Anomaly detection device and anomaly detection method
The abnormality determination device calculates an evaluation value for the entire series of operations to detect machine abnormalities, overcoming the challenge of transient phenomena during operation start, ensuring consistent detection regardless of the machine's state.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- OMRON CORP
- Filing Date
- 2022-04-01
- Publication Date
- 2026-05-15
AI Technical Summary
Existing abnormality detection devices for machines, such as machine tools, struggle to detect abnormalities during transient phenomena in the waveform of target signals, particularly at the start of operations.
An abnormality determination device that calculates an evaluation value for the entire series of operations based on acquired signals, determining the machine's abnormality by comparing the calculated value to a threshold, regardless of the operating state.
Enables consistent abnormality detection in machines by ignoring transient phenomena, allowing for reliable determination of machine health across various operating states.
Smart Images
Figure 0007859150000002 
Figure 0007859150000003 
Figure 0007859150000004
Abstract
Description
Technical Field
[0001] The present disclosure relates to an abnormality determination device and an abnormality determination method for determining an abnormality of a machine including a machine tool and a processing machine.
Background Art
[0002] Patent Document 1 discloses an abnormality detection device for a machine tool. In the abnormality detection device, an evaluation value during normal operation of the machine tool is used to compare with a preset normal range and an evaluation value during operation of the machine tool, and when the evaluation value during operation deviates from the normal range, the machine tool is determined to be abnormal. The evaluation value is obtained by quantifying the waveform change of a target signal, which is a signal of a specific period during the operation of the machine tool among the signals output from the sensor.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] In the abnormality detection device, usually, the abnormality of the machine tool is determined with a period in which the waveform is in a steady state as a specific period. Therefore, in the abnormality detection device, it may be difficult to detect an abnormality of the machine tool in an operating state where a transient phenomenon occurs in the waveform of the target signal (for example, immediately after the start of a series of operations of the machine tool).
[0005] The present disclosure aims to provide an abnormality determination device and an abnormality determination method capable of determining an abnormality of a machine regardless of the operating state.
Means for Solving the Problems
[0006] An abnormality determination device according to one aspect of the present disclosure is An acquisition unit that acquires signals corresponding to the operation of a machine, including a machine tool and a processing machine, which can repeatedly perform a series of operations, A calculation unit that calculates an evaluation value of the entire series of operations based on the signals acquired during an operation period set according to the length of the series of operations, A determination unit that determines whether the machine tool is abnormal or not based on the calculated evaluation value, Equipped with, The determination unit determines that the machine is malfunctioning if the calculated evaluation value is greater than the threshold.
[0007] An abnormality determination method in one aspect of this disclosure is: A signal corresponding to the operation of a machine, including a machine tool and a processing machine capable of repeatedly performing a series of operations, is acquired. Based on the signals acquired during the operating period set according to the length of the series of operations, an evaluation value for the entire series of operations is calculated. If the calculated evaluation value is greater than the threshold, the machine is determined to be malfunctioning. [Effects of the Invention]
[0008] According to the abnormality detection device of the above embodiment, an abnormality detection device capable of determining abnormalities in a machine regardless of its operating state can be realized. [Brief explanation of the drawing]
[0009] [Figure 1] A block diagram showing an abnormality detection device according to one embodiment of the present disclosure. [Figure 2] A graph showing an example of the waveform of a signal output from a sensor that detects machine signals. [Figure 3] A flowchart illustrating the abnormality detection process performed by the abnormality detection device shown in Figure 1. [Figure 4] Figure 1 shows a graph illustrating an example of an evaluation value for the entire sequence of operations calculated based on signals output from sensors that detect signals from a normal machine, as determined by the abnormality detection device. [Figure 5]Figure 1 shows a graph illustrating an example of an evaluation value for the entire sequence of operations calculated based on signals output from sensors that detect abnormal machine signals using the abnormality detection device. [Figure 6] This graph shows an example of an evaluation value calculated based on signals output from a sensor that detects signals from a normal machine, using a conventional method. [Figure 7] This graph shows an example of an evaluation value calculated based on signals output from sensors that detect abnormal machine signals using conventional methods. [Modes for carrying out the invention]
[0010] An example of this disclosure is described below with reference to the attached drawings. The following description is essentially illustrative and is not intended to limit this disclosure, its applications, or its uses. The drawings are schematic, and the proportions of the dimensions, etc., do not necessarily correspond to those of reality.
[0011] An abnormality determination device 1 in one embodiment of the present disclosure, as shown in Figure 1, comprises an acquisition unit 10, a calculation unit 20, and a determination unit 30, and performs abnormality determination of a machine 100 that can be repeatedly executed as a series of operations. The machine 100 includes, for example, a machine tool that shapes a workpiece into a required shape and a processing machine that processes a material into a desired shape using a machine tool or the like. In this embodiment, the abnormality determination device 1 comprises a calculation control unit 40 and an output unit 50.
[0012] An anomaly detection device 1, as an example, comprises a CPU 2, a storage device 3, and a communication device 4. Each of the acquisition unit 10, calculation unit 20, determination unit 30, and calculation control unit 40 is a function realized, for example, by the CPU 2 executing a predetermined program. The storage device 3 is composed of, for example, ROM and RAM, and stores programs and the like necessary for performing anomaly detection. The communication device 4 is configured to communicate wirelessly or via wired connection with an external device (not shown).
[0013] The acquisition unit 10 acquires a signal corresponding to the operation of the machine 100, for example, via the communication device 4. Signals corresponding to the operation of the machine 100 include, for example, signals output from sensors that detect signals of the machine 100, signals output from sensors that detect the current supplied to the machine 100, and signals output from an encoder attached to a motor that drives the machine 100.
[0014] When the machine 100 uses a drill to make a hole in a metal workpiece, an example of the waveform of the signal output from a sensor that detects the signal of the machine 100 is shown in FIG. 2. In (1) and (3), a transient phenomenon occurs in which the waveform of the acquired signal temporarily increases due to, for example, switching of the driving state of the drill. In (2) and (4), the waveform of the acquired signal is in a steady state. (1) to (4) in FIG. 2 show the following operating states. (1) The operating state until the drill approaches and contacts the workpiece while being driven. (2) The operating state in which the drill moves to a predetermined position inside the workpiece after contacting the workpiece. (3) The operating state until the drill stops after reaching the predetermined position. (4) The operating state in which the drill is separated from the workpiece while stopped.
[0015] The calculation unit 20 calculates an "evaluation value of the entire series of operations of the machine 100 (hereinafter referred to as the evaluation value)" based on the signal acquired by the acquisition unit 10 during an operation period set according to the length of a series of operations of the machine 100. The evaluation value is, for example, an average value or an effective value calculated from the following formula 1. In the following formula 1, x i is the measured value of the sensor acquired in each operation constituting the series of operations, and n is the number of operations included in the series of operations.
Equation
[0016] The average value is the centroid of the entire data set and is calculated by summing the sensor measurements taken for each operation that makes up a series of operations and dividing by the number of operations that make up the series. The RMS value indicates the power consumption that would be achieved if all AC voltage and current were replaced with DC voltage and current, respectively, assuming that the AC voltage and current were replaced with what voltage and current would be equivalent.
[0017] The determination unit 30 makes an abnormality determination to determine whether the machine 100 is abnormal or not based on the evaluation value calculated by the calculation unit 20. Specifically, the determination unit 30 determines that the machine 100 is abnormal if the calculated evaluation value deviates from the normal range. Whether the calculated evaluation value deviates from the normal range is determined, for example, by whether the calculated evaluation value is greater than a threshold.
[0018] The threshold is set, for example, based on a pre-calculated normal evaluation value, or based on a pre-calculated normal evaluation value and a pre-calculated abnormal evaluation value. For example, the determination unit 30 sets the threshold to the 3-sigma value of the normal evaluation value, or any value between the normal evaluation value and the abnormal evaluation value (for example, an intermediate value). The normal evaluation value is the evaluation value when the machine 100 is made to perform a series of operations and no abnormality occurs in the machine 100, and the abnormal evaluation value is the evaluation value when the machine 100 is made to perform a series of operations and an abnormality occurs in the machine 100. When obtaining the normal evaluation value and the abnormal evaluation value, whether or not an abnormality has occurred in the machine 100 is determined, for example, by the user's visual inspection.
[0019] The calculation control unit 40 causes the calculation unit 20 to calculate an evaluation value at set intervals after the machine 100 has started a series of operations. For example, the calculation control unit 40 outputs a calculation signal to the calculation unit 20 to calculate an evaluation value at any timing after the machine 100 has started a series of operations. The calculation signal is output repeatedly, for example, until the machine 100 stops or until it is determined that the machine 100 is abnormal.
[0020] The output unit 50 outputs the result of the abnormality determination made by the determination unit 30 to the user, for example, via the communication device 4.
[0021] Referring to Figure 3, an example of the abnormality detection process performed by the abnormality detection device 1 will be explained. Here, we will explain the case where it is determined whether the calculated evaluation deviates from the normal range by whether the calculated evaluation value is greater than a threshold. The abnormality detection process is performed, for example, by the CPU 2 executing a predetermined program.
[0022] As shown in Figure 3, when the abnormality detection process is started, the abnormality detection device 1 determines whether the acquisition unit 10 is able to acquire signals from the machine 100 throughout the entire operating period (step S1).
[0023] If it is determined that a signal can not be obtained from machine 100 throughout the entire operating period, the abnormality detection process ends. If it is determined that a signal can be obtained from machine 100 throughout the entire operating period, the calculation unit 20 calculates an evaluation value for the entire series of operations based on the acquired signal (step S2).
[0024] Once the evaluation value for the entire series of operations is calculated, the determination unit 30 determines whether the calculated evaluation value is greater than a threshold (step S3). If it is determined that the calculated evaluation value is greater than a threshold, the determination unit 30 determines that the machine 100 is abnormal (step S4), and the output unit 50 outputs the determination result to the user, ending the abnormality determination process.
[0025] If the calculated evaluation value is not determined to be greater than the threshold, the process returns to step S1, where it is determined whether the acquisition unit 10 acquired a signal from the machine 100 throughout the entire operating period.
[0026] The anomaly detection device 1 can achieve the following effects:
[0027] Figures 4 to 7 show examples of evaluation values calculated based on signals output from a sensor that detects signals from machine 100 when machining a workpiece using an end mill. Figures 4 and 5 are examples of evaluation values calculated by the calculation unit 20 of the abnormality determination device 1. Figures 6 and 7 are examples of evaluation values calculated by a conventional method. In Figures 6 and 7, signals with a steady waveform are extracted from the output signals, and an evaluation value is calculated each time a signal is output. In other words, conventionally, evaluation value group B in Figures 6 and 7 is not calculated, and abnormality determination of machine 100 is performed using only evaluation value group A in Figures 6 and 7.
[0028] In contrast, the abnormality detection device 1 comprises an acquisition unit 10, a calculation unit 20, and a determination unit 30. The acquisition unit 10 acquires signals corresponding to the operation of the machine 100, which can repeatedly perform a series of operations. The calculation unit 20 calculates an evaluation value for the entire series of operations based on the signals acquired during an operation period set according to the length of the series of operations of the machine 100. The determination unit 30 makes an abnormality determination based on the calculated evaluation value to determine whether the machine 100 is abnormal or not. If the calculated evaluation value deviates from the normal range (for example, if the calculated evaluation value is greater than the threshold), the machine 100 is determined to be abnormal. In other words, as shown in Figures 4 and 5, the abnormality detection device 1 calculates the evaluation value C for the entire series of operations without extracting signals. With this configuration, the machine 100 can be constantly monitored without being affected by transient phenomena, so an abnormality detection device 1 can be realized that can determine abnormalities in the machine 100 regardless of its operating state.
[0029] The abnormality detection device 1 can optionally adopt one or more of the following configurations. In other words, one or more of the following configurations can be optionally deleted if they were included in the above embodiment, and optionally added if they were not included in the above embodiment. By adopting such a configuration, an abnormality detection device 1 can be realized that can more reliably determine abnormalities in the machine 100 regardless of its operating state.
[0030] The determination unit 30 determines that the machine 100 is abnormal if the calculated evaluation value is greater than the threshold.
[0031] The determination unit 30 sets a threshold based on a pre-calculated normal evaluation value, or based on both a pre-calculated normal evaluation value and a pre-calculated abnormal evaluation value.
[0032] The determination unit 30 sets the 3-sigma value of the normal evaluation value as the threshold.
[0033] The determination unit 30 sets an arbitrary value between the normal evaluation value and the abnormal evaluation value as a threshold.
[0034] An abnormality detection method in one aspect of this disclosure acquires signals corresponding to the operation of a machine 100, including machine tools and processing machines capable of repeatedly executing a series of operations. Based on the signals acquired during an operation period set according to the length of the series of operations, an evaluation value for the entire series of operations is calculated. If the calculated evaluation value is greater than a threshold, the machine 100 is determined to be abnormal. With this configuration, the machine 100 can be constantly monitored without being affected by transient phenomena, and abnormalities in the machine 100 can be determined regardless of its operating status.
[0035] The abnormality detection device 1 can also be configured as follows:
[0036] The determination unit 30 may determine the range of evaluation values for which an abnormality judgment is made based on pre-calculated normal evaluation values and pre-calculated abnormal evaluation values. With such a configuration, for example, the frequency of judgment results output to the user from the output unit 50 can be reduced.
[0037] The calculation control unit 40 can be omitted.
[0038] Thresholds are not limited to being set based on normal and abnormal evaluation values. For example, thresholds may be set by an operator manually entering them.
[0039] Having described in detail various embodiments of this disclosure with reference to the drawings above, we will now conclude by describing various aspects of this disclosure. Reference numerals are also included in the following description as an example.
[0040] An anomaly detection device 1 according to a first aspect of this disclosure is An acquisition unit 10 acquires signals corresponding to the operation of a machine, including a machine tool and a processing machine, which can repeatedly perform a series of operations. A calculation unit 20 calculates an evaluation value for the entire series of operations based on the signals acquired during an operation period set according to the length of the series of operations, A determination unit 30 performs an abnormality determination to determine whether the machine is abnormal or not based on the calculated evaluation value. Equipped with, The determination unit 30 determines that the machine is malfunctioning if the calculated evaluation value deviates from the normal range.
[0041] An anomaly detection device 1 in a second aspect of this disclosure is The determination unit 30 determines that the machine is malfunctioning if the calculated evaluation value is greater than the threshold.
[0042] An anomaly detection device 1 in a third aspect of this disclosure is If the evaluation value when the series of operations is performed and no abnormality occurs in the machine is defined as the normal evaluation value, and the evaluation value when the series of operations is performed and an abnormality occurs in the machine is defined as the abnormal evaluation value, The determination unit 30 sets the threshold value based on the normal evaluation value calculated in advance, or based on the normal evaluation value and the abnormal evaluation value calculated in advance.
[0043] An anomaly detection device 1 according to a fourth aspect of this disclosure is The determination unit 30 sets the 3-sigma value of the normal evaluation value as the threshold value.
[0044] An anomaly detection device 1 according to a fifth aspect of this disclosure is The determination unit 30 sets the value between the normal evaluation value and the abnormal evaluation value as the threshold value.
[0045] An anomaly detection device 1 according to the sixth aspect of this disclosure is If the evaluation value when the series of operations is performed and no abnormality occurs in the machine is defined as the normal evaluation value, and the evaluation value when the series of operations is performed and an abnormality occurs in the machine is defined as the abnormal evaluation value, The determination unit 30 determines the range of evaluation values for which abnormality is determined, based on the previously calculated normal evaluation values and the previously calculated abnormal evaluation values.
[0046] The method for determining abnormalities in the seventh aspect of this disclosure is: A signal corresponding to the operation of a machine, including a machine tool and a processing machine capable of repeatedly performing a series of operations, is acquired. Based on the signals acquired during the operating period set according to the length of the series of operations, an evaluation value for the entire series of operations is calculated. If the calculated evaluation value deviates from the normal range, the machine is determined to be malfunctioning.
[0047] By appropriately combining any embodiment or modification from the various embodiments or modifications described above, the effects of each can be achieved. Furthermore, combinations of embodiments with each other, combinations of examples with each other, and combinations of embodiments with examples are possible, as well as combinations of features from different embodiments or examples.
[0048] While this disclosure is adequately described in relation to preferred embodiments with reference to the accompanying drawings, various variations and modifications will be obvious to those skilled in the art. Such variations and modifications should be understood to be included within the scope of this disclosure as defined by the attached claims. [Industrial applicability]
[0049] The abnormality detection device described herein is not limited to machine tools and processing machines, but can also be applied to extruders, conveyors, pumps, fans, machining centers, and the like. [Explanation of Symbols]
[0050] 1 Abnormality determination device 2 CPU 3 Storage device 4. Communication equipment 10 Acquisition Department 20 Calculation Section 30 Judgment section 40 Calculation Control Unit 50 Output section 100 machines
Claims
1. An acquisition unit that acquires signals corresponding to the operation of a machine, including a machine tool and a processing machine, which can repeatedly perform a series of operations, A calculation unit that calculates an evaluation value of the entire series of operations based on the signals acquired during an operation period set according to the length of the series of operations, A determination unit that determines whether the machine is abnormal or not based on the calculated evaluation value. Equipped with, The determination unit determines that the machine is abnormal if the calculated evaluation value is greater than the threshold. If the evaluation value when the series of operations is performed and no abnormality occurs in the machine is defined as the normal evaluation value, and the evaluation value when the series of operations is performed and an abnormality occurs in the machine is defined as the abnormal evaluation value, The determination unit is an abnormality determination device that determines the range of the evaluation value for which an abnormality determination is made, based on the normal evaluation value and the abnormal evaluation value that have been calculated in advance.
2. If the evaluation value when the series of operations is performed and no abnormality occurs in the machine is defined as the normal evaluation value, and the evaluation value when the series of operations is performed and an abnormality occurs in the machine is defined as the abnormal evaluation value, The abnormality determination device according to claim 1, wherein the determination unit sets the threshold based on the normal evaluation value calculated in advance, or based on the normal evaluation value calculated in advance and the abnormal evaluation value calculated in advance.
3. The abnormality determination device according to claim 2, wherein the determination unit sets the 3-sigma value of the normal evaluation value as the threshold value.
4. The abnormality determination device according to claim 2, wherein the determination unit sets a value between the normal evaluation value and the abnormal evaluation value as the threshold value.
5. A signal corresponding to the operation of a machine, including a machine tool and a processing machine capable of repeatedly performing a series of operations, is acquired. Based on the signals acquired during the operating period set according to the length of the series of operations, an evaluation value for the entire series of operations is calculated. If the calculated evaluation value is greater than the threshold, the machine is determined to be abnormal. If the evaluation value when the series of operations is performed and no abnormality occurs in the machine is defined as the normal evaluation value, and the evaluation value when the series of operations is performed and an abnormality occurs in the machine is defined as the abnormal evaluation value, An abnormality determination method for determining the range of evaluation values for which an abnormality determination is made, based on the normal evaluation value and the abnormal evaluation value calculated in advance.