 and the load 13 .

(12) Furthermore, the sign judging device 100 includes an updating unit 109 that receives an input from the outside and updates the factor information when it is judged that the degree of similarity is not equal to or greater than the second threshold value.

The omen judging device 100 receives an input from the outside when the omen judging device 100 cannot identify a factor, and updates the factor information. Thereby, the sign determination device 100 can specify more factors.

(13) Furthermore, the specifying unit 108 of the sign judging device 100 collates the specified factor with time information associating the factor with the time until the abnormality occurs, and specifies the time again.

The sign determination device 100 specifies the time until abnormality occurs based on the time information. Thereby, the user of the omen determination device 100 can specify the time until at least one of the motor 11 and the load 13 becomes abnormal.

(14) Furthermore, the sign determination device 100 includes an output unit 110 that outputs the content identified by the identification unit 108 .

The sign determination device 100 outputs the cause, location, time, etc. specified by the specifying unit 108 . Thereby, the user of the sign judging device 100 can confirm the cause, location, time, etc. specified by the specifying unit 108 .

(15) Also, the omen judging system 1 is an abnormal omen judging system formed by an omen judging device 100 capable of communicating with the terminal device 200, wherein the terminal device has: a request section 2001 for requesting abnormal omen judgment about the motor 11 and the load 13 of the motor 11; wherein the omen judging device 100 is equipped with: a measurement result obtaining section 101 for obtaining the measurement result of the current flowing through the motor 11 according to a request from the terminal device 200; An analysis unit 102 that decomposes the frequency into frequency components; and a prediction unit 105 that determines whether there is a sign of abnormality in at least one of the motor 11 and the load 13 of the motor 11 based on the time-series data of the frequency components.

The sign determination system 1 can determine whether at least one of the motor 11 and the load 13 of the motor 11 has a sign of abnormality based on the frequency component of the electric current of the motor 11 . Furthermore, the omen judging device 100 can remotely determine whether or not there is the omen even if it is not installed in the vicinity of the motor 11 and the load 13 that are affected by vibration, temperature change, or the like.

(16) The method for judging the sign of the present disclosure includes: obtaining the measurement result of the current flowing through the motor 11; performing frequency analysis on the measurement result and decomposing it into frequency components; based on the time series data of the frequency component, judging whether there is a sign of abnormality in at least one of the motor 11 and the load 13 of the motor 11.

The user of the omen determination method can determine whether at least one of the motor 11 and the load 13 of the motor 11 has a sign of abnormality based on the frequency component of the electric current of the motor 11 by using the omen determination method. In addition, users of the omen judgment method can remotely determine whether there is the omen even if they are not installed around the motor 11 and the load 13 that are affected by vibration, temperature change, etc., by using the omen judgment method.

(17) The memory medium of the present disclosure stores a program that causes the computer to execute the following content: obtain the measurement result of the current flowing through the motor 11; perform frequency analysis on the measurement result and decompose it into frequency components; based on the time series data of the frequency components, determine whether there is any sign of abnormality in at least one of the motor 11 and the load 13 of the motor 11.

The user of the program can determine whether at least one of the motor 11 and the load 13 of the motor 11 has a sign of abnormality based on the frequency component of the current of the motor 11 by executing the program on the computer. In addition, the user of the program can determine whether there is the above-mentioned omen at a long distance even if the user of the program does not install the motor 11 and the load 13 that are affected by vibration, temperature change, etc., by executing the program on a computer.

(18) In the omen judgment method of the present disclosure, the terminal device 200 has: a request step for judging the omen that the motor 11 and the load requirement of the motor 11 are abnormal; wherein, the omen judging device 100 capable of communicating with the terminal device 200 has a measurement result acquisition step of obtaining the measurement result of the current flowing through the motor 11 according to a request from the terminal device 200; an analysis step of frequency analysis and decomposing the measurement result into frequency components; And the step of predicting whether at least one of the load 13 of the motor 11 has a sign of abnormality.

The user of the omen determination method can determine whether at least one of the motor 11 and the load 13 of the motor 11 has a sign of abnormality based on the frequency component of the electric current of the motor 11 by using the omen determination method. In addition, users of the omen judgment method can remotely determine whether there is the omen even if they are not installed around the motor 11 and the load 13 that are affected by vibration, temperature change, etc., by using the omen judgment method. [industrial availability]

According to at least one of the above aspects, it can be determined whether or not there is a sign of abnormality in at least one of the motor and the load of the motor.

1: Omen Judgment System 10: Power source 11: Motor 12: wire 13: load 14: axis 15: Bearing 16: Measuring device 17: Converter 100: Omen judgment device 101: Measurement Result Acquisition Department 102: Analysis Department 103: Image Generation Department 104: Detection Department 105: The first judgment department 106: Calculation department 107: The second judgment department 108: specific department 109: update department 110: output unit 111: memory department 112: Processing Department 200: terminal device 1100: computer 1110: Processor 1120: main memory 1130: storage 1140: interface 2001: Ministry of Requirements

[ Fig. 1 ] A diagram showing the structure of an omen judgment system according to a first embodiment. [ Fig. 2] Fig. 2 is a schematic block diagram showing the structure of the omen judging device according to the first embodiment. [ Fig. 3 ] A diagram showing an example of a video in the first embodiment. [FIG. 4] A diagram showing an example of factor information in the first embodiment. [FIG. 5] A diagram showing an example of time information in the first embodiment. [ Fig. 6] Fig. 6 is a flow chart showing an example of the operation of the sign judging system according to the first embodiment. [ Fig. 7] Fig. 7 is a diagram showing the structure of a sign judging system according to the second embodiment. [ Fig. 8 ] Fig. 1 showing an example of training data in the second embodiment. [ Fig. 9] Fig. 2 showing an example of training data in the second embodiment. [ Fig. 10] Fig. 10 is a flow chart showing an example of the operation of the sign judging system according to the second embodiment. [ Fig. 11] Fig. 11 is a diagram showing the structure of a sign judging system according to a fifth embodiment. [ Fig. 12 ] A schematic block diagram showing the structure of a computer of at least one embodiment.

100: Omen judgment device

101: Measurement Result Acquisition Department

102: Analysis Department

103: Image Generation Department

104: Detection Department

105: The first judgment department

106: Calculation department

107: The second judgment department

108: specific department

109: update department

110: output unit

111: memory department

Claims (17)

An omen determination device comprising: a measurement result acquisition unit that acquires a measurement result of a current flowing through a motor; an analysis unit that frequency-analyzes and decomposes the measurement result into frequency components; a prediction unit that determines whether there is a sign of abnormality in at least one of the motor and a load of the motor based on the time-series data of the frequency component; wherein the prediction unit includes learning to determine parameters using training data that uses the time-series data of the frequency component as input data and a result of judging whether the input data has the sign of the abnormality as output data. finished model. As the sign judging device described in Claim 1, wherein, the aforementioned analysis unit decomposes into complex numbers of the aforementioned frequency components by fast Fourier transform (FFT (Fast Fourier Transform)); the aforementioned prediction unit determines whether there is a sign of abnormality in at least one of the aforementioned motor and the load based on the time series data of each of the aforementioned complex number frequency components. The sign judging device according to claim 1 or claim 2, wherein the sign of abnormality of the motor determined by the prediction unit includes at least one of breakage of a rotor bar in the motor, a change in tension of a belt connecting the motor and the load, and a void. The omen judging device as described in Claim 1 has: An image generation unit that generates an image of time-series data representing the frequency components; wherein the prediction unit determines whether there is a sign of abnormality in at least one of the motor and the load based on the image. In the sign judging device as described in claim 4, wherein, the analysis unit decomposes into a plurality of the aforementioned frequency components by fast Fourier transform (FFT (Fast Fourier Transform)); and the aforementioned image represents time-series data of each of the aforementioned complex frequency components. The omen judging device as described in claim 4 or claim 5, wherein the aforementioned image further displays time-series data of energy values associated with the aforementioned frequency component; a detection unit that detects a change value of the aforementioned frequency component based on the aforementioned image; and the aforementioned prediction unit compares the aforementioned change value to a preset first threshold to determine whether there is the aforementioned omen. The sign judging device according to claim 6, wherein the image represents the frequency components as a color graph with time and frequency as axes, and the energy value is represented by color on the color graph. An omen judgment device comprising: a measurement result acquisition unit that acquires a measurement result of a current flowing through a motor; An analyzing part that performs frequency analysis on the aforementioned measurement results and decomposes them into frequency components; a predicting part that determines whether there is an omen of abnormality in at least one of the aforementioned motor and the load of the aforementioned motor based on the time series data of the aforementioned frequency components; The sign judging device according to claim 8, wherein the specifying unit, when judging that there is the sign, further specifies the cause based on the cause information linking the image and the cause of the abnormality. The sign judging device according to claim 9, which includes: a calculation unit that calculates a degree of similarity based on the feature value of the image associated with the factor and the feature value of the image generated by the image generation unit; a second determination unit that determines whether the degree of similarity is equal to or greater than a preset second threshold when the sign is determined to be present; The sign judging device according to claim 10, further comprising: an update unit that receives an input from outside and updates the factor information when it is judged that the degree of similarity is not equal to or greater than the second threshold. As the sign judging device described in Claim 9, its Herein, the specifying unit compares the specified factor with time information associating the factor with time until the abnormality occurs, and further specifies the time. The sign judging device according to claim 8, further comprising: an output unit that outputs the content identified by the identification unit. An omen judging system, which is an abnormal omen judging system formed by an omen judging device as described in claim 1 or 8 capable of communicating with a terminal device, wherein the terminal device has: a request unit for omen judgment about an abnormality in the load demand of the motor and the motor; wherein the omen judging device includes: the measurement result obtaining unit that obtains the measurement result of the current flowing through the motor according to the request from the terminal device; The said prediction part which judges whether there is a sign of abnormality in at least one of the said electric motor and the said load of the said electric motor according to the said time-series data. A method for judging an omen, using the omen judging device as described in Claim 1 or 8, comprising: obtaining the aforementioned measurement result of the aforementioned current flowing through the aforementioned motor; Based on the aforementioned measurement results, the aforementioned current is frequency-analyzed and decomposed into the aforementioned frequency components; and based on the aforementioned timing data of the aforementioned frequency components, it is determined whether there is a sign of abnormality in at least one of the aforementioned motor and the aforementioned load of the aforementioned motor. A memory medium that memorizes a program that causes a computer to use the omen judging device as described in claim 1 or 8 to execute the following program: obtain the aforementioned measurement result of the aforementioned current flowing through the aforementioned motor; based on the aforementioned measurement result, perform frequency analysis and decompose the aforementioned current into the aforementioned frequency components; based on the aforementioned timing data of the aforementioned frequency components, determine whether there is an omen of abnormality in at least one of the aforementioned motor and the aforementioned load of the aforementioned motor. A method for judging an omen, comprising a terminal device and the omen judging device as described in claim 1 or 8 capable of communicating with the terminal device, wherein the terminal device has: a request step for judging the omen of the abnormality of the load demand of the motor and the motor; wherein the omen judging device capable of communicating with the terminal device has the step of obtaining the measurement result of the measurement result of the current flowing through the motor according to the request from the terminal device; performing frequency analysis on the measurement result and decomposing it into the frequency components An analyzing step; a predicting step of determining whether there is a sign of abnormality in at least one of the motor and the load of the motor based on the time-series data of the frequency component.

Images