JP2011131371A - Machine tool, and method and program for determining number and arrangement of temperature measurement parts of machine tool - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a thermal displacement correction method for accurately correcting a thermal displacement by a method for identifying a thermal source in which the amount of thermal displacement is estimated by an inverse analysis in a machine tool. <P>SOLUTION: In a method for correcting a thermal displacement error of a driving system having a complicated structure, and estimating a thermal displacement amount, at first, a sensor having a higher usefulness is selected as a temperature measuring part out of sensors arranged in a structural body, then a thermal source region of a structural body to be measured is set, and the temperature of the structural body to be measures is increased. Next, a thermal inflow amount of the thermal source region is measured by an inverse analysis, and a correction formula of a thermal displacement having the thermal inflow amount as a parameter is established. The thermal displacement amount taking the thermal inflow amount to each thermal source into consideration, is then estimated, and the thermal displacement correction based on the thermal inflow amount is made. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a machine tool, a method and a program for determining the number and arrangement of temperature measuring units of a machine tool.

  In the drive system of machine tools, the structure of machine tools has become more complex with the increase in the number of axes, and the problem of thermal displacement, which is one of the obstacles to high accuracy and high efficiency of machine tools, is the site where machines are used. In recent years, the problem has become increasingly prominent.

  Conventional thermal displacement correction methods for machine tools include those described in Patent Documents 1 to 7 and Non-Patent Document 1.

  Patent Document 1 describes a method of correcting a thermal displacement in a machine tool with high accuracy by using a learning function of a neural network based on outputs of a plurality of temperature measuring means arranged in each part of the machine tool. . Patent Document 2 and Non-Patent Document 1 describe a method for correcting a thermal displacement error of a drive system of a machine tool without using a detector. Patent Document 3 and Patent Document 5 describe thermal displacement correction methods that take temperature changes into account. Patent Document 4 proposes a new method for obtaining an optimum target temperature for forced temperature control by an inverse method of a neural network and highly suppressing thermal deformation of a machine tool. Patent Document 6 describes a thermal displacement correction method that estimates the amount of thermal displacement based on the position information of the machine, the cutting edge position of the tool, or the workpiece position information in consideration of the temperature distribution of the structure of the machine tool. Furthermore, Patent Document 7 describes a thermal displacement correction method for estimating a thermal displacement amount at a machining point based on temperature measurements at a plurality of locations on a machine body.

Japanese Unexamined Patent Publication No. 6-8107 JP-A-7-299701 JP 2000-135653 A Japanese Patent Laid-Open No. 2001-138175 JP 2003-94290 A JP 2006-281420 A JP 2007-167966 JP

Hori 3 Kei, Nishiwaki Nobuhiko, Ishitomi Katsuya, Research on thermal deformation estimation of machine tools, Transactions of the Japan Society of Mechanical Engineers (C), 63, 608 (1997), pp.1391-1396.

  However, the prior art described in the above literature has room for improvement in the following points.

  First, in Patent Document 1, this method requires a data table for predicting the amount of thermal displacement of the machine tool corresponding to the output of the temperature measuring means, and data that can correspond to all temperature measurement values is prepared in advance. It is difficult to do.

  Secondly, in Patent Document 2 and Non-Patent Document 1, a heat distribution model including a drive system is formed on a numerical simulation, and the amount of heat supplied to the heat distribution model is estimated, whereby a correction value for thermal displacement is obtained. Therefore, if the amount of heat generated supplied to the heat distribution model cannot be estimated, a correction value for thermal displacement cannot be obtained. Also, no means for estimating the amount of heat generated supplied to the heat distribution model is shown.

  Thirdly, in Patent Document 3, it is necessary to obtain in advance the characteristics of the maximum thermal displacement amount according to the drive status of the drive mechanism and store it in the storage means, so the heat of the multi-axis control machine tool that has a complicated heat source distribution As a displacement correction method, sufficient accuracy is not always obtained.

  Fourthly, any of the methods disclosed in Patent Document 4, Patent Document 6, and Patent Document 7 is a method that focuses on the one-to-one relationship between temperature and machine tool, and is sufficient for thermal deformation of a multi-axis control machine tool having a complicated structure. It is not always possible to estimate with high accuracy.

  The present invention has been made in view of the above circumstances, and enables efficient estimation of the amount of heat inflow supplied to a heat distribution model of a machine tool, and enables thermal displacement correction based on the estimated amount of heat inflow. The purpose is to provide technology.

  According to the present invention, there is provided a machine tool that relatively moves a tool and a work material by a plurality of drive mechanisms to process the work material into a predetermined shape, and includes a plurality of temperature sensors provided in the machine tool, A temperature sensor selection unit that selects one of the temperature sensors that is used for temperature measurement as a temperature measurement unit, a heat source candidate region setting unit that sets a heat source candidate region in the machine tool, and heat from the heat source candidate region A heat inflow assumption unit that assumes an inflow amount, a temperature distribution estimation unit that generates an estimated temperature distribution of the machine tool based on the heat inflow amount assumed by the heat inflow amount assumption unit, and a temperature measurement unit The temperature detection value is compared with the temperature estimation value of the location corresponding to the temperature measurement unit in the temperature distribution estimated by the temperature distribution estimation unit, and the gap between the temperature detection value and the temperature estimation value is compared. A comparison / gap analysis unit that causes the heat inflow amount assumption unit to re-assum the heat inflow amount to a different value so that the heat inflow amount assumption unit and the distribution estimation unit repeatedly perform calculations. And an approximate data calculation unit that calculates an approximate heat inflow amount from the heat source candidate region and an approximate temperature distribution of the machine based on the result of the return calculation, and the temperature sensor selection unit includes the temperature sensor selection unit. Based on whether each is used, a plurality of combinations of temperature sensors are generated according to an orthogonal table, a combination evaluation value is calculated for each of the combinations, and a certain temperature sensor uses the combination evaluation value calculated for each combination The temperature sensor evaluation value of the temperature sensor is calculated by classifying it based on whether it is Selecting a temperature sensor used as a temperature measuring part Te, the machine tool is provided.

  According to this configuration, the heat inflow amount assumption is made to re-assum the heat inflow amount to another value so that the gap between the temperature detection value and the temperature estimation value becomes small, and the heat inflow amount assumption is made. The approximate heat inflow amount in which the estimated temperature value is close to the detected temperature value can be searched for by repeatedly causing the temperature distribution estimating unit and the temperature distribution estimating unit to perform the calculation. Therefore, if the machine tool of the present invention is used, based on the approximate heat inflow amount thus obtained, the approximate heat inflow amount is applied to the model data of the machine tool, and the thermal displacement data in the machine tool is obtained. It can be estimated, and accurate thermal displacement correction can be performed based on the thermal displacement data.

By the way, generally, as the number of temperature measuring units increases, the amount of calculation increases accordingly, and it takes time to calculate the approximate heat inflow amount. Therefore, it is desirable to reduce the number of temperature measuring units as much as possible. On the other hand, if the number of temperature measurement units is too small, the accuracy of the approximate heat inflow amount may be deteriorated, so the number of temperature measurement units cannot be reduced unnecessarily. Therefore, a method for determining the number and arrangement of the temperature measurement units so as to increase the accuracy of the approximate heat inflow amount while minimizing the number of temperature measurement units is desired.
According to the machine tool of the present invention, the temperature sensor selection unit calculates the temperature sensor evaluation value indicating the usefulness of each temperature sensor, and selects the temperature sensor to be used as the temperature measurement unit based on the temperature sensor evaluation value. By installing a number of temperature sensors in advance and using the most useful temperature sensor as a temperature measurement unit and searching for approximate heat inflow, the number of temperature measurement units can be minimized. However, the accuracy of the approximate heat inflow amount can be increased.

Further, when the ambient temperature of the machine tool changes or the operating conditions of the machine tool are changed, the thermal environment to which the machine tool is exposed changes. When the thermal environment changes, the optimal number and arrangement of temperature measurement units may also change, but in machine tools where the number and arrangement of temperature measurement units are fixed, the temperature measurement unit will respond to changes in the thermal environment. The number and arrangement cannot be changed.
According to the machine tool of the present invention, a large number of temperature sensors are installed in advance, and a highly useful one of these temperature sensors is used as the temperature measurement unit. The temperature sensor to be used can be changed, and it is possible to flexibly cope with changes in the thermal environment.

  Note that the above-described device is one embodiment of the present invention, and the device of the present invention may be any combination of the above components. The method, system, computer program, recording medium, etc. of the present invention have the same configuration.

It is a perspective view for demonstrating the structural example of the multi-axis control machine tool which concerns on one Embodiment of this invention. It is a perspective view for demonstrating the cross slide model of the drive mechanism of the multi-axis control machine tool which concerns on one Embodiment of this invention. It is a conceptual diagram for demonstrating the concept of the thermal displacement correction method of the multi-axis control machine tool which concerns on one Embodiment of this invention. It is the figure which showed the installation place of the data logger (temperature sensor) in the peripheral part in the drive mechanism of the cross slide model of the multi-axis control machine tool which concerns on one Embodiment of this invention. It is a functional block diagram for demonstrating the structure of the multi-axis control machine tool which concerns on one Embodiment of this invention. It is a flowchart which shows the step which the temperature sensor selection part of the multi-axis control machine tool which concerns on one Embodiment of this invention performs. An orthogonal table _{L 12} used in the description of step SA1 in FIG. An orthogonal table _{L 16} used in the description of step SA1 in FIG. It is a flowchart for demonstrating the detail of step SA2 of FIG. 10 is a table showing examples of combination evaluation values calculated in step SB8 of FIG. 9. It is a flowchart for demonstrating the detail of step SA3 of FIG. It is a table | surface which shows the example of the temperature sensor evaluation value calculated by step SC5 of FIG. It is a flowchart for demonstrating the detail of the 1st method of the selection process of the temperature sensor in step SA4 of FIG. It is a table | surface which shows an example of the combination of the temperature sensor produced | generated by step SD1 of FIG. It is a graph which shows an example of the combination evaluation value calculated by step SD2 of FIG. It is a flowchart for demonstrating the detail of the 2nd method of the selection process of the temperature sensor in step SA4 of FIG. It is a functional block diagram which shows the detailed structure of the thermal displacement correction apparatus of the multi-axis control machine tool which concerns on one Embodiment of this invention. 1 is an enlarged functional block diagram for explaining in detail a configuration of a part of a multi-axis control machine tool according to an embodiment of the present invention. FIG. It is a flowchart for demonstrating operation | movement of the multi-axis control machine tool which concerns on one Embodiment of this invention. It is a flowchart for demonstrating the detail of step SF1 of FIG. It is a flowchart for demonstrating the detail of step SF3 of FIG. It is a perspective view which shows the model used in Example 1 of this invention. It is a top view which shows arrangement | positioning of the evaluation point set in Example 1 of this invention. It is a temperature distribution map calculated with the model of Example 1 of this invention. An orthogonal table L ₁₂ used in Example 1 of the present invention. It is a table | surface which shows the combination evaluation value calculated in Example 1 of this invention. In Example 1 of this invention, it is a graph which shows the assumption heat | fever inflow amount set last, when a basic evaluation point is f12. In Example 1 of this invention, it is a graph which shows the assumption heat | fever inflow amount set last, when a basic evaluation point is f15. In Example 1 of this invention, it is a table | surface which shows the evaluation point evaluation value calculated when the basic evaluation point is f12. In Example 1 of this invention, it is a table | surface which shows the evaluation point evaluation value calculated when the basic evaluation point is f15. In Example 1 of this invention, when a basic evaluation point is f12, it is a table | surface which shows several combinations of the evaluation point produced | generated by step SD1. In Example 1 of this invention, when a basic evaluation point is f15, it is a table | surface which shows several combinations of the evaluation point produced | generated by step SD1. In Example 1 of this invention, when a basic evaluation point is f12, it is a graph which shows the combination evaluation value calculated about each combination by step SD3. In Example 1 of this invention, when a basic evaluation point is f15, it is a graph which shows the combination evaluation value calculated about each combination by step SD3. It is an experimental conceptual diagram showing the experimental purpose of Example 2 of the present invention. It is the graph which illustrated the temperature measurement result of the temperature sensor in the peripheral part in the drive mechanism of the cross slide model of the multi-axis control machine tool used in Example 2 of the present invention. In the cross-slide model drive mechanism of the multi-axis control machine tool used in Example 2 of the present invention, the cross-slide model estimated based on the heat source distribution of the optimum condition calculated by the inverse analysis from the temperature measurement result of the temperature sensor It is a 3D analysis figure which illustrated the temperature distribution of the drive mechanism. It is a perspective view which shows the position of the cross slide model used in Example 3 of this invention, and the temperature sensor on this model. It is a perspective view which shows the position of the heat source candidate area | region in the cross slide model used in Example 3 of this invention. It is a graph which shows the relationship of the combination of the temperature sensor based on Example 3 of this invention, and the heat inflow amount from each heat source candidate area | region. It is a graph which shows the influence which each temperature sensor has on S / N ratio based on Example 3 of this invention. It is a perspective view which shows the position of the evaluation point on the cross slide model used in Example 4 of this invention, and this model. It is a temperature distribution figure of the correct answer in each heat source pattern based on Example 4 of this invention. It is a graph which shows the value of the heat inflow amount (heat flux) from each heat source candidate area | region in each step of repetition, and the value of an objective function about the combination a of the heat source pattern 1 and the evaluation point based on Example 4 of this invention. It is a temperature distribution figure which shows the heat flux search analysis result in the heat source pattern 1 based on Example 4 of this invention. It is a temperature distribution figure which shows the heat flux search analysis result in the heat source pattern 2 based on Example 4 of this invention. It is a temperature distribution figure which shows the heat flux search analysis result in the heat source pattern 3 based on Example 4 of this invention. It is a temperature distribution figure which shows the heat flux search analysis result in the heat source pattern 4 based on Example 4 of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same reference numerals are given to the same components, and the description will be omitted as appropriate.

1. FIG. 1 illustrates an example of the structure of a machine tool that relatively moves a tool and a work material by a plurality of drive mechanisms according to the present embodiment to process the work material into a predetermined shape. It is a perspective view for doing. A spindle for rotating the workpiece is set on the left and right. In particular, the tool post 211 is provided with a swivel axis in addition to three axes in the XYZ-axis direction, and multi-axis control machining is possible. .

  FIG. 2 is a perspective view for explaining an inclined support body which is a cross slide model of the drive mechanism of the multi-axis control machine tool according to the embodiment. Here, the cross slide 200 of the present embodiment is a component used on the bed 212 (shown in FIG. 1) of the machine tool, and moves back and forth along a slide way (not shown) on the bed 212. A kind of moving saddle.

  As shown in the figure, the reciprocating slide ways 207a and 207b are provided in parallel on the inclined surface 205 of the cross slide. On the reciprocating slide ways 207a and 207b, a tool post 211 (shown in FIG. 1) for holding a cutting tool for processing a work material is installed. The tool post 211 is configured to be movable in an oblique direction (restricted YZ axis direction) along the reciprocating slide ways 207a and 207b. Therefore, the tool post 211 (shown in FIG. 1) moves in the front-rear and left-right directions (XY directions) with respect to the bed 212 (shown in FIG. 1) described above, and the oblique direction (restricted) of the tool post relative to the cross slide 200 In other words, the combination of the movement in the YZ-axis direction) can be freely moved in the XYZ-axis direction in the three-dimensional space.

  FIG. 3 is a conceptual diagram for explaining the concept of the thermal displacement correction method of the multi-axis control machine tool according to the embodiment. In this thermal displacement correction method for a multi-axis control machine tool, first, the machine tool operator generates heat based on his / her own experience among the outer surfaces of the machine tool components (for example, the outer surface of the inclined support 200). A heat source candidate area is set in a place (for example, a sliding part or a rotating shaft) that is likely to be a source. Subsequently, among a plurality of temperature sensors provided on the peripheral portion of the outer surface (for example, the outer surface of the inclined support body 200) of the component of the machine tool that is the structure to be measured, a temperature sensor selection unit 204 described later. The temperature rise is measured by the one selected as the temperature measuring unit by. Next, the heat inflow amount from the heat source candidate region is calculated by inverse analysis from the temperature rise history obtained from the temperature rise measurement of the structure to be measured. And based on this heat inflow amount, the thermal displacement data of the machine tool provided with the cross slide model which is a structure to be measured is estimated. In a machine tool having a cross slide model, which is a structure to be measured, if such a series of operations are performed for several drive patterns, a thermal displacement correction formula for correcting the thermal displacement generated based on the amount of heat inflow Can be established. Finally, the thermal displacement amount is estimated in consideration of the heat inflow amount to each heat source by an established formula peculiar to each machine tool having a cross slide model as a structure to be measured in this way. It becomes possible to correct the thermal displacement based on the amount of displacement.

  Therefore, according to this thermal displacement correction method for a multi-axis control machine tool, the following can be realized which has been conventionally desired in the field of machine tools but difficult to realize. That is, first, the final shape of the work material is input as CAD data, and a machine tool drive pattern is calculated by a computer. In the calculated drive pattern, for example, assuming that the rotational speed of the shaft is defined as a rotation / minute and the stroke of the rail is defined as b times / minute, the inflow heat amount is Q1 = cJ / minute, Q2 = dJ / minute, etc. Inflow heat quantity is estimated by conversion using a data table stored in advance in a computer. If the thermal displacement correction formula for correcting the thermal displacement generated based on the heat inflow amount has already been established, the heat displacement amount Δd can be obtained from the inflow heat amount estimated in this way. High-precision machining is performed by performing correction based on the thermal displacement amount Δd.

  FIG. 4 is a diagram illustrating an example of an installation location of a data logger (temperature sensor) installed at the peripheral portion of the cross slide model of the multi-axis control machine tool according to the embodiment. In this figure, an installation place of a data logger (temperature sensor) in an experiment conducted in an example described later is shown as an example. Among the data loggers (temperature sensors) in this figure, the one selected by the temperature sensor selection unit 204 described later is used as the temperature measurement unit. During installation, the outside air is shut off by a heat-insulating putty so that the surface temperature of the cross slide model can be measured. In this figure, 15 data loggers are installed.

2. Embodiment of machine tool 2-1. Functional Block FIG. 5 is a functional block diagram for explaining the configuration of the multi-axis control machine tool according to the embodiment invented based on the above idea. The multi-axis control machine tool 1000 is a machine tool 1000 that relatively moves a tool and a work material by a plurality of drive mechanisms 300 to process the work material into a predetermined shape. The machine tool 1000 as a whole receives a plurality of temperature sensors 202 for measuring the temperature of the structure to be measured and various data inputs from the operator of the machine tool, and the structure to be measured based on the input data. An input reception / model data generation unit 101 that generates model data, a temperature sensor selection unit 204 that selects a temperature sensor 202 to be used for temperature measurement as a temperature measurement unit, and a temperature distribution measured in this manner Is a machine tool comprising: a thermal displacement correction device 100 that calculates thermal displacement correction data based on the thermal displacement correction data; and a plurality of drive mechanisms 300 that correct the drive pattern based on the thermal displacement correction data derived as described above. .

2-2. First, the temperature sensor 202 constituting the machine tool 1000 will be described below. This multi-axis control machine tool 1000 is provided with a temperature sensor 202 for measuring the temperature at the peripheral edge of the outer surface of the structure 200 to be measured. The temperature sensor 202 is brought into close contact with the peripheral portion of the outer surface of the structure 200 to be measured, and functions as a data logger that measures and records the surface temperature. The temperature sensor 202 converts an electrical signal corresponding to the measured and recorded temperature data into temperature data in a format that can be read by a computer, and outputs the temperature data. Among the plurality of temperature sensors 202, one that is used as a temperature measurement unit is selected by the temperature sensor selection unit 204, and an output from the selected temperature sensor 202 is transmitted to the thermal displacement correction device 100.

2-3. Input Reception / Model Data Generation Unit Next, the input reception / model data generation unit 101 constituting the machine tool 1000 will be described. The input reception / model data generation unit 101 receives various data inputs from an operator of the machine tool, and generates measured structure model data based on the input data.

  The input reception / model data generation unit 101 includes a heat source candidate region setting unit 104 for setting, as a heat source candidate region, a region that is likely to be a heat source in the three-dimensional structure of the machine tool. The heat source candidate area setting unit 104 may receive data of a heat source candidate area input by the operator via the input unit 102 such as a keyboard, or may be set in advance by calling from the network 112, the server 114, or the like. You may read the data regarding a heat source candidate area | region. The data received in this way is transferred to the measured structure model data generation unit 120 described later.

  The input reception / model data generation unit 101 is suitable for use as a target for various 3D simulations by a computer by combining various information on the three-dimensional structure, physical properties, temperature sensor position, and heat source candidate region of the structure to be measured. A measured structure model data generation unit 120 that generates measured structure model data is provided so as to be data. In addition, in order to obtain and transfer these various types of information to the measured structure model data generation unit 120 from the outside, the input reception / model data generation unit 101 sets the heat source candidate region setting conditions from the outside. The heat source candidate region setting unit 104 for receiving, the temperature sensor position data receiving unit 116 for receiving the position information of the temperature sensor installed in the structure to be measured, and the structure to be measured for receiving data relating to the three-dimensional structure and material properties of the structure to be measured. A structure / physical property data receiving unit 118 is also provided. Then, the three-dimensional structure / physical properties / temperature of the structure to be measured acquired from the outside via the heat source candidate region setting unit 104, the temperature sensor position data receiving unit 116, and the structure / physical property data receiving unit 118 of the structure to be measured. Various information about the sensor position / heat source candidate region is collected in the structure model data generation unit 120 to be measured, and these data are combined with each other so as to be data suitable for use as various 3D simulation targets by a computer. In addition, measured structure model data is generated. The measured structure model data generated in this way is stored in the measured structure model data storage unit 122.

2-4. Temperature Sensor Selection Unit Next, the temperature sensor selection unit 204 will be described below. The multi-axis control machine tool 1000 is provided with a temperature sensor selection unit 204 that selects a temperature sensor 202 to be used for temperature measurement as a temperature measurement unit. Hereinafter, the operation of the temperature sensor selection unit will be described in detail with reference to the flowchart shown in FIG. Although not specified in the following description, various types of input or generated data are stored in a predetermined storage unit so that they can be used for subsequent data processing. Each of the following steps is realized by the CPU executing a predetermined computer program stored in the main memory or the external storage device.

2-4-1. Generation of a plurality of combinations of temperature sensors (step SA1)
First, in step SA1, the temperature sensor selection unit 204 generates a plurality of combinations of temperature sensors according to the orthogonal table based on whether or not each of the temperature sensors 202 is used. An orthogonal table is a table having the property that all combinations of levels appear for the same two factors (columns) the same number of times. In this embodiment, since it is a problem whether to use a certain temperature sensor 202, a two-level orthogonal table is used. Examples of the two-level orthogonal table include L ₄ , L ₈ , L ₁₂ , L ₁₆ , L ₃₂ , L ₆₄ , and L ₁₂₈ . An orthogonal table of L ₁₂ and L ₁₆ is shown in FIGS. Orthogonal table of L ₃₂ is ₍₁₎ each row of the orthogonal table _{L 16} and two rows, add the 16th column consisting 01010101 .... 0101 (2) on the right side of the 15 column, (3) the It can be created by adding a column consisting of the exclusive OR of the 16th column and the 1st to 15th columns after the 17th column. Orthogonal table of L ₆₄ performs the same operation on the orthogonal table L _32, orthogonal table of L ₁₂₈ can be created by performing the same operation on the orthogonal table L _64. The maximum number of temperature sensors 202 that can be assigned to the L ₁₂ , L ₁₆ , L ₃₂ , L ₆₄ , and L ₁₂₈ orthogonal tables is 11, 15, 31, 63, and 127, respectively. Therefore, the orthogonal table to be used may be selected according to the number of temperature sensors 202 considered by the temperature sensor selection unit 204. For example, if the number of the temperature sensor selection unit 204 to consider the temperature sensor 202 is 11 or less, it is preferable to use an orthogonal table of L _12.

  Although the temperature sensor selection unit 204 may assign all the temperature sensors 202 to the orthogonal table, the temperature measurement unit is always used for the temperature sensors 202 (hereinafter referred to as “basic temperature sensors”) that are clearly necessary. In other words, only the other temperature sensors 202 may be assigned to the orthogonal table. In this case, the number of temperature sensors 202 to be considered can be reduced, and the calculation time can be shortened. As the basic temperature sensor, it is preferable to select the temperature sensor 202 arranged at a position where the amount of temperature rise in the reference temperature distribution calculated in step SB2 described later is the smallest. In this case, even when the number of temperature sensors 202 to be used is small, it is clear from Example 1 described later that a relatively good result can be obtained.

The temperature sensor 202 is assigned to each column of the orthogonal table. For _example, when using orthogonal table of _{L 12,} allocates up to 11 temperature sensors 202 in column A through K. When the number of temperature sensors 202 to be assigned is less than 11, for example, 8, the number is assigned to columns A to H. Each row of the orthogonal table indicates a combination of the temperature sensors 202. For _example, the first row to the 12th row of the orthogonal table of _{L 12} represents a combination of the temperature sensor 202 of the first to 12. “0” and “1” in the orthogonal table mean that the temperature sensor 202 is “not used” and “used”, respectively. For _example, line 10 of the orthogonal table of _{L 12} is a column ABCHIK is 1, the remaining columns is zero. Thus, this row shows a combination of temperature sensors 202 that uses the temperature sensor 202 assigned to column ABCHIK. Thus, by using the orthogonal table, multiple combinations of temperature sensors 202 are generated based on whether each of the temperature sensors 202 is used.
Note that step SA1 may be performed at any stage before the first combination is selected in step SB3 to be described later. Therefore, step SA1 may be performed before starting step SA2, and step SB3 is executed after step SA2 is started. You may go before you do.

2-4-2. Combination evaluation value calculation processing (step SA2)
Next, in step SA2, the temperature sensor selection unit 204 calculates a combination evaluation value for each of a plurality of combinations of the temperature sensors 202 created in step SA1. The combination evaluation value is an index for determining the quality of the combination of the temperature sensors 202, and the calculation method is not particularly limited.

In one example, the combination evaluation value sets a reference heat inflow amount from the heat source candidate region, generates a reference temperature distribution in the heat source candidate region based on the reference heat inflow amount, and assumes the assumed heat from the heat source candidate region. An inflow amount is set, an estimated temperature distribution of the machine tool is generated based on the assumed heat inflow amount, a temperature reference value corresponding to a temperature sensor included in the combination in the reference temperature distribution, and the estimated The assumed heat inflow amount is set so that a gap between the temperature reference value and the temperature estimated value is reduced by comparing the temperature distribution with the temperature estimated value corresponding to the temperature sensor included in the combination in the temperature distribution. Repeat the calculation by resetting to another value, and the difference between the final estimated temperature distribution obtained as a result of the above repeated calculation and the reference temperature distribution or the difference between the final assumed heat inflow and the reference heat inflow It can be calculated by evaluating. According to this method, the combination evaluation value is calculated by a method similar to the method used when the approximate data calculation unit 134, which will be described later, calculates the approximate heat flow rate. Therefore, the temperature is calculated using the combination evaluation value calculated by this method. If the sensor is selected, the approximate data calculation unit 134 can calculate the approximate heat flow rate with high accuracy.
Hereinafter, the calculation process of the combination evaluation value will be described in more detail with reference to the flowchart of FIG.

(1) Setting of reference heat inflow amount (step SB1)
In step SB1, the temperature sensor selection unit 204 sets the heat inflow amount flowing from each heat source candidate region. Here, the heat inflow amount to be set is used to create a reference temperature distribution that serves as a reference when performing a comparison / gap analysis in step SB6 described later. The heat inflow amount set here is hereinafter referred to as “reference heat inflow amount”. The reference heat inflow amount can be set in consideration of the specifications of a motor or the like serving as a heat source, or can be set to an approximate heat inflow amount previously calculated by an approximate data calculation unit 134 described later. In the latter case, the approximate heat inflow amount previously calculated by the approximate data calculation unit 134 and the approximate heat inflow amount to be calculated by the approximate data calculation unit 134 are considered to be relatively close to each other. By setting the approximate heat inflow amount previously calculated by the calculation unit 134 as the reference heat inflow amount, the approximate data calculation unit 134 can calculate the approximate heat flow rate with higher accuracy. Also, when the ambient temperature of the machine tool changes or the operating conditions of the machine tool change, the thermal environment to which the machine tool is exposed may change. To cope with such a change in the thermal environment, In this case, the reference heat inflow amount is set to the approximate heat inflow amount previously calculated by the approximate data calculation unit 134. The trouble of setting the heat inflow amount can be saved. Further, since the approximate heat inflow amount calculated immediately before by the approximate data calculating unit 134 is a value reflecting the change in the thermal environment, the approximate heat flow rate is set to the reference heat inflow amount due to the change in the thermal environment. It can respond reliably. When there are a plurality of heat source candidate regions, a reference heat inflow amount is set for each heat source candidate region.

(2) Calculation of reference temperature distribution (step SB2)
In step SB2, the temperature sensor selection unit 204 calculates a reference temperature distribution. The reference temperature distribution is the calculation algorithm such as the finite element analysis method, how the temperature distribution of the structure to be measured becomes after a predetermined time when the reference heat inflow flows into the structure model data to be measured. It is the temperature distribution of the structure to be measured obtained by calculating using.

(3) Selection of combination (steps SB3, SB9, SB10)
In step SB3, the temperature sensor selection unit 204 selects one of a plurality of combinations of the temperature sensors generated in step SA1. Specifically, for example, to select the first row orthogonal table L ₁₂ shown in FIG. In this case, since the values of all the columns are 0, only the basic temperature sensor becomes the temperature measuring unit. If there is no basic temperature sensor, the second row may be selected as the first combination.
The temperature sensor selection unit 204 calculates a combination evaluation value by executing Step SB4 to Step SB8 for the combination of temperature sensors selected here. In order to calculate combination evaluation values for all of the plurality of combinations of the temperature sensors generated in step SA1, the temperature sensor selection unit 204 confirms whether or not combination evaluation values have been calculated for all combinations in step SB9. If an uncalculated combination exists (N in Step SB9), the next combination is selected in Step SB10. When there is no uncalculated combination (Y in step SB9), the temperature sensor selection unit 204 completes the combination evaluation value calculation process.

(4) Setting of assumed heat inflow amount (step SB4)
In step SB4, the temperature sensor selection unit 204 assumes a heat inflow amount flowing from each heat source candidate region. The heat inflow assumed here is used to calculate an estimated temperature distribution, which serves as a reference when performing a comparison / gap analysis in step SB6, which will be described later, in step SB5. The heat inflow assumed here is hereinafter referred to as “assumed heat inflow”. The assumed heat inflow amount is repeatedly set until the result of the comparison / gap analysis in step SB6 satisfies the end condition defined in step SB7. The method for setting the initial value of the assumed heat flow rate is not particularly limited, but in view of the purpose of calculating a combination evaluation value that reflects the quality of the combination of temperature sensors, the assumed heat flow rate is different from the reference heat inflow amount. It is preferable that the value is not more than ½ or twice the reference heat inflow amount. This is because if the assumed heat inflow amount and the reference heat inflow amount are too close, it is difficult for the combination evaluation value to reflect the quality of the combination of the temperature sensors.

(5) Calculation of estimated temperature distribution (step SB5)
In step SB5, the temperature sensor selection unit 204 calculates an estimated temperature distribution. Estimated temperature distribution is a calculation algorithm such as the finite element analysis method that shows the temperature distribution of the measured structure after a lapse of a predetermined time when the assumed heat inflow flows into the measured structure model data. It is the temperature distribution of the structure to be measured obtained by calculating using.

(6) Comparison / gap analysis (step SB6)
In step SB6, the temperature sensor selection unit 204 performs comparison / gap analysis. Here, the contrast / gap analysis is an analysis of the magnitude of the difference between the reference temperature distribution generated in step SB2 and the estimated temperature distribution generated in step SB5. Specifically, this analysis is performed by (1) the temperature (hereinafter referred to as the temperature on the reference temperature distribution generated in step SB2) at each location corresponding to the temperature sensor included in the combination of temperature sensors selected in step SB3 or SB10. , Referred to as “temperature reference value”) and the temperature on the estimated temperature distribution generated in step SB5 (hereinafter referred to as “temperature estimated value”), and (2) for each location corresponding to the temperature sensor The difference between the temperature reference value and the temperature estimated value can be calculated, and (3) a gap evaluation value based on the sum of absolute values or the sum of squares of residuals for the calculated difference can be calculated. In one example, the gap evaluation value can be calculated based on the following formula (1). In Expression (1), T _i is a temperature estimation value, T _s is a temperature reference value, and n is the number of temperature sensors included in the combination of temperature sensors selected in Step SB3 or SB10. .

(7) End condition determination (step SB7)
In step SB7, the temperature sensor selection unit 204 determines whether or not the repetition process of steps SB4 to SB6 can be ended. This determination can be made based on various criteria. For example, (a) whether the gap evaluation value exceeds the threshold, (b) whether the number of repetitions is equal to or greater than the threshold, (c) whether the calculation time is equal to or greater than the threshold, etc. Any of the above may be adopted as the termination condition, and a combination thereof may be adopted as the termination condition.
If the end condition is not satisfied (N in Step SB7), the process returns to Step SB4 so that the sum of absolute values or the sum of residual squares of the difference between the temperature reference value and the temperature estimated value becomes small (in other words, The value of the assumed heat flow rate is changed so that the value of the above equation (1) is maximized), and the processes after step SB5 are performed again. If the end condition is satisfied (Y in step SB7), the iterative process is ended and the process proceeds to step SB8.

(8) Calculation of combination evaluation value (step SB8)
In step SB8, the temperature sensor selection unit 204 calculates a combination evaluation value. The combination evaluation value may be a value that serves as an index for determining whether the combination of temperature sensors is good or bad. Specifically, the combination evaluation value is calculated by, for example, (1) calculating the difference between the temperature reference value and the estimated temperature value at each location corresponding to the temperature sensor used for calculating the combination evaluation value, and (2) calculating the difference. Can be calculated by calculating a value based on a sum of absolute values or a sum of squared residuals. The combination evaluation value can be set to a value determined by the following formula (2), for example. In Expression (2), T _i is a temperature estimation value, T _s is a temperature reference value, and m is the number of temperature sensors used for calculating the combination evaluation value.

The value of n in equation (1) differs for each combination of temperature sensors, but the value of m in equation (2) is the same value for any combination of temperature sensors, and is always used for calculating the combination evaluation value. The same temperature sensor is used. Thereby, a plurality of combinations of temperature sensors can be compared on the same basis. It is preferable to use all temperature sensors installed in the machine tool 1000 for the calculation of the combination evaluation value. This is because the combination evaluation value can be calculated with higher accuracy. An example of the calculated combination evaluation value is shown in FIG.
The method for calculating the combination evaluation value is not limited to the one shown here. For example, (1) for all heat source candidate regions, the assumed heat inflow amount set last in step SB4 and the reference heat set in step SB1. You may calculate by calculating the difference with inflow, and (2) calculating the value based on the sum of the absolute value about the calculated difference, or the residual sum of squares.

  Thereafter, as described in the section “(3) Selection of combinations (steps SB3, SB9, SB10)”, has the temperature sensor selection unit 204 calculated combination evaluation values for all combinations in step SB9? If there is an uncalculated combination (N in Step SB9), the next combination is selected in Step SB10. When there is no uncalculated combination (Y in step SB9), the temperature sensor selection unit 204 completes the combination evaluation value calculation process.

2-4-3. Calculation of temperature sensor evaluation value (step SA3)
Next, in step SA3, the temperature sensor selection unit 204 calculates a temperature sensor evaluation value. The temperature sensor evaluation value is a value serving as an index indicating the usefulness of the temperature sensor 202. By classifying the combination evaluation values calculated for each combination of the temperature sensors 202 based on whether or not a certain temperature sensor is used, the temperature sensor evaluation value of the temperature sensor can be calculated.
Hereinafter, the method for calculating the temperature sensor evaluation value will be described more specifically with reference to the flowchart shown in FIG. 11 using the example of the combination evaluation value shown in FIG.

(1) Selection of the first temperature sensor (step SC1)
First, in step SC1, the temperature sensor selection unit 204 selects the first temperature sensor 202 for calculating the temperature sensor evaluation value. For example, the temperature sensor A in FIG. 10 is selected.

(2) Classification of temperature sensor combinations (step SC2)
Next, in step SC2, the temperature sensor selection unit 204 classifies the combination of temperature sensors into “none group” and “present group” depending on whether or not the selected temperature sensor is included. For example, taking temperature sensor A as an example, in the combination of temperature sensors in the first to sixth rows, the column of temperature sensor A is 0. Therefore, these combinations include temperature sensor A. It is not classified and is classified as a “none group”. Further, in the combinations of the temperature sensors in the seventh to twelfth rows, the column of the temperature sensor A is 1, so that these combinations include the temperature sensor A, and are classified into the “present group”. Is done.

(3) Calculation of no representative value (step SC3)
Next, in step SC3, the temperature sensor selection unit 204 calculates a representative value (none representative value) of the combination evaluation value for the combination of temperature sensors belonging to the none group. The representative value is a value obtained by performing a predetermined arithmetic process on the combination evaluation value for the combination of temperature sensors belonging to the no group, for example, an average value or a total value. Taking the case where the representative value is an average value and the selected temperature sensor is the temperature sensor A as an example, the combination of the temperature sensors in the first row to the sixth row belongs to the “none group”. The average value of the combination evaluation values for the combination is the representative value without any value. Since the combination evaluation values of the temperature sensor combinations in the first to sixth rows are 25.8, 57.4, 51.3, 30.9, 44.6, and 47.9, respectively, the average value thereof is 43.0. Therefore, the representative value without the temperature sensor A is 43.0.

(4) Calculation of presence representative value (step SC4)
Next, in step SC4, the temperature sensor selection unit 204 calculates a representative value (representative representative value) of the combination evaluation value for the combination of temperature sensors belonging to the present group. The representative value is a value obtained by performing predetermined arithmetic processing on a combination evaluation value for a combination of temperature sensors belonging to a certain group, for example, an average value or a total value. Taking the case where the representative value is an average value and the selected temperature sensor is the temperature sensor A as an example, the combination of the temperature sensors in the seventh row to the twelfth row belongs to the “present group”. There is an average value of the combination evaluation values for the combination, which is a representative value. Since the combination evaluation values of the combinations of the temperature sensors in the seventh row to the twelfth row are 36.7, 41.0, 49.9, 38.4, 36.5, and 54.1, respectively, the average value thereof is 42.8. Therefore, the representative value of the temperature sensor A is 42.8.

(5) Calculation of temperature sensor evaluation value (step SC5)
Next, in step SC5, the temperature sensor selection unit 204 calculates a temperature sensor evaluation value based on the difference between the representative representative value and the representative representative value. The temperature sensor evaluation value may be a value that reflects the difference between the representative value with and without the representative value. For example, a value obtained by “(present representative value) − (present representative value)” or “(present representative value) ) / (None representative value) ”. For example, if the temperature sensor evaluation value is “(present representative value) − (no representative value)” and the selected temperature sensor is temperature sensor A, the representative value of temperature sensor A is 42.8, and the temperature Since the representative value without sensor A is 43.0, the temperature sensor evaluation value is -0.2.

(6) Selection of next temperature sensor (step SC6, step SC7)
Next, in step SC6, the temperature sensor selection unit 204 confirms whether or not the temperature sensor evaluation values have been calculated for all the temperature sensors, and when there are uncalculated temperature sensors (N in step SC6). Then, the next temperature sensor is selected (step SC7), and the process returns to step SC2 to calculate the temperature sensor evaluation value for the selected temperature sensor. If there are no uncalculated temperature sensors remaining (Y in step SC6), the temperature sensor evaluation value calculation process ends.

  With respect to the temperature sensors A to K, the absence representative value, the presence representative value, and the temperature sensor evaluation value calculated according to the above-described method are shown in FIG. The average value was used as the representative value, and the temperature sensor evaluation value was calculated by “(present representative value) − (absent representative value)”. The temperature sensor evaluation value is a value serving as an index indicating whether or not the representative value is improved when the temperature sensor is present. The larger the value, the higher the usefulness of the temperature sensor. . In the case of FIG. 12, when the temperature sensors are arranged in descending order from the highest temperature sensor evaluation value, the temperature sensors are arranged in the order of I, J, F, E, G, C, H, A, K, D, and B. .

2-4-4. Temperature sensor selection process (step SA4)
In step SA4, the temperature sensor selection unit 204 selects a temperature sensor based on the temperature sensor evaluation value. Although the selection method of a temperature sensor is not specifically limited, For example, it can carry out by the two methods shown below.

(1) First Method of Selecting Temperature Sensor A first method of selecting a temperature sensor will be described below with reference to the flowchart of FIG. In the first method, a plurality of combinations of temperature sensors are generated in advance, a combination evaluation value is calculated for each combination, and a temperature sensor is selected based on the calculated combination evaluation value.

(1-1) Generating a plurality of combinations of temperature sensors (step SD1)
In step SD1, the temperature sensor selection unit 204 generates a plurality of combinations of temperature sensors by adding temperature sensors to be used in order from the one having the highest evaluation based on the temperature sensor evaluation value. Taking the temperature sensor evaluation values shown in FIG. 12 as an example, first, when the temperature sensors are arranged in descending order from the one having the highest temperature sensor evaluation value, and the temperature sensors used are increased one by one, the table shown in FIG. 14 is obtained. The first to twelfth rows of this table correspond to a plurality of combinations of temperature sensors. For example, in the sixth row, since the columns of the temperature sensors I, J, F, E, and G are 1 and the remaining columns are 0, the sixth row includes the temperature sensors I, J, F, E, and G. Means a combination. Moreover, although not shown in FIG. 14, when there exists a basic temperature sensor, a basic temperature sensor is added to all the combinations. In the case of the first row in FIG. 14, the sensor to be used is only the basic temperature sensor. However, even when only one sensor is used in this way, it is referred to as “combination” in this specification for convenience. Use terminology. In addition, when there are a plurality of temperature sensors having the same evaluation based on the temperature sensor evaluation value, a combination of the temperature sensors may be generated by alternatively selecting the sensors. For example, in the example of FIG. 12, the temperature sensors C and H and the temperature sensors D and K have the same temperature sensor evaluation value. Therefore, even if a combination that uses one of the temperature sensors C and H is generated. Good.

(1-2) Calculate a combination evaluation value for each combination (step SD2)
In step SD2, the temperature sensor selection unit 204 calculates combination evaluation values for all combinations generated in step SD1. The combination evaluation value can be calculated by the method described in steps SB3 to SB10.
FIG. 15 shows an example of a combination evaluation value calculated for a combination obtained by adding a basic temperature sensor to each of the 12 combinations shown in FIG. In the example of FIG. 15, combination evaluation values are shown for each of the combinations of the first row to the twelfth row of FIG. 14. In the first row combination, the combination evaluation value is relatively small, but the combination evaluation value increases every time a temperature sensor is added up to the fourth row combination. In the combination of the fifth row, the combination evaluation value is slightly lower than the combination of the fourth row, but in the combination of the sixth row, the combination evaluation value rises again, and is the maximum value among all the combinations. It has become.

(1-3) Select a combination of temperature sensors based on the combination evaluation value (step SD3)
In step SD3, the temperature sensor selection unit 204 selects a combination of temperature sensors based on the combination evaluation value. The selection method is not particularly limited. For example, the combination having the highest evaluation based on the combination evaluation value may be selected without considering the number of use of the temperature sensor, and the total evaluation of the number of use of the temperature sensor and the combination evaluation value may be selected. A combination may be selected according to. According to the former selection method, the sixth row combination is selected in the example of FIG. In this case, the basic temperature sensor and the temperature sensors I, J, F, E, and G are selected as the temperature measurement unit. Further, according to the latter selection method, there may be a case where the combination in the fourth row is selected in which the number of temperature sensors used is small and the combination evaluation value is relatively high. In this case, the basic temperature sensor and the temperature sensors I, J, and F are selected as the temperature measurement unit.

(2) Second Method of Selecting Temperature Sensor A second method of selecting the temperature sensor will be described below using the flowchart of FIG. In the second method, a combination of temperature sensors is generated one by one, a combination evaluation value is calculated each time the combination is generated, and a combination of temperature sensors is generated when the value of the combination evaluation value satisfies a predetermined condition. When finished, select the temperature sensor. According to this method, since the number of combinations for calculating the combination evaluation value can be reduced as compared with the first method, the calculation time can be shortened.

(2-1) Generate a combination of temperature sensors (Step SE1, Step SE5)
In step SE1, the temperature sensor selection unit 204 generates the first combination of temperature sensors. For example, if there is a basic temperature sensor, the first combination of temperature sensors generates a combination of the basic temperature sensors first, and if there is no basic temperature sensor, the combination of the temperature sensors having the highest temperature sensor evaluation value. Is generated. The number of temperature sensors included in the first combination of temperature sensors may be one or two or more. For example, in the case of two, it can be a combination of a basic temperature sensor and a temperature sensor with the highest temperature sensor evaluation value, or a combination of two temperature sensors selected in descending order of the temperature sensor evaluation value.

  Further, in step SE5, the temperature sensor selection unit 204 generates a combination of temperature sensors by adding temperature sensors to be used in order from the one having the highest evaluation based on the temperature sensor evaluation value. For example, in the case of the example of FIG. 14, temperature sensors are added in the order of temperature sensors I, J, F, E, G, C, H, A, K, D, and B. The number of temperature sensors added may be one by one or two or more. When the first combination of the temperature sensors is the combination of the first row of FIG. 14 and one temperature sensor is added one by one, in step SE5, the second row, the third row,... A combination of temperature sensors is generated.

(2-2) Calculation of combination evaluation value (step SE2)
In step SE2, the temperature sensor selection unit 204 calculates a combination evaluation value for the combination generated in step SE1 or step SE5. The combination evaluation value can be calculated by the method described in steps SB4 to SB8.

(2-3) Comparison of combination evaluation values (step SE3)
In step SE3, the temperature sensor selection unit 204 compares the combination evaluation values before and after adding the temperature sensor. This step is not executed for the first combination generated in step SE1, but is executed only when the combination evaluation value is calculated for the combination generated in step SE5 in step SE2. For example, in the case of the example in FIG. 14, when the combination of the third row is generated in step SE5, the combination evaluation value of the combination of the second row and the combination evaluation value of the combination of the third row are compared in step SE3. .

(2-4) Evaluation of combination evaluation value (steps SE4 and SE5)
In step SE4, the temperature sensor selection unit 204 determines whether or not the decrease in evaluation due to the combination evaluation value before and after adding the temperature sensor exceeds the threshold value. If the threshold is not exceeded (N in step SE4), a temperature sensor is added to generate a new combination in step SE5, and the process returns to step SE2. If the threshold value is exceeded (Y in step SE4), the addition of the temperature sensor is terminated and the process proceeds to step SE6.
For example, in the example of FIGS. 14 and 15, the combination evaluation value increases every time a temperature sensor is added up to the combination of the fourth row, but the combination evaluation value of the combination of the fifth row is the fourth row. The value is lower than the combination evaluation value of the combination. If the difference between these combination evaluation values is smaller than a preset threshold value, the determination at step SE4 is N, and the process proceeds to step SE5 to continue adding the temperature sensor. If the difference between these combination evaluation values is larger than a preset threshold value, the determination at step SE4 is Y, and the addition of the temperature sensor is terminated.

  The threshold value may be 0, but in this case, the addition of the temperature sensor will be terminated even if the combination evaluation value slightly decreases, so that the combination evaluation value will increase greatly when a further temperature sensor is added. Even in this case, since the temperature sensor is not added, the finally obtained combination evaluation value may be a relatively low value. Therefore, for example, there is a possibility that the combination evaluation value can be set to a higher value by setting the threshold value to 10% or the like of the combination evaluation value for the combination before adding the temperature sensor (this threshold value is It may be 5% or 20%.) In the case of the example of FIG. 15, by adding another temperature sensor without ending the addition of the temperature sensor when the combination evaluation value of the combination of the fifth row is lower than that of the fourth row, In the combination of the sixth row, a very high combination evaluation value is obtained.

(2-5) Selection of combination (step SE6)
In step SE6, the temperature sensor selection unit 204 selects the combination having the maximum combination evaluation value from among the combinations of temperature sensors for which the combination evaluation value has already been calculated. For example, in the example of FIG. 14 and FIG. 15, when the addition of the temperature sensor is finished when the combination evaluation value of the fifth row is lower than that of the fourth row, the combination of the first to fourth rows The evaluation values are compared, and the combination having the maximum combination evaluation value (in this case, the combination in the fourth row) is selected. Since the combination of the fourth row includes temperature sensors I, J, and F, the temperature sensors I, J, and F are selected as the temperature measurement unit (if there is a basic temperature sensor, Temperature sensors I, J, and F are selected as temperature measuring units).

2-5. Thermal Displacement Correction Device Next, the thermal displacement correction device 100 constituting the machine tool 1000 will be described below with reference to the block diagram of FIG.

2-5-1. Temperature Data Receiving Unit The thermal displacement correction apparatus 100 includes a temperature data receiving unit 108 that receives an output from the temperature sensor 202 selected by the temperature sensor selecting unit 204. The temperature data received by the temperature data receiving unit 108 is stored in the temperature data storage unit 110.

2-5-2. Heat Inflow Amount Assumption Unit This thermal displacement correction device 100 is provided with a heat inflow amount assumption unit 124 that assumes the heat inflow amount from the heat source candidate region. This heat inflow assumption unit 124 reads the measured structure model data stored in the measured structure model data storage unit 122 described above, and is assumed in the measured structure model data based on a predetermined rule. The amount of heat inflow is linked to the data of the heat source candidate area and written. Thus, the structure model data to be measured in which the heat inflow amount assumed based on the predetermined rule is written is transferred to the temperature distribution estimation unit 126 described later.

2-5-3. Temperature Distribution Estimation Unit This thermal displacement correction device 100 is provided with a temperature distribution estimation unit 126 that generates an estimated temperature distribution of the machine tool based on the heat inflow amount assumed by the heat inflow amount assumption unit as described above. It has been. The temperature distribution estimation unit 126 receives the measured structure model data in which the heat inflow assumed based on a predetermined rule is written from the heat inflow assumption unit 124, and in the case of the assumed heat inflow, The temperature distribution of the structure to be measured is estimated by using a calculation algorithm such as a finite element analysis method to determine the temperature distribution of the structure to be measured after a predetermined time has elapsed. The estimated temperature distribution is stored in the temperature distribution storage unit 128.

2-5-4. Comparison / Gap Analysis Unit In this thermal displacement correction device 100, temperature detection values in a plurality of temperature measurement units detected by the temperature measurement unit stored in the temperature data storage unit 110 and temperature distribution storage are stored. The gap between the temperature detection value and the temperature estimation value is reduced by comparing the temperature estimation value of the location corresponding to the temperature measurement unit in the model data of the structure to be measured stored in the unit 128. As described later, a comparison / gap analysis unit 130 for performing analysis is provided. That is, the comparison / gap analysis unit 130 sets the heat inflow amount to a predetermined rule with respect to the heat inflow amount assumption unit 124 described above so that the gap between the detected temperature value and the estimated temperature value becomes small. Based on this, a signal is transmitted to instruct another value to be re-assumed. Then, the heat inflow amount assumption unit 124 to which such a signal has been transmitted generates heat inflow amount assumption data that is reasserted to another value based on a predetermined rule, and transmits the heat inflow amount assumption data to the temperature distribution estimation unit 126. become. On the other hand, the comparison / gap analysis unit 130 transmits to the heat inflow amount assumption unit 124 and the temperature distribution estimation unit 126 a signal for further repeatedly performing various calculations described so far. Therefore, the temperature distribution estimation unit 126 to which the signal is transmitted is re-asserted to another value based on the heat inflow assumption data re-asserted to another value received from the heat inflow assumption unit 124. In the case of heat inflow, calculate the temperature distribution of the structure under measurement using a calculation algorithm such as the finite element analysis method after a lapse of a predetermined time, and then calculate the temperature distribution of the structure under measurement again. presume. The estimated temperature distribution is stored in the temperature distribution storage unit 128. Thereafter, such calculation is repeated so that the gap between the detected temperature value and the estimated temperature value becomes small.

2-5-5. Repeat Calculation Determination Unit Furthermore, the thermal displacement correction apparatus 100 is provided with a repetition calculation determination unit 132. This iterative calculation determination unit 132 is repeated by the heat inflow amount assumption unit 124 and the temperature distribution estimation unit 126 so that the gap between the temperature detection value and the temperature estimation value becomes small according to the instruction of the comparison / gap analysis unit 130. The calculation result is received every time via the comparison / gap analysis unit 130. Then, the iterative calculation determination unit 132, when the end of the iterative calculation is permitted based on a predetermined rule, for example, because the gap becomes sufficiently small, This is transmitted to the data calculation unit 134. On the other hand, if it is not possible to determine whether to end the repeated calculation, the heat inflow assumption unit 124 and the temperature distribution estimation unit 126 continue to repeat the calculation.

2-5-6. Approximate Data Calculation Unit This thermal displacement correction apparatus 100 is provided with an approximate data calculation unit 134. When the approximate data calculation unit 134 receives a signal for instructing the end from the iterative calculation determination unit 132, based on the result of the iterative calculation, the approximate heat inflow amount and the work flow from the heat source candidate region with the smallest gap described above. Approximate data consisting of combination data of the approximate temperature distribution of the machine is calculated. Then, the approximate data calculated in this way is stored in the approximate data storage unit 136.

2-5-7. Thermal Displacement Correction Data Generation Unit And this thermal displacement correction device 100 is provided with a thermal displacement correction data generation unit 139. This thermal displacement correction data generation unit 139 is based on the approximate data acquired from the approximate data storage unit 136 and is required to cancel the thermal displacement amount derived at each position of the three-dimensional structure of the structure to be measured. The amount is calculated, and the thermal displacement correction data written by attaching the thermal displacement correction amount to each position of the three-dimensional structure of the structure to be measured is generated. The thermal displacement correction data obtained in this way is output to the drive mechanism of the multi-axis control machine tool outside the thermal displacement correction device 100 via the output unit 138 of the thermal displacement correction device 100, and is subjected to Used to improve the machining accuracy of the cutting material. Alternatively, the thermal displacement correction data obtained in this manner is output to a server, network, printer, or the like outside the thermal displacement correction device 100 via the output unit 138 of the thermal displacement correction device 100, so that the operator May be visually observed as image data on a screen or print on paper.

2-6. FIG. 18 is an enlarged functional block diagram for explaining in detail the configuration of a part of the multi-axis control machine tool according to the embodiment. First, the data in the temperature distribution storage unit 128 and the temperature data storage unit 110 are transferred to the comparison / gap calculation unit 130. In the comparison / gap analysis unit 130, the residual square sum calculation unit 304 calculates the residual square sum of the data calculated by the difference data calculation unit 302 and passes it to the iterative calculation determination unit 132. Based on the data stored in the threshold storage unit 308, the iterative calculation determination unit 132 determines the residual square sum and the threshold value in the residual square sum / threshold magnitude determination unit 306. This iterative calculation determination unit 132 is repeated by the heat inflow amount assumption unit 124 and the temperature distribution estimation unit 126 so that the gap between the temperature detection value and the temperature estimation value becomes small according to the instruction of the comparison / gap analysis unit 130. The calculation result is received every time via the comparison / gap analysis unit 130. Then, the iterative calculation determination unit 132, when the end of the iterative calculation is permitted based on a predetermined rule, for example, because the gap becomes sufficiently small, This is transmitted to the data calculation unit 134. On the other hand, if it is not possible to determine whether to end the repeated calculation, the heat inflow assumption unit 124 and the temperature distribution estimation unit 126 continue to repeat the calculation.

2-6-1. Details of the Comparison / Gap Analysis Unit As shown in FIG. 6, the comparison / gap analysis unit 130 obtains a difference data calculation unit 302 that obtains a difference by associating data corresponding to each other and a difference data calculation unit 302. There is provided a residual sum of squares calculation unit 304 for calculating the residual sum of squares of the difference of the plurality of terms.

  The comparison / gap calculation unit 130 receives data from the external temperature data storage unit 110 and the temperature distribution storage unit 128 as shown in the figure. Then, in the comparison / gap analysis unit 130, the corresponding data acquired from the temperature data storage unit 110 and the temperature distribution storage unit 128 are associated with each other by the difference data calculation unit 302, and a difference is extracted for each item. . The difference data extracted in this way is transferred from the difference data calculation unit 302 to the residual sum of squares calculation unit 304, and the residual sum of squares of the difference of the obtained multiple terms is calculated. Then, the obtained data regarding the residual sum of squares is passed to the iterative calculation determination unit 132.

2-6-2. Details of the Iterative Calculation Determining Unit The iterative calculation determining unit 132 includes a threshold value storage unit 308 that stores a threshold value of the residual sum of squares for use in determining whether to continue the iterative calculation, and the threshold value storage unit 308. A residual sum of squares / threshold magnitude determination unit 306 is provided that determines the sum of the squares of the residuals and the magnitude between the thresholds using the threshold values acquired from the above. That is, in the iterative calculation determination unit 132, based on the data stored in the threshold storage unit 308, the residual sum of squares obtained from the residual sum of squares calculation unit 304 by the residual square sum / threshold magnitude determination unit 306 Whether the threshold value is called from the threshold value storage unit 308 or not is determined. Then, the determination result of the residual sum of squares / threshold magnitude determination unit 306 thus obtained is passed to the comprehensive determination unit 318.

  In addition, the iterative calculation determination unit 132 includes a time calculation unit 314 that measures the calculation time required for the repetitive calculation, a number calculation unit 316 that counts the number of times the repetitive calculation is performed, and the calculation time and An upper limit storage unit 312 that stores a preset upper limit for the number of calculations is provided. In addition, the iterative calculation determination unit 132 includes the calculation time and the number of calculations acquired from the time calculation unit 314 and the number of times calculation unit 316, and the calculation time and the upper limit of the number of calculations acquired from the upper limit value storage unit 312. In comparison, a time / number of times upper limit attainment determination unit 310 for determining the size of these is provided. That is, in the repetitive calculation determination unit 132, based on the data stored in the upper limit value storage unit 312, the time / number upper limit reaching determination unit 310 calculates the calculation time / time obtained from the time calculation unit 314 and the number calculation unit 316. A determination is made as to whether the number of calculations is the calculation time acquired from the upper limit storage unit 312 and the upper limit of the number of calculations. Then, the determination result of the time / times upper limit reaching determination unit 310 obtained in this way is passed to the comprehensive determination unit 318.

  In addition, the iterative calculation determination unit 132 acquires the determination result of the residual sum of squares / threshold magnitude determination unit 306 and the determination result of the time / number of times upper limit determination unit 310, and comprehensively determines these determination results. And a comprehensive determination unit 318 for determining whether or not to perform repeated calculation. For example, the overall determination unit 318 acquires the determination result of the residual sum of squares / threshold magnitude determination unit 306 and the determination result of the time / number of times upper limit determination unit 310, and both of these determination results are obtained. If the determination result indicates that it is not necessary to end the repeated calculation, the comprehensive determination unit 318 similarly gives the comprehensive determination result (continuation determination result) that it is not necessary to end the repeated calculation. Then, the continuation determination result is transmitted from the comprehensive determination unit 318 to the heat inflow amount assumption unit 124. Then, as already described, the heat inflow amount assumption unit 124 generates heat inflow amount assumption data re-assumed to another value based on a predetermined rule, and transmits it to the temperature distribution estimation unit 126, and thereafter The above repeated calculation will continue.

  On the other hand, if at least one of these determination results is a determination result that the repeated calculation should be terminated, the comprehensive determination unit 318 gives a comprehensive determination result (end determination result) that the repeated calculation should be ended. I will give you. The end determination result is transmitted to the approximate data calculation unit 134. Then, as already described, the approximate data calculation unit 134 is a combination of the approximate heat inflow amount from the heat source candidate region and the approximate temperature distribution of the machine tool, in which the gap is minimized based on the result of the repeated calculation. Approximate data consisting of data is calculated.

2-7. Operation of Multi-axis Control Machine Tool Hereinafter, the operation of the multi-axis control machine tool according to the present embodiment will be described. FIG. 19 is a flowchart for explaining the operation of the multi-axis control machine tool according to this embodiment.

2-7-1. Input reception / model data generation process (step SF1)
In step SF1, the input reception / model data generation unit 101 receives various data inputs from the operator of the machine tool, and generates measured structure model data based on the input data. Hereinafter, the input reception / model data generation process will be described in detail with reference to the flowchart of FIG.

(1) Setting of heat source candidate area (step SG1)
In step SG1, the heat source candidate region setting unit 104 accepts the setting of the heat source candidate region. Specifically, for example, when a series of operations is started by turning on the power of the machine tool 1000, an operator of the machine tool 1000 is displayed on a liquid crystal screen attached to the machine tool 1000. A screen that prompts the user to set, as a heat source candidate region, a region that is likely to be a heat source in the three-dimensional structure of the machine tool based on his / her past experience is displayed. Then, the operator of this machine tool uses an input device attached to the machine tool 1000 and based on his / her past experience, an area that is likely to be a heat source in the three-dimensional structure of the machine tool 1000 is selected as a heat source candidate. It will be set as an area.

(2) Temperature sensor position data input (step SG2)
In step SG2, the temperature sensor position data receiving unit 116 receives an input of the position of the temperature sensor 202. Specifically, for example, an area in which the temperature sensor 202 is installed in the three-dimensional structure of the machine tool 1000 is set for the operator of the machine tool 1000 on a liquid crystal screen attached to the machine tool 1000 or the like. A screen prompting you to do so is displayed. Then, the operator of the machine tool sets an area where the temperature sensor 202 is installed using an input device attached to the machine tool 1000 or the like. Alternatively, if the region where the temperature sensor is installed is a fixed region every time, the machine tool 1000 records the region where the temperature sensor prepared in advance is recorded instead of the above screen display. A data table or the like may be read. In any case, the position of the temperature sensor 202 is input in this way.

(3) Input of structure / physical property data of structure to be measured (step SG3)
In step SG3, the structure / physical property data receiving unit 118 to be measured receives an input of structure / physical property data of the structure to be measured. Specifically, a screen prompting the operator of the machine tool 1000 to input the three-dimensional structure data and material property data of the machine tool 1000 is displayed on a liquid crystal screen attached to the machine tool 1000 or the like. . Then, the operator of the machine tool inputs the three-dimensional structure data and material property data of the machine tool 1000 using an input device attached to the machine tool 1000 or the like. Alternatively, if the three-dimensional structure data and the material property data of the machine tool 1000 are the same three-dimensional structure data and material property data every time, the machine tool 1000 uses the prepared three-dimensional structure data and You may read 3D CAD data (with material physical property data) etc. which recorded material physical property data. In any case, the three-dimensional structure data and the material property data are input in this way.

(4) Generation of structure model data to be measured (step SG4)
In step SG4, the measured structure model data generation unit 120 generates measured structure model data. Specifically, the measured structure model data generation unit 120 aggregates information about the heat source candidate region, the temperature sensor position, and the structure / physical property data of the measured structure input in steps SG1 to SG3. The measured structure model data is generated by combining these data with each other so as to be data suitable for use as a target of various 3D simulations by a computer.

2-7-2. Selection process of temperature sensor used as temperature measuring unit (step SF2)
In step SF2, the temperature sensor selection unit 204 selects the temperature sensor 202 to be used as the temperature selection unit from the temperature sensors 202 installed in the machine tool 1000. This selection can be performed by the method described in FIG. 6 and “2-4. Temperature sensor selection unit”.

2-7-3. Thermal displacement correction process (step SF3)
In step SF3, the thermal displacement correction apparatus 100 performs a thermal displacement correction process. Hereinafter, the thermal displacement correction process will be described in detail with reference to the flowchart of FIG.

(1) Obtaining the temperature of the machine tool from the temperature measurement unit (step SH1)
In step SH1, the temperature data receiving unit 108 acquires the temperature of the machine tool in the temperature selection unit selected by the temperature sensor selection unit 204.

(2) Setting of assumed heat inflow (step SH2)
In step SH2, the heat inflow amount assumption unit 124 assumes the heat inflow amount flowing from each heat source candidate region. The heat inflow assumed here is used to calculate an estimated temperature distribution, which becomes a reference when performing a comparison / gap analysis in step SH4, which will be described later, in step SH3. The heat inflow assumed here is hereinafter referred to as “assumed heat inflow”. The assumed heat inflow amount is repeatedly set until the result of the comparison / gap analysis in step SH4 satisfies the end condition defined in step SH5. The method for setting the initial value of the assumed heat flow rate is not particularly limited. However, in view of the purpose of accurately estimating the heat inflow amount flowing from the heat source candidate region, the heat amount actually generated by each heat source candidate region is determined in the past. It is preferable to set a value that is as close to the true value as possible as an initial value by inferring from the above knowledge and motor specifications.

(3) Calculation of estimated temperature distribution (step SH3)
In step SH3, the temperature distribution estimation unit 126 calculates an estimated temperature distribution. Estimated temperature distribution is a calculation algorithm such as the finite element analysis method that shows the temperature distribution of the measured structure after a lapse of a predetermined time when the assumed heat inflow flows into the measured structure model data. It is the temperature distribution of the structure to be measured obtained by calculating using.

(4) Comparison / gap analysis / end determination (steps SH4, SH5)
In step SH4, the comparison / gap analysis unit 130 compares the temperature detection value detected by the temperature measurement unit with the estimated temperature distribution of the structure to be measured after the predetermined time elapses. The gap is analyzed so that the gap between the temperature estimates becomes small.
Subsequently, in step SH5, the iterative calculation determination unit 132 checks whether or not the end condition of the iterative calculation is satisfied. If it is determined that it is satisfied (Y in step SH5), the iterative calculation ends. Then, the process proceeds to step SH6. On the other hand, if it is determined that the end condition is not satisfied (N in step SH5), the process returns to step SH2 and the calculation is repeated.

(5) Calculation of approximate data (step SH6)
In step SH6, the approximate data calculation unit 134 calculates the approximate heat inflow amount from the heat source candidate region and the approximate temperature distribution of the machine tool where the gap is minimized based on the results of the repeated calculations performed so far. Approximate data consisting of combination data is calculated.

(6) Generation of thermal displacement correction data (step SH7)
In step SH7, the thermal displacement correction data generation unit 139 is required to cancel the thermal displacement amount derived for each position of the three-dimensional structure of the structure to be measured based on the approximate data calculated in step SH6. The correction amount is calculated, and the thermal displacement correction data is generated by linking the thermal displacement correction amount to each position of the three-dimensional structure of the structure to be measured.

(7) Data output (step SH8)
In step SH8, the output unit 138 outputs the thermal displacement correction data generated in step SH7 to the drive mechanism 300 of the multi-axis control machine tool outside the thermal displacement correction device 100.

2-7-4. Machine tool operation (step SF4)
In step SF4, the machine tool 1000 performs thermal displacement correction according to the data output in step SH8, and processes the workpiece with the drive mechanism 300.

2-7-5. Thermal displacement recorrection (step SF5)
The machine tool 1000 continues the operation of step SF4 for a while, but if the operation time becomes longer, the thermal displacement correction data output in step SH8 is appropriate in the thermal displacement correction data output in step SH8 due to the accumulation of heat generated during processing and the change in ambient temperature. May not be possible. Accordingly, in step SF5, the machine tool 1000 confirms whether or not the thermal displacement re-correction condition is satisfied. If satisfied (Y in step SF5), the machine tool 1000 returns to step SF3 and again heats up. Displacement correction processing is performed. If not satisfied (N in step SF5), the process returns to step SF4 and the operation is continued. The conditions for thermal displacement re-correction are as follows: (1) When the processing time exceeds the threshold, (2) When the temperature change at the temperature measurement unit exceeds the threshold, (3) The change in ambient temperature exceeds the threshold Cases.

2-7-6. Temperature sensor reselection (step SF6)
In addition, when the accumulation of heat generated during processing and the change in the ambient temperature are large, the accuracy of thermal displacement correction may be improved by changing the temperature sensor used as the temperature measurement unit. Therefore, in step SF6, the machine tool 1000 confirms whether or not the temperature sensor reselection condition is satisfied. If satisfied (Y in step SF6), the machine tool 1000 returns to step SF2 to again check the temperature. Select the temperature sensor to be used as the measurement unit. If not satisfied (N in step SF5), the process returns to step SF4 and the operation is continued. The conditions for re-selecting the temperature sensor are as follows: (1) When the processing time exceeds the threshold, (2) When the temperature change in the temperature measurement unit exceeds the threshold, (3) The change in ambient temperature exceeds the threshold Cases. Since the temperature sensor reselection does not need to be performed as frequently as the thermal displacement recorrection, it is preferable that the temperature sensor reselection condition is set to be gentler than the thermal displacement recorrection condition.

2-8. Effects of Multi-axis Control Machine Tool Hereinafter, effects of the multi-axis control machine tool according to the present embodiment will be described.
The machine tool 1000 according to the present embodiment is a machine tool 1000 that processes a work material into a predetermined shape by relatively moving a tool and the work material by a plurality of drive mechanisms. Here, the machine tool 1000 includes a plurality of temperature sensors 202, a temperature sensor selection unit 204 that selects a temperature sensor to be used as a temperature measurement unit, and a heat source candidate region setting unit that sets a heat source candidate region in the machine tool 1000. 104 is provided. Further, the machine tool 1000 is provided with a heat inflow amount assumption unit 124 that assumes the heat inflow amount from the heat source candidate region. Further, the machine tool 1000 is provided with a temperature distribution estimation unit 126 that estimates the temperature distribution of the machine tool 1000 based on the heat inflow amount assumed by the heat inflow amount assumption unit 124. In addition, the machine tool 1000 compares the temperature detection value by the temperature measurement unit with the temperature estimation value at the location corresponding to the temperature measurement unit in the temperature distribution estimated by the temperature distribution estimation unit 126. In order to reduce the gap between the temperature detection value and the estimated temperature value, the heat inflow amount assumption unit 124 reassumes the heat inflow amount to another value, and the heat inflow amount assumption unit 124 and the temperature distribution estimation unit 126. A comparison / gap analysis unit 130 is provided for repeatedly performing the calculation. The machine tool 1000 is provided with an approximate data calculation unit 134 that calculates the approximate heat inflow amount from the heat source candidate region and the approximate temperature distribution of the machine tool 1000 based on the result of the repeated calculation.

  Since the machine tool 1000 according to the present embodiment has such a configuration, the temperature sensor selection unit calculates a temperature sensor evaluation value indicating the usefulness of each temperature sensor, and the temperature measurement unit is based on the temperature sensor evaluation value. As a temperature sensor to be used is selected, a number of temperature sensors are set in advance, and among these temperature sensors, a highly useful one is used as a temperature measurement unit to search for approximate heat inflow, The accuracy of the approximate heat inflow amount can be increased while minimizing the number of temperature measuring units.

  Further, in the machine tool 1000 according to the present embodiment, the heat inflow amount assumption unit 124 is reasserted to another value so that the gap between the detected temperature value and the estimated temperature value becomes small. By causing the heat inflow amount assumption unit 124 and the temperature distribution estimation unit 126 to repeatedly calculate, an approximate heat inflow amount in which the temperature estimated value is close to the temperature detection value can be searched. Therefore, if the machine tool 1000 of the present embodiment is used, based on the approximate heat inflow amount obtained in this way, the approximate heat inflow amount is applied to the model data of the machine tool 1000, so that the thermal displacement in the machine tool 1000 is achieved. Data can be estimated, and accurate thermal displacement correction can be performed based on the thermal displacement data.

  Further, in the machine tool 1000 according to the present embodiment, the comparison / gap analysis unit 130 causes the heat inflow amount assumption unit 124 so that the residual sum of squares between the temperature detection value and the temperature estimation value becomes small. And a residual sum of squares calculation unit 304 that causes the temperature distribution estimation unit 126 to perform calculation repeatedly.

  Since the machine tool 1000 according to the present embodiment has such a configuration, the residual sum of squares calculator 304 appropriately sets the gap between the detected temperature value and the estimated temperature value as the residual sum of squares. And can be evaluated efficiently. Therefore, it is possible to search for the approximate heat inflow amount that makes the estimated temperature value close to the detected temperature value appropriately and efficiently.

  Furthermore, in the machine tool 1000 according to the present embodiment, the residual sum of squares calculation unit 304 assumes the heat inflow amount so that the residual sum of squares between the detected temperature value and the estimated temperature value is smaller than a predetermined threshold value. And a residual sum of squares / threshold magnitude determination unit 306 that causes the unit 124 and the temperature distribution estimation unit 126 to perform calculation repeatedly.

  Since the machine tool 1000 according to the present embodiment has such a configuration, the calculation is continued until the residual sum of squares becomes smaller than a predetermined threshold value, so that a certain level of accuracy is ensured in the approximate calculation. In addition, since iterative calculation can be terminated when the residual sum of squares becomes smaller than a predetermined threshold value, approximate calculation can be efficiently performed without extra calculation.

  Furthermore, in the machine tool 1000 according to the present embodiment, the residual sum of squares calculation unit 304 repeatedly performs calculations on the heat inflow amount assumption unit 124 and the temperature distribution estimation unit 126 until the number of repeated calculations reaches a predetermined number. The number calculation part 316 to be performed is included.

  Since the machine tool 1000 according to the present embodiment has such a configuration, it is possible to end the repeated calculation when the number of repeated calculations reaches a predetermined number of times. Therefore, the machine tool 1000 can be efficiently performed without performing unnecessary calculations. Approximate calculation can be performed.

  In the machine tool 1000 according to the present embodiment, the residual sum of squares calculation unit 304 causes the heat inflow amount assumption unit 124 and the temperature distribution estimation unit 126 to wait until the time required for repeated calculation reaches a predetermined length. A time calculation unit 314 that repeatedly performs calculation is provided.

  Since the machine tool 1000 according to the present embodiment has such a configuration, it is possible to finish the repeated calculation when the time required for the repeated calculation reaches a predetermined length. And approximate calculation can be performed efficiently.

  Furthermore, in the machine tool 1000 according to the present embodiment, the thermal displacement correction apparatus 100 includes a heat source candidate area setting unit 104 that receives an input of setting conditions for a heat source candidate area by an operator of the machine tool 1000. This makes it possible to set conditions based on the experience of the operator, etc., so it is only necessary to narrow down the heat source candidate area and make assumptions about the heat inflow only within a limited area. Since it can be approximated well, as a result, thermal displacement correction can be performed efficiently and accurately.

2-9. Other Configurations Embodiments of the present invention have been described above with reference to the drawings. However, these are examples of the present invention, and various configurations other than those described above can be employed.

  For example, in the machine tool 1000 according to the present embodiment, a detailed description is omitted, but the heat inflow amount assumption unit 124 randomly assumes an initial value of the heat inflow amount in the heat source candidate region set as described above. It may have a random numerical assumption part (not shown). In this way, since the initial value of the heat inflow amount is assumed at random, and the calculation can be repeatedly performed many times, the possibility of the initial value of any heat inflow amount can be simulated in detail. As a result, it is possible to calculate the approximate heat inflow amount with high accuracy. This also applies to the step of setting the assumed heat inflow amount in step SB4.

  Furthermore, in the machine tool 1000 according to the present embodiment, although detailed description is omitted, the heat inflow amount assumption unit 124 assumes the heat inflow amount assumed in the previous calculation based on the determination result of the repeated calculation determination unit 132. May be randomly varied to have a random variation part (not shown) that re-assums the next heat inflow amount. In this way, based on the determination result of the repeated calculation determination unit 132, the heat inflow amount assumed in the previous calculation is randomly changed in a form based on a predetermined rule, and the next heat inflow amount is re-assumed. In addition, it is possible to perform exhaustive and exploratory simulations about the possibility of the initial value of any heat inflow by repeatedly performing heuristic calculations over and over, resulting in a highly accurate approximate heat inflow. Can be calculated efficiently. This also applies to the step of setting the assumed heat inflow amount in step SB4.

  Further, in the machine tool 1000 according to the present embodiment, although detailed description is omitted, the temperature distribution estimation unit 126 is assumed to be measured structure model data including information on the three-dimensional structure and physical properties of the machine tool 1000. You may have the finite element analysis part (not shown) which estimates temperature distribution by finite element analysis by applying heat inflow. In this way, by estimating the temperature distribution by finite element analysis, a simple calculation formula is created for each block obtained by finely dividing the three-dimensional structure of the machine tool 1000, and the calculation result is expressed as a boundary condition between the blocks. By exchanging with each other, it is possible to improve the calculation efficiency while maintaining the accuracy of the calculation as a whole. Therefore, it is possible to perform an analysis with higher accuracy and efficiency compared to other analysis methods. This also applies to the temperature distribution generation step in steps SB2 and SB5.

  Further, in the machine tool 1000 according to the present embodiment, although the description is omitted in detail, the approximate data calculation unit 134 selects the combination of the heat inflow amount and the temperature distribution that minimizes the gap among the series of results of the repeated calculation. The minimum gap selection unit (not shown) may be selected as the approximate heat inflow amount and the approximate temperature distribution. In this way, by repeating the calculation over and over again, we simulated the possibility of the initial values of all heat inflows and calculated the approximate heat inflow and approximate temperature distribution of many combinations. By selecting a combination that minimizes the gap from among them, it is possible to calculate a combination of approximate heat inflow and temperature distribution with high accuracy as a result.

  Further, the machine tool 1000 according to the present embodiment omits a detailed description, but based on the approximate temperature distribution calculated by the approximate data calculation unit 134, a thermal displacement estimation unit that estimates the thermal displacement data of the machine tool 1000. (Not shown) may further be provided. As described above, if the heat displacement at each position of the machine tool 1000 based on the estimated heat inflow amount is estimated after the heat inflow amount is efficiently estimated, the heat inflow amount is based on the estimated heat displacement. Thermal displacement correction becomes possible.

  Furthermore, although detailed description of the machine tool 1000 according to the present embodiment is omitted, movement target position data of the moving body based on the thermal displacement data of the machine tool 1000 estimated by a thermal displacement estimation unit (not shown). It further has a thermal displacement correction data generation unit 139 for calculating the correction amount. As described above, by calculating the correction amount of the movement target position data of the moving body based on the thermal displacement data, the thermal displacement correction based on the estimated thermal displacement becomes possible, and the accuracy can be improved regardless of the temperature change. High work material processing is possible.

  The machine tool 1000 according to the present embodiment further includes an axial feed correction unit (not shown) that corrects the axial feed of the cutting unit based on the correction amount calculated by the thermal displacement correction data generation unit 139. . In this way, it is possible to correct the axial feed of the cutting portion based on the calculated correction amount, and thus it is possible to process the work material with high accuracy.

  In the above embodiment, the heat source candidate region setting unit 104 sets a heat source candidate region with one type of heat source pattern, and the temperature sensor selection unit 204 calculates a temperature sensor evaluation value for this heat source pattern, and the calculated temperature sensor Although the temperature sensor is selected based on the evaluation value, it is difficult to set one type of heat source pattern when the heat source position cannot be specified. In such a case, the heat source candidate region setting unit 104 sets heat source candidate regions with a plurality of heat source patterns based on the orthogonal table, and the temperature sensor selection unit 204 calculates a temperature sensor evaluation value for each heat source pattern, It is preferable that the temperature sensor used as the temperature measuring unit is selected based on the average value. By adopting such a configuration, even when the heat source position cannot be specified, the number and arrangement of the temperature measuring units can be appropriately determined.

3. Embodiment of Method for Determining Number and Arrangement of Temperature Measuring Units of Machine Tool So far, the embodiment of the machine tool has been described so far, but the technical idea embodied by the temperature sensor selecting unit 204 is a machine tool Even in a computer or the like away from the computer, it can be embodied in a similar form. In this respect, the present invention provides a method for determining the number and arrangement of temperature measuring parts of a machine tool in a computer or the like separate from the machine tool. The following method can be realized by a computer program that causes a computer to execute each step.

  A method for determining the number and arrangement of temperature measuring units of a machine tool according to an embodiment of the present invention is a method of machining a work material into a predetermined shape by relatively moving a tool and the work material by a plurality of drive mechanisms. A method for determining the number and arrangement of temperature measuring units of a machine, based on the step of setting a heat source candidate region and a plurality of evaluation points on a model of the machine tool, and whether to use each of the evaluation points A step of generating a plurality of combinations of evaluation points according to an orthogonal table, a step of calculating a combination evaluation value for each of the above combinations, and whether or not a certain evaluation point is used for the above combination evaluation value calculated for each combination Calculating the evaluation point evaluation value of the evaluation point by classifying based on the evaluation point, and the evaluation point Comprising the step of selecting an evaluation point of placing the temperature measurement section based on bets evaluation value.

This method should be executed according to the description in “2-3. Input reception / model data generation unit” and “2-4. Temperature sensor selection unit” with “temperature sensor” replaced by “evaluation point”. Can do. That is, in the above description of the machine tool, the position where the temperature sensor is actually installed in the machine tool is set as the position of the temperature sensor. However, in the method of the present embodiment, instead of setting the position of the temperature sensor. Then, a point that is a candidate for a position where the temperature measurement unit is arranged is set as an evaluation point at an arbitrary position, and the subsequent steps are executed.
That is, the input reception / model data generation unit 101 receives the input of the positions of a plurality of evaluation points instead of the position of the temperature sensor, and in step SA1, generates a plurality of combinations of evaluation points according to the orthogonal table. Then, a combination evaluation value is calculated for each generated combination. In step SA3, an evaluation point evaluation value is calculated for each evaluation point. In step SA4, an evaluation point is selected based on the evaluation point evaluation value. The selected evaluation point is a position where the temperature measurement unit is arranged. For example, if evaluation points I, J, F, E, and G are selected in step SA4, temperature sensors are arranged at these evaluation points to form a temperature measurement unit.

  The temperature measuring unit whose number and arrangement are determined by such a method is a machine tool that relatively moves a tool and a work material by a plurality of drive mechanisms to process the work material into a predetermined shape. A heat source candidate region setting unit for setting a heat source candidate region in the machine tool, a heat inflow amount assumption unit for assuming a heat inflow amount from the heat source candidate region, and the heat inflow amount. A temperature distribution estimation unit that generates an estimated temperature distribution of the machine tool based on the heat inflow assumed by the assumption unit, a temperature detection value by the temperature measurement unit, and a temperature distribution estimated by the temperature distribution estimation unit Of these, the heat inflow amount is input to the heat inflow amount assumption unit so that the gap between the temperature detection value and the temperature estimation value is reduced by comparing with the temperature estimation value of the portion corresponding to the temperature measurement unit. Temporary to another value And a comparison / gap analysis unit that causes the heat inflow amount assumption unit and the temperature distribution estimation unit to repeatedly calculate, and an approximate heat inflow amount from the heat source candidate region and the work based on the result of the iterative calculation. The present invention can be employed in a machine tool including an approximate data calculation unit that calculates an approximate temperature distribution of a machine.

In the machine tool described in “2. Embodiment of Machine Tool”, a large number of temperature sensors are installed in the machine tool in advance, and a highly useful temperature sensor is selected as the temperature measurement unit from among the temperature sensors. For this reason, it is possible to respond flexibly to changes in the thermal environment to which the machine tool is exposed.
On the other hand, in the method of the present embodiment, a highly useful evaluation point is selected in advance on a computer or the like, a temperature sensor is installed at the evaluation point, and the temperature sensor is used as a temperature measuring unit. Therefore, there is an advantage that the number of installed temperature sensors can be minimized.

  EXAMPLES Hereinafter, although an Example demonstrates this invention further, this invention is not limited to these.

<Example 1>
In Example 1, a plurality of evaluation points were installed on the model, an evaluation point evaluation value was calculated for each evaluation point, and an evaluation point for arranging the temperature measurement unit was selected based on the calculated evaluation point evaluation value. Details of the present embodiment will be described below.

In this example, a model having the shape shown in FIG. 22 was used. This model has a shape in which a cube shape having a side of 10 mm protrudes from a corner of a rectangular parallelepiped of 50 mm × 50 mm × 20 mm. The heat source Q1 is disposed on the upper surface of the cube, and the heat inflow amount from Q1 to the model is 0.25 W (heat flux: 2500 W / m ² ). The heat source Q2 is an area having a width of 10 mm around the cube, and the heat inflow amount from Q2 to the model is 0.75 W (heat flux: 2500 W / m ² ). The heat source Q3 is a region in which Q1 and Q2 are projected on the lower surface, and the heat inflow amount from Q3 to the model is 1.00 W (heat flux: 2500 W / m ² ). The symmetry plane shown in FIG. 22 was a heat insulation plane, and the plane other than the symmetry plane and the heat source was a heat dissipation plane (heat transfer coefficient: 10 W / (m ² k)). Eleven evaluation points (f03, f05, f13, f14, f30, f31, f32, f33, f34, f35, f36) were set on the upper surface of the model at the positions indicated by the circles in FIG. Further, basic evaluation points f12 and f15 were set at positions indicated by triangles in FIG. In the simulation of this example, one of f12 and f15 was used as the basic evaluation point.
When the temperature distribution was calculated by simulation for the model created under the above conditions, the results shown in FIG. 24 were obtained. In the positions indicated by MAX and MIN in FIG. 24, the temperature rise amounts were the highest (32.995K) and the lowest (30.800K), respectively.
The steps so far correspond to step SF1 in FIG. 19 and part of step SF2 (steps SB1 to SB2 (step SA2) shown in FIG. 9).

Then, according to the orthogonal table L ₁₂ shown in FIG. 25, to create various combinations of 11 evaluation points described above. First row to the 12th row of the orthogonal table of L ₁₂ represents a combination of the first to twelfth evaluation point. “0” and “1” in the orthogonal table mean “do not use” and “use” the evaluation points, respectively. For _example, line 10 of the orthogonal table of _{L 12} is a column ABCHIK is 1, the remaining columns is zero. Thus, this row shows a combination of evaluation points that use the evaluation point assigned to column ABCHIK. This step corresponds to step SA1 in FIG.

Next, according to steps SB3 to SB10 (step SA2) of FIG. 9, a combination evaluation value was calculated for each combination. In step SB4, the assumed heat flow rate was set so as to maximize the gap evaluation value shown in the following formula (3) (to minimize the difference between T _i and T _s ). In Equation (1), T _i is the temperature (temperature estimated value) on the estimated temperature distribution generated in step SB5, and T _s is the temperature (temperature reference value) on the reference temperature distribution generated in step SB2. N is the number of evaluation points included in the combination of evaluation points selected in step SB3 or SB10.

  This calculation was performed using an optimization method called PAO in commercially available software called iSIGHT (manufactured by Engineers Japan, version 10.0). In any of Q1 to Q3, the maximum value of the assumed heat flow rate was set to 2.0, and the initial value and the minimum value were set to 0.1. The optimization calculation was completed after 3 hours.

Moreover, the combination evaluation value was computed according to following formula (4), for example. In Equation (4), T _i is a temperature estimated value, T _s is a temperature reference value, and m is the number of evaluation points used for calculating a combined evaluation value. For the calculation of the combination evaluation value, 14 evaluation points f03, f05, f11, f12, f13, f14, f15, f30, f31, f32, f33, f34, f35, and f36 shown in FIG. 23 were used.

  FIG. 26 shows the combination evaluation values calculated when f12 is the basic evaluation point and f15 is the evaluation point. FIG. 27 shows the assumed heat inflow amount set last when f12 is set as the basic evaluation point, and FIG. 28 shows the assumed heat inflow amount set last when f15 is set as the basic evaluation point. Comparing FIG. 26 with FIG. 27 and FIG. 28, it can be seen that the higher the combined evaluation value, the closer the value of the assumed heat flow rate is to the “solution”. For example, when the basic evaluation point is f12, the combination of the second row and the twelfth row has a very large combination evaluation value, but the values of the assumed heat flow rates Q1 to Q3 are also close to “solution”. On the other hand, in the combination of the fourth row, the values of the assumed heat flow rates Q1 to Q3 are greatly deviated from “solution” when the combination evaluation value is relatively small.

  Next, an evaluation point evaluation value was calculated for each evaluation point according to steps SC1 to SC7 (step SA3) in FIG. FIG. 29 shows the result when f12 is the basic evaluation point, and FIG. 30 shows the result when f15 is the basic evaluation point. The average value was adopted as the representative value, and the evaluation point evaluation value was defined by “(present value) − (representative value)”.

  Next, according to step SD1 (step SA4) in FIG. 13, a plurality of combinations of evaluation points was generated by adding evaluation points to be used in order from the one having the highest evaluation point evaluation value. The results are shown in FIG. 31 (f12 is the basic evaluation point) and FIG. 32 (f15 is the basic evaluation point).

  Next, a combination evaluation value was calculated for s00 to s11 according to step SD2 (step SA4). The results are shown in FIG. 33 (f12 is the basic evaluation point) and FIG. 34 (f15 is the basic evaluation point).

Next, according to step SD3 (step SA4), the combination with the highest combination evaluation value was selected. When f12 is a basic evaluation point, the combination evaluation value of job s05 is the highest, so this combination was selected. Since this combination includes evaluation points f12, f34, f35, f31, f30, and f32, the temperature measurement unit is arranged at the position of this evaluation point. When f15 is the basic evaluation point, the combination evaluation value of job s05 is the highest, and this combination was selected. Since this combination includes evaluation points f15, f31, f36, f13, f03, and f34, the temperature measurement unit is arranged at the position of this evaluation point. Note that when f15 is a basic evaluation point, the job s03 and the job s05 have substantially the same combination evaluation value, so the job s03 can be selected in consideration of the small number of evaluation points. Since the combination of job s03 includes evaluation points f15, f31, f36, and f13, the temperature measurement unit can be arranged at the position of this evaluation point.
Further, comparing FIG. 33 with FIG. 34, the combination evaluation value is very low when the number of evaluation points is small in FIG. 33, but the combination evaluation value is high even when there are few evaluation points in FIG. I understand that. The difference between FIG. 33 and FIG. 34 is the difference whether the basic evaluation point is f12 or f15, and the evaluation point f15 is an evaluation arranged at the position where the temperature rise amount in the temperature distribution of FIG. 24 is the smallest. It is a point. This result shows that a favorable result can be obtained with a small number of measurement points by selecting an evaluation point or a temperature sensor arranged at a position where the temperature change is small as a basic evaluation point or a basic temperature sensor.
With the above method, the number and arrangement of the temperature measurement units could be determined.

<Example 2>
In Example 2, when all the temperature sensors installed in the machine tool were selected as the temperature measurement unit, the heat inflow amount from the heat source to the model was estimated. Details will be described below.
FIG. 35 is an experimental conceptual diagram showing the experimental purpose of this example. As shown in the figure, in the experiment carried out in this example, an experimental value obtained by measuring the temperature of the cross slide in a temperature-controlled room and an amount of temperature increase after a lapse of a certain time and a heat source are assumed and after a lapse of a certain time. By comparing the analytical values of the temperature rise at each point by analyzing the temperature rise of the sample, and by searching repeatedly for the heat source (temperature rise) that matches the experimental temperature measurement results, the heat source of the cross slide model The distribution was estimated. Specific experimental procedures and experimental results will be described below.

  Two silicon rubber heaters as known heat sources were brought into close contact with the sliding surface with heat transfer grease on the inclined surface of the cross slide model and fixed to the bakelite plate with a vise. Subsequently, temperature sensors (temperature sensors) are installed at 15 points on the cross slide model (hereinafter referred to as temperature sensors 1, 2... 15), and heating conditions are 150 W × 2 in a temperature-controlled room with a room temperature of 21 ° C. and a humidity of 55%. Each was heated in 8 hours, and the temperature rise history meter 15 points were examined over time. The result is shown in FIG.

  FIG. 36 is a graph illustrating the temperature measurement result of the temperature sensor at the peripheral edge in the drive mechanism of the cross slide model of the multi-axis control machine tool used in this example. That is, FIG. 36 is a temperature-time plot when a heating experiment is performed. As shown in FIG. 36, the temperature sensor 15 close to the silicon rubber heater showed a large temperature rise. Conversely, the temperature sensor 4 and the like were far from the installed silicon rubber heater, so the temperature rise was small. The amount of temperature increase after 8 hours was as shown in Table 1 and Table 2 below.

From the measurement results of the temperature rises at these 15 points, the heat source was estimated by inverse analysis. The method of reverse analysis is shown below. Assuming that the temperature rise history measured by the temperature sensor has nine heat sources on the cross slide model as shown in FIG. 36, the optimization algorithm is compared with the temperature rise history of the model in which heat flows from them. Search for a combination of heat fluxes that minimizes the sum of squares of the residual, which is the objective function of (Equation A) described later, by comparing with the time history temperature analysis performed with the combination of heat fluxes selected in. did. The physical properties of the cross model are gray cast iron, the density is 7500 kg / m ³ , the thermal conductivity is 50 W / (m · K), the specific heat is 0.55 kJ / (kg · K), and it is applied to the heat dissipation surface other than the heat source. The heat transfer coefficient is 10 W / (m ² · K). Here, all surfaces except for the surface to which the heat flux was applied were heat radiation surfaces. That is, heat flux = (incoming heat amount W) / (area mm ² ).

  Subsequently, in the same experiment, this time, the number of measurement points was increased to 17 points, and the temperature rise amount after a certain time of 17 points of the temperature measurement results tc_04 to tc_o in the constant temperature room was optimized. As shown in Table 3 below, the temperature rises at each point after the elapse of a predetermined time in the time history temperature analysis at 17 points of the heat flux combinations selected by the algorithm, kp99 to kp43.

Here, the objective function is the sum of squared residuals δ ² = (kp99−tc — 04) ² + (kp42−tc — 05) ² +.
+ (Kp43−tc_o) ² (Formula A)
A combination of heat fluxes that minimizes δ ² was searched. When the experimental value completely matches, δ ² = 0. It took about 3 minutes to analyze the temperature of one heat flux combination. We searched for a combination of heat fluxes that minimizes the objective function through repeated calculations of this temperature analysis. In this iterative calculation time, about 500 hours were repeated for about 24 hours. The results are shown in Table 4 below. In addition, FIG. 37 shows the temperature distribution of the cross slide model under the optimum conditions in the iterative calculation. Therefore, the inflow of the known heat source can be almost estimated from the heat source distribution in the optimum condition.

  FIG. 37 is a 3D analysis diagram showing the temperature distribution of the cross slide model under the optimum conditions. That is, FIG. 37 shows the cross slide estimated based on the optimal heat source distribution calculated by inverse analysis from the temperature measurement result of the temperature sensor in the drive mechanism of the cross slide model of the multi-axis control machine tool used in the example. It is a 3D analysis figure which illustrated temperature distribution of the drive mechanism of a model. From this analysis diagram, it can be seen that the heat inflow amount of the known heat source can be almost estimated from the heat source distribution under the optimum conditions by the method of this embodiment.

<Example 3>
FIG. 38 shows a schematic diagram of a multi-machine cross-slide model used for examining the arrangement of the temperature sensor in the heat inflow estimation of the machine tool body. In this example, the cross slide model was handled alone, and the arrangement and number of temperature sensors were examined based on the quasi-steady temperature rise experiment results. The temperature increase of the cross slide model was measured by installing temperature sensors at 15 points (t1 to t15) in FIG. A silicon rubber heater with an output of 300 W was used as the surface heat source.

  FIG. 39 shows a heat source candidate area of the cross slide model used for estimating the heat inflow amount of the machine tool body. In consideration of the movement of the cross slide in the actual machine, eight heat source candidate regions (surfaces A and C to I) were set in addition to the surface (surface B) on which the silicon rubber heater was installed.

  The estimation of the heat inflow of the machine tool body by inverse analysis was performed as follows. First, the temperature rise of the machine was measured with a temperature sensor installed on the machine tool. Next, a quasi-stationary temperature rise analysis of the aircraft is performed on the assumed heat inflow by the finite element method, and the objective function δ defined by the equation (5) is based on the analysis result and the quasi-stationary temperature rise measurement result. The amount of heat inflow that minimizes the amount was calculated using an optimization method.

Here, n is the number of temperature sensors, _Tan is a quasi-steady temperature rise amount of the temperature sensor obtained by finite element analysis with respect to the assumed heat inflow amount, and T _ex is a measured value of the corresponding quasi-steady temperature rise amount. It is.

  Table 5 shows combinations of temperature sensors used for studying the influence of the arrangement and number of temperature sensors on the estimation accuracy of the heat inflow amount.

The selection of the combination of temperature sensors, utilizing orthogonal table L ₈ of 2n series of experimental design. The circles in Table 5 indicate that the quasi-steady temperature rise amount of the temperature sensor is taken into account in the calculation of δ in the equation (5), and x indicates that it is not taken into account. In addition, the influence of the temperature sensor on the estimation accuracy of the heat inflow amount was examined using the temperature sensors at seven locations in Table 5. Further, the measured value of the quasi-steady temperature rise amount used for estimating the heat inflow amount of the cross slide model is No. 1 in Table 5. 1-No. It was a thing 8 hours after a heating start with respect to the combination of 8 temperature sensors.

  In this analysis, the heat inflow to each surface heat source shown in FIG. 39 was obtained from the combination of heat fluxes that minimizes δ within the heat flux range shown in Table 6. For optimization, a method combining linear programming, sequential quadratic programming, Downhill Simplex method, and genetic algorithm was used. The search time for the optimum solution was 6 hours.

<Results and discussion>
40 shows No. 1 shown in Table 5. 1-No. 8 shows the estimation results of heat inflow for each combination of temperature sensors. The amount of heat inflow shown in FIG. 40 was determined by multiplying the value of the optimum heat flux solution by the area of the heat source candidate region. The estimation result of the heat inflow amount of the heat source candidate region B where the silicon rubber heater was installed was closest to the output of 300 W. 1 and the heat inflow amount was 251.7 W. However, as shown in FIG. The δ of 1 was considerably larger than other temperature sensor combinations. As the cause of this, no. It is conceivable that the number of temperature sensors used in 1 was 7, which was larger than 3 in other combinations. Here, no. 1-No. No. 8 in order to find the most suitable one for estimating the heat inflow amount. 1-No. 8 Analyzing the quasi-steady temperature rise of the cross slide model using the optimal solution of the heat inflow for each combination, and measuring the quasi-steady temperature rise at the 15 locations shown in FIG. 38 given by equation (6) Δ using the value was determined.

  As shown in FIG. 40, Δ is the smallest because the heat inflow amount of the heat source candidate region B in which the silicon rubber heater is installed can be estimated most accurately. 1. Next is No. 4 and the estimated value was 244.1W. As can be seen from the bar graph in FIG. 1, no. In the case of 4, the amount of heat inflow in the heat source candidate regions other than B, which was an actual heat source, was generally small compared to the case of the combination of other temperature sensors.

  By the way, as shown in Table 5, the number of temperature sensors is No. Except for 1 of 7, all others are 3. However, even if the number of temperature sensors is the same, there is a large difference in Δ in FIG. Therefore, in order to find a temperature sensor for estimating the heat inflow amount with high accuracy, whether or not each temperature sensor is considered is determined by using the SN ratio η of the desired characteristic of quality engineering given by Equation (7). The effect of the inflow on the estimation accuracy was examined.

  FIG. 41 shows a signal-to-noise ratio factor effect diagram of each temperature sensor obtained by the equation (7). Judging from the S / N ratio, in this analysis, taking into account the quasi-steady temperature rise amount in the calculation of δ for any temperature sensor led to an improvement in the estimation accuracy of the heat inflow amount. In particular, at t14 and t15, which are temperature sensors in the vicinity where the silicon rubber heater was installed in the quasi-steady temperature rise experiment, the S / N ratio was significantly affected although the respective quasi-steady temperature rise amounts were not taken into consideration. This is because No. includes both t14 and t15. 1, no. In the case of the combination of the temperature sensors 4, this agrees well with the fact that Δ shown in FIG. 40 is small.

  As described above, in order to increase the accuracy of the estimation of the heat inflow amount of the machine tool body, the arrangement and the number of temperature sensors were examined using the orthogonal table of the experimental design method. Evaluating the factorial effect of each temperature sensor on the estimation accuracy of heat inflow based on the search result of the optimal solution of heat inflow is the arrangement of the temperature sensor that accurately estimates the heat inflow of the machine tool body. Useful for determining numbers.

<Example 4>
In the embodiments so far, the temperature sensor installation method when the heat source position is specified has been examined. In Example 4, the temperature sensor installation method when the heat source position cannot be specified is examined.

In the present embodiment, a model having a shape as shown in FIG. 42 is used, and seven surface heat sources 3, 7, 12, 21, 32, 33, 34 and evaluation points kp38, 51, 52, 55, 299, 303, 324, 359, 5, 287, 288 were set.
Next, as shown in Table 7, the surface heat source 32, 33, 34 based on whether dissipating or from which a heat source in accordance with orthogonal table L _4, were prepared four types of heat sources pattern. When estimating the heat inflow amount with each heat source pattern, a surface heat source other than the surface number indicated as “radiation” in Table 7 was used as the heat source. For example, in the heat source pattern 1, the surface heat sources 3, 7, 12, and 21 are heat sources, and in the heat source pattern 2, the surface heat sources 3, 7, 12, 21, 33, and 34 are heat sources.

Next, as shown in Table 8, the evaluation point kp5,287,288 based on or not included or included in the calculation of the objective function in accordance with orthogonal table L _4, and four create a combination of the evaluation points. For each heat source pattern, the amount of heat inflow from each heat source candidate region is estimated based on the combination of these evaluation points.

  Next, by performing an unsteady temperature rise analysis for each of the heat source patterns 1 to 4, a correct temperature distribution shown in FIG. 43 was obtained. In this analysis, a study was made on the temperature rise 600 seconds after the start of the heating time. In this case, the analysis time of the temperature rise analysis was about 60 seconds. In the heat flux search analysis, the amount of heat inflow from the heat source candidate region is estimated so that the temperature at the evaluation point matches the temperature in the correct temperature distribution.

Next, a heat flux search analysis was performed under the conditions shown in Table 9 so as to minimize the objective function shown in Equation (8). In Equation (8), _Tan is the temperature rise value at each evaluation point analyzed using the heat flux value selected by the optimization method, and T _ex is the correct temperature rise value at each evaluation point. n is the number of evaluation points. In this heat flux search analysis, the heat inflow to each heat source candidate region shown in Table 7 was obtained from the combination of heat fluxes that minimizes δ within the heat flux range shown in Table 9. For optimization, Pointer installed in the optimization software iSIGHT FD 3.1, which combines linear programming, sequential quadratic programming, Downhill Simplex method, and genetic algorithm, was used. The search time for optimization was 1 hour.

  In the present embodiment, for each of the four heat source patterns, the heat inflow amount from the heat source candidate region is estimated using four sets of evaluation point patterns, so that a total of 16 heat flux search analyzes are performed. In this heat flux search analysis, iterative calculation is performed. As an example, for heat source pattern 1 and evaluation point combination a, the amount of heat inflow (heat flux) from each heat source candidate region and the objective function at each iteration step The values are shown in FIG. Referring to FIG. 44, it can be seen that the heat inflow amount from each heat source candidate region greatly increases and decreases at each step, but the objective function steadily decreases as the step increases.

When the heat flux search analysis is completed and the estimated value of the heat inflow from each heat source candidate area is determined, the temperature distribution of the cross slide model is obtained based on this estimated value. Δ _all was calculated based on (9). In calculating δ _all , data of all evaluation points was used. Then, based on the calculated δ _all , the SN ratio η defined by the equation (10) was calculated.

The results obtained for the heat source pattern 1 are shown in Table 10 and FIG.

The results obtained for the heat source pattern 2 are shown in Table 11 and FIG.

The results obtained for the heat source pattern 3 are shown in Table 12 and FIG.

The results obtained for the heat source pattern 4 are shown in Table 13 and FIG.

  Next, each evaluation point was scored for each heat source pattern. Hereinafter, the scoring method will be specifically described using the heat source pattern 1. As shown in Table 8, the evaluation point “5” is not included in the calculation for the combination of evaluation points “a” and “b”, and is included in the calculation for the combination of evaluation points “c” and “d”. The evaluation “×” on the right side of the evaluation point “5” in the following Table 14 means the result when the evaluation point “5” is not included in the calculation, and the SN ratio value in that case is shown in Table 10. In other words, the average (5.46) of the SN ratio (−0.76) when the combination of the evaluation points is “a” and the SN ratio (11.684) when the combination is “b”. Similarly, the evaluation “◯” means the result when the evaluation point “5” is included in the calculation, and the value of the SN ratio in that case is as follows. It is the average (6.08) of the SN ratio (3.89) in the case of “d” and the SN ratio (8.26) in the case of “d”. The average difference (increment) is 0.61. The larger this increment is, the greater the degree of improvement of the result by including the evaluation point “5” in the calculation.

When the SN ratio increments for the evaluation points “287” and “288” were calculated by the same method, they were 8.41 and 4.04, respectively.
Since it can be said that the larger the increment is, the more preferable the evaluation point is, the score of the evaluation point was set to 3, 2, and 1 in order from the largest increment.
For the heat source patterns 2 to 4, the evaluation points were scored in the same manner. If the heat source pattern changes, the number of evaluation points also changes. For example, the evaluation point “5” has a score of 1 in the heat source pattern 1, but has a score of 3 in the heat source pattern 2.

If the heat source pattern is fixed, the score of the evaluation point for the heat source pattern is obtained, and the evaluation point to be included in the calculation can be determined based on the score. In many cases, it is difficult to precisely determine the position of the heat source, and the number of evaluation points is often obtained in consideration of fluctuations in the position of the heat source.
Therefore, in this example, as shown in Table 15, the scores of the respective evaluation points in the four heat source patterns were averaged, and the average value was used as the score of the evaluation points. According to Table 15, the average score of the evaluation points “287” was 2.5, and the average scores of the evaluation points “5” and “288” were both “1.75”. From this result, it is possible to conclude that the evaluation point “287” is the most important evaluation point when the variation of the heat source position is taken into consideration.
The evaluation points “5” and “288” have the same average score. In such a case, an evaluation point having a high score in the heat source pattern close to the operation mode of the actual machine can be determined as a more important evaluation point. For example, when the heat source pattern in the operation mode of the actual machine is closer to the heat source pattern 1, the evaluation point “288” can be said to be more important. On the other hand, the heat source pattern in the operation mode of the actual machine is closer to the heat source pattern 2 In this case, it can be said that the evaluation point “5” is more important.

In this embodiment, the combination of heat source patterns and the combination of evaluation points are created using the orthogonal table L ₄ , but these combinations may be created using an orthogonal table other than the orthogonal table L ₄ . Further, the orthogonal table used for creating the combination of heat source patterns and the orthogonal table used for combining the evaluation points may be the same or different.

  Further, in this example, after scoring the evaluation points for each heat source pattern, the final evaluation points were evaluated based on the average score, but the SN ratio increment for each heat source pattern The final evaluation point may be evaluated based on the average value. Further, various temperature sensor evaluation values described in “(5) Calculation of temperature sensor evaluation value (step SC5)” may be used instead of the SN ratio increment.

  In the above, this invention was demonstrated based on the Example. It is to be understood by those skilled in the art that this embodiment is merely an example, and that various modifications are possible and that such modifications are within the scope of the present invention.

DESCRIPTION OF SYMBOLS 100 Thermal displacement correction | amendment apparatus 101 Input reception and model data generation part 102 Input part 104 Heat source candidate area setting part 110 Temperature data storage part 114 Server 116 Temperature sensor position data reception part 118 Measured structure structure / physical property data reception part 120 Measured Structure model data generation unit 122 Structure model data storage unit to be measured 124 Heat inflow assumption unit 126 Temperature distribution estimation unit 128 Temperature distribution storage unit 130 Contrast / gap analysis unit 132 Repetition calculation determination unit 134 Approximate data calculation unit 136 Approximate data Storage unit 138 Output unit 139 Thermal displacement correction data generation unit 200 Inclined support (cross slide model (cross slide))
201 Temperature Measurement Unit 202 Temperature Sensor 204 Temperature Sensor Selection Unit 205 Cross Slide Inclined Surface 207a Reciprocating Slide Way
207b Reciprocating slide way 211 Tool post 212 Bed 300 Multi-axis control machine tool drive mechanism 302 Difference data calculation unit 304 Residual square sum calculation unit 306 Residual square sum / threshold magnitude determination unit 308 Threshold storage unit 310 Time / times upper limit reached Determination unit 312 Upper limit storage unit 314 Time calculation unit 316 Number calculation unit 318 General determination unit 1000 Machine tool

Claims (10)

A machine tool that relatively moves a tool and a work material by a plurality of drive mechanisms to process the work material into a predetermined shape,
A plurality of temperature sensors provided in the machine tool;
A temperature sensor selection unit that selects, as a temperature measurement unit, one used for temperature measurement among the temperature sensors;
A heat source candidate area setting unit for setting a heat source candidate area in the machine tool;
A heat inflow assumption section that assumes the heat inflow from the heat source candidate region;
A temperature distribution estimation unit that generates an estimated temperature distribution of the machine tool based on the heat inflow amount assumed by the heat inflow amount assumption unit;
The detected temperature value and the estimated temperature value are compared by comparing the detected temperature value by the temperature measuring unit with the estimated temperature value of the location corresponding to the measured temperature portion of the temperature distribution estimated by the estimated temperature distribution unit. The heat inflow amount assumption unit re-assums the heat inflow amount to another value so that the gap between the heat inflow amount assumption unit and the temperature distribution estimation unit is repeatedly calculated. A gap analysis section;
Based on the result of the repeated calculation, comprising an approximate data calculation unit for calculating an approximate heat inflow amount from the heat source candidate region and an approximate temperature distribution of the machine tool,
The temperature sensor selection unit generates a plurality of combinations of temperature sensors according to an orthogonal table based on whether each of the temperature sensors is used, calculates a combination evaluation value for each of the combinations, and is calculated for each combination. Further, the temperature sensor evaluation value of the temperature sensor is calculated by classifying the combination evaluation value based on whether or not a certain temperature sensor is used, and the temperature sensor is used as a temperature measuring unit based on the temperature sensor evaluation value Choose a machine tool. The selection of the temperature sensor generates a plurality of combinations of temperature sensors by adding temperature sensors to be used in descending order of evaluation based on the evaluation value of the temperature sensor, and sets a combination evaluation value for each of the generated combinations. The machine tool according to claim 1, wherein the machine tool is calculated and selected by selecting a combination of temperature sensors based on an evaluation based on the calculated combination evaluation value. The machine tool according to claim 2, wherein selection of the combination of the temperature sensors is performed by selecting a combination having the highest evaluation based on the calculated combination evaluation value. The temperature sensor is selected by generating a combination of temperature sensors by adding the temperature sensors to be used in order from the one with the largest temperature sensor evaluation value, calculating the combination evaluation value for the generated combination, and adding the temperature sensor. The combination evaluation values before and after the comparison are compared, and when the decrease in the evaluation due to the combination evaluation value after the addition of the temperature sensor exceeds a predetermined threshold, the addition of the temperature sensor is finished and the combination evaluation value is maximized. The machine tool according to claim 1, wherein the machine tool is selected. The combination evaluation value is
Set a reference heat inflow amount from the heat source candidate area,
Generate a reference temperature distribution in the heat source candidate region based on the reference heat inflow amount,
Set the assumed heat inflow from the heat source candidate area,
Generating an estimated temperature distribution of the machine tool based on the assumed heat inflow,
Contrast the temperature reference value of the location corresponding to the temperature sensor included in the combination in the reference temperature distribution and the temperature estimated value of the location corresponding to the temperature sensor included in the combination in the estimated temperature distribution, The hypothetical heat inflow amount is reset to another value and repeatedly calculated so that the gap between the temperature reference value and the temperature estimated value becomes small,
Any one of claims 1 to 4 calculated by evaluating a difference between a final estimated temperature distribution obtained as a result of the iterative calculation and a reference temperature distribution or a difference between a final assumed heat inflow amount and a reference heat inflow amount. A machine tool according to claim 1. The machine tool according to claim 5, wherein the temperature sensor is a basic temperature sensor that is arranged at a position where the amount of temperature rise is the smallest in the reference temperature distribution. The machine tool according to claim 5 or 6, wherein the reference heat inflow amount is set to an approximate heat inflow amount previously calculated by an approximate data calculation unit. The heat source candidate region setting unit sets a plurality of heat source patterns that combine a plurality of heat source candidate regions based on an orthogonal table,
The said temperature sensor selection part calculates the temperature sensor evaluation value about each heat source pattern, and selects the temperature sensor used as a temperature measurement part based on the average value, The heat sensor pattern as described in any one of Claims 1-7. Machine Tools. A method for determining the number and arrangement of temperature measuring parts of a machine tool that relatively moves a tool and a work material by a plurality of drive mechanisms to process the work material into a predetermined shape,
Setting a heat source candidate region and a plurality of evaluation points on the model of the machine tool;
Generating a plurality of combinations of evaluation points according to an orthogonal table based on whether to use each of the evaluation points;
Calculating a combination evaluation value for each of the combinations;
Calculating the evaluation point evaluation value of the evaluation point by classifying the combination evaluation value calculated for each combination based on whether or not an evaluation point is used;
A method for determining the number and arrangement of temperature measuring units of a machine tool, comprising: selecting an evaluation point at which a temperature measuring unit is arranged based on the evaluation point evaluation value. A program for determining the number and arrangement of temperature measuring parts of a machine tool for processing a work material into a predetermined shape by relatively moving a tool and the work material by a plurality of drive mechanisms,
Setting a heat source candidate region and a plurality of evaluation points on the model of the machine tool;
Generating a plurality of combinations of evaluation points according to an orthogonal table based on whether to use each of the evaluation points;
Calculating a combination evaluation value for each of the combinations;
Calculating the evaluation point evaluation value of the evaluation point by classifying the combination evaluation value calculated for each combination based on whether or not an evaluation point is used;
A program for determining the number and arrangement of temperature measuring units of a machine tool, causing a computer to execute a step of selecting an evaluation point for arranging a temperature measuring unit based on the evaluation point evaluation value.