Disconnection detection device

Abstract

To provide a disconnection detection device capable of detecting disconnection of a strand of a cable even in a state where a management object device is actually operated.SOLUTION: A disconnection detection device 1 for detecting a disconnection progress state of a conductor of a cable includes a data acquisition processing unit 21 that acquires operation information data 31 indicating a temporal change of an operation of a management target device 11, and resistance value data 32 indicating a resistance value of the conductor that changes due to the operation of the management target device 11. And a disconnection detection processing part 23 for detecting the disconnection of the strand or the disconnection progress state of the conductor. The acquisition of the operation information data 31 and the resistance value data 32 by the data acquisition processing part 21 is performed during the actual operation of the management target device 11.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a wire breakage detection device. [Background technology]

[0002] In cables wired to moving parts of equipment such as industrial robots, repeated bending and twisting of the moving parts gradually causes the wires that make up the conductor to break, eventually leading to a break in the conductor. Therefore, it is desirable to detect a break in a wire and replace the cable before the entire conductor breaks.

[0003] When the number of broken wires that make up the conductor is small, the change in the conductor's resistance is so slight that it is difficult to distinguish it from noise during detection. For this reason, in the past, when the conductor's resistance rose by several tens of percent from its initial value, it was often determined that a complete break was imminent and that a break had occurred at this point. However, in this case, there was little time (or number of operations) between the detection of a break and the complete break of the entire conductor, which could lead to the entire conductor breaking before the cable could be replaced, causing equipment such as industrial robots to stop operating.

[0004] Patent Document 1 proposes a method for detecting wire breakage based on the results of measuring the resistance of the conductor when the cable is operated (bent, twisted, etc.) at a constant cycle and performing frequency analysis on the fluctuations in the resistance. This method is highly sensitive, so it is possible to detect wire breakage even when, for example, only one wire is broken. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent Publication No. 2021-162570 Summary of the Invention [Problem to be solved by the invention]

[0006] Users (equipment users) of industrial robots and other devices and manufacturers (equipment manufacturers) of such devices perform maintenance of such devices. During device maintenance, the operation of the industrial robot or other device to be managed is stopped for inspection, but it is desirable to be able to perform maintenance even while the device is in operation to prevent the device from suddenly stopping due to a malfunction during operation. In particular, it is desirable to be able to detect whether a wire in a cable mounted on an industrial robot or other device is broken while the device is in operation.

[0007] SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a wire break detection device that can detect a break in a cable wire even when a device to be managed is in operation. [Means for solving the problem]

[0008] In order to solve the above-mentioned problems, the present invention provides a wire break detection device that detects the progression of a wire break in a cable having a conductor composed of a plurality of wires and that is subjected to strain due to operation based on the actual operation of a device to be managed, the wire break detection device comprising: a data acquisition processing unit that acquires operation information data that indicates changes in the operation of the device to be managed over time and resistance value data that indicates the resistance value of the conductor that changes due to the operation of the device to be managed; an analysis processing unit that performs frequency analysis of the operation information data and the resistance value data over a predetermined period of time; a database that associates the results of the frequency analysis of both by the analysis processing unit; and a wire break detection processing unit that detects a wire break or the progression of a wire break in the conductor based on the results of the frequency analysis of both, wherein the operation information data and the resistance value data are acquired by the data acquisition processing unit during the actual operation of the device to be managed.

[0009]

[0010] [Effects of the Invention]

[0011] According to the present invention, it is possible to provide a disconnection detection device that can detect a disconnection in a wire of a cable even when a device to be managed is in operation. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a schematic configuration diagram of a wire break detection device according to an embodiment of the present invention; [Figure 2] 10(a) to 10(c) are diagrams illustrating the analysis process. [Figure 3] 1A is a flow diagram of a disconnection detection method according to an embodiment of the present invention, and FIG. 1B is a flow diagram of an analysis process. [Figure 4] 10(a) to 10(c) are diagrams showing an example of changes in index value I. FIG. [Figure 5] FIG. 10 is a diagram showing an example of a change in an index value Ix. [Figure 6] FIG. 10 is a flowchart of an analysis process according to another embodiment of the present invention. [Figure 7] 10A and 10B are diagrams illustrating disconnection detection for a plurality of moving parts. [Figure 8] FIG. 10 is a flowchart of an analysis process according to a modified example of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] [Embodiment Mode] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

[0014] FIG. 1 is a schematic diagram of a wire breakage detection device 1 according to the present embodiment. The wire breakage detection device 1 is a device that detects wire breakage in a metal wire constituting a conductor of a cable 10 when the cable 10 is subjected to bending, twisting, or other motions based on the actual operation of the device 11. The device 11 has a movable part 11b, and the cable 10 is wired so as to pass through the movable part 11b. By operating the movable part 11b of the device 11, the cable 10 is repeatedly bent, twisted, or the like. FIG. 1 illustrates a case where the device 11 to which the cable 10 is wired is an industrial robot 11a. The conductor in the cable 10 to be detected for wire breakage is a stranded conductor formed by twisting together a plurality of wires.

[0015] Hereinafter, "breakage of a wire" refers to a breakage of each of the wires constituting the conductor of the cable 10. Furthermore, "the progression of a breakage of the conductor" refers to the breakage rate, or the percentage of the wires constituting the conductor that have broken due to repeated bending, twisting, and other actions applied to the cable 10, i.e., the number of broken wires among the total number of wires constituting the conductor. The progression of a breakage of the conductor can be expressed as the ratio (%) of the number of broken wires to the total number of wires constituting the conductor. For example, when the breakage rate of a wire constituting the conductor is 100%, it indicates that all of the wires constituting the conductor are broken. In other words, the progression of a breakage of the conductor can be said to be the progression of a breakage of the cable 10.

[0016] Furthermore, "conductor breakage" means that the conductor breakage rate has reached a preset rate, reaching a state where it is determined that the conductor has broken. For example, the state where it is determined that the conductor has broken is set as when the resistance increase rate of the conductor constituting cable 10 (the rate of increase in resistance relative to the initial resistance value of the conductor) exceeds 20%, and the conductor breakage rate (= the progress of the conductor breakage) at this time is, for example, 80% or more. Conductor breakage indicates the time when it is determined that cable 10 has reached the end of its life (= cable life), and serves as a guide for recommending replacement of cable 10.

[0017] A robot control device 12 for controlling the industrial robot 11a is connected to the industrial robot 11a. The robot control device 12 is equipped with an operation control unit 12a that controls the operation of the industrial robot 11a (operation of the movable part 11b). The robot control device 12 stores an operation program (not shown) that presets the order of operation of the movable part 11b, the angle of operation, speed, etc., and the operation control unit 12a operates the industrial robot 11a in accordance with the operation program.

[0018] The robot control device 12 is also equipped with a resistance value detection unit 12b that detects the resistance value of the conductor of the cable 10. The detection method of the resistance value detection unit 12b is not particularly limited. That is, the resistance value detection unit 12b may have any configuration as long as it can detect the resistance value of the conductor. For example, it may be configured to calculate the resistance value from the ratio of the voltage applied to the conductor to the current flowing through the conductor. The resistance value detection unit 12b does not have to be installed in the robot control device 12. For example, it may be installed in a unit (module) dedicated to resistance value detection that is disposed between the industrial robot 11a and the wire-breakage detection device 1, or it may be installed in the wire-breakage detection device 1 or the industrial robot 11a. By installing the resistance value detection unit 12b in the unit (module), the industrial robot 11a, the resistance value detection unit 12b, and the robot control device 12 can be connected to detect the resistance value of the conductor of the cable 10 without requiring complex modifications to the structures of the industrial robot 11a and the robot control device 12, respectively. The resistance value detected by the resistance value detector 12b is transmitted to the disconnection detector 1 and stored as resistance value data 32 in the memory 3 of the disconnection detector 1.

[0019] (Disconnection detection device 1) The wire break detection device 1 includes a control unit 2 and a storage unit 3. A display unit 4 is connected to the wire break detection device 1, and is configured to display various information on the display unit 4, such as operation information data 31 and resistance value data 32 (described later), or the results of wire break detection. The wire break detection device 1 is also provided with an input device 5 such as a keyboard, and various settings and the display contents of the display unit 4 can be operated by inputting information from the input device 5. The display unit 4 may be configured as a touch panel display so that it also serves as the input device 5. Furthermore, the display unit 4 and the input device 5 do not have to be connected to the wire break detection device 1 by wire, but may be connected wirelessly. In this case, the display unit 4 and the input device 5 may be, for example, a smartphone or a tablet.

[0020] The control unit 2 has a data acquisition processing unit 21, an analysis processing unit 22, a disconnection detection processing unit 23, and a cable life prediction processing unit 24. The data acquisition processing unit 21, the analysis processing unit 22, and the disconnection detection processing unit 23 are realized by appropriately combining a processing element such as a CPU, memories such as RAM and ROM, software, an interface, a storage device, etc.

[0021] (Data acquisition processing unit 21) The data acquisition processing unit 21 performs data acquisition processing to acquire operation information data 31, which is data indicating changes over time in information related to the operation of the device 11, and resistance value data 32, which is data on the resistance value of a conductor that changes over time due to the operation of the device 11. The operation of the device 11 is, for example, the operation of the movable part 11b when the device 11 is in actual operation. The data acquisition processing corresponds to the data acquisition step of the present invention. The data acquisition processing unit 21 acquires the operation information data 31 and the resistance value data 32 during the actual operation of the device 11.

[0022] In this embodiment, angle information of the movable part 11b is used as information related to the operation of the movable part 11b. That is, in this embodiment, the operation information data 31 is data indicating a change in the angle α of the movable part 11b over time. The angle information of the movable part 11b may be acquired from an operation program stored in the robot control device 12, or may be acquired by a sensor attached to the movable part 11b of the industrial robot 11a or by detecting the drive current of the motor that drives the movable part 11b. Furthermore, the resistance value data 32 may be received from the resistance value detection unit 12b of the robot control device 12.

[0023] (Analysis processing unit 22) The analysis processing unit 22 performs an analysis process to analyze the resistance value data 32 based on the motion information data 31 acquired by the data acquisition processing unit 21. The analysis process corresponds to the analysis step of the present invention. In this embodiment, the analysis processing unit 22 calculates, for each motion state (here, angle α), an average value of the resistance value during a predetermined period when the motion state is achieved.

[0024] 2(a) to (c) are diagrams for explaining the analysis process. As shown in FIG. 2(a), when the industrial robot 11a is in operation, the movable part 11b moves intermittently at various speeds. In the analysis process, the analysis is performed on only the data of a predetermined period (for example, 30 minutes) of the operation information data 31 and the resistance value data 32. The period to be analyzed can be set appropriately taking into consideration the operation cycle of the movable part 11b, etc. When the angle α of the movable part 11b is an arbitrary angle α k Since the time when the angle α of the movable part 11b is equal to or smaller than the arbitrary angle α k The resistance value r when nk (n is an integer equal to or greater than 1). In the example of Figure 2(a), the angle α k Therefore, the resistance value r corresponding to times t1, t2, . . . , t7 is 1k ,r 2k ,···,r 7k Then, the analysis processing unit 22 extracts the average value r of the extracted resistance values ​​using the following formula (1): k Ask for.

number

[0025] In this embodiment, the average value r k From this, angle α k The index value Δr is the normalized value obtained by subtracting the average value at k That is, the angle α k The average value r0 at 0 is calculated using the following formula (2), and the index value Δr k asked for.

number

[0026] Similarly, the index value Δr k By calculating the angle α as shown in Figure 2(b), k The index value Δr for kThe resistance value of the conductor changes due to the influence of strain caused by the movement of the movable part 11b even when there is no break in the wire. In the illustrated example, the angle α k When the angle is ±90 degrees, the resistance value of the conductor increases due to the influence of strain. When a break occurs in a wire constituting the conductor, the distance between the wires at the break point changes according to the movement of the movable part 11b, and the resistance value of the conductor fluctuates accordingly. For example, as shown in FIG. 2(b), if the first of the multiple wires constituting the conductor is broken, which is considered to be an initial break, the resistance value of the conductor increases with the occurrence of the initial break. As a result, when a break occurs in a wire, the resistance value of the conductor increases at a predetermined angle α k The resistance value (index value Δr k ) becomes larger. Such changes in resistance are very small and are usually difficult to observe because they are buried in noise components. However, the average resistance value over a certain long period of time, r k and then find the angle α k The difference between the average resistance value r0 at k By doing so, the index value Δr k The analysis processing unit 22 extracts the change in the resistance value of the conductor due to the breakage of the wire from the change in the angle α k Index value Δr for each k The calculation results are stored in the storage unit 3 as analysis data 33.

[0027] Figure 2(c) shows the time variation of the graph in Figure 2(b), and the index value Δr k In the illustrated example, the index value Δr k It is believed that an initial break has occurred at this point. Basically, when a break occurs, the change in the resistance value of the conductor (index value Δr k ) becomes larger. After that, as the wire breakage progresses (i.e., the ratio of wire breakage increases), the change in resistance value (index value Δr k ) is expected to become larger.

[0028] (Disconnection detection processing unit 23) The disconnection detection processing unit 23 performs a disconnection detection process to detect a disconnection in the wires that make up the conductor based on the analysis results (analysis data 33) from the analysis processing unit 22. The disconnection detection process corresponds to the disconnection detection step of the present invention. In this embodiment, the disconnection detection processing unit 23 uses the index value Δr k Based on this, breaks in the wires that make up the conductor are detected.

[0029] The disconnection detection processing unit 23 detects the predetermined angle α k The index value Δr k It is determined whether the resistance value of the conductor has become larger than a preset threshold value. By appropriately setting this threshold value, it is possible to detect an early stage of disconnection (when the wires start to disconnect) or to detect that a predetermined disconnection has progressed. For example, in order to detect that the disconnection has progressed two weeks before all of the wires constituting the conductor have completely disconnected, operation information data 31 and resistance value data 32 from the state of the conductor when it is new until it has completely disconnected are obtained in advance, and a time change such as that shown in FIG. 2(c) is created. Then, using the data transition in FIG. 2(c), the angle α k The index value Δr k is set as the threshold value for detecting a disconnection. k The index value Δr k becomes larger than a preset threshold value, it is possible to detect whether a break has occurred in the wire or whether the conductor has reached a predetermined break progression state. k The index value Δr k becomes larger than a preset threshold value, the disconnection detection processing unit 23 stores the result of the disconnection detection (determination result) in the storage unit 3 as disconnection detection data 34. A plurality of threshold values ​​may be set in stages, and the index value Δr k By comparing the signal with a plurality of threshold values, it may be possible to detect how far the disconnection has progressed.

[0030] However, the disconnection detection processing unit 23 may be configured to detect a disconnection in the wire by using machine learning based on the analysis data 33. That is, the disconnection detection processing unit 23 may use a trained model created in advance to detect a disconnection in the wire from the analysis results for which a disconnection is to be detected. The trained model may be the analysis results when there is no disconnection in the wire (the above-mentioned index value Δr k ) is used. More specifically, for example, anomaly detection using an autoencoder may be used to detect the presence or absence of a disconnection. When anomaly detection using an autoencoder is performed, the angle α when there is no disconnection is previously input to the autoencoder. k Index value Δr for each k The index value Δr in normal state is learned repeatedly. k In the autoencoder that has been repeatedly trained, the index value Δr k Even if you input Δr k is output as a predicted value. Therefore, the input index value Δr k If any abnormality is included in the input value, a deviation (error) will occur between the input value and the predicted value. Therefore, by determining whether the degree of this deviation is greater than a preset threshold, it is possible to detect the onset of a wire breakage. Furthermore, by appropriately adjusting the threshold, it is also possible to detect whether the conductor has reached a predetermined breakage progression state. Note that wire breakage can be detected using various machine learning algorithms, such as support vector machines, in addition to anomaly detection using an autoencoder.

[0031] In this embodiment, multiple industrial robots 11a (and robot control devices 12) are connected to the wire breakage detection device 1, and the multiple industrial robots 11a detect wire breakage in the conductor of the cable 10. Therefore, it is possible to efficiently collect operation information data 31 and resistance value data 32, and efficiently collect data for machine learning, and to perform learning using a large amount of data to create a trained model. This improves the prediction accuracy of the trained model. Note that in the example of FIG. 1, a robot control device 12 is provided for each industrial robot 11a, but multiple industrial robots 11a may also be configured to be controlled by a single robot control device 12. Furthermore, the wire breakage detection device 1 may be equipped with the functionality of the robot control device 12.

[0032] (Cable life prediction processing unit 24) The cable life prediction processing unit 24 calculates the index value Δr k Based on the value of , a cable life prediction process is performed to predict the time until the conductor breaks, i.e., the time until the cable 10 reaches the end of its life (hereinafter referred to as cable life). The cable life prediction processor 24 may predict the cable life taking into consideration the progress of the break detected by the break detection processor 23 and the operating conditions of the movable part 11b. For example, the time change as shown in FIG. 2(c) is acquired in advance, and the current index value Δr is calculated by referring to this time change. k The time required for reaching the cable lifespan may be predicted from the time corresponding to the cable lifespan. The prediction result of the cable lifespan prediction process is stored in the storage unit 3 as lifespan prediction data 35. Note that the cable lifespan prediction processor 24 is not essential and can be omitted.

[0033] (Disconnection detection method) FIG. 3(a) is a flow diagram of a disconnection detection method according to this embodiment. The flow in FIG. 3(a) starts when the device 11 starts operating. As shown in FIG. 3, first, in step S1, the data acquisition processing unit 21 starts acquiring operation information data 31 and resistance value data 32 from the resistance value detection unit 12b. Thereafter, in step S2, the analysis processing unit 22 determines whether unanalyzed resistance value data 32 exists in the memory unit 3 (for example, whether 30 minutes of unanalyzed data exists). Note that in step S2, it may also be determined whether a predetermined time has elapsed since the previous analysis (for example, whether 30 minutes have elapsed).

[0034] 3(b), in the analysis process, in step S31, the analysis processing unit 22 calculates the angle α k In step S32, the resistance value is extracted for each angle αk using the above equations (1) to (3). k Then, in step S33, the analysis processing unit 22 calculates the index value Δr k The calculation result is stored in the storage unit 3 as analysis data 33, and then the process returns to step S4 in FIG. 3(a).

[0035] In step S4, a disconnection detection process is performed. In the disconnection detection process, the disconnection detection processing unit 23 detects the index value Δr k Based on the calculation results (analysis data 33), it is detected whether a break has occurred in the wire (whether a break has begun to occur in the wire). By appropriately adjusting the threshold value, it is also possible to detect whether the conductor has reached a desired progress state of breakage. The detection results are stored in the memory unit 3 as breakage detection data 34. The detection results may also be displayed on the display 4, or an alert may be issued when a breakage is detected.

[0036] Thereafter, in step S5, it is determined whether the operation of the device 11 has been completed. If the determination in step S5 is NO, the process returns to step S2. If the determination in step S5 is YES, in step S6, data acquisition by the data acquisition processing unit 21 is completed, and then the process ends. Although not shown in the figure, a cable life prediction process may be performed after step S6.

[0037] (Other embodiments) In the above embodiment, the angle α k Index value Δr for each k However, the present invention is not limited to this, and other index values ​​can also be used. Below, an embodiment using frequency analysis will be described.

[0038] In this embodiment, the analysis processing unit 22 performs frequency analysis of the motion information data 31 and the resistance value data 32 for a predetermined period. More specifically, a fast Fourier transform (FFT) is performed on each of the motion information data 31 and the resistance value data 32 for a predetermined period (e.g., 30 minutes). By performing a fast Fourier transform on the motion information data 31 (here, data on the change in angle α over time), the angle α(t) is converted into the following equation (4). Note that ω in equation (4) na is the angular frequency, α ωn is the angular frequency ω na (n is an integer equal to or greater than 1).

number

[0039] Similarly, by performing a fast Fourier transform (FFT) on the resistance value data 32, the resistance value r(t) is converted into the following equation (5). mr is the angular frequency, r ωm is the angular frequency ω mr (m is an integer equal to or greater than 1).

number

[0040] Then, the analysis processing unit 22 extracts an arbitrary frequency component α ωn and an arbitrary frequency component r of resistance value data 32 ωm Multiplying this gives the index value I = α ωn ×r ωm Here, both frequency components α ωn ,r ωm was multiplied by all combinations to obtain n×m index values ​​I. However, this is not limiting, and only combinations that are likely to cause a change when a wire is broken may be picked out and used as the index value I.

[0041] As shown in FIG. 4(a), when the wires constituting the conductor are not broken, the strain applied to the conductor by the movement of the movable part 11b causes a na :ω mr When the angle between the conductor and the cable 10 reaches a certain ratio (for example, a ratio of integers), the index value I increases. Specifically, when the movable part 11b is operated so as to repeatedly bend the cable 10 left and right, the strain of the conductor increases when the angle reaches a maximum left and right in one cycle of operation, and therefore the resistance value increases twice in one cycle of operation. In such a case, ω na :ω mr The index value I for the combination of 1:2 is large even when the wire is not broken.

[0042] Furthermore, as shown in Figure 4(b), when a break occurs in the wire that makes up the conductor, ω na :ω mr When the cable 10 is repeatedly bent left and right, the distance between the wires at the break point increases when the cable is bent in one direction, so the resistance value increases. na :ω mr The index value I of the combination where the ratio ω = 1:1 is large. Note that the ratio ω of the index value I that increases when the wire is broken varies depending on the type of movement of the movable part 11b (twisting, U-shaped bending, etc.) and the layout of the cable 10. na :ω mrIn addition, for example, the index value I, which has increased due to strain, may become even larger when a wire breaks. As wire breakage progresses (i.e., as the number of broken wires increases), the index values ​​I for various combinations are expected to increase, as shown in Figure 4(c). Note that the graphs shown in Figures 4(a) to (c) are plotted against the angular frequency ω on the horizontal axis. na The frequency component α of the angle corresponding to ωn and the angular frequency ω on the vertical axis mr The frequency component of the resistance value r corresponding to ωm The magnitude (intensity) of the index value I multiplied by the two angular frequencies ω na ,ω mr The intersections are shown in different shades.

[0043] The wire breakage detection processing unit 23 detects a break in the wires that make up the conductor based on the results of frequency analysis of both the operation information data 31 and the resistance value data 32 by the analysis processing unit 22. Here, the wire breakage detection processing unit 23 detects a break in the wires based on the index values ​​I of all combinations (n ​​× m) obtained by calculation by the analysis processing unit 22. The wire breakage detection processing unit 23 uses, for example, a trained model created in advance to detect a break in the wires from the n × m index values ​​I that are targets for wire break detection. Here, the trained model can be an autoencoder that has repeatedly learned the index values ​​I when there is no break in the conductor, and wire breakage detection can be performed using anomaly detection using an autoencoder, as in the above embodiment.

[0044] Here, n × m index values ​​I are used, but since the number of data may become relatively large and the calculation load may become heavy, in order to reduce the number of data, it is also possible to use a combination of index values ​​I that are significantly affected by wire breakage. As mentioned above, the effect of wire breakage is determined by the frequency ratio (ω na :ω mr ) becomes larger at a certain ratio (for example, an integer equal to or greater than 1). Therefore, the analysis processing unit 22 adds up the products of the frequency components of the resistance value data 32 whose frequency is x times (where x is an integer equal to or greater than 1) the frequency of the operation information data 31 to obtain the index value I xis calculated by the following formula (6), and the index value I obtained by the calculation is x The disconnection of the wire may be detected based on the above.

number

[0045] Figure 5 shows the index value I x 1 shows an example of the change in the index value I1. In the example shown, even when there is no break in the wire, the index value I2 is relatively large due to the influence of the strain applied to the conductor. When an initial break occurs (at the position indicated by the symbol A), the index value I1 increases. As time passes and the break in the conductor progresses (the number of broken wires increases), the other index values ​​I3 and I4 also increase. In this way, these index values ​​I x It is possible to detect the progression of a conductor breakage using this.

[0046] 6 is a flow diagram of analysis processing in another embodiment. As shown in FIG. 6, first, in step S34, frequency analysis is performed on the resistance value data 32, and in step S35, frequency analysis is performed on the operation information data 31. Then, in step S36, n×m index values ​​I are calculated, and in step S36, the calculation results of the index values ​​I are stored in the storage unit 3 as analysis data 33.

[0047] Furthermore, if the device 11 has multiple moving parts, it is possible to perform a similar analysis for each moving part and detect disconnection. For example, as shown in Figure 7, there are two moving parts 11b and 11c, and the angle of one moving part 11b is α and the angle of the other moving part 11c is β. When the data (operation information data 31) of the angles α and β of both moving parts 11b and 11c are subjected to fast Fourier transform, the following equations (7) and (8) are obtained.

number

[0048] The analysis processing unit 22 calculates the frequency component α of the angle α of the movable part 11b. ωn and frequency component r of resistance value rωm and are multiplied by all combinations to get the index value I α =α ωn ×r ωm The analysis processing unit 22 also calculates the frequency component β of the angle β of the movable part 11c. ωn and frequency component r of resistance value r ωm and are multiplied by all combinations to get the index value I β =β ωn ×r ωm Calculate the following.

[0049] The disconnection state of the conductor in the movable part 11b moving at the angle α is expressed by the frequency component α ωn Frequency component r of resistance r related to ωm The disconnection state of the conductor in the movable part 11c moving at an angle β appears as the frequency component β of the angle β. ωn Frequency component r of resistance r related to ωm Therefore, the frequency component r ωm frequency component α ωn The index value I calculated by multiplying α (i.e., frequency component α ωn Frequency components related to r ωm Index value I with emphasis on α ) can be used to detect the breakage of the wire or the progress of the breakage of the conductor in the movable part 11b moving at an angle α. ωm The frequency component β ωn The index value I calculated by multiplying β (i.e., frequency component β ωn Frequency components related to r ωm Index value I with emphasis on β ) can be used to detect the breakage of the wire or the progression of the breakage of the conductor in the movable part 11c moving at the angle β. In this way, when there are multiple movable parts 11b, 11c, the frequency component r of the resistance value r related to the operation of each movable part 11b, 11c can be detected. ωm Index value I with emphasis on α ,I β By determining the above, it becomes possible to detect the breakage of the wire or the progress of the breakage of the conductor for each of the movable parts 11b and 11c.

[0050] (Actions and Effects of the Embodiments) As described above, the wire break detection method according to this embodiment includes a data acquisition process for acquiring operation information data 31, which is data showing changes over time in information related to the operation of device 11, and resistance value data 32, which is data on the resistance value of the conductor that changes over time due to the operation of device 11; an analysis process for analyzing the resistance value data 32 based on the operation information data 31 acquired in the data acquisition process and acquiring an index value for detecting a wire break; and a wire break detection process for detecting a wire break based on the index value acquired in the analysis process.

[0051] By analyzing the resistance data 32 based on the operation information data 31, it is possible to detect resistance values ​​that change depending on the operation, and based on the detection results, it is possible to detect a wire break even when the device 11 equipped with the cable 10 is in operation. For example, the device 11, such as an industrial robot 11a used in a factory, spends far more time in operation than in operation when maintenance or other tasks can be performed. Therefore, by making data collected during operation available, more data can be collected, improving the accuracy of wire break detection using machine learning.

[0052] (Variation) In the above embodiment, disconnection detection is performed using the motion information data 31 and the resistance value data 32. However, if the motion information data 31 cannot be acquired for some reason, it is also possible to generate (or predict) the motion information data 31 from the resistance value data 32 and perform disconnection detection. For example, the frequency analysis results of the resistance value data and the frequency analysis results of the motion information data may be stored in advance in a database in association with each other, data closest to the frequency analysis results of the measured resistance value data may be extracted from the database, and the index value I may be calculated using the frequency analysis results of the motion information data included in the extracted data. In this case, as shown in FIG. 8, instead of step S35 in FIG. 6, a process of extracting the frequency analysis results of the motion information data from a past database (DB) may be performed in step S38.

[0053] Furthermore, when the conductor is not broken, the strain applied to the conductor by the movement of the movable part 11b causes ω na :ω mr The index value I becomes larger when the ratio (for example, the ratio of integers) of the strain coefficients becomes larger, and this tendency continues to some extent even if the wire breakage progresses. Therefore, the index value I becomes larger due to the influence of strain. na :ω mr If the ratio is confirmed in advance, it is possible to generate the result of frequency analysis of the operation information data 31 using the result of frequency analysis of the resistance value data 32. In other words, it is possible to detect a disconnection without acquiring the operation information data 32.

[0054] More specifically, when calculating the index value I, the angular frequency ω na The frequency component α corresponding to ωn When is unknown, ω as shown in Figure 4(a) na =ω mr ω such as / 2 na :ω mr If you check the ratio of these in advance, you can get the angular frequency ω mr The frequency component r corresponding to ωn α ωn For example, when a bending motion in which the resistance value increases twice in one cycle of motion is applied, ωna :ω mr The index value I of the combination where ω = 1:2 is large. na =ω mr The index value I of the combination of α ωn Instead of r ωn / 2 Using this, the index value I can be calculated using the following equation (9). I=r ωn / 2 ×r ωm ···(9) However, both frequency components r ωn ,r ωm It is possible to obtain n×m index values ​​I by multiplying all combinations of these, or to appropriately select only combinations that are likely to cause changes when a wire is broken and use these as index values ​​I.

[0055] (Summary of the embodiment) Next, the technical ideas grasped from the above-described embodiments will be described by using the reference numerals and the like in the embodiments. However, the reference numerals and the like in the following description do not limit the components in the claims to the members and the like specifically shown in the embodiments.

[0056] [1] A wire break detection method for detecting that a wire of a cable (10) having a conductor composed of a plurality of wires and wired to a device (11) is broken due to the operation of the device (11), the wire break detection method comprising: a data acquisition step of acquiring operation information data (31) which is data showing changes over time in information related to the operation of the device (11) and resistance value data (32) which is data on the resistance value of the conductor which changes over time due to the operation of the device (11); an analysis step of analyzing the resistance value data (32) based on the operation information data (31) acquired in the data acquisition step to acquire an index value for detecting a wire break; and a wire break detection step of detecting a wire break based on the index value acquired in the analysis step.

[0057] [2] In the analysis step, for each of the operation states, the average value (r k) and normalizing the average value to obtain the index value, and in the wire break detection process, a wire break is detected based on the index value obtained in the analysis process.

[0058] [3] The wire break detection method according to [1] or [2], wherein in the analysis step, a frequency analysis is performed on the operation information data (31) and the resistance value data (32) for a predetermined period, and the index value is obtained based on the results of the frequency analysis, and in the wire break detection step, a wire break is detected based on the index value obtained in the analysis step.

[0059] [4] The wire break detection method according to [3], wherein in the analysis step, an arbitrary frequency component of the operation information data (31), which is the result of the frequency analysis, is multiplied by an arbitrary frequency component of the resistance value data (32) to calculate the index value (I), and in the wire break detection step, a wire break is detected based on the index value in the analysis step.

[0060] [5] In the analysis step, based on the result of the frequency analysis, the index value (I) is calculated by adding the products of frequency components whose frequency is x times (where x is an integer equal to or greater than 1) the frequency of the resistance value data (32) with respect to the operation information data (31). x ) and in the disconnection detection step, a disconnection of the wire is detected based on the index value in the analysis step.

[0061] [6] A wire break detection method according to any one of [1] to [5], wherein in the wire break detection process, a trained model is created by previously training using training data including the index value when there is no break in the wire, and a wire break is detected from the index value that is the target for wire break detection.

[0062] [7] A wire breakage detection device (1) having a conductor composed of a plurality of wires, which detects that the wire of a cable (10) wired to a device (11) is broken due to the operation of the device (11), the wire breakage detection device (1) comprising: a data acquisition processing unit (21) that acquires operation information data (31) that is data showing changes over time in information related to the operation of the device (11) and resistance value data (32) that is data on the resistance value of the conductor that changes over time due to the operation of the device (11); an analysis processing unit (22) that analyzes the resistance value data (32) based on the operation information data (31) acquired by the data acquisition processing unit (21) and acquires an index value for detecting a wire breakage; and a wire breakage detection processing unit (23) that detects a wire breakage based on the index value acquired by the analysis processing unit (22).

[0063] [8] A wire breakage detection device (1) having a conductor composed of a plurality of wires in a cable (10) wired to a device (11) for detecting a wire breakage due to the operation of the device (11), the wire breakage detection device (1) comprising: a data acquisition processing unit (21) for acquiring resistance value data (32), which is data on the resistance value of the conductor that changes over time due to the operation of the device (11); an analysis processing unit (22) for generating operation information data (31), which is data showing changes over time in information related to the operation of the device, from the resistance value data (32), analyzing the resistance value data (32) based on the operation information data (31), and acquiring an index value for detecting a wire breakage; and a wire breakage detection processing unit (23) for detecting a wire breakage based on the index value in the analysis processing unit (22).

[0064] (Addendum) Although the embodiments of the present invention have been described above, the invention according to the claims is not limited to the above-described embodiments. It should be noted that not all of the combinations of features described in the embodiments are necessarily essential to the means for solving the problems of the invention. Furthermore, the present invention can be appropriately modified and implemented within the scope of its spirit. [Explanation of symbols]

[0065] 1...Disconnection detection device 2...Control unit 21...Data acquisition processing unit 22...Analysis processing unit 23...Disconnection detection processing unit 3...Storage section 31...Operation information data 32...Resistance value data 33...Analysis data 34...Disconnection detection data 10...Cable 11...device 11a...Industrial robots 11b...Movable part 12a...Motion control section 12b...Resistance value detection section

Claims

1. A wire break detection device for detecting a wire break progression state of a cable having a conductor composed of a plurality of wires, the conductor being subjected to strain due to actual operation of a managed device, a data acquisition processing unit that acquires operation information data indicating a change over time in the operation of the managed device and resistance value data indicating a resistance value of the conductor that changes depending on the operation of the managed device; an analysis processing unit that performs frequency analysis of both the motion information data and the resistance value data for a predetermined period of time; a database in which the results of the frequency analysis of both the signals by the analysis processing unit are associated with each other; a disconnection detection processing unit that detects a disconnection of the wire or a disconnection progression state of the conductor based on the results of the frequency analysis of both the wires; and the data acquisition processing unit acquires the operation information data and the resistance value data while the managed device is in operation. Disconnection detection device.

2. the managed device is an industrial robot, The operation information data is data indicating a change over time in the angle α of the movable part of the industrial robot. The wire breakage detection device according to claim 1 .

3. The resistance value is detected by a resistance value detection unit (module) connected to the industrial robot, and is calculated from the ratio of the voltage applied to the conductor to the voltage flowing through the conductor. The wire breakage detection device according to claim 2 .

4. The detection by the disconnection detection processing unit utilizes machine learning. The wire breakage detection device according to claim 1 .

Images