Analysis apparatus, analysis method and program

Abstract

To provide an analysis apparatus, an analysis method and a program that analyze the degree of damage to a fiber rope in a non-destructive manner.SOLUTION: An analysis apparatus 10 comprises: an analysis unit 11 that analyzes a degree of damage to a fiber rope using an invariant model which indicates a relationship among a plurality of time-series data that is based on measurement data from one or more sensors attached to the fiber rope or in the vicinity thereof.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an analysis device, an analysis method, and a program. [Background technology]

[0002] Patent Document 1 describes a technology in which sounds emitted by a monitored object are input into a model to detect abnormalities in the monitored object. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2023 / 105546 Summary of the Invention [Problem to be solved by the invention]

[0004] However, there has been a problem in that it is difficult to analyze the degree of damage to fiber ropes non-destructively.

[0005] The present disclosure has been made to solve such problems, and aims to provide an analysis device, an analysis method, and a program for non-destructively analyzing the degree of damage to a fiber rope. [Means for solving the problem]

[0006] The analysis device according to the first aspect of the present disclosure includes an analysis unit that analyzes the degree of damage to the fiber rope using an invariant model that shows the relationship between multiple time series data based on measurement data from one or more sensors attached to the fiber rope or its surroundings.

[0007] An analysis method according to a second aspect of the present disclosure analyzes the degree of damage to a fiber rope using an invariant model that shows the relationship between multiple time series data based on measurement data from one or more sensors attached to the fiber rope or its surroundings.

[0008] A program according to a third aspect of the present disclosure causes a computer to execute a process of analyzing the degree of damage to a fiber rope using an invariant model that shows the relationship between multiple time series data based on measurement data from one or more sensors attached to the fiber rope or its surroundings. [Effects of the Invention]

[0009] According to the present disclosure, an analysis device, an analysis method, and a program for non-destructively analyzing the degree of damage to a fiber rope can be provided. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram showing a configuration of an analysis device according to the present disclosure. [Figure 2] 1 is a flowchart showing the flow of an analysis method according to the present disclosure. [Figure 3] FIG. 1 is a diagram illustrating a configuration of an analysis device according to the present disclosure. [Figure 4] FIG. 1 is a diagram for explaining an invariant model according to the present disclosure. [Figure 5] FIG. 1 is a diagram for explaining a method for generating time-series data according to the present disclosure. [Figure 6] 1 is a graph showing a load-time curve and the number of AE events when a tensile load is applied to a fiber rope according to the present disclosure. [Figure 7] 1 is a graph showing a load-time curve and the number of AE events when a tensile load is applied to a fiber rope according to the present disclosure. [Figure 8] 1 is a graph showing a load-time curve and the number of AE events when a tensile load is applied to a fiber rope according to the present disclosure. [Figure 9] 10 is a graph showing the change in anomaly score over time when a tensile load is applied to a fiber rope according to the present disclosure. [Figure 10] 1 is a graph showing the number of AE events when a load is repeatedly applied to a fiber rope according to the present disclosure. [Figure 11] 1 is a graph showing the abnormality score and the number of AE events when a load is repeatedly applied to a fiber rope according to the present disclosure. [Figure 12] 1 is a graph showing the abnormality score and the number of AE events when a load is repeatedly applied to a fiber rope according to the present disclosure. [Figure 13] FIG. 2 is a block diagram illustrating an example of a hardware configuration of an analysis apparatus according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] Embodiment 1 An example of the configuration of the analysis device 10 will be described below with reference to FIG. 1. The analysis device 10 may be a computer device that operates when a processor executes a program stored in a memory. The analysis device 10 may also be an information processing device, such as a server device. The analysis device 10 may also be composed of multiple computer devices. In this case, the components or functions that make up the analysis device 10 may be distributed across the multiple computer devices. The multiple computers may be connected via a network or directly via a cable or the like.

[0012] The analysis device 10 includes an analysis unit 11. The analysis unit 11 may be software or a module that performs processing by a processor executing a program stored in a memory, or may be hardware such as a circuit or a chip.

[0013] The analysis unit 11 analyzes the degree of damage to the fiber rope using an invariant model. The invariant model indicates the relationship between multiple time series data based on measurement data from one or more sensors attached to the fiber rope or its surroundings. The invariant model may also indicate the relationship between features of the multiple time series data.

[0014] The flow of the analysis method will be described with reference to Fig. 2. In the analysis method, the analysis unit 11 analyzes the degree of damage to the fiber rope using an invariant model (step S11).

[0015] The analysis device 10 analyzes the degree of damage to the fiber rope using an invariant model, and therefore can analyze the degree of damage to the fiber rope non-destructively.

[0016] Embodiment 2 An example of the configuration of the analysis device 100 will be described with reference to Fig. 3. The analysis device 100 is a specific example of the analysis device 10. The analysis device 100 includes a tester 110, AE sensors 121-122, a microphone 130, amplifiers 141-143, an A / D (Analog-Digital) converter 150, and a computer 160. Alternatively, the A / D converter 150 may be built into the computer 160.

[0017] The fiber rope 20 is attached to the testing machine 110. The fiber rope 20 is a rope made of fiber. The testing machine 110 applies a tensile load, a repeated load, or the like to the fiber rope 20. The fiber rope 20 used for training the invariant model may be different from the fiber rope 20 to be analyzed. Furthermore, multiple fiber ropes 20 may be used for training. By training the time-series measurement data of multiple fiber ropes 20, the performance of the invariant model can be improved. The fiber rope 20 may or may not be attached to the testing machine 110 and subjected to a tensile test.

[0018] The AE sensors 121-122 are attached near where the fiber rope 20 is connected to the testing machine 110 (sometimes referred to as the "root") and at the center of the fiber rope 20, respectively, and measure signals due to elastic waves generated in the fiber rope 20. The AE sensors 121-122 measure signals due to elastic waves generated by material deformation, transformation, transition, breakage, etc. When the AE sensors 121-122 are not distinguished from one another, they may be simply referred to as AE sensors 120. The microphone 130 is attached to the periphery of the fiber rope 20 and measures audible sound signals from the fiber rope 20. The AE sensors 121-122 and the microphone 130 are attached at different positions in the extension direction of the fiber rope.

[0019] The frequency range of the microphone 130 is lower than the frequency range of the AE sensor 120. By using both the AE sensor 120 and the microphone 130, the analysis device 100 can use information from a wider frequency range.

[0020] A common problem of the AE sensor 120 and the microphone 130 is that they are susceptible to noise from the surrounding environment and electrical noise. The second embodiment can solve this problem by using an invariant model.

[0021] The amplifiers 141 to 143 respectively amplify the measurement signals of the AE sensor 121, the AE sensor 122, and the microphone 130. The A / D converter 150 acquires the measurement signals amplified by the amplifiers 141 to 143.

[0022] The computer 160 is, for example, a PC (Personal Computer) and includes a processor and memory (not shown). The computer 160 includes a storage unit 161, a model generation unit 162, and an analysis unit 163. The storage unit 161 is a storage medium accessible by the processor and stores a program. The processor may execute a program loaded into the memory to realize the functions of the model generation unit 162 and the analysis unit 163. The storage unit 161 also stores an invariant model (described later).

[0023] First, an exemplary explanation of a general invariant model will be provided with reference to FIG. 4. FIG. 4 includes time-series data measured by sensors 1 to 4 under normal conditions as training data 30. The invariant model 40 represents the invariant relationship between multiple time-series data. In the relationship diagram representing the invariant model 40, "1" to "4" correspond to sensors 1 to 4, respectively. After generating the invariant model 40, the actual measurement data 50 of sensor 1 is compared with the predicted value calculated using the invariant model 40 from the measurement values of sensors 2 to 4. Similarly, the measurement data (measurements) of sensors 2 to 4 are compared with the predicted value. The graph below shows the time changes in the measurement values and predicted values of any of sensors 1 to 4. An anomaly score is calculated based on the maximum difference (also known as residual) between the measurement value and the predicted value. The anomaly score corresponds to the degree of damage to the fiber rope 20.

[0024] 3, the model generation unit 162 generates time-series data for each frequency band from the time-series measurement data of each AE sensor 120 attached to the fiber rope 20 in a normal state using a Fourier transform. Similarly, the model generation unit 162 generates time-series data for each frequency band from the time-series measurement data of the microphone 130 attached to the periphery of the fiber rope 20 in a normal state using a Fourier transform. Then, the model generation unit 162 generates an invariant model that learns the invariant relationship between the generated multiple pieces of time-series data. Then, the model generation unit 162 stores the invariant model in the storage unit 161.

[0025] It can also be considered that the number of sensors in the invariant model is increased by generating time-series data for each frequency band using a Fourier transform. Figure 5 illustrates an example of a method for generating time-series data. The upper diagram shows the time-series measurement data (raw data) measured by the AE sensor 120 or the microphone 130. The broken diagram shows the results of performing an FFT (Fast Fourier Transform) on the measurement data for each time interval. The FFT results corresponding to different time intervals are arranged vertically. Furthermore, as shown in the lower diagram, the FFT results are averaged for each frequency domain. The width of the frequency domain is, for example, 100 Hz. The averaged values across multiple time intervals are combined to generate time-series data for that frequency domain.

[0026] Referring to FIG. 3, the analysis unit 163 analyzes the damage level of the fiber rope 20 to be analyzed using an invariant model. Specifically, the analysis unit 163 generates time-series data for each frequency band from the time-series measurement data of each AE sensor 120 attached to the fiber rope 20 to be analyzed using a Fourier transform. Similarly, the analysis unit 163 generates time-series data for each frequency band from the time-series measurement data of a microphone 130 attached to the periphery of the fiber rope 20 to be analyzed using a Fourier transform. The analysis unit 163 then analyzes the damage level from the generated multiple pieces of time-series data using an invariant model. Specifically, the analysis unit 163 calculates the damage level (also referred to as an anomaly score) based on the maximum value of the difference between the actual value of the time-series data and the predicted value predicted from other time-series data using the invariant model. The damage level may be, for example, the sum of the above-mentioned maximum values in the multiple pieces of time-series data.

[0027] Next, with reference to FIGS. 6, 7, 8, and 9, the results of an experiment in which a tensile load was applied to the fiber rope 20 will be described.

[0028] The upper graph in Figure 6 shows the load-time curve when a tensile load is applied to six fiber ropes 20, which are accepted test specimens. The tensile condition is 100 mm / min (displacement control). The lower graph in Figure 6 shows the number of AE events when a tensile load is applied to six fiber ropes 20, which are accepted test specimens. An AE event means that an AE wave with an amplitude above the threshold is detected by the AE sensor 120. The number of AE events at the time of breakage varies from test specimen to test specimen, but the timing at which the number of AE events increases is consistent between test specimens.

[0029] The upper graph in Figure 7 shows the load-time curves when a tensile load is applied to five fiber ropes 20 that serve as abrasion test specimens. The number of abrasion cycles differs between the abrasion test specimens. The lower graph in Figure 7 shows the number of AE events when a tensile load is applied to six fiber ropes 20 that serve as abrasion test specimens. As can be seen from the upper graph, the breaking load decreases as the number of abrasion cycles increases. As can be seen from the lower graph, the timing at which the number of AE events increases is roughly consistent between the test specimens, regardless of the number of abrasion cycles.

[0030] The upper graph in Figure 8 shows the load-time curve when a tensile load is applied to four fiber ropes 20 that are S-bend test specimens. The number of S-bends varies between the S-bend test specimens. The lower graph in Figure 8 shows the number of AE events when a tensile load is applied to four fiber ropes 20 that are S-bend test specimens. Referring to the upper graph in Figure 8, it was confirmed that the breaking load decreases as the number of bends increases. Referring to the lower graph in Figure 8, there is variation in the number of AE events, which is thought to be due to the timing at which the test started.

[0031] Figure 9 shows the calculation results of the abnormality scores based on the experimental results. The results for the period (0 to 37 s) in which no AE events occurred in the above-mentioned accepted test specimen were used as learning data. The upper graph in Figure 9 is a graph showing the abnormality scores (degree of damage) when a tensile load was applied to the above-mentioned worn test specimen. The abnormality scores at the time of breakage for worn test specimens with different numbers of wears fall within a certain range, and the changes in the abnormality scores for worn test specimens with different numbers of wears are similar to each other. Therefore, it is believed that breakage of the fiber rope 20 can be predicted using the abnormality scores.

[0032] The lower graph in Figure 9 is a graph showing the abnormality scores (degree of damage) when a tensile load is applied to the above-mentioned S-bend test specimen. The abnormality scores (degree of damage) at break for S-bend test specimens with different S-bend counts fall within a certain range, and the changes in the abnormality scores for S-bend test specimens with different S-bend counts are similar to each other. Therefore, the abnormality scores can be used to predict breakage of a fiber rope.

[0033] Next, with reference to FIGS. 10, 11, and 12, the results of an experiment in which a load was repeatedly applied to the fiber rope 20 will be described.

[0034] The upper graph in Figure 10 shows the change over time in the number of AE events when a repeated load was applied to the abrasion test specimen. The lower graph in Figure 10 shows the change over time in the number of AE events when a repeated load was applied to the S-bend test specimen. The maximum load was 15% of the standard strength of the fiber rope 20, the minimum load was 2% of the standard strength, the frequency was 0.125 Hz, and the number of repetitions was 75 times (10 minutes). No AE events occurred in the accepted test specimen, a small number of AE events occurred in the abrasion test specimen, and an increase in the number of AE events was confirmed in the S-bend test specimen.

[0035] 11 and 12 show the calculation results of the anomaly scores. The measurement data of the accepted specimens was used as training data.

[0036] The upper graph in Fig. 11 shows the change over time in the anomaly score (degree of damage) when a load is repeatedly applied to a wear test piece. The lower graph in Fig. 11 shows the change over time in the number of AE events when a load is repeatedly applied to a wear test piece. Because a small peak in the anomaly score exists before the time when an increase in AE events is detected, the analyzer 100 may be able to detect damage that cannot be detected from the number of AE events.

[0037] The upper graph in Fig. 12 shows the change over time in the anomaly score (degree of damage) when a repeated load is applied to an S-bend test piece. The lower graph in Fig. 12 shows the change over time in the number of AE events when a repeated load is applied to an S-bend test piece. A small peak in the anomaly score occurs before the time when the number of AE events increases, and the analysis device 100 may be able to detect damage that cannot be detected from the number of AE events.

[0038] Since the damage mechanism of fiber ropes is unknown and there are no standards for replacement timing, a technology for non-destructively analyzing the damage level of fiber ropes has been desired. In the second embodiment, the damage level of fiber ropes can be analyzed non-destructively by using an invariant model.

[0039] FIG. 13 is a block diagram showing an example configuration of the analysis device 10 and computer 160 (hereinafter referred to as the analysis device 10, etc.) described in the above embodiment. Referring to FIG. 13, the analysis device 10, etc. includes a network interface 1001, a processor 1002, and a memory 1003. The network interface 1001 may be used to communicate with a network node. The network interface 1001 may include, for example, a network interface card (NIC) that complies with the IEEE 802.3 series. IEEE stands for Institute of Electrical and Electronics Engineers.

[0040] The processor 1002 reads and executes software (computer programs) from the memory 1003 to perform the processes of the analysis device 10 and the like described using flowcharts in the above-described embodiments. The processor 1002 may be, for example, a microprocessor, an MPU, or a CPU. The processor 1002 may include multiple processors.

[0041] The memory 1003 is configured by a combination of volatile memory and non-volatile memory. The memory 1003 may include storage located remotely from the processor 1002. In this case, the processor 1002 may access the memory 1003 via an I / O (Input / Output) interface (not shown).

[0042] 13, memory 1003 is used to store software modules. Processor 1002 reads and executes these software modules from memory 1003, thereby performing the processing of analysis device 10 and the like described in the above-described embodiment.

[0043] As explained using FIG. 13, each of the processors possessed by the analysis device 10 etc. in the above-described embodiments executes one or more programs including a group of instructions for causing a computer to perform the algorithm explained using the drawings.

[0044] In the above examples, the program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored on a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable medium or tangible storage medium includes random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray® disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable medium or communication medium includes electrical, optical, acoustic, or other forms of propagated signals.

[0045] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0046] Each drawing is merely an example for describing one or more embodiments. Each drawing may relate not only to one particular embodiment, but also to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0047] Some or all of the elements (e.g., configurations and functions) described in Supplementary Notes 2 to 5 that are dependent on Supplementary Note 1 may also be dependent on Supplementary Notes 6 and 10 in the same dependency relationship as Supplementary Notes 2 to 5. Some or all of the elements described in any Supplementary Note may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0048] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes. (Appendix 1) an analysis unit that analyzes the degree of damage to the fiber rope using an invariant model that indicates the relationship between multiple time-series data based on measurement data from one or more sensors attached to the fiber rope or its periphery; An analytical device comprising: (Appendix 2) The analysis unit further generates time series data for each frequency band from the measurement data of each sensor using Fourier transform, thereby generating the plurality of time series data. 2. The analytical device of claim 1. (Appendix 3) The one or more sensors include an Acoustic Emission (AE) sensor attached to the fiber rope and a microphone attached to the periphery of the fiber rope. 3. The analytical device of claim 1 or 2. (Appendix 4) The one or more sensors are attached at different positions in the extension direction of the fiber rope. 4. The analytical device of claim 3. (Appendix 5) Repeated loads are applied to the fiber rope. 2. The analytical device of claim 1. (Appendix 6) The damage level of the fiber rope is analyzed using an invariant model that indicates the relationship between a plurality of time series data based on measurement data from one or more sensors attached to the fiber rope or its surroundings. Analysis method. (Appendix 7) Furthermore, the plurality of time series data are generated by generating time series data for each frequency band from the measurement data of each sensor using Fourier transform. Analytical method described in Appendix 6. (Appendix 8) The one or more sensors include an Acoustic Emission (AE) sensor attached to the fiber rope and a microphone attached to the periphery of the fiber rope. The analytical method described in Appendix 6 or 7. (Appendix 9) The one or more sensors are attached at different positions in the extension direction of the fiber rope. Analytical method described in Appendix 8. (Appendix 10) A program that causes a computer to execute a process for analyzing the degree of damage to a fiber rope using an invariant model that shows the relationship between multiple time series data based on measurement data from one or more sensors attached to the fiber rope or its surroundings. [Explanation of symbols]

[0049] 10, 100 analyzer 11, 163 Analysis Department 110 Testing Machine 120, 121, 122 AE sensors 130 microphones 141~143 Amplifier 150 A / D converter 160 Computers 161 Storage section 162 Model Generation Unit 20 Fiber rope 30 training data 40 Invariant Model 1001 Network Interface 1002 processor 1003 memory

Claims

1. an analysis unit that analyzes the degree of damage to the fiber rope using an invariant model that indicates the relationship between a plurality of time series data based on measurement data from one or more sensors attached to the fiber rope or its periphery; An analytical device comprising:

2. The analysis unit further generates time series data for each frequency band from the measurement data of each sensor using Fourier transform, thereby generating the plurality of time series data. The analytical device of claim 1 .

3. The one or more sensors include an acoustic emission (AE) sensor attached to the fiber rope and a microphone attached to the periphery of the fiber rope. The analytical device according to claim 1 or 2.

4. The one or more sensors are attached at different positions in the extension direction of the fiber rope. The analytical device according to claim 3 .

5. Repeated loads are applied to the fiber rope. The analytical device according to claim 1 or 2.

6. The degree of damage to the fiber rope is analyzed using an invariant model that indicates the relationship between a plurality of time series data based on measurement data from one or more sensors attached to the fiber rope or its periphery. Analysis method.

7. Furthermore, the plurality of time series data are generated by generating time series data for each frequency band from the measurement data of each sensor using Fourier transform. The analytical method according to claim 6.

8. The one or more sensors include an acoustic emission (AE) sensor attached to the fiber rope and a microphone attached to the periphery of the fiber rope. The analytical method according to claim 6 or 7.

9. The one or more sensors are attached at different positions in the extension direction of the fiber rope. The analytical method according to claim 8.

10. A program that causes a computer to execute a process for analyzing the degree of damage to a fiber rope using an invariant model that shows the relationship between multiple time series data based on measurement data from one or more sensors attached to the fiber rope or its surroundings.

Images