Temperature interpolation device

Abstract

The present invention provides a temperature interpolation device that can appropriately interpolate temperature data during a period when temperature data is missing during the use of a machine tool. The temperature interpolation device comprises: a temperature data reading and writing unit that continuously records time information and temperature data in a non-volatile storage medium during the use of an object device whose temperature data is to be read, and compares the time information recorded in the non-volatile storage medium at a predetermined interval to determine whether data is missing within the predetermined time interval; and a past temperature estimation unit that uses the data recorded in the non-volatile storage medium and the calculation parameters recorded in the calculation parameter setting unit to estimate and interpolate the missing data using a temperature interpolation formula when the temperature data reading and writing unit determines that data is missing.

Description

Technical Field

[0001] The invention relates to a temperature interpolation device. Background Art

[0002] Conventionally, a method for correcting thermal displacement is known (for example, see Patent Document 2). Patent Document 2 states the following: "The method for correcting thermal displacement of a machine tool according to the present invention is characterized in that a feed instruction of an NC device is corrected using a correction value, thereby correcting the tool position during workpiece processing, wherein the correction value is calculated based on a historical value of measurement data of a machine temperature and a delay value of thermal movement."

[0003] In addition, a thermal displacement correction method for a tool is known (for example, see Patent Document 3). Patent Document 3 states the following: "The thermal displacement correction method for a tool according to the invention of claim 1 is characterized in that a first machine tool temperature near a spindle bearing and a second machine tool temperature at a location with high thermal stability of the machine tool are measured, the thermal displacement amount of the tool is estimated based on the difference between the first machine tool temperature and the second machine tool temperature according to a predetermined first-order delay equation, and the tool position is corrected using the estimated correction amount."

[0004] However, in the above-described conventional technology, when temperature data is lost due to a power outage, a network failure, or the like, it is desired to be able to refer to the lost past temperature data.

[0005] On the other hand, a method for interpolating missing data is known (for example, see Patent Document 1). Patent Document 1 states the following: "As a countermeasure to prevent a decrease in the estimation accuracy of the degradation state, when a missing record d occurs in the temperature history data D1, it is also considered to use the temperature Tb of the secondary battery 110 in the normal period before and after the missing period to interpolate the temperature Tb during the missing period. There are various methods for interpolating data. Figure 4 An example of linear interpolation is shown in .

[0006] Prior art literature

[0007] Patent Literature

[0008] Patent Document 1: Japanese Patent Application Publication No. 2018-120785

[0009] Patent Document 2: Japanese Patent Application Publication No. 2017-144527

[0010] Patent Document 3: Japanese Patent Application Publication No. 2004-148443 Summary of the invention

[0011] Problem that the invention aims to solve

[0012] A temperature interpolation device is desired that can appropriately interpolate temperature data during a period in which temperature data is missing during use of a machine tool.

[0013] Solutions for solving problems

[0014] (1) One embodiment of the present invention relates to a temperature interpolation device, comprising: a calculation parameter setting unit, which records calculation parameters of a temperature interpolation formula, wherein the temperature interpolation formula is pre-selected in order to interpolate missing temperature data; a temperature data read / write unit, which continuously records time information and temperature data in a non-volatile storage medium during use of an object device whose temperature data is to be read, and compares the time information recorded in the non-volatile storage medium at a predetermined time interval to determine whether data is missing within a predetermined time interval; and a past temperature estimation unit, which, when the temperature data read / write unit determines that data is missing, uses the data recorded in the non-volatile storage medium and the calculation parameters recorded in the calculation parameter setting unit to estimate and interpolate the missing data using the temperature interpolation formula.

[0015] Effects of the Invention

[0016] According to one embodiment, it is possible to provide a temperature interpolation device that can appropriately interpolate temperature data during a period in which temperature data is missing during use of a machine tool. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a diagram showing the structure of the first embodiment.

[0018] Figure 2 It is a block diagram showing the temperature interpolation device, the thermal displacement correction device, and the numerical control device according to the first embodiment.

[0019] Figure 3 This is a diagram for explaining the missing of temperature data to be interpolated in the first embodiment.

[0020] Figure 4 This is a diagram for explaining how acquired temperature data is recorded in the nonvolatile memory in the first embodiment.

[0021] Figure 5 This is a diagram for explaining the determination of whether or not temperature data is missing in the first embodiment.

[0022] Figure 6 It is a diagram for explaining the estimation and interpolation of missing data in the first embodiment.

[0023] Figure 7 It is a diagram for explaining the thermal displacement correction based on the temperature data recorded in the nonvolatile memory, which is performed by the correction displacement amount calculation unit in the first embodiment.

[0024] Figure 8 This is a diagram for explaining that the time constant varies depending on the position where the temperature sensor is mounted in the second embodiment.

[0025] Fig. 9 This is a graph for explaining the difference in temperature rise and temperature drop due to the difference in time constant in the second embodiment.

[0026] Fig.10 This is a diagram for explaining calculation of parameters of the temperature rise curve in the second embodiment.

[0027] Description of Reference Numerals

[0028] 61: non-volatile memory (non-volatile storage medium); 100: temperature interpolation device; 102: temperature data reading and writing unit; 103: past temperature estimation unit; 104: calculation parameter setting unit. DETAILED DESCRIPTION

[0029] Below, refer to Figures 1 to 7 A temperature interpolation device 100 according to a first embodiment of the present invention will be described.

[0030] Figure 1 1 is a diagram showing the entire structure including the machine tool 1 . Figure 2 is a block diagram showing the temperature interpolation device 100 , the thermal displacement correction device 60 , and the numerical control device 70 . Figure 3 It is a diagram illustrating the missing of temperature data to be interpolated. Figure 4 It is a diagram for explaining recording of acquired temperature data in the nonvolatile memory 61 . Figure 5 This is a diagram for explaining the determination of whether or not temperature data is missing. Figure 6 is a graph illustrating the estimation and interpolation of missing data. Figure 7 1 is a diagram for explaining thermal displacement correction based on temperature data recorded in the nonvolatile memory 61 by the correction displacement amount calculation unit.

[0031] First, the overall outline of the temperature interpolation device 100 is described. The temperature interpolation device 100 is included in a structure having a thermal displacement correction function, which estimates the thermal displacement of the machine tool 1 based on the thermal displacement correction model and temperature data, and adds the correction amount corresponding to the thermal displacement to the axis movement amount. In this structure, for example, machine learning is used to create a thermal displacement correction model, and the correction amount is calculated based on the thermal displacement correction model and action state data (temperature data, etc.). In addition, when the difference between the estimated value and the measured value is not less than a threshold value, the thermal displacement correction model is calculated again.

[0032] Specifically, if Figure 1 As shown, the machine tool 1 is provided with a temperature sensor 2 and a displacement sensor 3 to detect temperature and displacement. Here, the machine tool 1 includes various machine tools, such as a machining center (with x-axis, y-axis, and z-axis), a lathe (with only x-axis and z-axis), and the like.

[0033] The detected temperature and displacement are used as learning data 4, and correction model calculation 5 is performed. Correction model calculation 5 is a calculation process for determining a thermal displacement correction model 5-3 by machine learning. Specifically, learning software 5-1 is a calculation process for making a thermal displacement correction model 5-3 by performing machine learning 5-2 based on the learning data 4. The learning software 5-1 can be executed on a predetermined computer. In addition, the learning data 4 can be data acquired in advance before the operation.

[0034] Next, the obtained thermal displacement correction model 5-3 is used to perform the correction amount calculation and output 6 in the actual working process. The correction amount calculation and output 6 can be performed by, for example, a CNC (Computerized Numerical Control) device 6-2. The CNC device 6-2 calculates and outputs the instruction (axis movement amount) for the machine tool 1. At this time, the correction amount is calculated using the thermal displacement correction model 5-3, and the correction amount is added to the axis movement amount. The axis movement amount corrected in this way is output to the machine tool 1. As a result, the machine tool 1 performs the action according to the axis movement amount after the correction 6-3. On the other hand, the machine tool 1 is equipped with a temperature sensor 2, and the temperature acquisition unit 6-1 reads the temperature from the temperature sensor 2 and supplies the temperature to the CNC device 6-2. Based on the supplied temperature, the CNC device 6-2 performs the correction amount calculation and output 6 using the thermal displacement correction model 5-3. Next, the same process is repeated. In this way, work that is not easily affected by heat can be performed.

[0035] More specifically, if Figure 2 As shown, the thermal displacement correction function is implemented by the thermal displacement correction device 60 and the numerical control device 70.

[0036] The thermal displacement correction device 60 includes a nonvolatile memory 61 as a nonvolatile storage medium, a correction displacement amount calculation unit 62, and a correction execution unit 63. In addition, the numerical control device 70 includes an axis control unit 71.

[0037] Each time the temperature sensor 2 reads and acquires temperature data at each time, the nonvolatile memory 61 records the temperature data and the time information in a correlated form. In addition, when the thermal displacement correction function is executed, a thermal displacement correction model for estimating thermal displacement and operation state data associated with thermal displacement are recorded.

[0038] The correction displacement amount calculation unit 62 calculates the correction amount based on the operation state data (temperature data, etc.) and the correction model, and supplies the correction amount to the correction execution unit 63. The correction execution unit 63 supplies the correction amount based on the temperature to the axis control unit 71.

[0039] The axis control unit 71 is a part that outputs instructions to the machine tool 1. Basically, it supplies instructions based on the machining program to the machine tool 1 of the system. At this time, it is executed using the axis control information of the system. In addition, it receives the correction amount based on the temperature from the correction execution unit 63, adds the correction amount to the instruction, and then supplies the instruction to the machine tool 1.

[0040] The temperature interpolation device 100 is configured to interpolate missing data when there is a gap in the temperature data read by the temperature sensor 2 of the machine tool 1 at each time, and output the interpolated temperature data together with the data at the time when the temperature sensor 2 reads the temperature data to the thermal displacement correction device 60.

[0041] Specifically, if Figure 2 As shown, the temperature interpolation device 100 includes a time information generating unit 101 , a temperature data reading and writing unit 102 , a past temperature estimating unit 103 , and a calculation parameter setting unit 104 .

[0042] It is known that the temperature read by the temperature sensor 2 at a predetermined portion of the machine tool 1 in this embodiment changes in accordance with a cubic equation. Based on this, the temperature data in the interval where the temperature data is missing is interpolated as follows.

[0043] First, for example Figure 3 As shown in FIG. 1 , when the temperature data in the interval from immediately after the temperature starts to drop to immediately after the temperature starts to rise (the central "data missing interval") is missing due to power outage, network failure, etc., a third-order equation having the temperature data as a prescribed parameter in the interval is obtained using the Lagrange interpolation formula. Then, the obtained third-order equation is used to interpolate the value of the temperature data in the interval where the temperature data is missing.

[0044] Specifically, the Lagrange interpolation formula for n sets of data (x1, y1), (x2, y2), (x3, y3), ..., (xn, yn) at time x and temperature y is provided by the following formula.

[0045] [Formula 1]

[0046]

[0047] [Formula 2]

[0048]

[0049] Here, if there are 4 points (4 groups) of data, a cubic equation can be obtained. In addition, it is preferable that the time of each data is separated from each other by a certain degree of time. If the time is too close, the temperature change is small. In addition, if the time is too far, the reliability of the temperature data decreases. Therefore, the calculation parameter setting unit 104 records the following interpolation requirements as calculation parameters used in the interpolation of temperature data.

[0050] [Requirements for interpolation]

[0051] Interpolation formula: Lagrange interpolation formula

[0052] Number of times: 3 times

[0053] Number of data used for interpolation: 4

[0054] The distance between the data used for interpolation: 30 minutes

[0055] Each time the temperature sensor 2 reads and acquires temperature data at each time, the time information generating unit 101 generates time information. The temperature data reading and writing unit 102 records the acquired temperature data and the time information generated by the time information generating unit 101 in the nonvolatile memory 61 of the thermal displacement correction device 60 in a form of association. Specifically, for example Figure 4 As shown, the temperature sensor 2 reads and acquires the temperature at intervals of one minute, and the temperature data reading and writing unit 102 sets the time and the temperature in a one-to-one relationship and records them in the nonvolatile memory 61 .

[0056] In addition, at the same time, the temperature data reading and writing unit 102 periodically compares the time information on the non-volatile memory 61 to determine whether missing data has occurred. Figure 5As shown, the temperature data reading and writing unit 102 reads two data closest to the current time (hereinafter referred to as "latest data" and "next latest data") from the non-volatile memory 61 and compares the time. If the time interval between the two read data is separated from the specified temperature data acquisition interval by more than a specified time, it is determined that data is missing ( Figure 3 The intervals with missing data are shown in Figure 5 As shown, an estimated temperature generation instruction is sent to the past temperature estimation unit 103.

[0057] More specifically, in Figure 4 , Figure 5 In the non-volatile memory 61, the temperature is read and acquired by the temperature sensor 2 at intervals of 1 minute, and the time when the temperature is read and the temperature when it is read are set as a group in a one-to-one relationship by the temperature data reading and writing unit 102 and recorded in the non-volatile memory 61. Here, there is no time information and temperature data in the 5 minutes between the time "2018 / 10 / 2319:02" and the time "2018 / 10 / 2319:07". In this way, when the temperature data is missing in the 5 minutes separated by more than 2 minutes exceeding the 1 minute interval as the specified interval, the temperature data reading and writing unit 102 Figure 5 As shown, the temperature data reading and writing unit 102 determines that data loss has occurred, and sends an estimated temperature generation command to the past temperature estimating unit 103 .

[0058] As the judgment formula at this time, for example, the following formula is used.

[0059] [Formula 3]

[0060] T t -T t-1 ≥SamplingRate+thresh

[0061] (T t : latest data time, SamplingRate: data acquisition interval, thresh: arbitrary threshold)

[0062] When the temperature data reading and writing unit 102 determines that data loss has occurred, Figure 6As shown, the past temperature estimation unit 103 estimates the temperature data at the missing time using the Lagrange interpolation formula using the remaining time and temperature data in the nonvolatile memory 61 and the calculation parameters recorded in the nonvolatile memory 61. In addition, the past temperature estimation unit 103 instructs the temperature data reading and writing unit 102 to record the estimated temperature data in the nonvolatile memory 61. As a result, the temperature data reading and writing unit 102 records the temperature data in the nonvolatile memory 61 in the form of interpolating the temperature data at the missing time.

[0063] With the above structure, for the temperature obtained by the temperature sensor 2, it is detected that the data obtained from the temperature sensor 2 is missing within the interval of the specified time, and the temperature data within the interval of the missing specified time is interpolated. Figure 7 As shown, the correction displacement amount calculation unit 62 estimates the thermal displacement based on the temperature data obtained from the temperature sensor 2 and the temperature data interpolated by the temperature interpolation device 100, and the thermal displacement correction model, and the correction execution unit 63 performs a thermal displacement correction function of adding a correction amount corresponding to the thermal displacement to the axis movement amount.

[0064] The present embodiment described above has the following effects.

[0065] In this embodiment, the temperature data reading and writing unit 102 continuously records the time information and the temperature data in the non-volatile memory 61 during the use of the machine tool 1 as the object device from which the temperature data is read, and compares the time information recorded in the non-volatile memory 61 at specified intervals to determine whether data loss occurs within the specified time interval.

[0066] Thus, during use of machine tool 1 , temperature data in a period (interval) in which temperature data is missing can be interpolated.

[0067] In addition, in the present embodiment, the past temperature estimation unit 103 estimates the missing data based on the temperature data using the Lagrange interpolation formula using the calculation parameters as the temperature interpolation formula.

[0068] Thus, when the temperature rise and temperature drop follow a cubic curve, the temperature data can be interpolated with high accuracy in consideration of the thermal characteristics of the portion of the machine tool 1 where the temperature sensor 2 is provided, resulting in more accurate thermal displacement correction.

[0069] Next, a second embodiment of the present invention will be described.

[0070] In the second embodiment, the difference from the first embodiment is that when the time constant of the temperature change is large depending on the position where the temperature sensor 2 is set in the machine tool 1, the missing data is interpolated using an exponential function. The other structures are the same as those of the first embodiment, so the description of the structures that are the same as those in the first embodiment is omitted.

[0071] Figure 8 This is a diagram for explaining that the time constant varies depending on the position where the temperature sensor 2 is mounted. Fig. 9 This is a graph explaining the difference in temperature rise and temperature drop caused by the difference in time constants. Fig.10 It is a diagram for explaining calculation of parameters of the temperature rise curve.

[0072] like Figure 8 As shown, in the machine tool 1, there is a portion 11 with a large time constant and a portion 12 with a small time constant. Fig. 9 As shown, the shape of the curve representing the transition of temperature change greatly varies depending on the size of the time constant, and in particular, when the user of the machine tool 1 stops using the machine tool 1 and then restarts using the machine tool 1, the displacement value obtained by the displacement sensor 3 deviates. Therefore, in order to estimate the missing temperature data with high accuracy, it is necessary to consider the time constant.

[0073] Specifically, the temperature rise data of each of the plurality of temperature sensors 2 of the machine tool 1 is acquired in advance, and the decay constant λ is obtained. Fig.10 As shown in FIG. 1 , when the saturation temperature value is set to 100% and the time until the temperature reaches 63.2% is set to τ during temperature rise, the decay constant λ which is a positive number can be obtained from the following relational expression.

[0074] λ=1 / τ

[0075] The obtained λ is recorded in advance in the calculation parameter setting unit 104 in a form associated with each temperature sensor 2 .

[0076] It is known that the change in temperature read by the temperature sensor 2 at a predetermined portion of the machine tool 1 in this embodiment where the temperature sensor 2 is provided and where the time constant is large is generally an exponential function of the temperature change. For example, when the temperature drops, the process is mathematically represented by the following differential equation.

[0077] dN / dt=-λN(t)

[0078] Where N(t) is the attenuation at time t. When this differential equation is solved, it becomes the following exponential function.

[0079] N(t)=Noe -λt

[0080] Among them, N 0 is the initial value. The differential equation is transformed into the following.

[0081] N(t)=Noe -λt +N ∞

[0082] Among them, N ∞ Set as the limit value for temperature.

[0083] The initial value N is defined as follows 0 and the limit value N ∞ .

[0084] N 0 = Temperature of the latest data - Temperature of the next latest data

[0085] N ∞ =The temperature of the next latest data (the temperature of the next new data after the latest data).

[0086] The present embodiment described above has the following effects.

[0087] In this embodiment, the past temperature estimation unit 103 uses an exponential function as a temperature interpolation formula in the temperature rise based on the temperature rise data acquired in advance to estimate the missing data based on the temperature data. The exponential function uses calculation parameters based on the time from 0°C as the start temperature value of the rise until a temperature that is a specified proportion to the saturation temperature value is reached.

[0088] Thereby, not only in the case of a decrease (decrease) in temperature but also in the case of an increase (rise), the missing temperature data can be estimated as a smooth curve.

[0089] The present embodiment has been described above. The above embodiment is a preferred embodiment, but the present invention is not limited to the above embodiment and can be implemented in various modified forms. For example, the present invention can be implemented in a modified form as described below.

[0090] That is, the structures of the calculation parameter setting unit, the temperature data reading and writing unit, the past temperature estimation unit, etc. are not limited to the structures of the calculation parameter setting unit 104, the temperature data reading and writing unit 102, the past temperature estimation unit 103, etc. in the above-mentioned embodiment. In addition, the structures of the thermal displacement correction device, the numerical control device, etc. for executing the thermal displacement correction function are not limited to the structures of the thermal displacement correction device 60 having the non-volatile memory 61, the correction displacement amount calculation unit 62 and the correction execution unit 63, the numerical control device 70 having the axis control unit 71, etc. in the above-mentioned embodiment. In addition, as the temperature interpolation formula, the Lagrange interpolation formula is used, but it is not limited to this. In addition, the temperature data follows a cubic function and an exponential function, but it is not limited to this. In addition, in the above-mentioned embodiment, as a non-volatile storage medium, the non-volatile memory 61 is used, but it is not limited to this. In addition, in the above-mentioned embodiment, the rise start temperature value is 0°C, but it is not limited to this.

Claims

1. A temperature interpolation device for a machine tool, comprising: A calculation parameter setting unit, which records calculation parameters of a temperature interpolation formula, wherein the temperature interpolation formula is pre-selected to interpolate the missing temperature data; a temperature data reading and writing unit which, during use of the machine tool from which the temperature data is read, continuously records the time information and the temperature data in a non-volatile storage medium, respectively, and compares the time information recorded in the non-volatile storage medium at a predetermined time interval to determine whether data loss occurs within the predetermined time interval; and A past temperature estimation unit, which uses the data recorded in the nonvolatile storage medium and the calculation parameters recorded in the calculation parameter setting unit to estimate and interpolate the missing temperature data using the temperature interpolation formula when the time interval between the two temperature data closest to the current time is determined by the temperature data reading and writing unit to be more than the specified time interval and the temperature data is missing, wherein: The interpolated missing temperature data is used to correct the thermal displacement of the machine tool.

2. The temperature interpolation device for a machine tool according to claim 1, wherein: The past temperature estimation unit estimates the missing data based on the temperature data using a Lagrange interpolation formula using the calculation parameters as the temperature interpolation formula.

3. The temperature interpolation device for a machine tool according to claim 1, wherein: The past temperature estimation unit estimates missing data based on the temperature rise data acquired in advance by using an exponential function as the temperature interpolation formula, wherein the exponential function uses the calculation parameters based on the time from the temperature value at the start of the rise until the temperature reaches a temperature that is a specified proportion relative to the saturation temperature value.

Images