JP7613084B2 - Processing Monitoring System - Google Patents

Description

The present invention relates to a processing monitoring system.

In recent years, additive manufacturing (AM), which creates products of desired shapes from raw materials such as metals and resins, has become widespread, and as a result, new challenges have arisen in cutting processes that are carried out as a post-process. The first challenge is that the material properties of products made by additive manufacturing (additively manufactured products) are different from those of materials made by conventional manufacturing methods (strength, etc.). A similar challenge is also expected for products made of composite materials formed by mixing multiple materials. Even when multiple materials are mixed, the properties of the composite material formed may be unknown. In such a situation, it is not easy to set the processing conditions for the workpiece, and in order to optimize the cutting conditions, it is necessary to determine the conditions through repeated trial and error, which increases the processing time.

The second challenge is that additively manufactured products often have complex shapes, and the unevenness of the layered surfaces can be large, making it difficult to accurately predict cutting resistance values before machining. As a result, the frequency of chipping and breakage of machining tools and destruction of the additively manufactured products that are the processed products increases. In addition, because the cutting state of the tool and workpiece (work material) assumed before machining cannot be maintained, the frictional forces and heat conditions on the machined surface cannot be controlled, and the processing quality must be checked more frequently.

As such, especially in the processing of additively manufactured products, the amount of work required to adjust cutting conditions, reprocess and reshape increases, resulting in increased processing time. For this reason, there is a growing need to understand the characteristics, cutting resistance and processing quality of the workpiece during processing, and to use this information to reduce processing time and stabilize processing quality. These needs exist in all processing technologies, regardless of additively manufactured products, and if they can be solved, they could be useful and widely applicable. The type and characteristics of the material of the workpiece, cutting resistance, processing quality, etc. are collectively referred to as "characteristic information of the workpiece." Characteristic information of the workpiece is a parameter that depends on the material of the workpiece. Furthermore, "processing quality" is an evaluation of the shape of the workpiece after processing, such as the thickness of the modified part, the dimensions after processing, and surface roughness.

For example, Patent Document 1 below is an example of a technology for predicting machining dimensions during machining. Patent Document 1 discloses a method for predicting machining dimensions (machining quality) by calculating the average value, maximum value, minimum value, standard deviation, integral value, etc. of a specific machining section as feature values using data on the machining section during machining of the machine tool, data indicating the operating state of the machine tool, and data acquired by sensors (current sensor, vibration sensor, temperature sensor) installed on the machine tool, and inputting these values into a model constructed in advance.

Patent document 2 discloses an AC motor monitoring device that monitors an AC motor driven by a drive power supply, which performs coordinate conversion of the current detection values using at least two-phase current detection values of the AC motor detected by a current detection means for the AC motor, converts the current detection values into DC amounts based on the phase of the coordinate-converted current detection values, and outputs a feature value based on the DC amount. Patent document 2 describes calculating the DC amount from the current of an AC motor connected to a machine (such as a machine tool), extracting feature values (changes over time or values or changes indicating abnormal operation) from this DC amount, and monitoring the behavior of the AC motor.

International Publication No. 2019/239606 International Publication No. 2019/049188

However, while the techniques described in Patent Documents 1 and 2 above can appropriately modify the processing conditions based on the operation of the machine tool during processing, they do not disclose any means for appropriately modifying the processing conditions while taking into account the material properties of the workpiece during processing (the material's processing temperature dependence and the cutting thickness (removal volume) dependence). As described above, particularly in the processing of products made of composite materials and having complex shapes such as additively manufactured products, it is more difficult to predict the processing temperature changes and processing volume in advance than in the processing of products made of a single material with few uneven surfaces, and these change from moment to moment even during processing, so it is estimated that it is difficult to improve the prediction accuracy of the characteristic information of the workpiece (particularly the material properties and type of material) using conventional methods.

Therefore, the object of the present invention is to provide a processing monitoring system that accurately predicts information about a workpiece, such as the processing quality, material characteristics, type, and cutting resistance of the workpiece, during processing.

One aspect of the present invention for achieving the above-mentioned object is a machining monitoring system that includes a machine tool having a tool for machining a workpiece and an AC motor (same meaning as the motor in the figure) for driving the tool, a signal conversion unit that converts the AC current of the AC motor into a DC current, an information extraction unit that extracts at least one of machining speed information and cutting thickness information from the DC current (taking into account machining temperature dependency and cutting thickness (removal volume) dependency), and a workpiece characteristic information prediction unit that inputs the above information into a pre-constructed model and predicts at least one of the machining quality, material characteristics, type, and cutting resistance.

More specific configurations of the present invention are described in the claims.

The present invention provides a processing monitoring system that accurately predicts information about a workpiece, such as the processing quality, material characteristics, type, and cutting resistance of the workpiece during processing.

Issues, configurations, and advantages other than those described above will become clear from the description of the embodiments below.

Configuration diagram of a machining monitoring system according to a first embodiment Schematic diagram showing the relative positions of the tool and the workpiece during milling Schematic diagram showing the relative positions of the tool and the workpiece during turning FIG. 2 is a flow diagram of the process performed in the signal conversion unit of FIG. Flow diagram of the process performed in the cutting speed and cutting thickness information extraction unit Graph showing example waveforms and average values of the spindle torque current estimate Ia and the spindle motor speed estimate Wi, divided at 0.1 second intervals. A scatter diagram showing the relationship between the average value of the estimated spindle motor speed (W1_ave) and the cutting speed (command value) when machining a certain material, and the relationship between the average value of the estimated spindle torque current (Ia_ave) and the feed amount per revolution (command value) Flow diagram of the process performed in the workpiece characteristic information prediction unit A graph showing a comparison between the actually measured cutting resistance value and the cutting resistance value calculated using formula (6) A graph showing an example of a region (model) for classifying two workpiece materials A and B, calculated by SVM, with cutting speed information and cutting thickness information on two axes. Configuration diagram of a machining monitoring system according to a second embodiment Configuration diagram of a machining monitoring system according to a third embodiment FIG. 1 is a diagram showing an example of information indicating the type of material of a workpiece. FIG. 11 is a configuration diagram of a machining monitoring system according to a fourth embodiment of the present invention.

A representative embodiment of the present invention will be described in detail. Reference symbols in the drawings are merely illustrative of what is included in the concept of the component to which they are attached.

Figure 1 is a configuration diagram of a machining monitoring system in Example 1. In this example, a milling machine (milling processing device) and a lathe are assumed as machine tools, and a machining monitoring system using cutting speed and cutting thickness information extracted from the phase current of the spindle motor is described.

As shown in FIG. 1, the monitoring system 100a of the first embodiment includes a machine tool 5, a signal conversion unit 6, a cutting speed/feed rate information extraction unit 7, a workpiece characteristic information prediction unit 8, and a notification device 9.

The machine tool 5 has a spindle motor 2 that drives a tool to machine a workpiece. In this embodiment, monitoring is performed using phase currents (motor currents) acquired from the spindle motor 2 (spindle motor) that rotates the tool and workpiece, so only the spindle motor 2 is shown in the machine tool 5. A typical machine tool also has multiple servo motors installed to move the workpiece and tool to any position, but these are omitted in this diagram.

Here, the positional relationship between the spindle and the workpiece will be explained. FIG. 2 is a schematic diagram showing the positional relationship between the tool and the workpiece in milling, and FIG. 3 is a schematic diagram showing the positional relationship between the tool and the workpiece in turning. As shown in FIG. 2, in milling, the tool 20 rotates around the spindle in a rotation direction 23 to machine the workpiece 21. The tool 20 or the workpiece 21 can be moved in the X-axis, Y-axis, and Z-axis directions by a servo motor (not shown). Also, as shown in FIG. 3, in turning, the workpiece 21 is machined by the tool 20 as it rotates around the spindle in a rotation direction 23. The tool 20 or the workpiece 21 can be moved in the X-axis, Y-axis, and Z-axis directions by a servo motor (not shown).

The procedure for acquiring current from the spindle motor 2 in FIG. 1 will be described below. The phase currents are acquired by the current sensors 3 and 4. Here, in the figure, two-phase currents are acquired for one motor, but two or more phases are sufficient, and three-phase currents may be acquired. In this embodiment, the explanation will be given assuming that the currents of the U and V phases out of the three phases UVW are acquired. In addition, although FIG. 1 shows a diagram in which currents are acquired outside the inverter 1, information from a current sensor installed inside the inverter 1 may be used. The acquired current information is input to the signal conversion unit 6 and converted into a DC amount. Here, the DC amount refers to a signal in which the frequency component of the motor rotation period is removed and the time series data is a constant value if there is no influence of load fluctuation, rotation speed fluctuation, or inverter carrier frequency (conversely, if there is load or rotation speed fluctuation, the DC amount will fluctuate in accordance with the period). The DC amount of the spindle motor current output from the signal conversion unit 6 is input to the cutting speed/cutting thickness information extraction unit 7, which extracts information (feature amount) related to the cutting speed and cutting thickness during cutting. The cutting speed information and cutting thickness information are then input to the workpiece characteristic information prediction unit 8, which estimates the processing quality, material (particularly the material characteristics and material type), and cutting resistance (collectively defined as the workpiece characteristic information). The estimated workpiece characteristic information is input to the notification device 9. From here, each of the processing blocks 6 to 9 will be explained in detail.

First, the signal conversion unit 6 will be described in detail. Fig. 4 is a flow diagram of the process performed in the signal conversion unit of Fig. 1. Fig. 4 shows a DC conversion process called coordinate conversion using a current phase. First, the acquired U-phase current Iu and V-phase current Iv are input to the W-phase current generation unit 10. This process is performed when the measured current is two-phase, and is not necessary when all three phases are acquired by current sensors. The W-phase current generation unit 10 performs the process shown in the following formula (1).
Iw=-(Iu+Iv)...(1)
The three-phase current is input to a three-phase to two-phase converter 11, which performs the processing shown in the following equations (2) and (3). As a result, Iα and Iβ are obtained as orthogonal two-axis current components based on the α axis and the β axis.
Iα=(2/3) {Iu-Iv/2-Iw/2}...(2)
Iβ=(1/√(3)) {Iv−Iw} …(3)
Thereafter, a phase required for coordinate transformation (conversion to DC) is calculated by a phase calculation unit 15. The phase calculation unit 15 is composed of an instantaneous phase calculation unit 12, a PLL unit (phase storage unit) 13, and an integration unit 14. Among these, the instantaneous phase calculation unit 12 performs the process shown in the following equation (4) to calculate an instantaneous phase θi*.
θi*=tan-1(Iβ/Iα)...(4)
The instantaneous phase θi ^* passes through a feedback loop consisting of a PLL unit 13 and an integrator 14, and finally the integrator 14 generates a coordinate transformation phase θi. This value becomes an estimate of the motor rotor position. In this process, the PLL unit 13 provides an electrical angle estimate Wi (motor speed estimate) as a DC amount. Since this feedback loop has the role of removing noise, if noise in the instantaneous phase θi ^* is not a concern, it is not necessary to pass the instantaneous phase θi* through the feedback loop.

The orthogonal two-axis current components Iα and Iβ based on the α-axis and β-axis calculated by the three-phase to two-phase conversion unit 11 and the coordinate transformation phase θi calculated by the phase calculation unit 15 are input to the coordinate transformation unit 16, and the coordinate transformation is performed by the process shown in the following equation (5).
Ia=Iα・cos(θi)+Iβ・sin(θi)…(5)
This coordinate transformation can be said to be a transformation of a current (AC) in a fixed coordinate system into a current (DC) in a rotating coordinate system. The calculated DC value Ia in the rotating coordinate system and the electrical angle estimate value Wi (motor speed estimate) output from the phase calculation unit 15 are DC amounts. Note that the DC amount Ia changes in a similar manner to the temporal change of the torque current used in motor control, and therefore is described as a torque current estimate in this specification.

Next, the details of the cutting speed/cutting thickness information extraction unit 7 will be described with reference to FIG. 5. FIG. 5 is a flow diagram of the processing performed by the cutting speed/cutting thickness information extraction unit. As shown in FIG. 5, the cutting speed information is calculated by taking the average value of the estimated spindle motor speed value Wi, and the cut thickness information is calculated by taking the average value of the estimated spindle torque current value Ia. The average value is calculated by dividing the input time series data for Ia and Wi into certain time intervals and dividing the total value within that interval by the number of data points.

Figure 6 is a graph showing an example of waveforms and average values of the torque current estimate Ia of the spindle and the motor speed estimate Wi of the spindle, divided by 0.1 seconds. Here, we will explain why the cutting speed and cutting thickness information can be calculated by the average value of Ia and Wi. First, the cutting speed is the peripheral speed of the rotating tool blade or the workpiece, and is proportional to the rotation speed of the tool or the workpiece. In other words, if wear on the tool blade is not taken into consideration, it can be seen that the electrical angle estimate Wi can be used as an alternative value for the cutting speed. On the other hand, the cutting thickness is a value that correlates with the thickness of the chips, and in milling, it increases with an increase in the feed rate per blade, and in turning, it increases with an increase in the feed rate per rotation. Here, the feed rate is a value that indicates how much the tool has moved relative to the workpiece, and increasing the feed rate for the spindle that rotates the tool or workpiece leads to an increase in the load because a force acts in the direction in which the tool and the workpiece are pressed against each other. Therefore, the average torque current estimate Ia of the main shaft motor correlates with the cutting thickness.

Figure 7 is a scatter plot showing the relationship between the average estimated spindle motor speed (W1_ave) and the cutting speed (command value) when turning a certain material, and the relationship between the average estimated spindle torque current (Ia_ave) and the feed amount (command value) per revolution and the command value (set value) for the cutting thickness. As shown in Figure 7, there is a positive correlation between the cutting speed and W1_ave, and between the cutting thickness (feed amount) and Ia_ave, confirming that the above-mentioned content is correct.

Furthermore, the workpiece characteristic information prediction unit 8 in FIG. 1 will be described with reference to FIG. 8. FIG. 8 is a flow diagram of the process performed by the workpiece characteristic information prediction unit 8. As shown in FIG. 8, the workpiece characteristic information prediction unit 8 predicts at least one of the processing quality, material characteristics or type, and cutting resistance (workpiece characteristic information) by inputting at least one of the calculated cutting speed information and cutting thickness information into a model constructed in advance. When predicting characteristic information of multiple workpieces, multiple models are prepared. Examples of models include a regression equation, a cluster model, and a classification model.

Here, an example of predicting cutting resistance using a regression equation is shown. For predicting cutting resistance, it is possible to use an equation obtained by modifying the Kronenberg theoretical equation shown in equation (6). Here, F is the cutting resistance value, A, b, α, β, and γ are coefficients, a is the cutting depth, V is the cutting speed, and f is the chip thickness (feed amount), and α, β, and γ are determined in advance from the cutting resistance value obtained in advance, and the cutting speed information and the chip thickness information extracted from the phase current.
[Equation 6]
F={(A*ab)*Vβ*fγ}=α*Vβ*fγ…(6)
That is, if the values calculated by the cutting speed/cutting thickness information extraction unit 7 are input to V and f in formula (6), the cutting resistance value can be calculated. FIG. 9 is a graph showing a comparison between the measured cutting resistance value and the cutting resistance value calculated using formula (6). If the plot is on the dashed line in the figure, it indicates that the predicted value and the measured value match. As can be seen from the figure, the cutting resistance calculated using formula (6) can be predicted with an average measurement error of about 10%. Note that formula (6) is only an example, and any formula containing terms for cutting speed information and cutting thickness information can be used. Furthermore, the regression formula can also be used to predict processing quality such as the thickness of the processing-affected layer and the residual stress of the processing surface.

As an example other than the regression model, a classification model for classifying the type of workpiece is constructed using a Support Vector machine (SVM) or the like. Figure 10 is a graph showing an example of a region (model) for classifying two types of workpiece materials A and B calculated by SVM, with cutting speed information and cutting thickness information on two axes. The line in the figure indicates the separation boundary surface calculated by SVM, and the dark black (material A) and gray (material B) square marks indicate separate material data. As can be seen from the figure, a classification region (classification model) is created to effectively separate the dark black and gray data, and in this example, a model that can obtain a classification accuracy rate of 90% or more has been constructed. When actually using this model, the cutting thickness information and cutting speed information when processing an unknown material are input, and by determining which region the data is located in, the workpiece characteristic information prediction unit 8 outputs whether it is material A or B.

The workpiece characteristic information prediction unit 8 may output A or B as described above, or may output the class probability for each material (for example, in a two-material classification model, the probability of material A is x [%], and the probability of material B is 100-x [%]). Using the class probability makes it possible to grasp in more detail the material characteristics of a metal additively manufactured product constructed by mixing multiple materials (such as the degree to which the characteristics of material A are expressed). Note that in this example, a model that classifies only two materials has been explained, but it is also possible to create a model that can classify more than two materials.

Finally, the notification device 9 in Figure 1 will be explained. The notification device 9 may, for example, display the predicted results of cutting resistance, cutting quality, and the material of the workpiece directly on a screen, or may sound an alarm, turn on a warning light, or display text such as a warning on the screen if the cutting resistance or quality is above (or below) a certain value. In addition, the predicted results of cutting resistance and the like may be notified to an NC control device that controls the machine tool, and used as an input value for a system that changes the operation command of the machine tool based on the cutting resistance value.

In this way, by using Example 1, a model can be constructed that can predict the characteristic information of the workpiece with high accuracy from the phase current of the spindle motor during machining. This makes it possible to shorten the time required to determine cutting conditions, shorten the time required for rework due to machining failures, and shorten the inspection time required to stabilize machining quality, thereby contributing to shortening the cutting processing time.

Conventionally, it was possible to modify the processing conditions during processing, taking into account the state of the machining device, but it was not possible to modify the processing conditions while taking into account the characteristics of the material of the workpiece during processing, as in this embodiment. The present invention is useful when the material of the workpiece is unknown, particularly in the processing of additively manufactured products.

Figure 11 is a block diagram of a machining monitoring system according to a second embodiment. In the second embodiment, another machining monitoring system is described that uses at least one of the motor current of the feed axis in the main force direction (set as the Y-axis direction in this example) and the chip temperature, together with the cutting speed and chip thickness information extracted from the phase current of the spindle motor, in order to improve the prediction accuracy of the material of the workpiece. Note that in this embodiment, the explanation of the same functional blocks as in the first embodiment is omitted.

As shown in FIG. 11, the difference between this embodiment and embodiment 1 is that a torque current estimate is calculated from the phase current of the Y-axis motor 206 acquired by the current sensors 207 and 208 (DC signal conversion unit (Y-axis) 213), the torque current estimate is averaged (average value calculation unit 214), and the average value and the chip temperature are input to the workpiece characteristic information prediction unit 211 together with the cutting speed information and the chip thickness information.

In FIG. 11, both the average value of the Y-axis torque current estimate and the chip temperature are input to the workpiece characteristic information prediction unit 211, and the following explanation will be given on the assumption that the two additional signals are used, but it is also possible to use either one of the signals. The other processing block contents are the same as those explained in FIG. 1. The following will explain the processing blocks and signals that have been added or changed.

First, the DC conversion unit (Y-axis) 213 added in Example 1 performs the same processing as that described in FIG. 4, and the average calculation unit 214 performs the same processing as that described in FIG. 5. Next, the chip temperature is a value obtained by measuring the temperature of the chips generated during processing, for example, with a high-speed thermo camera or a radiation thermometer. Note that the hardness or color of the chips may be used instead of the chip temperature. This is because the hardness and color change depending on the chip temperature. As a result of the inventor's investigation, it was found that information regarding the material of the workpiece can be obtained more accurately by using the chip temperature, hardness, or color rather than the temperature of the tool or workpiece.

The workpiece characteristic information prediction unit 211 constructs a classification model as a four-dimensional space by adding the two dimensions of the average Y-axis torque current and the chip temperature to the two-dimensional space shown in FIG. 10. This improves the prediction accuracy of material characteristics and types. This is because the thermal conductivity characteristics of each material can be understood from the chip temperature information, and the average Y-axis torque current can be used to understand the load in the principal force direction during machining, making it possible to more accurately understand information such as the hardness of the material.

In this way, by using Example 2, the prediction accuracy of the material of the workpiece can be improved, which contributes to shortening the cutting processing time more than Example 1.

Figure 12 is a configuration diagram of a machining monitoring system according to a third embodiment. In the third embodiment, another machining monitoring system is described that uses information indicating the type of material of the workpiece in combination with information on the cutting speed and cutting thickness extracted from the phase current of the spindle motor in order to improve the prediction accuracy of the cutting resistance and machining quality of the cutting process. Note that in this embodiment, explanations of the same functional blocks as those in the first embodiment are omitted.

As shown in FIG. 12, the difference between this embodiment and the first embodiment is that the material of the workpiece is predicted by inputting information indicating the type of material of the workpiece in addition to cutting speed information and cutting thickness information to the workpiece characteristic information prediction unit 108.

FIG. 13 is a diagram showing an example of information indicating the type of material of the workpiece. The information indicating the type of material of the workpiece in FIG. 12 is, for example, a list of numbers as shown in FIG. 13. Each digit of the number is called a dummy variable, and only one dummy variable is represented as one (1), and the rest are represented as zero (0). When processing material A, the number "100...0" is input, with only dummy variable A being 1. This number is input, for example, by the worker performing the processing to the monitoring device. Then, the list of numbers made of this dummy variable is used as in formula (7) by adding it to formula (6) shown in embodiment 1. Formula (7) represents the case where four materials are used. F is the cutting resistance prediction value, α, β, γ, δ, ε, and ζ are coefficients, V is cutting speed information, f is cutting thickness (feed amount) information, DA, DB, and DC are dummy variables, and α, β, γ, δ, ε, and ζ are determined from the cutting resistance value, cutting speed information, and cutting thickness information obtained in advance. Here, although equation (7) targets four materials, only three dummy variables are used. This is because the time when DA, DB, and DC are all zero can be treated as a prediction equation for material D, and three dummy variables are sufficient.
F=α*Vβ*fγ*(δ*DA+ε+DB+ζ*DC)…(7)
In this way, by adding information indicating the type of workpiece material to the equation as a dummy variable, the coefficients δ, ε, and ζ applied to the dummy variables correspond to the machinability and machinability of each material (material properties taking into account hardness, thermoelectric coefficient, etc.), making it possible to predict the resistance value more accurately.

In other words, by using Example 3, the prediction accuracy of cutting resistance and processing quality can be improved, which contributes to shortening the cutting processing time more than Example 1.

Figure 14 is a configuration diagram of a machining monitoring system according to a fourth embodiment of the present invention. In the fourth embodiment, another machining monitoring system is described, which uses workpiece information indicating the type of material of the workpiece predicted from the phase current of the spindle motor in combination with information on the cutting speed and cutting thickness extracted from the phase current of the spindle motor in order to improve the prediction accuracy of cutting resistance and machining quality. Note that in this embodiment, explanations of the same functional blocks as those in the third and fourth embodiments are omitted.

As shown in FIG. 14, the difference between this embodiment and the first and third embodiments is that the present embodiment includes the workpiece characteristic information prediction units 308 and 310, and the results predicted by the workpiece characteristic information prediction unit (workpiece material) 308 are used as information indicating the type of material of the workpiece used in the workpiece characteristic information prediction unit (cutting resistance or machining quality) 310. Note that, although information on the chip temperature and the Y-axis torque current average value is input to the workpiece characteristic information prediction unit (workpiece material) 308 shown in FIG. 14, these two signals may be omitted (configuration of the first embodiment). The output of the workpiece characteristic information prediction unit (workpiece material) 308 may be output in the form of dummy variables shown in FIG. 13, or may be a list of numbers by assigning class probability numbers to each dummy variable. The output workpiece material information indicating the type of material of the workpiece is input to the workpiece characteristic information prediction unit (cutting resistance or machining quality) 310. The processing of the workpiece characteristic information prediction unit (cutting resistance or processing quality) 310 is the same as in Example 4.

In this way, by using Example 4, it is possible to improve the prediction accuracy of cutting resistance and processing quality without manually inputting information indicating the type of material of the workpiece, which contributes to a greater reduction in cutting processing time than Example 3.

As described above, the present invention has been shown to provide a processing monitoring system that can accurately predict information about a workpiece, such as the processing quality, material characteristics, and cutting resistance of the workpiece during processing.

The present invention is not limited to the above-described embodiments, but includes various modified examples. For example, the above-described embodiments have been described in detail to make the present invention easier to understand, and the present invention is not necessarily limited to those having all of the configurations described. It is also possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

1,101,201,301... Main shaft inverter, 2,102,202,302... Main shaft motor, 205,311... Y-axis inverter, 206,312... Y-axis motor, 3,4,103,104,203,204,207,208,303,304,313,314... Current sensor, 5,105,305... Machine tool, 6,106,2 09, 213, 306, 315...signal conversion unit, 7, 107, 210, 307...cutting speed/cutting thickness information extraction unit, 214, 316...average value calculation unit, 8, 108, 211, 308, 310...workpiece characteristic information prediction unit, 9, 109, 212, 309...notification device, 100a, 100b, 100c, 100d...machining monitoring system.

Claims (7)

a machine tool having a tool for machining a workpiece and an AC motor for driving the tool;
A signal conversion unit that converts the AC current of the AC motor into a DC current;
an information extracting unit that extracts at least one of information on a cutting speed and information on a chip thickness from the DC amount;
a workpiece characteristic information prediction unit that inputs the information into a pre-constructed model and predicts at least one of the processing quality, the characteristics or type of the material of the workpiece, and the cutting resistance;
Equipped with
The DC amount is a main shaft motor speed estimate value and a main shaft motor torque current estimate value,
the information on the cutting speed is an average value of the estimated spindle motor speeds;
The cutting thickness information is an average value of the main shaft motor torque current estimate.
A processing monitoring system comprising: a machine tool having a tool for machining a workpiece and an AC motor for driving the tool;
A signal conversion unit that converts the AC current of the AC motor into a DC current;
an information extracting unit that extracts at least one of information on a cutting speed and information on a chip thickness from the DC amount;
a workpiece characteristic information prediction unit that inputs the information into a pre-constructed model and predicts at least one of the processing quality, the characteristics or type of the material of the workpiece, and the cutting resistance;
Equipped with
When predicting the characteristics or type of the material of the workpiece, an average value of the torque current estimate value of the AC motor responsible for the movement in the principal force direction and information on the chip temperature are input to the pre-constructed model together with information from the information extraction unit;
When predicting the processing quality and the cutting resistance, information indicating a type of material of the workpiece is input into the pre-constructed model together with information from the information extraction unit;
A processing monitoring system comprising: 3. The machining monitoring system according to claim 1, wherein the signal conversion unit uses measurement values of current sensors installed in at least two phases of a spindle motor that rotates the tool or the workpiece. 4. The machining monitoring system according to claim 3 , wherein the signal conversion unit calculates the DC amount by using a current phase calculated from the two-phase currents. 5. The machining monitoring system according to claim 1, wherein the information indicating the characteristics or type of the material of the workpiece is a predicted value of the material of the workpiece predicted from the AC current. The machining monitoring system according to any one of claims 1 to 5 , wherein the pre-constructed model is a regression model, a cluster model or a classification model . a machine tool having a tool for machining a workpiece and an AC motor for driving the tool;
A signal conversion unit that converts the AC current of the AC motor into a DC current;
an information extracting unit that extracts at least one of information on a cutting speed and information on a chip thickness from the DC amount;
a workpiece characteristic information prediction unit that inputs the information into a pre-constructed model and predicts at least one of the processing quality, the characteristics or type of the material of the workpiece, and the cutting resistance;
Equipped with
The information indicating the characteristics or type of the material of the workpiece is a predicted value of the material of the workpiece predicted from the AC current.
A processing monitoring system comprising:

Images