JP7221459B1 - Thermal displacement model learning device, thermal displacement estimation device, machining system, and machining method - Google Patents

Abstract

A thermal displacement model learning device (2) performs learning in which input data including at least temperature data representing the time-series temperature of the machine tool and thermal displacement data representing the time-series amount of thermal displacement of the machine tool are associated with each other by time. A learning data acquisition unit (21) that acquires data for learning, a data set generation unit (22) that generates a data set by extracting a part of the learning data from the learning data, and a machine tool using the data set a learning unit (23) for generating a learned thermal displacement model for estimating the amount of thermal displacement from the input data of the data set generation unit (22) for at least two of the plurality of data sets In the data set, the data set is generated such that the time lengths of the extracted time intervals are different from each other.

Description

The present disclosure relates to a thermal displacement model learning device, a thermal displacement estimation device, a machining system, and a machining method for correcting thermal displacement of machine tools.

A machine tool is a processing device that removes and processes an object called a workpiece into a desired shape by driving a feed shaft to change the relative position between the object and the tool. Machine tools represented by milling machines, lathes, and the like mount a tool or an object to be machined on a spindle, and perform machining while rotating the spindle. A machine tool includes an actuator for driving a feed shaft and a spindle, and a servo amplifier for generating electric power for driving the actuator. A machine tool equipped with an actuator performs positioning by measuring the position of the feed shaft with a position detector and controlling the actual position to match the commanded position.

When a machine tool processes a workpiece, heat is generated in the machine tool. For example, actuators and servo amplifiers generate heat when converting electric power into power. Further, in processing such as cutting and polishing, heat is generated due to material deformation, friction, and the like. Furthermore, the friction generated when the feed shaft is operated also generates heat. When this heat is transferred to the machine tool structure such as the column and bed, the temperature of the machine tool structure rises and thermal expansion occurs. The deformation of structures due to this thermal expansion is called thermal displacement. In some cases, the temperature of the structure does not change evenly, and in such cases, different thermal displacements occur depending on the position in the structure, which may cause displacement other than simple expansion and contraction such as collapse, distortion, and twisting of the structure. . It is known that such thermal displacement lowers the positioning accuracy of the machine tool, causing an error in the relative position between the tool and the workpiece, and lowering the machining accuracy.

Patent Literature 1 discloses a technique for estimating the amount of thermal displacement from the measured data through machine learning using measured data obtained during machining by a machine tool and actual measured values of the amount of thermal displacement. disclosed. Specifically, the machine learning device disclosed in Patent Document 1 includes a measurement data acquisition unit that acquires a measurement data group, a thermal displacement amount acquisition unit that acquires an actual measurement value of a thermal displacement amount, and inputs the measurement data group. The thermal displacement is calculated based on the measurement data group by performing machine learning using a storage unit that stores training data by associating the measured values of the thermal displacement as data with each other as labels, and by performing machine learning using the training data. and a calculation formula learning unit for setting an amount prediction calculation formula. The storage unit stores the measurement data group as teacher data for a predetermined period.

JP 2019-111648 A

However, according to the conventional technique described above, if the period during which the storage unit stores the teacher data is not appropriate for the thermal displacement that occurs in the machine tool, the accuracy of estimating the amount of thermal displacement that occurs in the machine tool decreases. There was a problem that there is a case to do. For example, the time required for heat from a certain heat source to cause thermal displacement is affected by the positional relationship with other heat sources, the ambient environment, and the like. For this reason, if the storage period is too short, measurement data that affects thermal displacement may not be included in the stored teacher data during learning, and the accuracy of estimating the amount of thermal displacement may decrease. In addition, if the storage period is too long, the accuracy of estimating the amount of thermal displacement may decrease due to erroneous learning of measurement data that is included in the data and has a small contribution to thermal displacement during learning. be.

The present disclosure has been made in view of the above, and an object thereof is to obtain a thermal displacement model learning device capable of improving the estimation accuracy of the amount of thermal displacement.

In order to solve the above-described problems and achieve an object, a thermal displacement model learning device of the present disclosure includes input data including at least temperature data representing time-series temperature of a machine tool, and time-series thermal displacement of the machine tool. A learning data acquisition unit that acquires learning data in which thermal displacement data representing a quantity is associated with time, a data set generation unit that generates a data set by extracting a part of the learning data from the learning data, a learning unit that uses the dataset to generate a trained thermal displacement model for estimating the amount of thermal displacement from the input data of the machine tool, the dataset generating unit generating at least The two data sets are characterized in that the time lengths of the extracted time intervals are different from each other.

Advantageous Effects of Invention According to the present disclosure, it is possible to improve the estimation accuracy of the amount of thermal displacement.

1 is a diagram showing the configuration of a processing system according to a first embodiment; FIG. 1 is a diagram showing a functional configuration of a machine tool according to Embodiment 1; FIG. FIG. 2 is a diagram showing an example of installation positions of temperature sensors in the machine tool according to the first embodiment; Explanatory diagram of an example of changes in the operating state of a machine tool 1 is a diagram showing a functional configuration of a thermal displacement model learning device according to a first embodiment; FIG. Explanatory diagram of the first example of the data set generation method Explanatory diagram of the second example of the method of generating the dataset Explanatory diagram of the third example of the data set generation method Explanatory diagram of the learning method of the thermal displacement model Illustration of neural network Flowchart for explaining the operation of the thermal displacement model learning device shown in FIG. The figure which shows the structure of the processing system concerning Embodiment 2. FIG. 2 is a diagram showing the functional configuration of a machine tool according to a second embodiment; FIG. 10 is a diagram showing a functional configuration of a thermal displacement estimating device according to a second embodiment; Flowchart for explaining the operation of the machining system shown in FIG. Diagram showing an example of hardware configuration FIG. 10 is a diagram showing a functional configuration of a thermal displacement model learning device according to a modified example of the second embodiment; The figure which shows the structure of the processing system concerning Embodiment 3. The figure which shows the structure of the processing system concerning Embodiment 4.

A thermal displacement model learning device, a thermal displacement estimating device, a machining system, and a machining method according to embodiments of the present disclosure will be described below in detail with reference to the drawings.

Embodiment 1.
FIG. 1 is a diagram showing the configuration of a processing system 100-1 according to the first embodiment. A machining system 100 - 1 has a machine tool 1 and a thermal displacement model learning device 2 . The machine tool 1 is a processing device that removes and processes an object to be processed using a tool. The thermal displacement model learning device 2 learns a thermal displacement model for estimating the amount of thermal displacement from input data based on the thermal displacement data, temperature data, and machine state data acquired from the machine tool 1 .

FIG. 2 is a diagram showing the functional configuration of the machine tool 1 according to the first embodiment. A machine tool 1 is equipped with a tool T for machining a work W, and includes a feed shaft 11 that changes the relative position between the tool T and the work W, a feed shaft motor 12 that is an actuator that drives the feed shaft 11, A spindle 13 that rotates a tool T or a workpiece W, a spindle motor 14 that is an actuator that drives the spindle 13, a control device 15 that controls the machine tool 1, and a cooling device for suppressing heat generation when the actuator is driven. , a temperature sensor TS for measuring the temperature of the machine tool 1, and a displacement sensor DS for measuring the relative position between the tool T and the work W.

A displacement sensor DS is installed in the machine tool 1 so as to detect the relative positions of the tool T and the work W. As shown in FIG. The displacement sensor DS measures the relative position and outputs the measured relative position to the thermal displacement model learning device 2 as thermal displacement data. The displacement sensor DS may be a sensor capable of detecting displacement in at least one axial direction, and may be, for example, a triangulation type laser sensor or a TOF (Time Of Flight) type laser sensor. It should be noted that a configuration may be adopted in which a plurality of orthogonal displacement sensors are installed to detect displacements in a plurality of axial directions.

A temperature sensor TS is installed in the machine tool 1 . The temperature sensor TS detects temperature and outputs the detected temperature to the thermal displacement model learning device 2 as temperature data. Although one temperature sensor TS is illustrated in FIG. 2, the processing system 100-1 preferably includes a plurality of temperature sensors TS. It is desirable to install the temperature sensor TS at a place where temperature change is estimated to be large and where it is estimated to have a large effect on thermal displacement. Moreover, it is desirable to install the temperature sensor TS outside the machine tool 1 so that the temperature of the space in which the machine tool 1 is installed can be measured.

FIG. 3 is a diagram showing an example of the installation position of the temperature sensor TS in the machine tool 1 according to the first embodiment. The machine tool 1 shown in FIG. 3 is a milling machine type processing apparatus, and a Y-axis feed shaft motor 12 and a table TA are installed on a bed 17, and a work W is placed on the table TA. In the machine tool 1 , the X-axis and Z-axis feed shaft motors 12 , the main shaft 13 , and the main shaft motor 14 are installed on the column 18 . Black circles in FIG. 3 indicate an example of the installation position of the temperature sensor TS. As described above, the installation location of the temperature sensor TS is preferably a location where temperature change is estimated to be large and a location where thermal displacement is estimated to have a large effect. An example of a location where temperature change is estimated to be large is drive mechanisms such as the feed shaft motor 12 for driving the X, Y, and Z axes and the main shaft motor 14 for rotating the main shaft 13 . An example of a location that is estimated to have a large effect on thermal displacement is the structure of machine tool 1 such as bed 17 , column 18 , and feed screw 19 . The temperature sensors TS are preferably installed at the vertices of the structure, points between vertices, inside the structure, and the like.

Returning to the description of FIG. The peripheral device 16 is a cooling device, and is installed to cool the machine tool 1 to prevent thermal displacement. The peripheral device 16 exchanges heat between the structure of the machine tool 1 , the work W and the tool T, and the outside air outside the machine tool 1 . For example, the peripheral device 16 includes a coolant device that injects cutting oil for cooling the workpiece W, a pump that circulates a coolant that cools the heat exchange pipes installed in the machine tool 1, and the like. The peripheral device 16 outputs operation information of the peripheral device 16 to the control device 15 .

The operation information of the peripheral device 16 includes at least an operating state indicating whether or not the peripheral device 16 is operating. The operation information of the peripheral device 16 may include power consumption, power of the peripheral device 16, etc. in addition to the operating state. Here, the operating state included in the operation information of the peripheral device 16 can be expressed numerically. For example, an operating state of "1" represents "a state in which the peripheral device 16 is operating", and an operating state of "0" represents "a state in which the peripheral device 16 is stopped".

A machining program is a series of commands given to the control device 15 so that the feed shaft 11, the spindle 13 and the peripheral equipment 16 of the machine tool 1 perform desired operations. Specifically, the machining program can include information representing a position command for the feed shaft 11, a speed command for the feed shaft 11, a rotation speed command for the main shaft 13, an operation and stop command for the peripheral device 16, and the like.

The control device 15 reads the machining program and analyzes commands written in the machining program. The control device 15 also generates a command position and a command speed for the feed shaft 11 to cause the feed shaft 11 to operate as desired, and gives an operation amount to the feed shaft motor 12 . The control device 15 also generates a command speed for the main shaft 13 and gives an operation amount to the main shaft motor 14 in order to cause the main shaft 13 to operate as desired.

Furthermore, the control device 15 outputs machine state data representing the operating state of the machine tool 1 to the thermal displacement model learning device 2 . The machine state data is data representing the operating state of the machine tool 1 , and indicates, for example, the operating state of the peripheral device 16 that adjusts the temperature of the machine tool 1 and the operating state of the actuators provided in the machine tool 1 . Specifically, the machine state data includes at least one or more types of information among a commanded position for the feed shaft 11 , a commanded speed for the feed shaft 11 , a commanded speed for the spindle 13 , and operation information of the peripheral device 16 .

Further, when a position detector is installed in the machine tool 1, the control device 15 controls the difference between the actual position detected by the position detector and the command position or the difference between the actual speed and the command speed to be zero. Feedback control may be performed. In this case, the machine state data can include information on one or more of the actual position of the feed shaft 11, the actual speed of the feed shaft 11, and the actual speed of the main shaft 13.

Here, the heat generated inside and outside the machine tool 1 and the transfer of heat will be described. Part of the electrical energy given by the controller 15 to drive the feed shaft motor 12 and the spindle motor 14 becomes heat loss. Furthermore, heat loss occurs due to mechanical friction during operation of the feed shaft 11 and the main shaft 13 . These heat losses have the effect of increasing the temperature of the structure of the machine tool 1, but are also cooled by the peripheral equipment 16 at the same time. Furthermore, processing heat is generated when the tool T and the work W are processed. Furthermore, if there is a temperature difference between the main body of the machine tool 1 and the surrounding environment, the temperature of the structure of the machine tool 1 changes with time according to the temperature difference. Heat generation, heat dissipation, and forced cooling generated inside and outside the machine tool 1 described above cause temperature changes in the structure, resulting in thermal expansion and thermal strain in the structure. Such thermal changes in the structure result in machining errors between the tool T and the work W. FIG. In the present disclosure, machining errors between the tool T and the work W caused by thermal changes in the machine tool 1 are called thermal displacement.

Next, the operating state of the machine tool will be explained. FIG. 4 is an explanatory diagram of an example of changes in the operating state of the machine tool 1. As shown in FIG. FIG. 4 shows changes in operating conditions from when the machine tool 1 is powered on, operates according to the machining program, to when the power is shut off.

FIG. 4 shows an example of transition over time between the operating states and machining program execution states of the spindle 13, the feed shaft 11, and the peripheral device 16, which are components of the machine tool 1. FIG. The horizontal axis in FIG. 4 represents time, and t0 to t9 represent times. Details from time t0 to time t9 will be described later. The machining program execution state indicates whether the machine tool 1 is running the program or stopped. Program operation means that the machine tool 1 operates according to commands written in a machining program. The machine tool 1 performs a programmed operation for a certain machining, and when the programmed operation ends, an operator performs work such as replacement of the workpiece W. FIG. After that, the work is carried out in such a flow that the program operation for the next machining is carried out.

In the example shown in FIG. 4, the machine tool 1 is powered on at time t0 and powered off at time t9. A period from time t0 to time t1 represents a standby state until the operation according to the machining program is started. Also, the period from time t8 to time t9 represents the period from the end of the operation according to the machining program until the power supply of the machine tool 1 is cut off.

Further, as indicated by the machining program execution state, the program operation is performed according to the machining program #1 from time t1 to time t5, and the program operation is performed according to the machining program #2 from time t6 to time t8. When these machining programs are executed, the spindle speed, which is the rotation speed of the spindle 13, the feed speed, which is the speed of the feed shaft 11, and the operating state of the peripheral device 16 change with time.

“Main shaft speed” in FIG. 4 indicates the rotational speed of the main shaft 13 . Here, in FIG. 4, the state in which the main shaft 13 is rotating at 1000 revolutions per minute is indicated by "S1000". As shown by the spindle speed in FIG. 4, the spindle motor 14 rotates at a rotational speed of 1000 rpm from time t1 to time t4, and at a rotational speed of 3000 rpm from time t4 to time t5. From time t6 to time t8, they rotate at a rotational speed of 3000 revolutions per minute. When the main shaft 13 rotates, heat is generated from the structure of the main shaft motor 14 and the main shaft 13 . The higher the spindle speed, the greater the amount of heat generated by the spindle 13 . Since the spindle speed changes according to the instructions of the machining program, the amount of heat generated by the spindle 13 also changes according to the machining program.

“Feeding speed” in FIG. 4 indicates the moving speed of the feed shaft 11 . Here, in FIG. 4, the state in which the feed shaft 11 moves 100 mm per minute is expressed as "F100". As shown by the feed speed in FIG. 4, the feed shaft motor 12 rotates at a rotational speed that drives the feed shaft 11 at 100 mm per minute from time t1 to time t3, and at a rotational speed that drives the feed shaft 11 at 300 mm per minute between time t3 and time t5. 11, and rotates at a rotational speed for operating the 300 mm feed shaft 11 between time t6 and time t8. When the feed shaft 11 operates, heat is generated from the structure of the feed shaft motor 12 and the feed shaft 11 . The higher the feeding speed, the greater the amount of heat generated. Since the feed speed changes according to the instructions of the machining program, the amount of heat generated by the feed shaft 11 also changes according to the machining program.

"Operating state of peripheral device" in FIG. 4 indicates whether the peripheral device 16 for cooling the machine tool 1 is in operation or stopped. The peripheral device 16 does not always operate during machining, but operates and stops according to instructions written in the machining program. 4, the peripheral device 16 is in operation from time t2 to time t5, the peripheral device 16 is stopped from time t5 to time t7, and the peripheral device 16 is stopped from time t7 to time t8. Peripheral device 16 is activated. When the peripheral device 16 operates, heat is transferred from the machine tool 1 to the environment outside the machine tool 1, and the machine tool 1 is cooled. Since the operating state of the peripheral device 16 changes according to the command of the machining program, the cooling state of the peripheral device 16 also changes according to the machining program.

As described with reference to FIG. 4, the heat generated by the feed shaft 11 and the spindle 13 and the state of cooling by the peripheral device 16 are not constant. Change.

In addition, there is a period of time, such as between time t5 and time t6, during which the operation according to the machining program is not performed. Between time t5 and time t6, for example, the machine tool 1 may be stopped due to setup work, such as the operator attaching/detaching the workpiece W or cleaning the inside of the machine tool 1. be. When the machine tool 1 is stopped in this manner, the operating conditions of the spindle 13, the feed shaft 11, and the peripheral equipment 16 are different from those during program operation. show.

As described above, the speed of the feed shaft 11 and the main shaft 13 and the operating state of the peripheral device 16 affect the occurrence of thermal displacement. In the present embodiment, each section in which the operating state of the machine tool 1 is different, such as the section from time t0 to time t1 and the section from time t1 to time t2, is defined as an operation unit. The time length of the operation unit is variable because it changes depending on the timing of the command described in the machining program.

Although FIG. 4 shows an example in which the machine tool 1 operates two different machining programs, the machining program may be executed at least once, and the types of machining programs and the number of executions are not limited to this example. .

Further, the machine tool 1 may be turned off and then turned on again to run the machining program. In this case, a time interval from when the power is turned off until when it is turned on again is defined as one operation unit.

Next, the functional configuration of the thermal displacement model learning device 2 will be described. FIG. 5 is a diagram showing the functional configuration of the thermal displacement model learning device 2 according to the first embodiment. The thermal displacement model learning device 2 uses data output by the machine tool 1 during operation to learn a thermal displacement model for estimating the amount of thermal displacement from input data including at least temperature data.

The thermal displacement model learning device 2 has a learning data acquisition unit 21 , a data set generation unit 22 , a learning unit 23 and a model storage unit 24 .

The learning data acquisition unit 21 acquires thermal displacement data, temperature data, and machine state data output by the machine tool 1, and generates learning data by temporally synchronizing the acquired data. Here, the temperature data and machine condition data are examples of information included in the input data. The input data includes at least temperature data. The learning data is time-series data in which thermal displacement data, temperature data, and machine state data are arranged at regular time intervals. Here, the fixed time period represents a sampling period preset in the learning data acquisition unit 21 . The learning data acquisition unit 21 receives data of time intervals corresponding to at least one driving operation unit from the machine tool 1 and generates learning data. The learning data acquisition unit 21 outputs the generated learning data to the data set generation unit 22 .

The data set generation unit 22 detects a specific time based on the operating state of the machine tool 1, and extracts thermal displacement data, temperature data, and machine state data from the learning data based on the detected time. Then, the extracted data are put together as one set of data sets, and a plurality of sets of data sets are output to the learning unit 23 . The specific time detected by the data set generation unit 22 is hereinafter referred to as an extraction reference time. The extraction reference time will be explained using FIG. The data set generator 22 detects each time from t1 to t8 as the time when the operating state of the machine tool 1 changes. In FIG. 4, the time at which any one of the spindle speed, the feed rate, the change in the operating state of the peripheral device 16, and the start and end of the machining program changes is set as the extraction reference time. is not limited to these examples. For example, the timing of calling and ending subprograms, the timing of replacing the tool T, and the like can also be regarded as changes in the operating state of the machine tool 1 . Further, when the timing of changing the operating state is specified in advance in the machining program, the data set generation unit 22 may be configured to read this information. Note that the data set generation unit 22 may additionally set a new extraction reference time by temporally interpolating the two extraction reference times.

The data set generation unit 22 generates a data set by extracting part of the learning data from the learning data based on the extraction reference time.

FIG. 6 is an explanatory diagram of a first example of a data set generation method. A first example of a method for the data set generation unit 22 to generate a data set from learning data will be described with reference to FIG. FIG. 6 shows data sets #a1 to #an extracted from the learning data when the extraction reference time t1 is the initial reference time for data extraction. The input data #a1 to #an are temperature data and machine condition data with the extraction reference time t1 as the initial time. Assuming that the i-th input data is input data #ai, the time interval of input data #ai is temporally extended backward by a predetermined time extension width T1 from the end of the time interval of input data #ai−1. ing. The time length of the input data #a2 is longer than the time length of the input data #a1 by the time extension width T1. Also, the time length of the input data #a3 is longer than the time length of the input data #a2 by the time extension width T1. That is, when the time width of the input data #a1 is T, the time width of the input data #ai is "T+(i-1)T1". The input data #a1 to #an are the data of the time interval "T+(i−1)T1" starting from the extraction reference time t1. The teacher data #a1 to #an are thermal displacement data at the end times of the input data #a1 to #an, respectively. The data set generator 22 generates a plurality of data sets #a1 to #an from one extraction reference time t1, as shown in FIG. Through the above processing, the dataset generation unit 22 generates datasets such that the time lengths of the extracted time intervals of the plurality of datasets #a1 to #an generated from one extraction reference time t1 are different. be able to. The data set generation unit 22 performs the above process for each of the plurality of extraction reference times included in the learning data. Note that, like data set #an, the end time may exceed t2, which is the next extraction reference time after extraction reference time t1.

FIG. 7 is an explanatory diagram of a second example of the data set generation method. A second example of a method of generating a dataset from learning data by the dataset generator 22 will be described with reference to FIG. FIG. 7 shows data sets #b1 to #bn extracted from the learning data when the extraction reference time t2 is the end reference time for data extraction. The input data #b1 to #bn are temperature data and machine condition data with the extraction reference time t2 as the end time. Assuming that the i-th input data is input data #bi, the time interval of input data #bi is temporally extended backward by a predetermined time extension width T1 from the beginning of the time interval of input data #bi-1. ing. As in the first example, the time width of the input data #bi is "T+(i-1)T1", where T is the time width of the input data #b1. The input data #b1 to #bn are data of a time interval of each time width "T+(i-1)T1" ending at the extraction reference time t2. The teacher data #b1 to #bn are thermal displacement data at the end times of the input data #b1 to #bn, respectively. The data set generator 22 generates a plurality of data sets #b1 to #bn from one extraction reference time t2, as shown in FIG. Through the above processing, the dataset generation unit 22 generates datasets such that the time lengths of the extracted time intervals of the plurality of datasets #b1 to #bn generated from one extraction reference time t2 are different. be able to. The data set generation unit 22 performs the above process for each of the plurality of extraction reference times included in the learning data. Note that the initial time may be earlier than t1, which is the extraction reference time before the extraction reference time t2, as in the data set #bn.

FIG. 8 is an explanatory diagram of a third example of the data set generation method. In a third example, the dataset generation unit 22 generates datasets so that the set of data included in all datasets uniformly includes the entire acquired data. Specifically, the data set generation unit 22 generates a data set by dividing the period from the extraction reference time t1 to the extraction reference time t2 next to the extraction reference time t1 into n equal intervals. n is a natural number. FIG. 8 shows data sets #c1 to #cn extracted from the learning data between the end extraction reference time t2 when the extraction reference time t1 is the initial reference time for data extraction. The input data #c1 is temperature data and machine condition data with the extraction reference time t1 as the initial time and the time t1+(t2-t1)/n as the final time. Input data #c2 is temperature data and machine condition data whose initial time is t1+(t2-t1)/n and whose final time is t1+2×(t2-t1)/n. Input data #ci is temperature data and machine condition data whose initial time is time t1+(i-1)*(t2-t1)/n and whose end time is time t1+i*(t2-t1)/n. The teacher data #c1 to #cn are thermal displacement data at the end times of the input data #c1 to #cn, respectively. Through the above processing, the dataset generation unit 22 can generate a dataset so that the input data uniformly includes the period from the extraction reference time t1 to the time t2 without duplication. The data set generation unit 22 executes the above process for each of the plurality of extraction reference times included in the learning data. Note that, in the third example, the time lengths of the extracted time intervals are uniform for the data sets #c1 to #cn generated between the extraction reference times t1 and t2. However, as shown in FIG. 4, the time lengths between adjacent extraction reference times are not uniform, and the time lengths of the time intervals extracted when the data set generation unit 22 generates data sets are equal to two adjacent time intervals. It is determined depending on the length of time between extraction reference times. Therefore, by performing the above processing for each of all extraction reference times, a plurality of data sets with different time lengths of the extracted time intervals are generated as a whole.

In the first to third examples of the above data set generation method, teacher data #a1 to #an, #b1 to #bn, #c1 to #cn are input data #a1 to #an, #b1 Although the thermal displacement data for one hour corresponding to the end times of ~#bn and #c1 to #cn are used, the time length of the teacher data is not limited to one hour, and may be time-series data. For example, teacher data #a1-#an, #b1-#bn, #c1-#cn are time series having the same time length as input data #a1-#an, #b1-#bn, #c1-#cn. It may be data.

In the first example of the above data set generation method, time interval #a1 corresponding to input data #a1 is selected n times from input data #a1 to input data #an and adopted as input data. . On the other hand, the number of times the extension section #a2 is used as input data is (n-1) times, and the number of times the extension section #an is used as input data is only one time. Similarly, in the second example, time interval #b1 corresponding to input data #b1 is selected n times from input data #b1 to input data #bn and used as input data. On the other hand, the number of times that the extended section #b2 is used as input data is (n-1) times, and the number of times that the extended section #bn is used as input data is only one time. In other words, a plurality of pieces of input data generated using one extraction reference time as a temporal reference includes sections with a large number of extractions and sections with a small number of extractions. When generating a data set from each extraction reference time, the data set generation unit 22 equalizes the number of extractions of each time interval and extension interval over the entire learning data. Specifically, when all sets of data sets generated based on each extraction reference time are combined, if the variation in the number of extractions in the time interval is greater than or equal to the allowable value, the data set generation unit 22 performs histogram equalization. Execute the process. Here, the histogram equalization process can adopt a known algorithm. Through such processing, the data set generation unit 22 can equalize the bias in the number of data extractions over the entire learning data. As a result, the entire set of data sets generated by the data set generation unit 22 does not extract only a specific section from the learning data, and a plurality of data that uniformly reflects the driving state included in the learning data. set can be generated. Note that although the equalization of the number of extractions focused on the input data has been described here, the number of extractions may be equalized by focusing on the teacher data.

As described above, in the first to third examples of the data set generation method, the time length of the extracted time interval differs based on the extraction reference time at which the operating state of the machine tool 1 changes. Multiple datasets can be generated. Further, in the third example, the data sets are generated so that the set of data included in all the data sets uniformly includes the entire acquired data. As a result, when the learning unit 23, which will be described below, learns the thermal displacement model, the entire learning data can be uniformly targeted for learning, and over-learning in which the model matches only part of the data can be prevented. is suppressed, and the generalization performance of the thermal displacement model is further improved.

In addition, in the third example of the above-described data set generation method, an example was given in which the interval between two adjacent extraction reference times was divided into n equal intervals. The same effect as in the third example can be obtained by dividing the input data so that there is no duplication of input data with a time length that is not constant.

Next, the learning unit 23 will be explained. FIG. 9 is an explanatory diagram of the learning method of the thermal displacement model. The learning unit 23 has a function of learning a thermal displacement model for estimating the amount of thermal displacement from the input data using the input data and teacher data included in the dataset generated by the dataset generating unit 22 . The learning unit 23 internally has a thermal displacement model that outputs the amount of thermal displacement in time series when the input data included in the data set is input in time series. The learning unit 23 optimizes the internal parameters of the thermal displacement model so that the difference between the estimated value of the thermal displacement amount, which is the output of the thermal displacement model, and the teacher data included in the data set is minimized. As a result, the learning unit 23 identifies a thermal displacement model capable of estimating the amount of thermal displacement from the input data, and outputs it as a learned thermal displacement model.

The thermal displacement model learning device 2 is characterized in that the data set output by the data set generator 22 has a variable length. That is, in at least two data sets out of the plurality of data sets output by the data set generator 22, the time lengths of the extracted time intervals are different. Therefore, the learning unit 23 uses a thermal displacement model corresponding to variable-length input data. The learning unit 23 learns the amount of thermal displacement by, for example, so-called supervised learning according to a neural network. Here, supervised learning refers to a method in which learning data, which is a set of input and result data, is given to a learning device to learn features in the learning data and infer the result from the input. Here, an example in which the learning unit 23 learns a thermal displacement model using an RNN (Recurrent Neural Network) will be described.

A RNN is a neural network having a structure in which the output of an intermediate layer is recursively input to the layer that output the result. The learning unit 23 can use, for example, a neural network having the structure shown in FIG. FIG. 10 is an explanatory diagram of a neural network. The neural network shown in FIG. 10 is an RNN whose input layer consists of n inputs from x1 to xn, whose hidden layer consists of m nodes from h1 to hm, and whose output layer consists of one output of y1. . Here, the upper diagram in FIG. 10 shows the state of the neural network at time t−1, and the lower diagram in FIG. 10 shows the state of the neural network at time t. The hidden layer at time t outputs to the output layer the weighted sum of the value calculated from the input to the input layer at time t and the term calculated from the value input from the hidden layer at time t−1. Here, let _xt be the vector that bundles the input layers x1 to xn at time t, and _ht be the vector that bundles the intermediate _layers h1 to hm at time t. be able to. In equation (1), if the dimension of vector _xt is n and the dimension of vector _ht is m, W is an n×m-dimensional linear transformation matrix. Also, R denotes an M×M-dimensional linear transformation matrix, b denotes a bias vector, and g denotes an activation function.

_ht =g( _Wxt +Rht _-1 +b) (1)

The learning unit 23 uses the diachronic error backpropagation method to adjust the weighting of the neural network so that the amount of thermal displacement, which is the output, matches the thermal displacement data.

Although FIG. 10 shows an example of a neural network composed of an n-dimensional input layer, an m-dimensional intermediate layer, and a one-dimensional output layer, the dimensions of each layer are not limited to this example and can be any dimension. can be done. Also, although FIG. 10 shows an example of a neural network having one intermediate layer, the number of intermediate layers may be plural.

In addition, in order to improve the generalization performance of the thermal displacement model, the learning unit 23 employs methods such as “dropout”, which randomly excludes neurons during learning, and “early stopping”, which monitors errors and terminates learning early. may be used.

In the above, an example using Elman net among RNNs was explained, but other known RNNs such as LSTM (Long Short Term Memory) or GRU (Gated Recurrent Unit) are used to build and learn a thermal displacement model. may be performed. As another method, the learning unit 23 may perform learning by regression analysis using an autoregressive model.

FIG. 11 is a flow chart for explaining the operation of the thermal displacement model learning device 2 shown in FIG. The thermal displacement model learning device 2 can generate a learned thermal displacement model by executing the processing shown in FIG. 11 .

The learning data acquisition unit 21 of the thermal displacement model learning device 2 acquires thermal displacement data, temperature data, and machine state data from the machine tool 1 (step S101). The learning data acquisition unit 21 associates the acquired thermal displacement data, temperature data, and machine state data with time to generate learning data (step S102).

The dataset generation unit 22 generates a plurality of datasets from the learning data (step S103). The learning unit 23 uses the data set to optimize the internal parameters of the thermal displacement model so that the amount of thermal displacement that is output when the input data included in the data set is input matches the teacher data included in the data set. By doing so, the thermal displacement model is learned (step S104).

As described above, according to Embodiment 1, the dataset generation unit 22 generates multiple types of variable-length datasets, and extracts each other in at least two datasets among the plurality of datasets. A thermal displacement model is trained using data sets with different time lengths for time intervals. As a result, the thermal displacement model learning device 2 can learn a thermal displacement model using a data set generated using a time interval having an appropriate length of time for thermal displacement occurring in the machine tool 1. , it becomes possible to improve the estimation accuracy of the amount of thermal displacement. Here, the thermal displacement model learning device 2 can improve the estimation accuracy of the thermal displacement amount without depending on the operation duration time of the machine tool 1 .

The machine tool 1 operates according to a machining program whose time length is variable according to the finished shape of various workpieces W and whose operation pattern is different. In order to generate a thermal displacement model capable of accurately estimating the amount of thermal displacement, the data set used for learning must be obtained from the machining program and the length of time must be variable.

In Embodiment 1, the data set used by the learning unit 23 for learning the thermal displacement model is generated based on the time when the operating state of the machine tool 1 operated based on the machining program changes. Therefore, the thermal displacement model learning device 2 according to the first embodiment can learn the thermal displacement model using a data set based on the temporal changes in the operating state of the machine tool 1 during actual machining. Therefore, the thermal displacement model can learn the input/output relationship reflecting the operating state of the machine tool.

Further, when using the first example and the second example of the above-described data set generation method, the data set generation unit 22 generates a time that is a multiple of the predetermined time extension width T1 from the time defined as the extraction reference time. A data set can be generated by extracting data with a width from the training data.

Furthermore, since the thermal displacement model is a temporally recursive model, in addition to the input data that was input at the time when the estimated value of the thermal displacement was calculated, information included in the input data past that time Learning can be performed so as to calculate the amount of thermal displacement that is also affected by Therefore, the thermal displacement model learning device 2 can learn not only the instantaneous temperature data and machine state data for calculating the thermal displacement, but also the effects of changes in these data over time.

Embodiment 2.
FIG. 12 is a diagram showing the configuration of a processing system 100-2 according to the second embodiment. A machining system 100-2 according to the second embodiment includes a machine tool 1A, a thermal displacement model learning device 2, and a thermal displacement estimating device 3. The machining system 100-2 has a thermal displacement estimating device 3 in addition to the configuration of the machining system 100-1 according to the first embodiment, and has a machine tool 1A instead of the machine tool 1 of the machining system 100-1. .

The processing performed by the thermal displacement model learning device 2 is the same as that of the first embodiment except that the learned thermal displacement model is output to the thermal displacement estimation device 3, so the explanation is omitted here. The thermal displacement estimating device 3 acquires a learned thermal displacement model generated by the thermal displacement model learning device 2 and input data for inference including at least temperature data representing the time-series temperature of the inference target machine tool 1A. do. The thermal displacement estimating device 3 uses the thermal displacement model to generate an estimated value of the amount of thermal displacement of the inference target machine tool 1A from the input data. The thermal displacement estimating device 3 outputs the estimated value of the amount of thermal displacement to the machine tool 1A.

In addition to the functions of the machine tool 1, the machine tool 1A has a function of outputting temperature data and machine state data to the thermal displacement estimating device 3, and based on the estimated value of the amount of thermal displacement output by the thermal displacement estimating device 3, and a function of correcting thermal displacement.

FIG. 13 is a diagram showing a functional configuration of machine tool 1A according to the second embodiment. The configuration of the machine tool 1A is the same as that of the machine tool 1 except that it has a control device 15A instead of the control device 15 of the machine tool 1. FIG. The control device 15A corrects the operation amount of at least one of the main shaft motor 14 that drives the main shaft 13 and the feed shaft motor 12 that drives the feed shaft 11 based on the estimated value of the thermal displacement amount output by the thermal displacement estimating device 3. to compensate for thermal displacement. The machine tool 1A performs removal machining using the corrected operation amount.

FIG. 14 is a diagram showing the functional configuration of the thermal displacement estimation device 3 according to the second embodiment. The thermal displacement estimation device 3 has an inference data acquisition unit 31 , a thermal displacement model acquisition unit 32 , and an inference unit 33 .

The inference data acquisition unit 31 acquires the temperature data and the machine state data from the inference target machine tool 1A, and provides the inference unit 33 with the temperature data and the machine state data that are time-synchronized as data for inference. Output. The inference data includes at least temperature data, and is generated so that the learned thermal displacement model will have the same combination of data as the input data used during learning. Further, the inference data is sequentially output in the same time cycle as the learning data generated by the thermal displacement model learning device 2 .

The thermal displacement model acquisition unit 32 acquires a learned thermal displacement model used for inference from the model storage unit 24 of the thermal displacement model learning device 2 . The thermal displacement model acquisition unit 32 outputs the acquired thermal displacement model to the inference unit 33 .

The inference unit 33 uses the learned thermal displacement model to output an estimated value of the amount of thermal displacement from the input data included in the inference data. The inference unit 33 outputs the estimated value of the amount of thermal displacement to the machine tool 1A. The inference unit 33 inputs the machine state data included in the inference data into the learned thermal displacement model with the same data structure and time period as during learning, so that the thermal displacement amount is calculated with the same time period as the input. An estimate can be calculated. The inference unit 33 outputs the calculated estimated value of the amount of thermal displacement to the controller 15A of the machine tool 1A at the same period as the input interval of the inference data.

The controller 15A of the machine tool 1A receives the estimated value of the amount of thermal displacement output by the inference unit 33, and corrects, for example, the position command of the feed shaft 11 so as to offset the thermal displacement indicated by the estimated value of the amount of thermal displacement. do. Then, the feed shaft motor 12, which is the actuator of the feed shaft 11, is operated using the corrected position command. Correction of the position command is updated at each inference data input interval.

FIG. 15 is a flow chart for explaining the operation of the processing system 100-2 shown in FIG. The thermal displacement model learning device 2 of the machining system 100-2 performs thermal displacement model generation processing (step S121). Note that the processing of step S121 corresponds to the processing of steps S101 to S104 shown in FIG. These processes are performed prior to performing processing for correcting thermal displacement.

The machine tool 1A starts program operation of machining for which thermal displacement is to be corrected (step S122). The processing after step S123 is performed during the processing of the target for thermal displacement correction.

The inference data acquisition unit 31 receives temperature data and machine state data that are sequentially output during machining of the machine tool 1A, and generates inference data from the received data (step S123). The inference data has the same data structure as the input data of the thermal displacement model generated in step S121. The generated inference data is sequentially output to the inference section 33 in the same time cycle as the learning data generated by the thermal displacement model learning device 2 .

The inference unit 33 uses the thermal displacement model to estimate the amount of thermal displacement from the inference data (step S124). The estimated value of the amount of thermal displacement is calculated in the same time period as the inference data generated in step S123, and output to the controller 15A of the machine tool 1A.

The control device 15A of the machine tool 1A uses the estimated value of the amount of thermal displacement to correct the operation amount of the actuator, thereby correcting the position command and performing machining (step S125). The machine tool 1A determines whether or not the programmed operation of machining for which thermal displacement is to be corrected has ended (step S126). If the program operation has not ended (step S126: No), the processing system 100-2 returns to step S123 and repeats the processing from step S123 to step S126. If the program operation has ended (step S126: Yes), the processing system 100-2 ends the process.

As described above, the machining system 100-2 according to the second embodiment has the thermal displacement estimating device 3 that estimates the amount of thermal displacement using a learned thermal displacement model. Based on the estimated value of the amount of thermal displacement, machine tool 1A corrects the operation amount of the actuator of machine tool 1A so as to offset the thermal displacement indicated by the estimated value.

In addition, the inference unit 33 can sequentially calculate an estimated value of the amount of thermal displacement for the sequentially input inference data, and output the estimated value to the machine tool 1A. Therefore, the amount of thermal displacement can be estimated no matter how long the machining program executed by the machine tool 1A takes. Furthermore, the estimated value of thermal displacement is calculated from the thermal displacement model, which is a temporally recursive model. It is possible to estimate the amount of thermal displacement including the influence of changes in the data over time. Therefore, the machine tool 1A can accurately correct the thermal displacement without depending on the elapsed time from the start of operation.

As another method, in step S126 shown in FIG. 15, it is determined whether or not the power to machine tool 1A is shut off, and machining system 100-2 continues to operate until power to machine tool 1A is shut off in step S123. You may return to and continue the process. In this case, it is possible to continue correcting the thermal displacement even when the program operation is continuously performed a plurality of times.

Here, the hardware configuration of the thermal displacement model learning device 2 will be described. The thermal displacement model learning device 2 of Embodiments 1 and 2 is realized by, for example, a computer system. The thermal displacement model learning device 2 may be realized by one computer system or may be realized by a plurality of computer systems. For example, the thermal displacement model learning device 2 may be realized by a cloud system. In the cloud system, it is possible to arbitrarily set the separation between the hardware of the computer system and the devices such as servers for each function. For example, one computer system may function as multiple devices, or multiple computer systems may function as one device.

A configuration example of a computer system that implements the thermal displacement model learning device 2 will be described. FIG. 16 is a diagram illustrating an example of a hardware configuration; As shown in FIG. 16, this computer system comprises a control section 101, an input section 102, a storage section 103, a display section 104, a communication section 105 and an output section 106, which are connected via a system bus 107. there is

In FIG. 16, the control unit 101 is, for example, a CPU (Central Processing Unit) or the like. The control unit 101 executes a thermal displacement model learning program in which each process performed by the thermal displacement model learning device 2 of the present embodiment is described. The input unit 102 is composed of, for example, a keyboard, mouse, etc., and is used by the user of the computer system to input various information. The storage unit 103 includes various memories such as RAM (Random Access Memory) and ROM (Read Only Memory) and storage devices such as hard disks, and stores programs to be executed by the control unit 101 and necessary information obtained in the course of processing. Stores data, etc. The storage unit 103 is also used as a temporary storage area for programs. The display unit 104 includes an LCD (Liquid Crystal Display) or the like, and displays various screens to the user of the computer system. The communication unit 105 is a communication circuit or the like that performs communication processing. The communication unit 105 may be composed of a plurality of communication circuits respectively corresponding to a plurality of communication schemes. The output unit 106 is an output interface that outputs data to an external device such as a printer or an external storage device.

Note that FIG. 16 is an example, and the configuration of the computer system is not limited to the example of FIG. For example, the computer system may not have output 106 . Moreover, when the thermal displacement model learning device 2 is implemented by a plurality of computer systems, all of these computer systems may not be the computer systems shown in FIG. For example, some computer systems may not include at least one of display 104, output 106, and input 102 shown in FIG.

Here, an example of the operation of the computer system until the thermal displacement model learning program in which the processing of the thermal displacement model learning device 2 of the first and second embodiments is described is ready for execution will be described. The computer system having the above-described configuration includes, for example, a CD (Compact Disc)-ROM drive or a DVD (Digital Versatile Disc)-ROM drive (not shown) from a CD-ROM or DVD-ROM set to a thermal displacement model learning program. is installed in the storage unit 103 . Then, when the thermal displacement model learning program is executed, the thermal displacement model learning program read out from the storage unit 103 is stored in a main storage area of the storage unit 103 . In this state, the control unit 101 executes processing as the thermal displacement model learning device 2 of the first and second embodiments according to the thermal displacement model learning program stored in the storage unit 103. FIG.

In the above description, a program describing the processing in the thermal displacement model learning device 2 is provided using a CD-ROM or DVD-ROM as a recording medium. For example, a program provided via a transmission medium such as the Internet via the communication unit 105 may be used depending on the capacity of the program to be used.

In the thermal displacement model learning program of the first and second embodiments, input data including at least temperature data representing the time-series temperature of the machine tool 1 and time-series thermal displacement amount of the machine tool 1 are stored in the computer. a step of acquiring learning data in which the represented thermal displacement data is associated with time; a step of generating a data set by extracting a part of the learning data from the learning data; and using the data set, the machine tool 1A and generating a learned thermal displacement model for estimating the amount of thermal displacement from the input data. Here, in the step of generating the datasets, the datasets are generated such that the time lengths of the extracted time intervals are different in at least two datasets among the plurality of datasets.

The learning data acquisition unit 21 shown in FIG. 5 is realized by the communication unit 105 shown in FIG. 16, and the model storage unit 24 shown in FIG. 5 is a part of the storage unit 103 shown in FIG. 5 are implemented by the control unit 101 shown in FIG.

As with the thermal displacement model learning device 2, the thermal displacement estimating device 3 is also realized by one or more computer systems. A program for the thermal displacement estimating device 3 to perform the operation described in the second embodiment is provided by a storage medium, a transmission medium, etc., and installed in a computer system, like the thermal displacement model learning program described above. As a result, the above-described operation of the thermal displacement estimating device 3 is realized.

The inference data acquisition unit 31 and the thermal displacement model acquisition unit 32 shown in FIG. 14 are implemented by the communication unit 105 shown in FIG. 16, and the inference unit 33 shown in FIG. Realized.

5 and 14 is an example, and if the processing systems 100-1 and 100-2 can perform the operations described above, the division of functions in each device is shown in FIG. and is not limited to the example shown in FIG. For example, as shown as a modified example below, the thermal displacement model learning device 2 and the thermal displacement estimating device 3 may be integrated so that the thermal displacement model learning device 2 also functions as the thermal displacement estimating device 3. good.

FIG. 17 is a diagram showing a functional configuration of a thermal displacement model learning device 2A according to a modification of the second embodiment. The thermal displacement model learning device 2A has the function of the thermal displacement estimation device 3 in addition to the function of the thermal displacement model learning device 2 . Since the function of each component has already been explained in FIG. 14, the explanation is omitted here. In the configuration shown in FIG. 17, when the functions of the thermal displacement model learning device 2A are realized by one computer system, the functions of the thermal displacement model acquisition unit 32 are realized by exchanging data inside the computer system.

Embodiment 3.
FIG. 18 is a diagram showing the configuration of a processing system 100-3 according to the third embodiment. A machining system 100-3 has a plurality of machine tools 1 and a thermal displacement model learning device 2B provided on a server. The thermal displacement model learning device 2B learns a thermal displacement model based on the thermal displacement data, temperature data, and machine state data received from the machine tools 1 . The function of the machine tool 1 is the same as that of the first embodiment, and the function of the thermal displacement model learning device 2B is the same as that of the first embodiment except that the thermal displacement models of a plurality of machine tools 1 are learned for each machine tool 1. is similar to

Embodiment 4.
FIG. 19 is a diagram showing the configuration of a processing system 100-4 according to the fourth embodiment. A machining system 100-4 has a plurality of machine tools 1, 1A, a thermal displacement model learning device 2B provided on a server, and a thermal displacement estimating device 3B.

The machine tool 1 is a processing device to be learned, and the feature information, which is the information indicating the features of the machine tool 1 including at least the model number of the machine tool 1, is output to the thermal displacement model learning device 2B. 1 is the same as the machine tool 1 of 1. Machine tool 1A is a processing device to be estimated, and is the same as in Embodiment 2, except that feature information, which is information indicating features of machine tool 1A including at least the model number of machine tool 1A, is output to thermal displacement estimating device 3B. It is the same as machine tool 1A.

The thermal displacement model learning device 2B is the same as the thermal displacement model learning device 2B of Embodiment 3 except that it receives the feature information from the machine tool 1, and generates learning data based on data received from a plurality of machine tools 1. and a model storage unit 24B that stores each of the plurality of thermal displacement models in association with the characteristic information of the machine tool 1 from which the learning data is obtained. The thermal displacement estimating device 3B has a thermal displacement model obtaining unit 32B instead of the thermal displacement model obtaining unit 32 of the thermal displacement estimating device 3B. The thermal displacement model acquisition unit 32B selects a thermal displacement model to be used by the inference unit 33 from among the plurality of thermal displacement models stored in the model storage unit 24B, based on the feature information received from the inference target machine tool 1A. It has a model selection unit 321 for selection.

Here, the feature information includes at least the model number, and in addition to the model number, can include information indicating the state of the machine tools 1, 1A, such as the environment of the factory where the machine tools 1, 1A are installed, installation conditions, and the like. The model selection unit 321 is a machine tool 1 having the same model number as the inference target machine tool 1A, and is a thermal displacement model generated based on data acquired from a machine tool 1 similar to the state of the inference target machine tool 1A. can be selected. As a result, even if there is no thermal displacement model generated from the data of the machine tool 1A itself to be inferred, the machine tool 1 having the same model number as the machine tool 1A and having a similar operating environment can be used. Using the thermal displacement model generated from the data of 1, it is possible to obtain an estimate of the amount of thermal displacement.

In the above description, the configuration in which the thermal displacement model acquisition unit 32B includes the model selection unit 321 has been described. Alternatively, an estimated value to be output may be selected based on characteristic information from among a plurality of estimated values of thermal displacement calculated using a plurality of types of thermal displacement models.

The configurations shown in the above embodiments are only examples, and can be combined with other known techniques, or can be combined with other embodiments, without departing from the scope of the invention. It is also possible to omit or change part of the configuration.

For example, the functions of the thermal displacement model learning device 2 and the thermal displacement estimation device 3 may be built into the machine tool 1 . When learning a thermal displacement model using data acquired from a plurality of machine tools 1 as in Embodiments 3 and 4, the thermal displacement model learning device 2B is used in the same area. Learning data may be acquired from a plurality of machine tools 1, or a thermal displacement model may be learned using learning data collected from a plurality of machine tools 1 operating independently in different areas. . It is also possible to add or remove the machine tool 1 from which data for learning is collected on the way. Furthermore, the thermal displacement model learning device 2 that has learned the thermal displacement model for a certain machine tool 1 is applied to a different machine tool 1, and the thermal displacement model for the different machine tool 1 is re-learned and updated. You may make it

1, 1A machine tool, 2, 2A, 2B thermal displacement model learning device, 3, 3B thermal displacement estimation device, 11 feed shaft, 12 feed shaft motor, 13 main shaft, 14 main shaft motor, 15, 15A control device, 16 peripheral device , 17 bed, 18 column, 19 feed screw, 21, 21B learning data acquisition unit, 22 data set generation unit, 23 learning unit, 24, 24B model storage unit, 31 inference data acquisition unit, 32, 32B thermal displacement model acquisition unit, 33 inference unit, 100-1, 100-2, 100-3, 100-4 processing system, 101 control unit, 102 input unit, 103 storage unit, 104 display unit, 105 communication unit, 106 output unit, 321 Model selector, DS Displacement sensor, T Tool, TA Table, TS Temperature sensor, W Work.

Claims (14)

Learning data for obtaining learning data in which input data including at least temperature data representing a time-series temperature of a machine tool and thermal displacement data representing a time-series thermal displacement amount of the machine tool are associated with each other for each time. an acquisition unit;
a dataset generation unit that generates a dataset obtained by extracting a part of the learning data from the learning data;
a learning unit that uses the data set to generate a learned thermal displacement model for estimating the amount of thermal displacement from the input data of the machine tool;
with
The data set generation unit generates the data sets such that at least two of the plurality of data sets have the same initial time or end time and different time lengths. Thermal displacement model learning device. 2. The thermal displacement model learning device according to claim 1, wherein said input data further includes machine state data representing an operating state of said machine tool in addition to said temperature data. The learning unit includes an estimated value of a thermal displacement amount that is an output of a thermal displacement model when the input data included in the data set is input, and the thermal displacement corresponding to the input data input to the thermal displacement model. The thermal displacement model learning device according to claim 1, wherein the learned thermal displacement model is generated by optimizing internal parameters of the thermal displacement model so as to minimize errors with data. . an inference data acquisition unit that acquires inference data, which is inference input data including at least temperature data representing time-series temperatures of the machine tool;
an inference unit that uses the thermal displacement model to output an estimated value of the amount of thermal displacement of the machine tool from the inference data input from the inference data acquisition unit;
The thermal displacement model learning device according to any one of claims 1 to 3, further comprising: 5. The thermal displacement model learning device according to claim 4, wherein the inference data further includes machine state data representing an operating state of the machine tool in addition to the temperature data. 3. The thermal displacement model learning device according to claim 2 , wherein the machine state data indicates an operating state of a peripheral device that adjusts the temperature of the machine tool and an operating state of an actuator provided in the machine tool. 6. The thermal displacement model learning device according to claim 5 , wherein the machine state data indicates an operating state of a peripheral device that adjusts the temperature of the machine tool and an operating state of an actuator provided in the machine tool. The learning data acquisition unit acquires the learning data from the plurality of machine tools,
The thermal displacement model learning device according to any one of claims 1 to 3, wherein the learning unit generates the thermal displacement model for each machine tool. The learning data acquisition unit acquires the learning data from the plurality of machine tools,
The learning unit generates the thermal displacement model for each of the machine tools, and the generated thermal displacement model and the machine tool from which the learning data used to generate the thermal displacement model are acquired. 4. The thermal displacement model learning device according to any one of claims 1 to 3, wherein feature information, which is information indicating features of the machine tool including at least a model number, is stored in association with the feature information. an inference data acquisition unit for acquiring inference data, which is inference input data including at least temperature data representing time-series temperatures of an inference target machine tool;
Using the thermal displacement model having the same model number as the machine tool to be inferred among the plurality of thermal displacement models, the thermal displacement of the machine tool to be inferred from the inference data input from the data acquisition unit for inference. an inference unit that outputs an estimate of the amount of displacement;
The thermal displacement model learning device according to claim 9 , further comprising: The characteristic information further includes information indicating operating conditions of the machine tool in addition to the model number,
The inference unit is a machine tool having the same model number as the machine tool to be inferred, and is generated based on the learning data acquired from a machine tool whose operating conditions are similar to the machine tool to be inferred. 11. The thermal displacement model learning device according to claim 10 , wherein the thermal displacement model is used to output an estimated value of the amount of thermal displacement of the inference target machine tool. A thermal displacement model acquisition unit that acquires the thermal displacement model that has been learned by the thermal displacement model learning device according to claim 1;
an inference data acquisition unit that acquires estimation data, which is inference input data including at least temperature data representing the time-series temperatures of the machine tool;
an inference unit that outputs an estimated value of the amount of thermal displacement from the input data input from the inference data acquisition unit, using the thermal displacement model acquired by the thermal displacement model acquisition unit;
A thermal displacement estimating device comprising: The thermal displacement model learning device according to claim 1 or the thermal displacement estimation device according to claim 12 ,
the machine tool;
with
The machine tool is
a tool;
a spindle that rotates the tool or the workpiece;
a feed axis that changes the relative position between the tool and the workpiece;
with
A machining system, wherein the tool removes and processes the workpiece. A work for removing and machining the workpiece with the tool, comprising a tool, a spindle that rotates the tool or the workpiece, and a feed shaft that changes the relative position between the tool and the workpiece. In the processing method using a machine,
a step of acquiring learning data in which input data including at least temperature data representing a time-series temperature of the machine tool and thermal displacement data representing a time-series thermal displacement amount of the machine tool are associated with each other for each time; ,
generating a data set by extracting a part of the learning data from the learning data;
using the data set to generate a trained thermal displacement model for estimating the amount of thermal displacement from the input data of the machine tool;
estimating the amount of thermal displacement from input data for inference including at least the temperature data, using the learned thermal displacement model;
correcting the operation amount of at least one of an actuator that drives the spindle and an actuator that drives the feed shaft of the machine tool using the estimated value of the thermal displacement;
a step of performing the removal processing using the corrected manipulated variable;
including
In the step of generating the data sets, the data sets are generated such that at least two of the plurality of data sets have the same initial time or end time and different lengths of time. and processing method.

Images