Defective determination device and computer-readable storage medium

Abstract

The deterioration determination device acquires a value of a load applied to a shaft of a machine tool as a sample, determines whether the sample is detected in a cutting time period or a non-cutting time period, calculates a probability density function of the sample in the non-cutting time period, calculates a feature quantity of the sample in the cutting time period, and determines deterioration of a tool based on a probability of the feature quantity occurring in the probability density function in the non-cutting time period.

Description

Technical Field

[0001] This disclosure relates to a damage assessment device and a computer-readable storage medium. Background Technology

[0002] Conventional devices for detecting or predicting tool breakage in machine tools involve multiplying a characteristic value of the load during the cutting period by a coefficient, calculating a threshold value for the change, and then determining the breakage. For example, Patent Document 1.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2004-130407 Summary of the Invention

[0006] The problem that the invention aims to solve

[0007] When a load is applied to detect breakage, trial processing is required; therefore, breakage cannot be determined from the initial processing.

[0008] In the field of numerical control devices, there is a desire to make the determination of damage more efficient.

[0009] Methods for solving problems

[0010] The breakage determination device includes: a data acquisition unit that acquires the load value applied to the axis of the machine tool as a sample; a determination unit that determines whether the sample acquired by the data acquisition unit was detected during the cutting time period or during the non-cutting time period; a probability calculation unit that calculates the probability density function of the sample during the non-cutting time period; a feature quantity calculation unit that calculates the feature quantity of the sample during the cutting time period; and a breakage detection unit that determines the tool breakage based on the probability of generating the feature quantity in the probability density function during the non-cutting time period. Attached Figure Description

[0011] Figure 1 This is a block diagram of the damage assessment device.

[0012] Figure 2 It is a graph representing the changes in commands during cutting.

[0013] Figure 3 It is a graph showing the non-cutting time period and the cutting time period during normal operation.

[0014] Figure 4 It is a graph showing the relationship between the probability density function and the characteristic quantity during the non-cutting time period.

[0015] Figure 5It is a graph showing the non-cutting time period and the cutting time period when the tool breaks down.

[0016] Figure 6 It is a graph showing the relationship between the probability density function and the characteristic quantity during the non-cutting time period.

[0017] Figure 7 This is a flowchart explaining the operation of the damage assessment device.

[0018] Figure 8 This is a curve showing the load under normal conditions.

[0019] Figure 9 It is a graph showing the load when the tool breaks down before the cutting time period.

[0020] Figure 10 This is a graph of the probability density function of samples in the non-cutting region.

[0021] Figure 11 It is a graph representing the relationship between the probability density function and the characteristic quantity.

[0022] Figure 12 This is a graph showing the load during tool breakage in the cutting period.

[0023] Figure 13 It is a graph representing the relationship between the probability density function and the characteristic quantity.

[0024] Figure 14 This is a block diagram of a modified example of a damage determination device.

[0025] Figure 15 This is a hardware structure diagram of the damage assessment device. Detailed Implementation

[0026] The damage determination device 100 will now be described. The damage determination device 100 is implemented by an information processing device such as a numerical control device and a PC (personal computer).

[0027] Figure 1 This is a block diagram of a damage determination device 100. The damage determination device 100 includes: a data acquisition unit 10, a determination unit 11, a probability calculation unit 12, a feature quantity calculation unit 13, and a damage detection unit 14.

[0028] The data acquisition unit 10 acquires at least one of the loads of the spindle or the feed axis. Hereinafter, the load value will be referred to as a sample.

[0029] The determination unit 11 determines whether the sample acquired by the data acquisition unit 10 was detected during the cutting time period or during the non-cutting time period. The cutting time period refers to the time during which the tool cuts the workpiece. The non-cutting time period is the time during which the tool does not cut the workpiece.

[0030] One method for determination is by detecting signals. For example, a signal that outputs a "cutting in progress" command from the numerical control device during the cutting time period. Figure 2 Examples of signals that indicate the "cutting in progress" command. Figure 2 The upper curve represents the load value. Figure 2 The lower curve represents the signal of the "cutting in progress" command. When the numerical control device analyzes the machining program and executes cutting commands such as G01, the signal for the "cutting in progress" command is "on". This signal can be used to determine whether it is a cutting or non-cutting period.

[0031] The probability calculation unit 12 calculates the probability density function of the samples obtained during the non-cutting time period. Figure 3 It is a graph representing the changes in a sample under normal conditions (when the tool is not damaged).

[0032] The feature calculation unit 13 calculates the feature quantities of the samples obtained during the cutting time period. Feature quantities are representative values ​​of the samples obtained during the cutting time period. Feature quantities include the mean, median, maximum, minimum, and most frequent value. Alternatively, feature quantities can also be moving averages, moving medians, moving maximums, moving minimums, and moving most frequent values. The type of feature quantity is not limited.

[0033] The damage detection unit 14 compares the probability density function of the samples during the non-cutting time period with the feature quantity of the samples during the cutting time period.

[0034] The load value during the cutting time period is unlikely to occur during the non-cutting time period. Therefore, if the probability of the characteristic quantity of the sample during the cutting time period is small enough, it is determined that the tool is not damaged.

[0035] For example, Figure 4 Point A represents the value of a characteristic quantity during the normal cutting time period. When the tool is not damaged, the probability of generating a value of the characteristic quantity during the non-cutting time period is infinitely close to zero. The damage detection unit 14 substitutes the value of the characteristic quantity into the probability density function, and determines that the tool is not damaged if the probability of generating the characteristic quantity is sufficiently small.

[0036] Figure 5 This is an example of tool breakage. In this case, the sample value for the cutting time period becomes lower. Figure 6 Point A represents the value of a characteristic quantity during the cutting period after tool breakage. In the case of tool breakage, the probability of generating a value of the characteristic quantity during the non-cutting period sometimes increases. The breakage detection unit 14 substitutes the value of the characteristic quantity into the probability density function, and determines that the tool is broken if the probability of generating a value of the characteristic quantity is sufficiently high.

[0037] The waveform of the load varies depending on the cause of the breakage, the raw material, the type of tool, and the timing of the breakage (during cutting or non-cutting). Therefore, multiple characteristic quantities are prepared to correspond to the changes in various waveforms.

[0038] Figure 7 This is a flowchart explaining the operation of the damage determination device 100.

[0039] Start the numerical control device, complete the preparation, and begin machining (step S1). The numerical control device moves the axis through rapid feed and rotates the spindle at the cutting start position.

[0040] The data acquisition unit 10 acquires the spindle load value as a sample. The determination unit 11 determines whether the sample acquired by the data acquisition unit 10 was detected during the cutting time period or during the non-cutting time period. The data acquisition unit 10 acquires the sample during the non-cutting time period (step S2).

[0041] The probability calculation unit 12 calculates the probability density function of the samples during the non-cutting time period (step S3). The data acquisition unit 10 acquires the samples during the cutting time period (step S4). The feature calculation unit 13 calculates the feature quantities of the samples during the cutting time period (step S5).

[0042] The breakage detection unit 14 substitutes the feature quantity into the probability density function and calculates the probability of generating the feature quantity value during the non-cutting time period (step S6). The breakage detection unit 14 compares the probability with a threshold (which can also be zero). If the probability is low enough, it is determined to be normal. Otherwise, if the probability is high enough, it is determined to be tool breakage (step S7).

[0043] As explained above, the damage determination device 100 of this embodiment calculates the probability density distribution of the sample during the cutting time period and the characteristic quantity of the non-cutting time period, calculates the probability of taking the value of the characteristic quantity during the non-cutting time period, and determines that there is no damage when the probability of the characteristic quantity of the cutting time period is generated during the non-cutting time period is sufficiently low.

[0044] The breakage determination device 100 of this embodiment determines breakage without trial processing, thus eliminating the need for trial processing and making the breakage determination process more efficient.

[0045] (Concrete example 1)

[0046] In Example 1, the method for determining tool breakage before the cutting time period is explained. Figure 8 This is the waveform of the load under normal conditions. Figure 9 It is the waveform of the load when the tool breaks down before the cutting time period.

[0047] like Figure 8As shown, the load under normal conditions increases at the start of cutting, remains roughly constant during cutting, and returns to the value during the non-cutting period when cutting ends. The non-cutting period maintains a roughly constant value.

[0048] If the tool breaks down before the cutting time period, such as Figure 9 As shown, the load during the non-cutting and cutting periods remained unchanged, maintaining the same value as during the non-cutting period.

[0049] In this example, the feature value is the maximum value of the sample during the cutting time period. If the probability of taking the feature value of the cutting time period during the non-cutting time period is less than about 0.03%, the breakage detection unit 14 determines that there is no breakage.

[0050] Figure 10 This is the probability density function of the samples in the non-cutting interval. In this example, the probability density function is a normal distribution with mean μ (=0.043) and variance σ (=0.043). The breakage detection unit 14 calculates the "3σ interval (99.7 percent)" for the non-cutting time period. The "3σ interval" for the non-cutting time period refers to the interval between the mean μ and ±3σ. The value of the "3σ interval" is generated with a probability of 99.7 percent. Figure 10 The values ​​are [-0.086, 0.172].

[0051] The damage detection unit 14 determines damage based on whether the characteristic quantity is included in the "3σ interval". For example... Figure 8 As shown, when the tool is not broken, the feature value (the maximum value of the sample in the cutting interval) is "2.714". Therefore, it is not included in the "3σ interval", and the probability of the feature value being generated during the non-cutting time period is sufficiently low. That is, the probability of the feature value "2.714" being generated during the non-cutting time period is less than 0.03%. Therefore, it can be determined that the tool is not broken.

[0052] like Figure 9 As shown, when the breakage occurs before the cutting time period, the load during the cutting time period becomes approximately the same as that during the non-cutting time period. If the characteristic value of the cutting time period (the maximum value of the sample during the cutting time period) is "0.047", then as... Figure 11 As shown, the value of the feature quantity "0.047" is contained within the "3σ interval". Therefore, the probability of generating the feature quantity (the maximum value of the sample during the cutting time period) during the non-cutting time period is sufficiently high. That is, the probability of generating the feature quantity value "0.047" during the non-cutting time period is higher than 0.03%. Therefore, it can be determined as tool breakage.

[0053] (Concrete example 2)

[0054] In Example 2, the method for determining tool breakage during the cutting time period is explained. Figure 12It is the waveform of the load when the tool breaks during the cutting period.

[0055] In the event of tool breakage during the cutting period, such as Figure 12 As shown, the load increases at the start of cutting, then gradually increases, rises sharply before failure, and returns to the value during the non-cutting period after failure.

[0056] In addition, the waveform of the load under normal conditions is similar to Figure 8 The same. Therefore, the probability density distribution of the non-cutting time period is also the same. Figure 10 equal.

[0057] In this example, the feature value is the minimum value of the sample during the cutting time period. If the probability of taking the feature value of the cutting time period during the non-cutting time period is less than 0.03%, the breakage detection unit 14 determines that there is no breakage.

[0058] When the tool breaks down during the cutting time interval, the feature quantity (the minimum value of the sample in the cutting interval) becomes "0.042". The value of the feature quantity "0.042" is as follows: Figure 13 The value is contained within the "3σ interval," therefore, the probability of generating the feature value (the minimum value of the sample during the cutting time period) during the non-cutting time period is sufficiently high. That is, the probability of generating the feature value "0.042" is higher than 0.03%. Therefore, it can be determined as tool breakage.

[0059] As explained above, in Specific Example 1, the maximum value of the sample during the cutting time period is used as the feature quantity, and in Specific Example 2, the minimum value of the sample during the cutting time period is used as the feature quantity to detect breakage under different conditions. In the breakage determination device 100 of this embodiment, breakage under various conditions can be detected by preparing multiple feature quantities.

[0060] (Modified example)

[0061] The modified example of the damage assessment device 100 combines two assessment methods. Regarding... Figure 14 The damage determination device 100 adds a change quantity unit 15, a second probability calculation unit 16, and a second damage detection unit 17 to the device. Figure 1 Damage determination device 100.

[0062] The change quantity section 15 calculates the change in the sample obtained during the cutting time period. The change in the sample refers to the increase or decrease in the sample value. The change quantity section 15 calculates the sample difference obtained by subtracting the values ​​of samples before or after a certain point in time from the value of the sample at that point in time.

[0063] The second probability calculation unit 16 calculates the probability density function of the change. The second probability calculation unit 16 performs preprocessing. In the preprocessing, a portion of the change is excluded from the sample. The second probability calculation unit 16 excludes values ​​with large absolute values. That is, values ​​with large changes are excluded regardless of whether they are positive or negative. The basis for exclusion can be the order of the magnitude of the change, or a threshold can be used.

[0064] The second damage detection unit 17 uses the probability density function of the sample's change to calculate the probability of an excluded value. If the probability of the excluded value is sufficiently small, the tool is determined to be damaged. That is, a portion of the change is excluded, and based on the probability density function calculated from the remaining change, if the probability of the excluded value is sufficiently small, the tool is determined to be damaged.

[0065] In the modified example, both the slope and the sample value can be used to determine the breakage, thus broadening the scope of breakage determination. Furthermore, by preparing multiple feature quantities, breakage under various conditions can be detected.

[0066] The hardware structure of the damage determination device 100 using the present disclosure will be described below. Figure 15 This is a hardware structure diagram of the damage assessment device 100. (For example...) Figure 15 As shown, the damage determination device 100 includes: a CPU 111 for overall control of the damage determination device 100, a ROM 112 for recording programs and data, and a RAM 113 for temporarily expanding data. The CPU 111 reads the system program recorded in the ROM 112 via the bus and determines the damage according to the system program.

[0067] The non-volatile memory 114, for example, uses a battery (not shown) for backup, and maintains its storage state even when the power supply to the damage determination device 100 is disconnected. The non-volatile memory 114 stores various data, such as programs read from the external device 120 via interfaces 115, 118, and 119, and operation inputs input via the input unit 30. The non-volatile memory 114 can store programs and data used to execute the damage determination device 100 of this embodiment.

[0068] Interface 115 is used to connect the damage determination device 100 and external devices 120 such as adapters. Programs and various parameters are read from the external device 120.

[0069] Interface 118 is used to connect the damage determination device 100 and the display unit 70, such as a liquid crystal display. The display unit 70 displays various data read from the memory, as well as data obtained as a result of executing a program, etc.

[0070] Interface 119 is used to connect the damage assessment device 100 to the input unit 30, such as a keyboard and a positioning device. The input unit 30 transmits instructions and data from the operator to the CPU 111 via interface 119.

[0071] This disclosure has been described in detail, but it is not limited to the various embodiments described above. These embodiments can be modified, supplemented, altered, or partially deleted in various ways without departing from the spirit of this disclosure or from the spirit of the disclosure derived from the claims and their equivalents. Furthermore, these embodiments can also be implemented in combination. For example, in the embodiments described above, the order of actions and the order of processes are shown as an example and are not limited thereto.

[0072] The following notes are also disclosed regarding the above-described embodiments and variations.

[0073] (Note 1)

[0074] The breakage determination device (100) includes: a data acquisition unit (10) that acquires the value of the load applied to the shaft of the machine tool as a sample; a determination unit (11) that determines whether the sample acquired by the data acquisition unit is a sample detected during the cutting time period or a sample detected during the non-cutting time period; a probability calculation unit (12) that calculates the probability density function of the sample during the non-cutting time period; a feature quantity calculation unit (13) that calculates the feature quantity of the sample during the cutting time period; and a breakage detection unit (14) that determines the breakage of the tool based on the probability of generating the feature quantity in the probability density function during the non-cutting time period.

[0075] (Note 2)

[0076] The determination unit (11) determines the cutting time period and the non-cutting time period based on the signal of the command controlling the machine tool.

[0077] (Note 3)

[0078] If the probability of generating the aforementioned feature is sufficiently small, the damage detection unit (14) determines that the tool is not damaged.

[0079] (Note 4)

[0080] The feature quantity calculation unit (13) calculates multiple feature quantities, and the damage detection unit uses the multiple feature quantities to detect damage under different conditions.

[0081] (Note 5)

[0082] The feature quantity is the maximum value of the sample in the cutting interval.

[0083] (Note 6)

[0084] The feature quantity is the minimum value of the sample in the cutting interval.

[0085] (Note 7)

[0086] Computer-readable storage media (112, 113, 114) store commands that cause one or more processors (111) to perform the following processes: obtain the value of the load applied to the axis of the machine tool as a sample, determine whether the sample was detected during the cutting time period or during the non-cutting time period, calculate the probability density function of the sample during the non-cutting time period, calculate the feature quantity of the sample during the cutting time period, and determine the tool breakage based on the probability of the feature quantity being generated in the probability density function during the non-cutting time period.

[0087] Explanation of reference numerals in the attached figures

[0088] 100 Damage Detection Device

[0089] 10 Data Acquisition Department

[0090] 11 Judgment Department

[0091] 12 Probability Calculation Department

[0092] 13 Characteristic Quantity Calculation Department

[0093] 14. Damage Inspection Department

[0094] 111 CPU

[0095] 112 ROM

[0096] 113 RAM

[0097] 114 Non-volatile memory.

Claims

1. A damage determination device, characterized in that, have: The data acquisition unit acquires the load values ​​applied to the machine tool's shafts as samples; The determination unit determines whether the sample acquired by the data acquisition unit was detected during the cutting time period or during the non-cutting time period. The probability calculation unit calculates the probability density function of the sample during the non-cutting time period; The feature calculation unit calculates the feature quantities of the sample during the cutting time period; The breakage detection unit determines tool breakage based on the probability of generating the feature quantity in the probability density function during the non-cutting time period.

2. The damage determination device according to claim 1, characterized in that, The determination unit determines the cutting time period and the non-cutting time period based on the signals of the commands controlling the machine tool.

3. The damage determination device according to claim 1, characterized in that, If the probability of generating the aforementioned characteristic quantity is sufficiently small, the damage detection unit determines that the tool is not damaged.

4. The damage determination device according to claim 1, characterized in that, The feature quantity calculation unit calculates multiple feature quantities. The damage detection unit uses the multiple feature quantities to detect damage under different conditions.

5. The damage determination device according to claim 1, characterized in that, The feature quantity is the maximum value of the sample in the cutting interval.

6. The damage determination device according to claim 1, characterized in that, The feature quantity is the minimum value of the sample in the cutting interval.

7. A computer-readable storage medium, characterized in that, The storage enables one or more processors to execute the following commands: The load value applied to the machine tool's shaft is obtained as a sample. Determine whether the sample was detected during the cutting period or during the non-cutting period. Calculate the probability density function of the samples during the non-cutting time period. Calculate the feature values ​​of the samples during the cutting time period. Tool breakage is determined based on the probability of generating the feature quantity in the probability density function during the non-cutting time period.

Images