Tool abnormality detection system

Abstract

A tool abnormality detection system that can make it difficult to erroneously determine tool abnormalities is provided. [Solution] The tool includes a detection unit 10 that detects sound waves generated when a workpiece 200 is machined using a tool 140, and a control unit 30 that determines wear on the tool 140 based on the sound waves detected by the detection unit 10. The tool includes a resonance unit 20 that generates resonance sound waves in resonance with wear sound waves generated when the tool 140 wears, and the detection unit 10 also detects the resonance sound waves from the resonance unit 20. The control unit 30 compares the intensity of a determination signal based on the sound waves detected by the detection unit 10 with a wear threshold, and determines that wear has occurred on the tool if the intensity of the determination signal is equal to or greater than the wear threshold.

Description

[Technical Field]

[0001] The present invention relates to a tool abnormality detection system. [Background technology]

[0002] Conventionally, tool abnormality detection systems for detecting abnormalities in tools used in machine tools have been proposed (see, for example, Patent Document 1). Specifically, this tool abnormality detection system includes a detection unit that detects sound waves generated when a workpiece is machined using a tool, and a control unit. The control unit extracts the intensity of sound waves in a predetermined frequency band to reduce the influence of noise sound waves caused by background noise, and then compares the intensity of the extracted frequency band with a threshold value to determine tool abnormality (i.e., wear). Examples of noise sound waves caused by background noise include the voices of workers monitoring the area around the machine tool, the operating sounds of other machine tools installed around the target machine tool, and chimes from the factory where the machine tool is installed. Examples of sound waves caused by background noise also include sound waves from equipment installed on the target machine tool, such as the operating sounds of a motor, an automatic tool changer (ATC), a coolant device, an oil mister collector, etc. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2007-222997 Summary of the Invention [Problem to be solved by the invention]

[0004] However, even in the tool abnormality detection system as described above, if noise sound waves due to background noise are included in the set frequency band, there is a possibility that an abnormality in the tool may be erroneously determined.

[0005] In view of the above, an object of the present invention is to provide a tool abnormality detection system that can make it difficult to erroneously determine an abnormality in a tool. [Means for solving the problem]

[0006] Claim 1 for achieving the above object provides a tool abnormality detection system, comprising: a detection unit (10) for detecting sound waves generated when a workpiece (200) is machined using a tool (140, 410); and a control unit (30) for determining tool wear based on the sound waves detected by the detection unit, and a resonance unit (20) for generating resonance sound waves by resonating with wear sound waves generated by tool wear, the detection unit also detects the resonance sound waves from the resonance unit, and the control unit includes a signal acquisition unit (301) for acquiring learning object data and monitoring object data based on the sound waves, a state observer generation unit (304) for generating a state observer using variables included in an input variable configuration, and a state observer generation unit (305) for generating a state observer using variables included in an input variable configuration. the normal model generation unit (306) that generates a threshold value by combining a first state observation value obtained by inputting the state observer and the learning target data and inputting the combined result into a competitive neural network; an abnormality degree calculation unit (308) that calculates the degree of abnormality by combining a second state observation value obtained by inputting the monitored data into the state observer and the monitored data and inputting the combined result into a competitive neural network; and a judgment unit (309) that obtains a judgment result by comparing the threshold value with the degree of abnormality, and the resonating unit is provided with a plurality of resonating units each having a different natural frequency, and each natural frequency is included in the range of abrasion sound waves that can be generated depending on the condition of the workpiece.

[0007] According to this, a resonating portion capable of resonating with abrasion sound waves is disposed, and abnormality determination of the tool is performed using learning target data and monitoring target data based on the sound waves, thereby making it possible to prevent erroneous determination.

[0008] The reference symbols in parentheses attached to each component indicate an example of the correspondence between the component and the specific components described in the embodiments described below. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a schematic diagram of a tool abnormality detection system according to a first embodiment. [Figure 2] FIG. 2 is a block diagram of a control unit. [Figure 3] FIG. 10 is a diagram illustrating the relationship between frequency and intensity. [Figure 4] 10 is a flowchart executed by a control unit. [Figure 5] FIG. 10 is a schematic diagram of a tool abnormality detection system according to a second embodiment. [Figure 6] FIG. 10 is a schematic diagram of a tool abnormality detection system according to a third embodiment. [Figure 7] FIG. 10 is a schematic diagram of a control unit in a fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the following embodiments, parts that are identical or equivalent to each other will be denoted by the same reference numerals.

[0011] (First embodiment) A tool abnormality detection system according to a first embodiment will be described with reference to Fig. 1. The tool abnormality detection system according to this embodiment is used in cutting work and detects abnormalities in the tool of a target machine tool. In this embodiment, an abnormality detection system that detects abnormalities in a drill used when drilling a workpiece as cutting work will be described as an example.

[0012] The target machine tool 100 is, for example, a machining center having a processing chamber. In this embodiment, the machine tool 100 includes a stage 110, a vise 120 placed on the stage 110 to fix a workpiece 200, a spindle holder 130, and a drill 140 attached to the spindle holder 130 to machine the workpiece 200. The machine tool 100 displaces the drill 140 to form a hole 201 in the workpiece 200. In this embodiment, the drill 140 corresponds to the tool. The workpiece 200 may be a material used in metal cutting, such as SUS or AL.

[0013] The tool abnormality detection system 1 includes a detection unit 10, a resonance unit 20, a control unit 30, and the like.

[0014] The detection unit 10 outputs a detection signal corresponding to the detected sound waves, and is composed of a highly waterproof and dustproof sound collecting microphone or the like, and is disposed near the workpiece 200 and the resonating unit 20 described below. As will be described later, the detection unit 10 of this embodiment detects machining sound waves generated when the workpiece 200 is cut, and resonance sound waves generated from the resonating unit 20. The machining sound waves of this embodiment include direct sound waves that are generated directly when the workpiece 200 is cut with the drill 140, as well as abrasion sound waves that are generated when the drill 140 wears and the workpiece 200 vibrates, as will be described later.

[0015] The resonating unit 20 is configured to be able to resonate with abrasion sound waves generated when the drill 140 wears and the workpiece 200 vibrates. Specifically, the resonating unit 20 is configured to have the same natural frequency as the workpiece 200 so as to be able to resonate with abrasion sound waves. For example, the resonating unit 20 is made of the same material and has the same shape as the workpiece 200. In this case, for example, the workpiece 200 before drilling is performed is used as the resonating unit 20. Alternatively, for example, the workpiece 200 after drilling is performed is used as the resonating unit 20. However, as long as the resonating unit 20 is configured to resonate with abrasion sound waves, it does not have to be made of the same material as the workpiece 200 or have the same shape as the workpiece 200. The natural frequencies of the workpiece 200 and the resonating portion 20 are derived, for example, by structural simulation or the impulse hammer method, but if the shape is a simple one such as a rectangular parallelepiped, they may also be derived based on mechanical engineering formulas, etc.

[0016] The resonating unit 20 is disposed in a position where it can resonate with the wear sound waves. For example, the resonating unit 20 is disposed near the workpiece 200, but may also be disposed inside the detection unit 10, the stage 110, the vise 120, the spindle holder 130, or the drill 140. The resonating unit 20 resonates with the wear sound waves and generates resonant sound waves having frequency components equivalent to those of the wear sound waves.

[0017] The control unit 30 is configured by a microcomputer or the like equipped with a CPU and a storage unit configured with a non-transitory physical storage medium such as a ROM, RAM, flash memory, or HDD. As shown in Fig. 2, the control unit 30 of this embodiment is equipped with an amplifier unit 31, a signal processing unit 32, a determination unit 33, and the like. Note that CPU stands for Central Processing Unit, ROM stands for Read Only Memory, RAM stands for Random Access Memory, and HDD stands for Hard Disk Drive. Storage media such as ROM are non-transitory physical storage media.

[0018] The amplifier 31 is connected to the detector 10 and amplifies the detection signal detected by the detector 10. The amplifier 31 may be connected to the detector 10 via wireless communication or via wired communication. The detection signal transmitted from the detector 10 is a signal based on processed sound waves and resonant sound waves, but also includes the influence of noise sound waves due to background noise.

[0019] The signal processing unit 32 derives a determination signal indicating the intensity in a predetermined frequency band. In this embodiment, the signal processing unit 32 is configured to include an FFT (abbreviation of Fast Fourier Transform) circuit unit that performs a Discrete Fourier Transform at high speed, a bandpass filter, etc. The predetermined frequency band is a predetermined range that includes the natural frequency of the workpiece 200.

[0020] Upon receiving the determination signal derived by the signal processing unit 32, the determination unit 33 compares the intensity of the determination signal with a wear threshold value to perform a wear determination as an abnormality determination for determining wear of the drill 140. Specifically, as shown in FIG. 3 , when the drill 140 is worn, the determination signal becomes equal to or greater than the wear threshold value at the natural frequency of the workpiece 200. In this case, in this embodiment, as will be described in detail later, when the drill 140 is worn, the workpiece 200 vibrates at its natural frequency, and the wear sound waves generated at this time resonate with the resonator 20, generating a resonant sound wave. Therefore, the intensity of the determination signal obtained when the drill 140 is worn can be increased compared to when the resonator 20 is not provided. Therefore, the influence of noise sound waves caused by background noise can be reduced, and erroneous determination of wear of the drill 140 can be suppressed.

[0021] Furthermore, in this embodiment, when the determination unit 33 determines that the drill 140 is worn in the wear determination, it then performs a damage determination to determine whether the intensity of the determination signal has become equal to or less than a damage threshold, which is smaller than the wear threshold. Specifically, when the drill 140 breaks, the abrasion between the drill 140 and the workpiece 200 decreases, the wear sound waves decrease, and the intensity of the determination signal decreases. Therefore, the determination unit 33 determines that the drill 140 has broken when the intensity of the determination signal becomes equal to or less than the damage threshold.

[0022] The determination unit 33 of this embodiment is connected to the notification unit 40, which has a display unit, an audio unit, etc., and when it determines that an abnormality such as wear or breakage has occurred in the drill 140, it transmits an abnormality signal indicating that an abnormality has occurred in the drill 140 to the notification unit 40. In this embodiment, when the determination unit 33 determines that the drill 140 is worn, it transmits a wear abnormality signal to the notification unit 40. When the determination unit 33 determines that the drill 140 is damaged, it transmits a damage abnormality signal to the notification unit 40. Then, the notification unit 40 issues a notification according to the content of the abnormality signal to notify the operator that an abnormality has occurred in the drill 140.

[0023] Determination unit 33 may operate independently of a numerical control device (i.e., an NC (abbreviation for Numerical Control) device) that is generally provided in machine tool 100, but may also send an abnormality signal to the numerical control device. When an abnormality signal is also sent to the numerical control device, the numerical control device may, for example, reset the set movement direction, movement amount, etc. based on wear of drill 140.

[0024] The above is the configuration of the tool abnormality detection system 1 in this embodiment. Next, we will explain the abnormality detection method using the tool abnormality detection system 1. In this embodiment, we will explain the abnormality detection method when drilling a hole in a workpiece 200 using the drill 140 as described above.

[0025] First, when drilling the workpiece 200, the workpiece 200 is fixed in the vise 120, and the spindle holder 130 is rotated and displaced downward to press the drill 140 against the workpiece 200, thereby forming a hole 201 in the workpiece 200. As the cutting edge of the drill 140 wears down due to continued drilling, the contact friction between the cutting edge of the drill 140 and the workpiece 200 increases at the bottom of the hole 201, increasing the cutting resistance.

[0026] When contact friction increases, it becomes difficult for the spindle holder 130, which displaces the drill 140 downward while rotating, to operate properly, causing the spindle holder 130 to vibrate slightly, causing the drill 140 to scrape against the side wall of the hole 201. When the drill 140 scrapes against the side wall of the hole 201, the workpiece 200 begins to vibrate freely due to its natural vibration. In other words, when the drill 140 wears, wear sound waves are generated from the workpiece 200 due to the wear of the drill 140.

[0027] In this embodiment, as described above, the resonating part 20 having the same natural frequency as the workpiece 200 is disposed in the vicinity of the workpiece 200. Therefore, the resonating part 20 generates a resonant sound wave in response to a wear sound wave.

[0028] The detector 10 detects sound waves and outputs a detection signal. Specifically, when the drill 140 is not worn, the detector 10 detects direct sound waves and noise sound waves, and when the drill 140 is worn, the detector 10 detects wear sound waves and resonance sound waves in addition to the direct sound waves and noise sound waves, and transmits a detection signal.

[0029] The determination unit 33 then compares the intensity of a determination signal based on the detection signal with a wear threshold to determine whether the drill 140 is worn. In this case, when the drill 140 is worn, the detection unit 10 detects resonance sound waves in addition to wear sound waves. Therefore, when the drill 140 is worn, the intensity of the determination signal becomes higher compared to when the resonance unit 20 is not provided. This reduces the influence of noise sound waves. Furthermore, when the determination unit 33 determines that the drill 140 is worn, it performs damage determination by comparing the intensity of the determination signal with a damage threshold.

[0030] The above is the abnormality detection method in this embodiment. Next, the operation performed by the control unit 30 will be described with reference to Fig. 4. The control unit 30 performs the following operation, for example, when machining of the workpiece 200 is started.

[0031] First, in step S101, the control unit 30 receives a detection signal from the detection unit 10 and generates a determination signal. Then, in step S102, the control unit 30 determines whether or not a wear abnormality signal has been transmitted. That is, the control unit 30 determines whether or not it has already determined that the drill 140 is in a worn state. Then, if the control unit 30 determines that a wear abnormality signal has been transmitted (i.e., step S102: YES), it performs the processes from step S105 onwards, which will be described later.

[0032] If the control unit 30 determines in step S102 that the wear abnormality signal has not been transmitted (that is, step S102: NO), it determines in step S103 whether the intensity of the determination signal is equal to or greater than the wear threshold value.

[0033] If the control unit 30 determines that the intensity of the determination signal is less than the wear threshold (i.e., step S103: NO), the drill 140 is not worn, and therefore the process ends. On the other hand, if the control unit 30 determines that the intensity of the determination signal is equal to or greater than the wear threshold (i.e., step 103: YES), it transmits an abnormal wear signal to the notification unit 40 in step S104. As a result, the notification unit 40 performs a process of notifying the operator that the drill 140 is worn.

[0034] After transmitting the wear abnormality signal in step S104, or when it is determined in step S102 that the wear abnormality signal has been transmitted (i.e., step S102: YES), the control unit 30 determines in step S105 whether the intensity of the determination signal is equal to or less than the damage threshold. When the control unit 30 determines that the intensity of the determination signal is equal to or less than the damage threshold (i.e., step S105: YES), the drill 140 is damaged, and therefore transmits a damage abnormality signal to the notification unit 40. As a result, the notification unit 40 performs processing to notify the operator that damage has occurred in the drill 140. On the other hand, when the control unit 30 determines that the intensity of the determination signal is greater than the damage threshold (i.e., step S105: NO), the processing ends.

[0035] According to the present embodiment described above, the resonator 20 capable of resonating with abrasion sound waves is provided, and the determination signal is based on the abrasion sound waves and the resonance sound waves. This allows the strength of the determination signal to be increased and the influence of noise sound waves to be reduced, thereby preventing erroneous determinations. Furthermore, since the tool abnormality detection system 1 of this embodiment can prevent erroneous determinations by increasing the strength of the determination signal using resonance sound waves, it can be applied even when the diameter of the hole 201 is small or when the abrasion sound waves are small due to the material of the workpiece 200.

[0036] (Modification of the first embodiment) A modification of the first embodiment will be described. In the first embodiment, the control unit 30 performs the processes of step S105 and thereafter after transmitting the wear abnormality signal in step S104. However, the control unit 30 may perform the processes of step S105 and thereafter, for example, when the wear abnormality signal is transmitted multiple times.

[0037] (Second embodiment) A second embodiment will be described. This embodiment differs from the first embodiment in that it detects abnormalities in a tool during turning, which is a cutting process. As the rest of the configuration is the same as the first embodiment, a description thereof will be omitted here.

[0038] The tool abnormality detection system 1 of this embodiment detects abnormalities in a tool that performs turning as a cutting process. To this end, the machine tool 100 of this embodiment is equipped with a cutting tool 410 as a tool for turning a workpiece 200, a vise 420 for holding the cutting tool 410, and the like, as shown in Fig. 5. The cutting tool 410 is rod-shaped with one end formed into a blade, and the other end opposite the one end is fixed by the vise 420.

[0039] As will be described in detail later, in this embodiment, when the cutting edge of the cutting tool 410 wears, the cutting tool 410 vibrates at its natural frequency, generating wear acoustic waves. For this reason, the detection unit 10 is disposed near the cutting tool 410. Note that Fig. 5 shows a state in which the cutting tool 410 is worn and vibrating.

[0040] The resonance unit 20 is disposed at a position where it can resonate with the wear sound waves, similar to the first embodiment. For example, the resonance unit 20 is disposed near the workpiece 200, but may also be disposed inside the cutting tool 410 or the vise 420.

[0041] Furthermore, since the tool bit 410 vibrates at its natural frequency when it wears down as described above, the resonating part 20 of this embodiment is made of a material having the same natural frequency as the tool bit 410. For example, the resonating part 20 uses an unused tool bit 410 that is the same as the tool bit 410 fixed to the vise 420.

[0042] As in the first embodiment, the control unit 30 compares the determination signal with a wear threshold to determine wear. The control unit 30 also compares the determination signal with a damage threshold to determine damage. However, in this embodiment, the control unit 30 (i.e., the signal processing unit 32) derives a determination signal in a predetermined frequency band that includes the natural frequency of the tool bit 410.

[0043] The above is the configuration of the tool abnormality detection system 1 in this embodiment. Next, we will explain the abnormality detection method using the above-mentioned tool abnormality detection system 1. In this embodiment, we will explain the abnormality detection method when turning the workpiece 200 using the cutting tool 410 as described above.

[0044] First, during turning, the rotating workpiece 200 is brought into contact with the tool bit 410 to form the desired shape. In this case, as the turning continues, the cutting edge of the tool bit 410 wears, the contact friction between the cutting edge and the workpiece 200 increases, resulting in higher turning resistance. When the force relationship between the tool bit 410 and the workpiece 200 becomes unbalanced and the relative velocity becomes zero, the tool bit 410 experiences a stick-slip phenomenon. The spring force of the tool bit 410 returning to its original position causes the tool bit 410 to begin free vibration due to its natural vibration. In other words, as the tool bit 410 wears, it generates wear acoustic waves specific to wear. In this embodiment, as described above, the resonating unit 20, which has the same natural frequency as the tool bit 410, is disposed near the workpiece 200. Therefore, the resonating unit 20 resonates with the wear acoustic waves to generate resonant sound waves.

[0045] The stick-slip phenomenon refers to an intermittent movement in which, when two objects in contact slide, the sliding motion is not continuous and smooth, but alternates between slipping and sticking.

[0046] When the bit 410 is not worn, the detection unit 10 detects direct sound waves and noise sound waves, and when the bit 410 is worn, the detection unit 10 detects wear sound waves and resonance sound waves in addition to the direct sound waves and noise sound waves and transmits a detection signal.

[0047] Then, the control unit 30 compares the intensity of the determination signal based on the detection signal with the wear threshold and the damage threshold to determine whether the tool bit 410 is abnormal, as in the first embodiment.

[0048] According to the present embodiment described above, the resonance portion 20 that resonates with abrasion sound waves is disposed, and therefore, the same effects as those of the first embodiment can be obtained.

[0049] (Third embodiment) A third embodiment will be described. This embodiment differs from the first embodiment in that a plurality of resonating parts 20 are arranged. As the rest of the configuration is the same as the first embodiment, a description thereof will be omitted here.

[0050] First, when drilling the workpiece 200 as in the first embodiment, the mass of the workpiece 200 may change due to the drilling, which may change the natural frequency of the workpiece 200. In other words, the wear sound waves generated from the workpiece 200 when the drill 140 is worn may change depending on the situation. In particular, when forming multiple holes 201 or a large hole 201, the change in the mass of the workpiece 200 becomes large, and the natural frequency of the workpiece 200 is likely to change.

[0051] For this reason, in this embodiment, as shown in Fig. 6, a plurality of resonating parts 20 are provided. Specifically, each resonating part 20 has a different natural frequency. However, each resonating part 20 has a natural frequency that corresponds to the natural frequency of the workpiece 200, which changes as the workpiece 200 is drilled. In other words, the natural frequency of each resonating part 20 is within the range of wear sound waves that can be generated depending on the condition of the workpiece 200.

[0052] The above is the configuration of the tool abnormality detection system 1 in this embodiment. Next, an abnormality detection method using the tool abnormality detection system will be described.

[0053] As in the first embodiment, when the drill 140 wears, wear sound waves are generated in the workpiece 200 due to the wear of the drill 140. In this case, the frequency of the wear sound waves changes depending on the mass of the workpiece 200. However, in this embodiment, multiple resonance parts 20 are provided, and each resonance part 20 has a different natural frequency. This makes it possible to prevent the problem of not generating resonance sound waves corresponding to wear sound waves.

[0054] According to the present embodiment described above, the resonance portion 20 that resonates with abrasion sound waves is disposed, and therefore, the same effects as those of the first embodiment can be obtained.

[0055] (1) In this embodiment, a plurality of resonating parts 20 are arranged, and the natural frequencies of the respective resonating parts 20 are set to different values. This prevents the occurrence of a problem in which a resonant sound wave corresponding to a wear sound wave is not generated.

[0056] (Fourth embodiment) A fourth embodiment will be described. In this embodiment, the configuration of the control unit 30 is changed compared to the first embodiment. As the rest is the same as the first embodiment, a description thereof will be omitted here. Note that in this embodiment, the wear determination in the first embodiment will be described, but damage determination may also be performed.

[0057] The control unit 30 of this embodiment determines wear of the drill 140 using the self-learning of an abnormality detection system already applied for by the present applicant in Japanese Patent Application Laid-Open No. 2020-129233.

[0058] As shown in FIG. 7 , the control unit 30 of this embodiment includes a signal acquisition unit 301, a learning target data storage unit 302, a monitoring target data storage unit 303, a state observer generation unit 304, a state observer information storage unit 305, a normal model generation unit 306, a normal model parameter storage unit 307, an abnormality degree calculation unit 308, a judgment unit 309, a factor analysis unit 310, and the like.

[0059] The signal acquisition unit 301 acquires learning target data and monitoring target data. In this embodiment, a detection signal based on a sound wave detected by the detection unit 10 is acquired as the monitoring target data and learning target data.

[0060] The learning target data may be acquired by downloading it from a database that stores reference values, rather than detection signals. Furthermore, the monitoring target data in this embodiment may be data to be monitored, and may or may not be used for other purposes. For example, the monitoring target data may also be data to be used as a learning target. Furthermore, the learning target data in this embodiment may be data to be used for learning, and may or may not be used for other purposes.

[0061] The learning object data storage unit 302 stores the learning object data acquired by the signal acquisition unit 301. The stored learning object data is then input to the normal model generation unit 306, which will be described later.

[0062] The monitoring target data storage unit 303 stores the monitoring target data acquired by the signal acquisition unit 301. The stored monitoring target data is then input to an abnormality probability calculation unit 308, which will be described later.

[0063] The learning target data storage unit 302 and the monitoring target data storage unit 303 are expected to be configured with a HDD, a flash memory, or the like, but may also be configured with a RAM, etc. Furthermore, the learning target data storage unit 302 and the monitoring target data storage unit 303 may be configured with either a volatile memory or a non-volatile memory.

[0064] The state observer generation unit 304 generates a state observer using variables included in the input variable configuration. The state observer is a value including a predetermined coefficient generated by linearly combining variables included in the input variable configuration with a preset mathematical expression. The state observer preferably reflects the state of the drill 140 whose abnormality is detected by the tool anomaly detection system 1, and preferably reflects wear of the drill 140. In this embodiment, the variables are variables corresponding to the learning target data and monitoring target data acquired by the signal acquisition unit 301, and the variable configuration is one variable or a combination of multiple variables.

[0065] The state observer information storage unit 305 stores the coefficients of the state observer calculated by the state observer generation unit 304. It also stores the variable configuration used to generate the state observer.

[0066] The state observer generation unit 304 may be provided when generating a state observer. That is, the state observer generation unit 304 may be configured to be separated from the control unit 30 after generating the state observer and storing the variable configuration of the state observer and the coefficients of the state observer in the state observer information storage unit 305.

[0067] The normal model generation unit 306 derives a first state observation value obtained by inputting the learning target data into the state observer. In this embodiment, the variable configuration and coefficients of the state observer are read from the state observer information storage unit 305, and the first state observation value is obtained by applying the learning target data read from the learning target data storage unit 302 to these variables. The normal model generation unit 306 then combines the derived first state observation value and the learning target data and inputs them into a competitive neural network (i.e., NN) to generate a normal model. For example, in this embodiment, the material of the drill 140, the diameter of the hole, etc. are input to the competitive neural network (NN) as learning target data.

[0068] When there are multiple combinations of input data attributes, such as measurement season and time, it is desirable to sample the initial values ​​given to the competitive neural network evenly or randomly. This allows for faster convergence of the neuron weight vectors during learning on the map in the competitive neural network.

[0069] Here, a competitive neural network is a network consisting of only an input layer and an output layer, and is composed of multiple input layer neurons and multiple output layer neurons that are fully connected to the input layer neurons.

[0070] Then, the normal model generation unit 306 calculates the degree of abnormality, which is the difference between the learning data and the first state observation value input to the competitive neural network and the neuron weight data of the winning unit, and calculates a wear-out threshold using the set of differences. For example, a constant multiple of the 99.9% quantile of the set of differences (i.e., absolute values) is used as the wear-out threshold.

[0071] The normal model parameter storage unit 307 stores the wear threshold value calculated by the normal model generation unit 306 .

[0072] The anomaly degree calculation unit 308 derives a second state observation value obtained by inputting the monitored data into the state observer. In this embodiment, the variable configuration and coefficients of the state observer are read from the state observer information storage unit 305, and the monitored data read from the monitored data storage unit 303 are applied to these variables to obtain a second state observation value. The anomaly degree calculation unit 308 then combines the derived second state observation value and the monitored data and inputs them into a competitive neural network (i.e., NN), thereby calculating the degree of anomaly using weight data of output layer neurons. For example, in this embodiment, the material of the drill 140, the diameter of the hole, etc. are input to the competitive neural network (NN) as monitored data.

[0073] The determination unit 309 performs wear determination by comparing the wear threshold value read from the normal model parameter storage unit 307 with the abnormality degree output from the abnormality degree calculation unit 308. Specifically, if the abnormality degree is equal to or greater than the wear threshold value, it determines that the drill 140 is worn, and if the abnormality degree is less than the wear threshold value, it determines that there is no wear.

[0074] When the determination unit 309 determines that wear exists, the factor analysis unit 310 identifies the cause of the wear using the second state observation value and the monitored data that caused the determination that wear exists. In this case, the factor analysis unit 310 may further use the second state observation value and the monitored data that caused the determination that wear exists as a reference to identify the cause of the abnormality using the second state observation value and the monitored data that are temporally before, after, or before and after the second state observation value and the monitored data that caused the determination that wear exists.

[0075] According to the present embodiment described above, the resonance portion 20 that resonates with abrasion sound waves is disposed, and therefore, the same effects as those of the first embodiment can be obtained.

[0076] (1) As in the present embodiment, the control unit 30 may perform self-learning to determine the wear of the drill 140. This allows for wear determination according to the situation, which is expected to further reduce erroneous determinations.

[0077] (Other embodiments) Although the present disclosure has been described with reference to the embodiments, it is understood that the present disclosure is not limited to the embodiments or structures. The present disclosure also encompasses various modifications and modifications within the scope of equivalents. In addition, various combinations and forms, as well as other combinations and forms including only one element, more than one element, or less than one element, are also within the scope and spirit of the present disclosure.

[0078] For example, in the first to third embodiments, wear determination and damage determination are performed, but it is also possible to perform only wear determination.

[0079] In the first, third, and fourth embodiments, the drill 140 is used as an example of the tool, and in the second embodiment, the cutting tool 410 is used as an example of the tool. However, the tool can be changed as appropriate, and may be an end mill, a grinding wheel, or the like.

[0080] The controller and methods described herein may be implemented by a special-purpose computer configured with a processor and memory programmed to perform one or more functions embodied in a computer program. Alternatively, the controller and methods described herein may be implemented by a special-purpose computer configured with a processor configured with one or more dedicated hardware logic circuits. Alternatively, the controller and methods described herein may be implemented by one or more special-purpose computers configured with a processor and memory programmed to perform one or more functions in combination with a processor configured with one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by a computer on a computer-readable non-transitory storage medium. [Explanation of symbols]

[0081] 10. Detection unit 20 Resonance part 30 Control Unit 140 Drill (Tool) 200 Work material

Claims

[Claim 1] A tool abnormality detection system, a detection unit (10) that detects sound waves generated when a workpiece (200) is machined using a tool (140, 410); a control unit (30) that determines wear of the tool based on the sound waves detected by the detection unit, a resonance part (20) that resonates with abrasion sound waves generated by the tool being worn to generate resonance sound waves, The detection unit also detects resonance sound waves from the resonance unit, The control unit a signal acquisition unit (301) that acquires learning target data and monitoring target data based on the sound waves; a state observer generation unit (304) that generates a state observer using variables included in the input variable configuration; a normal model generation unit (306) that generates a threshold value by combining a first state observation value obtained by inputting the learning object data into the state observer and the learning object data and inputting the combined result into a competitive neural network; an anomaly degree calculation unit (308) that calculates an anomaly degree by combining a second state observation value obtained by inputting the monitored data into the state observer and the monitored data and inputting the combined result into the competitive neural network; a determination unit (309) that obtains a determination result by comparing the threshold value with the abnormality degree, A tool abnormality detection system in which multiple resonating parts are provided, each with a different natural frequency, and each of the natural frequencies is included in the range of wear sound waves that can be generated depending on the condition of the workpiece.

Images