JP5399537B2 - Numerical controller - Google Patents

Description

  The present invention relates to a numerical control device used for various machine tools.

  Conventionally, in various machine tools, a workpiece is processed by numerically controlling a transfer device that transfers a workpiece and a tool for processing the workpiece according to a processing command (NC program). Patent Document 1 below discloses a numerical control device that performs numerical control of such a transfer device.

  In the numerical control device disclosed in Patent Document 1, a tool path representing a change in the position coordinate of a tool in a coordinate system fixed on a workpiece as a function of a parameter is obtained from a machining program, and the transfer path of the transfer device is determined from the tool path. A drive axis path for each drive axis is obtained, and a command pulse indicating a movement amount for each reference unit time (cycle time) in each drive axis is obtained based on the obtained drive axis path for each drive axis, and obtained. Each transfer device is controlled by outputting the command pulse to the motor of the transfer device corresponding to each drive shaft. In this numerical controller, block smooth interpolation and corner smooth interpolation are performed on the drive axis path, and the command pulse is obtained by pulse interpolation from the drive axis path after the smooth interpolation. The block smooth interpolation interpolates a drive axis path formed by connecting a large number of command points with straight lines into a smooth curve, and the corner smooth interpolation is positioned before and after a predetermined command point in the drive axis path. This corner portion is interpolated so that a corner portion where the amount of change in coordinates changes greatly is a corner that smoothly turns.

  However, in the technique described in Patent Document 1, in the case where unexpected abnormality data (noise or the like) in which the change amount of the position coordinates repeatedly changes suddenly at a minute interval is included in the drive axis path, It is difficult to smoothly interpolate the abnormal data portion. When such abnormal data is included in the drive shaft path and the transfer device is controlled according to the drive shaft path, the operation of the transfer device changes suddenly at the portion corresponding to the abnormal data, and the processing machine A mechanical shock occurs. For this reason, it is an important issue to correct the abnormal data in the drive shaft path.

  Therefore, Patent Document 2 below shows a numerical control device that can solve such a problem.

  In the numerical control device disclosed in Patent Document 2, post-interpolation acceleration / deceleration processing that can correct abnormal data that causes mechanical shock is performed. Specifically, in this numerical control device, a movement command to be output to the servo motor for each drive axis is obtained by performing interpolation processing (pulse interpolation) on a path obtained by converting the machining program into an execution format, and the obtained movement command Acceleration / deceleration processing is performed on According to this post-interpolation acceleration / deceleration processing, even if a portion corresponding to abnormal data in the drive axis path is included in the movement command obtained by pulse interpolation, a portion corresponding to the abnormal data in the movement command And the occurrence of mechanical shock can be suppressed.

JP 2008-225825 A JP 2010-140312 A

  As described above, the post-interpolation acceleration / deceleration processing shown in Patent Document 2 has an effect of suppressing the occurrence of mechanical shock caused by abnormal data, but on the other hand, it causes an error in the workpiece machining shape. There is a drawback of making it.

  Specifically, in the post-interpolation acceleration / deceleration shown in Patent Document 2, the part to which the acceleration / deceleration process is applied is not limited in the movement command, so the acceleration / deceleration process corresponds to the abnormal data in the movement command. It can also be applied to parts other than parts. For example, when the acceleration / deceleration processing is performed on a part of the movement command that corresponds to a part that actually instructs acceleration or deceleration of the tool or workpiece in the machining program, the tool or workpiece to be executed based on that part is processed. The acceleration or deceleration is alleviated, and as a result, an error occurs with respect to the machining shape indicated by the machining program in the shape of the actually machined workpiece. For this reason, it is required to reduce the machining shape error associated with such post-interpolation acceleration / deceleration processing.

  The present invention has been made in order to solve the above-described problem, and can suppress an error in a machining shape of a workpiece while suppressing occurrence of a mechanical shock due to abnormal data in a mechanical axis path. It is to provide a numerical control device.

  In order to achieve the above object, a numerical control device according to the present invention comprises a plurality of transfer devices for transferring a workpiece to process the workpiece using the workpiece or a tool for processing the workpiece as a target, The transfer device is provided in a machine tool having a support for supporting the object and a transfer unit for transferring the object by transferring the support in a specific machine axis direction. A numerical control device for performing numerical control of a transfer device, wherein a position in each machine axis direction to which each support is to be transferred is predetermined based on a processing command representing a movement trajectory of the tool during processing of the workpiece. A reference machine axis path deriving unit for deriving a reference machine axis path expressed as a function of a set time of the machine, and the reference machine axis for each of the machine axis directions derived by the reference machine axis path deriving unit A path conversion unit for deriving a post-conversion machine axis path for each of the machine axis directions by performing a specific conversion process on the machine, and the post-conversion machine axis for each of the machine axis directions derived by the path conversion unit An acceleration / deceleration processing unit that performs acceleration / deceleration processing for relaxing a change in the transfer speed of each of the supports obtained from the post-conversion mechanical axis path for the path, and after acceleration / deceleration processing by the acceleration / deceleration processing unit A transfer amount deriving unit for deriving a transfer amount of each of the supports per predetermined set cycle time in the direction of each of the machine axes from the post-processing machine axis path which is a machine axis pass of A transfer control unit that transfers the support corresponding to each transfer unit according to the transfer amount, and the specific conversion process performed by the path conversion unit includes the reference mechanical axis path On the other hand, a correction amount corresponding to the error between the mechanical axis path obtained when the acceleration / deceleration processing unit performs the same process as the acceleration / deceleration process and the reference mechanical axis path is calculated for each of the machine axis directions, A calculation process for calculating the converted machine axis path obtained by correcting the reference machine axis path in each machine axis direction in a direction in which the error is canceled by the calculated correction amount for each machine axis direction is included.

  In this numerical control device, the acceleration / deceleration processing unit changes the transfer speed of each support obtained from the post-conversion machine axis path with respect to the post-conversion machine axis path for each machine axis direction calculated by the path conversion unit. In order to perform acceleration / deceleration processing to alleviate this, even if abnormal data that causes mechanical shock is included in the machine axis path, the abnormal data is subjected to a gradual speed change that does not cause mechanical shock by the acceleration / deceleration processing. It can be modified as shown. Further, in this numerical control apparatus, an error between the mechanical axis path and the reference mechanical axis path obtained when the path conversion unit performs the same processing as the acceleration / deceleration processing by the acceleration / deceleration processing unit on the reference mechanical axis path. A converted mechanical axis in which a corresponding correction amount is calculated for each machine axis direction, and the reference machine axis path in each machine axis direction is corrected in a direction that cancels the error with the calculated correction amount for each machine axis direction. In order to perform calculation processing to calculate the path, the post-conversion machine axis path is a machine axis path that has been corrected in a direction that cancels the error by predicting in advance the error in each machine axis direction caused by the acceleration / deceleration processing by the acceleration / deceleration processing unit. It becomes. For this reason, when the acceleration / deceleration processing unit performs acceleration / deceleration processing on the converted machine axis path, errors in all the machine axis directions caused by the acceleration / deceleration processing can be suppressed. An error in the work shape of the workpiece caused by the deceleration process can be suppressed.

In the numerical control apparatus, when the acceleration / deceleration processing unit is set for the machine axis direction of the converted machine axis path to be subjected to the acceleration / deceleration process with the set time as t as the acceleration / deceleration process. A value of half of the constant is a, and a speed function that is a primary differential function of the converted mechanical axis path at each set time T in a section from t−a to t + a is P _B ′ (T). 1), where f (T) is a distribution function representing a specific distribution format for distributing the speed function P _B ′ (T), a speed that is a first-order differential function of the post-process mechanical axis path The function Q _B ′ (t) is calculated based on the following equation (2), and the post-processing mechanical axis path Q _B (t) is calculated by integrating the calculated speed function Q _B ′ (t). It is preferable to perform arithmetic processing.

According to this configuration, it is possible to prevent the set time from being delayed from the post-conversion mechanical axis path before the acceleration / deceleration process in the post-processing mechanical axis path Q _B (t). Specifically, suppose that the acceleration / deceleration processing unit distributes the speed at each set time in a section of the converted mechanical axis path into a distribution section having a section width corresponding to a time constant after that time, When the speed function is calculated by integrating the distributed values, and the acceleration / deceleration processing of the converted machine axis path is performed according to a calculation formula that further integrates the calculated speed function, the processed machine axis The set time in the path is delayed with respect to the set time in the post-conversion machine axis path due to the speed distribution format. On the other hand, in the present configuration, the speed function P _B ′ (T) at each set time T in the section extending from t−a to t + a is expressed as the set time T by the distribution function satisfying the above formula (1). Since the distribution is performed in the distribution section from T−a to T + a that spreads evenly in the front and rear as the center, the distribution value P _B ′ (T) · f () of the velocity function P _B ′ (T) distributed in the distribution section t-T) integral to speed function Q _B in accordance with the above formula (2) treatment after mechanical axis path calculated by further integrating the 'calculates (t), the calculated velocity function Q _B' (t) In Q _B (t), it is possible to prevent the set time from being delayed due to the speed distribution format as described above.

In this case, as the same process as the acceleration / deceleration process, the path conversion unit obtains a speed function that is a first-order differential function of the reference mechanical axis path at each set time T in the section from t−a to t + a. When P _A ′ (T) is set, a speed function Q _A ′ (t) that is a linear differential function of the mechanical axis path after performing the same processing as the acceleration / deceleration processing is calculated based on the following equation (3). Then, it is preferable to perform arithmetic processing for calculating the mechanical axis path Q _A (t) after performing the same processing as the acceleration / deceleration processing by integrating the calculated speed function Q _A ′ (t).

According to this configuration, when calculating the correction amount corresponding to the error between the mechanical axis path and the reference mechanical axis path obtained when the path conversion unit performs the same processing as the acceleration / deceleration processing by the acceleration / deceleration processing unit, The speed distribution format in the same process as the acceleration / deceleration process is simplified while simplifying the association of the set times of the points to be subtracted between the machine axis path and the reference machine axis path after the same process as the acceleration / deceleration process. The element of the phase error caused due to this can be excluded from the correction amount. Specifically, in the same process as the acceleration / deceleration process, the path conversion unit distributes the speed at each set time in a section of the reference mechanical axis path having a section width corresponding to the time constant after that time. When calculating the machine axis path after the processing according to the calculation formula that integrates each distributed value in the section, calculates the speed function by integrating the distributed values, and further integrates the calculated speed function Since the set time in the machine axis path after the processing is delayed with respect to the set time in the reference machine axis path, the reference machine axis path is subtracted from the processed machine axis path without considering the delay in the set time. When the correction amount corresponding to the error between the two machine axis paths is obtained, the phase error is included in the correction amount. In such a case, if the phase error element is excluded from the correction amount, a delay in the set time between the machine axis path after the processing and the reference machine axis path caused by the speed distribution format described above. It is necessary to obtain an error between the machine axis path after the process and the reference machine axis path after performing the association (time adjustment) for eliminating the above, and the process of the association is complicated. On the other hand, in this configuration, the speed function Q _A ′ (t) of the machine axis path after the processing is calculated based on the above formula (3), and the calculated speed function Q _A ′ (t) is integrated. In order to calculate the machine axis path Q _A (t) after the processing, according to the distribution function in the above equation (3), the speed function P of the reference machine axis path in the section instructing the transfer at a constant speed It is possible to prevent the set time from being delayed due to the distribution form of _A ′ (T). As a result, in the machine axis path Q _A (t) after the processing, the section instructing the transfer at a constant speed The coordinate values at the same time t coincide with the reference machine axis path. For this reason, when calculating the correction amount corresponding to the error between the processed machine axis path Q _A (t) and the reference machine axis path, the correspondence of the set times of the points to be subtracted between the two machine axis paths While simplifying the attachment, it is possible to eliminate the element of the phase error caused by the distribution form of the speed of the reference machine axis path from the correction amount.

In the configuration for performing the processing described above acceleration and deceleration processing section for calculating the post-processing machine axis path Q _{B (T)} as a deceleration process, the reference mechanical axis path derivation unit, said on the adjustments of the set time A time adjustment unit for deriving a reference machine axis path, the time adjustment unit being set to t as a set time before the time adjustment, and tj (t) as a set time after the time adjustment; the time constant of the axial direction and Tj, the time of the start of the start transfer of the object and t = 0, the time of the end point of the transfer stop of the object and t = te, satisfies the formula (1), wherein When the distribution function representing a specific distribution format is f (T), it is preferable to perform time adjustment based on the following equation. The following expression represents a function used for time adjustment in the form of a general expression.

  According to this configuration, the part corresponding to the acceleration period immediately after the start of the transfer of the object and the part corresponding to the deceleration period immediately before the stop of the transfer of the object, which occurs when the path conversion unit performs the same process as the acceleration / deceleration process. The deviation of the set time in can be eliminated. Specifically, the path conversion unit performs the same process as the acceleration / deceleration process on the reference machine axis path derived without the time adjustment based on the machining command path by the reference machine axis path deriving unit. If so, the set time from the reference machine axis path in the part corresponding to the acceleration period immediately after the start of the object transfer and the part corresponding to the deceleration period immediately before the object transfer stops in the machine axis path after the processing. When calculating the correction amount corresponding to the error between the machine axis path after this processing and the reference machine axis path, an appropriate correction amount can be calculated for each part corresponding to the acceleration period and the deceleration period. Disappear. On the other hand, according to this configuration, the time adjustment unit adjusts the time for each part corresponding to the acceleration period (0 ≦ t <Tj) and the deceleration period (te−Tj <t ≦ te). Thus, before the path conversion unit performs the same process as the acceleration / deceleration process, a reference machine axis path corrected in a direction to cancel the set time difference of each part corresponding to the acceleration period and the deceleration period caused by the process is obtained in advance. be able to. As a result, it is possible to prevent the set time from being shifted in each part corresponding to the acceleration period and the deceleration period of the mechanical axis path after the path conversion unit performs the same process as the acceleration / deceleration process. It is possible to calculate an appropriate correction amount for each part.

  In the configuration described above, the machine tool has a special command for inputting a special command for instructing an operation accompanied by a speed change of the target object separately from a normal transfer of the target object at the time of machining the workpiece. The transfer amount deriving unit determines the length of the set cycle time in a state immediately before the input of the special command in response to the input of the special command to the special command input device. Then, the length is changed according to the speed change of the object indicated by the special command, and the length of the support is changed to the corresponding machine axis direction of each support per set cycle time. It is preferable to derive the transfer amount from the post-processing machine axis path.

  According to this configuration, when a special command is input to the special command input device to cause the machine tool to perform an irregular operation with a change in the speed of the object, the length of the set cycle time common to each machine axis direction is reduced. The length in the state immediately before the input of the special command is changed to the length corresponding to the speed change instructed by the special command, and each set cycle time after the change calculated from the machine axis path after processing is changed. Each support is transferred according to the transfer amount of the support in the corresponding machine axis direction. For this reason, when an operation involving irregular speed changes of the object is executed separately from the normal transfer of the object at the time of machining the workpiece, the object deviates from the trajectory indicated by the machine axis path after processing. Can be prevented. Specifically, when changing the length of the common set cycle time in each machine axis direction according to the special command as in this configuration, it is calculated per set cycle time from the processed machine axis path. The amount of transfer of the support in each machine axis direction is calculated in a state in which the relative positional relationship between the machine axis directions defined by the post-processing machine axis path in each machine axis direction is maintained. . For this reason, when the corresponding support is transferred by the transfer unit according to the calculated transfer amount in each machine axis direction per set cycle time, it is defined by the post-process machine axis path in each machine axis direction. An operation involving a change in the speed of the object is performed in a state where the relative positional relationship between the machine axis directions is maintained. As a result, even if an irregular operation with a change in the speed of the object as described above is performed, it is possible to prevent the object from deviating from the locus indicated by the machine axis path after processing in each machine axis direction. it can.

  Moreover, in this configuration, from the trajectory indicated by the original machining path when the object performs an operation with a speed change according to the special command while obtaining the effect of suppressing the mechanical shock caused by the abnormal data by the acceleration / deceleration processing. It can be prevented from coming off. Specifically, if acceleration / deceleration processing after pulse interpolation is performed in order to suppress mechanical shock caused by abnormal data, the transfer pulse after the post-interpolation acceleration / deceleration processing is indicated. The movement trajectory of the object is deviated from the movement trajectory indicated by the original machining path due to an error caused by the post-interpolation acceleration / deceleration processing. In this case, when an operation involving a change in the speed of the object according to the special command is performed according to the transfer pulse after the acceleration / deceleration processing after interpolation, the object moves out of the locus indicated by the original machining path. It will be. On the other hand, in this configuration, while the acceleration / deceleration processing by the acceleration / deceleration processing unit as described above obtains the effect of suppressing the mechanical shock caused by abnormal data, the acceleration / deceleration processing is performed in the machine axis path after the acceleration / deceleration processing. Due to the effect of error correction by the conversion process of the previous path conversion unit, the trajectory indicated by the machine axis path after the acceleration / deceleration process can be prevented from deviating from the trajectory indicated by the machining path. Thus, it is possible to prevent the object from deviating from the trajectory indicated by the original machining path when performing an operation accompanied by a speed change according to the mechanical axis path after the acceleration / deceleration processing.

  As described above, according to the present invention, there is provided a numerical control device capable of suppressing an error in a workpiece machining shape while suppressing occurrence of a mechanical shock due to abnormal data in a machine axis path. be able to.

1 is a schematic side view of a machine tool to which a numerical control device according to an embodiment of the present invention is applied. It is a functional block diagram of a numerical control device and a special command input device according to an embodiment of the present invention. It is a functional block diagram of a post-processing machine axis path deriving unit and a memory. It is a figure which shows the locus | trajectory of the tool which the reference | standard machine axis path by an example instruct | indicates. When the tool trajectory indicated by the reference machine axis path is as shown in FIG. 4, the reference machine axis path, the machine axis path after the same process as the acceleration / deceleration process, and the machine axis after the conversion unit acceleration / deceleration process are performed. It is a schematic diagram which shows the image of the relative relationship with a path | pass (post-conversion machine axis path | pass). It is a flowchart which shows the numerical control process by the numerical control apparatus of one Embodiment of this invention. It is a flowchart which shows the content of the subroutine of step S2 which derives | leads-out the machine-axis path | pass after a process in FIG. It is a flowchart which shows the content of the subroutine of step S30 which derives | leads-out the machine axis path | pass after conversion in FIG. In the conversion part acceleration / deceleration process in one Embodiment of this invention, it is a figure which shows the image of distribution when the area width of the block used as the distribution object of the speed function of a reference | standard machine shaft path is larger than a time constant. In the conversion part acceleration / deceleration process in one Embodiment of this invention, it is a figure which shows the image of distribution when the area width of the block used as the distribution object of the speed function of a reference | standard machine shaft path is smaller than a time constant. In the conversion part acceleration / deceleration process in one Embodiment of this invention, it is a figure which shows the image of distribution when the area width of the block used as the distribution object of the speed function of a reference | standard machine shaft path is smaller than a time constant. It is a flowchart which shows the monitoring process of the special command by the acceleration / deceleration request | requirement monitoring part. It is a flowchart which shows the emergency stop process of a target object. It is a flowchart which shows the restart process of a target object. It is a flowchart which shows the speed change process of a target object. It is a flowchart which shows the content of the subroutine of step S54 which calculates the deceleration stop period time in FIG. 13, and the cycle time fluctuation function g (t) at the time of a stop. It is a figure which shows the cycle time fluctuation function at the time of stopping in the case of performing the emergency stop process of the target object which is moving at a constant speed. It is a figure which shows the cycle time variation function at the time of stopping in the case of performing an emergency stop process during acceleration of a target object. It is a figure which shows the cycle time fluctuation function at the time of stopping in the case of performing an emergency stop process during the deceleration of a target object. It is a figure for demonstrating the cycle time fluctuation function at the time of a stop which consists of three areas. It is a figure which shows the cycle time variation function at the time of a restart which consists of three areas in the case of performing the restart process of a target object. It is a figure which shows the cycle time variation function at the time of a restart which consists of two areas in the case of performing the restart process of a target object. It is a figure which shows the cycle time variation function at the time of a speed change in the case of performing the acceleration process of the target object which is moving at a fixed speed. It is a figure which shows the cycle time variation function at the time of a speed change in the case of performing the acceleration process of the target object further during acceleration of a target object. It is a figure which shows the cycle time variation function at the time of a speed change in the case of performing an acceleration process during the deceleration of a target object. It is a figure which shows the cycle time variation function at the time of a speed change in the case of performing the deceleration process of the target object which is moving at a fixed speed. It is a figure which shows the cycle time variation function at the time of a speed change in the case of performing a deceleration process during acceleration of a target object. It is a figure which shows the cycle time variation function at the time of a speed change in the case of performing the deceleration process of the target object further during deceleration of the target object.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  First, with reference to FIG.1 and FIG.2, the structure of the machine tool to which the numerical control apparatus 2 by one Embodiment of this invention is applied is demonstrated.

  A machine tool provided with a numerical control device 2 (see FIG. 2) according to the present embodiment is for cutting a workpiece 100 that is a workpiece, and includes a workpiece transfer device 102, a column 104, a tool 105, , A tool vertical transfer device 106, a tool first horizontal transfer device 108, a tool second horizontal transfer device 110, a spindle head 112, and a control panel 114. The workpiece transfer device 102, the tool vertical transfer device 106, the tool first horizontal transfer device 108, and the tool second horizontal transfer device 110 are each included in the concept of the transfer device of the present invention.

  The workpiece transfer device 102 is a device for transferring the workpiece 100 in the X-axis direction that is a direction perpendicular to the paper surface of FIG. The workpiece transfer device 102 includes a base 102a fixed on a predetermined installation location, a workpiece support 102b provided on the base 102a so as to be movable in the X-axis direction, and the workpiece support 102b. And a workpiece transfer section 102c (see FIG. 2) for transferring in the X-axis direction. The workpiece support 102b supports the workpiece 100, and the workpiece 100 is installed in a state of standing vertically on the workpiece support 102b. The workpiece transfer unit 102c has a servo motor as a drive source, and is operated by power generated by the servo motor to transfer the workpiece support 102b. The X-axis direction is included in the concept of the machine axis direction of the present invention, the workpiece support 102b is included in the concept of the support of the present invention, and the workpiece transfer unit 102c is It is included in the concept of the transfer part of the invention.

  The column 104 is erected at a position spaced apart from the installation location of the base 102a in the horizontal direction and in the direction perpendicular to the X axis, and extends in the vertical direction.

  The tool vertical transfer device 106 is provided in the column 104 and is a device for transferring a tool 105 for cutting the workpiece 100 in the Y-axis direction extending in the vertical direction. The tool vertical transfer device 106 is provided on the column 104 so as to be movable in the Y-axis direction, and is provided on the column 104. The vertical support 106a is transferred along the column 104 in the Y-axis direction. And a vertical transfer unit 106b (see FIG. 2). The vertical support 106a supports the tool first horizontal transfer device 108. The tool first horizontal transfer device 108 receives the tool 105 via the tool second horizontal transfer device 110 and the spindle head 112 as will be described later. In order to support, the vertical support 106a supports the tool 105 indirectly. The vertical transfer unit 106b has a servo motor as a drive source, and is actuated by power generated by the servo motor to transfer the vertical support 106a. The Y-axis direction is included in the concept of the machine axis direction of the present invention, the vertical support 106a is included in the concept of the support of the present invention, and the vertical transfer unit 106b is It is included in the concept of the transfer part of the invention.

  The tool first horizontal transfer device 108 is provided on the vertical support 106a and is a device for transferring the tool 105 in the W-axis direction extending perpendicular to both the X-axis and the Y-axis. The tool first horizontal transfer device 108 includes a first horizontal support 108a provided on the vertical support 106a so as to be movable in the W-axis direction, and a vertical support 106a. And a first horizontal transfer unit 108b (see FIG. 2) that transfers in the W-axis direction. The first horizontal support 108a supports the tool second horizontal transfer device 110, and the tool second horizontal transfer device 110 supports the tool 105 via the spindle head 112 as will be described later. The horizontal support 108a indirectly supports the tool 105. The first horizontal transfer unit 108b has a servo motor as a drive source, and is operated by power generated by the servo motor to transfer the first horizontal support 108a. The W-axis direction is included in the concept of the machine axis direction of the present invention, and the first horizontal support 108a is included in the concept of the support of the present invention. Reference numeral 108b is included in the concept of the transfer unit of the present invention.

  The tool second horizontal transfer device 110 is provided on the first horizontal support 108a, and is a device for transferring the tool 105 in the Z-axis direction parallel to the W-axis. The tool second horizontal transfer device 110 is provided on the first horizontal support member 110a provided on the first horizontal support member 108a so as to be movable in the Z-axis direction, and on the first horizontal support member 108a. It has the 2nd horizontal transfer part 110b (refer FIG. 2) which transfers the support body 110a to a Z-axis direction. The second horizontal support 110 a supports the spindle head 112 and supports the tool 105 via the spindle head 112. The second horizontal transfer unit 110b has a servo motor as a drive source, and is operated by power generated by the servo motor to transfer the second horizontal support 110a. The Z-axis direction is included in the concept of the machine axis direction of the present invention, and the second horizontal support 110a is included in the concept of the support of the present invention, and the second horizontal transfer unit 110b is included in the concept of the transfer unit of the present invention.

  The spindle head 112 is provided on the second horizontal support 110a so that the rotation axis thereof is parallel to the W axis and the Z axis. The spindle head 112 holds the tool 105 and rotates the tool 105 about its axis. The tool 105 cuts the workpiece 100 by being brought into contact with the workpiece 100 while being rotated by the spindle head 112.

  The control panel 114 has a function for controlling the driving of each of the transfer units 102c, 106b, 108b, 110b, driving of the spindle head 112, and other parts of the machine tool. The control panel 114 is electrically connected to the driving sources of the transfer units 102c, 106b, 108b, 110b and the spindle head 112.

  In addition, the control panel 114 includes a special command input device 122 (see FIG. 2). The special command input device 122 is a special command for instructing an operation accompanied by a speed change (acceleration or deceleration) of the workpiece 100 and the tool 105 in addition to the normal transfer of the workpiece 100 and the tool 105 when the workpiece 100 is processed. It is a device for inputting commands from the outside. Hereinafter, the workpiece 100 or the tool 105 that is the object of transfer by the transfer devices 102, 106, 108, and 110 is referred to as an object.

  Special command input device 122 includes a stop command input device 124, a restart command input device 126, and an override device 128.

  The stop command input device 124 is for inputting an emergency stop command for urgently decelerating and stopping the movement of the object. The emergency stop command is included in the concept of the special command of the present invention. The stop command input device 124 includes an emergency stop button 124a provided on the control panel 114, and a stop signal for transmitting an emergency stop signal to an acceleration / deceleration request monitoring unit 10 described later in response to the pressing of the emergency stop button 124a. A transmitter 124b. In the present embodiment, pressing the emergency stop button 124a corresponds to inputting an emergency stop command.

  The restart command input device 126 is for inputting a restart command for restarting and accelerating the movement of the moving object that has stopped moving. The restart command is included in the concept of the special command of the present invention. This restart command input device 126 transmits a restart signal to a later-described acceleration / deceleration request monitoring unit 10 in response to the restart button 126a provided on the control panel 114 and the restart button 126a being pressed. And a start signal transmitter 126b. In the present embodiment, pressing the restart button 126a corresponds to inputting a restart command.

  The override device 128 includes an instruction to increase the moving speed of the object and an acceleration command including information on an acceleration rate that is an increasing rate of the moving speed or an instruction to decrease the moving speed of the object and a decreasing rate of the moving speed. This is for inputting a deceleration command including information on the deceleration rate. The acceleration command and the deceleration command are included in the concept of the special command of the present invention.

  The override device 128 includes an override dial 128a provided on the control panel 114, and a speed change signal transmitter that transmits a speed change signal according to the operation direction and the operation amount of the override dial 128a to the acceleration / deceleration request monitor 10 described later. 128b.

  The override dial 128a is an operation unit that is operated when an acceleration command or a deceleration command is input. The override dial 128a is provided on the control panel 114 so as to be rotatable about its axis. It is possible to input an acceleration command or a deceleration command to the override device 128 by rotating the override dial 128a in either direction. In the present embodiment, the operation of rotating the override dial 128a to one side corresponds to the input of an acceleration command, and the operation of rotating the override dial 128a to the other side, which is the opposite side to the one side, is a deceleration command. Corresponds to input. The amount of rotation of the override dial 128a toward the one side corresponds to the acceleration rate of the moving object, and the amount of rotation of the override dial 128a toward the other side corresponds to the deceleration rate of the moving object.

  The speed change signal transmission unit 128b includes information on an override coefficient corresponding to the amount of rotation of the override dial 128a toward one side in response to the override dial 128a being rotated toward the one side (acceleration side). An override coefficient corresponding to the amount of rotation of the override dial 128a toward the other side in response to the change signal being transmitted to the acceleration / deceleration request monitoring unit 10 and the override dial 128a being rotated toward the other side (deceleration side). Is sent to the acceleration / deceleration request monitoring unit 10. The override coefficient corresponding to the rotation amount of the override dial 128a toward the one side corresponds to the acceleration rate, and the override coefficient corresponding to the rotation amount of the override dial 128a toward the other side corresponds to the deceleration rate. The override coefficient increases from 1 when the override dial 128a is turned to the acceleration side with reference to 1, and decreases from 1 when the override dial 128a is turned to the deceleration side.

  The numerical control device 2 according to the present embodiment is provided in the machine tool having the above-described configuration, and performs numerical control of each of the transfer devices 102, 106, 108, and 110. Next, the configuration of the numerical control device 2 of the present embodiment will be specifically described with reference to FIGS.

  As illustrated in FIG. 2, the numerical control device 2 includes a storage unit 4, a memory 5, and an arithmetic processing device 6.

  The storage unit 4 stores an NC program as a machining command. The NC program indicates a machining path representing the movement locus and transfer speed of the tool 105 when machining the workpiece 100.

  The memory 5 stores an override coefficient immediately before the special command is input to the special command input device 122, an override coefficient included in the speed change signal, a machining path, a command path, and various machine axis paths to be described later. The information is memorized.

  The arithmetic processing unit 6 obtains a machining path from the NC program stored in the storage unit 4 and calculates the transfer amount of each support 102b, 106a, 108a, 110a per predetermined set cycle time based on the machining path. Various processes such as transfer control of each transfer unit 102c, 106b, 108b, 110b and monitoring of the input of a special command to the special command input device 122 are performed. The arithmetic processing device 6 includes an acceleration / deceleration request monitoring unit 10, a calculation unit 12, and a transfer control unit 14 as functional blocks.

  The acceleration / deceleration request monitoring unit 10 monitors whether or not the special command is input to the special command input device 122.

  Specifically, the acceleration / deceleration request monitoring unit 10 transmits a stop signal whether or not a deceleration stop command is input to the stop command input device 124, that is, whether or not the emergency stop button 124a of the stop command input device 124 is pressed. Monitoring is performed by detecting an emergency stop signal transmitted from the unit 124b. The acceleration / deceleration request monitoring unit 10 sends an emergency stop request to the calculation unit 12 in response to a deceleration stop command input to the stop command input device 124, that is, an emergency stop signal is transmitted from the stop signal transmission unit 124b. To the transfer amount deriving unit 22.

  Further, the acceleration / deceleration request monitoring unit 10 transmits a restart signal indicating whether or not a restart command has been input to the restart command input device 126, that is, whether or not the restart button 126a of the restart command input device 126 has been pressed. Monitoring is performed by detecting a restart signal transmitted from the unit 126b. The acceleration / deceleration request monitoring unit 10 transmits an emergency stop signal from the stop signal transmitting unit 124b, that is, an emergency stop signal is transmitted from the stop signal transmitting unit 124b after an emergency stop command is input to the stop command input device 124. In response to the restart signal transmitted from the restart signal transmission unit 126b, a restart request is output to the transfer amount deriving unit 22 described later.

  Further, the acceleration / deceleration request monitoring unit 10 transmits a speed change signal transmitted from the speed change signal transmission unit 128b as to whether an acceleration command or a deceleration command is input to the override device 128, that is, whether the override dial 128a is rotated. Monitoring is done by detecting the signal. The acceleration / deceleration request monitoring unit 10 receives the speed change signal transmitted from the speed change signal transmission unit 128b and outputs a speed change request including information on the override coefficient included in the speed change signal to the transfer amount derivation unit 22 described later. To do.

  The calculation unit 12 derives a machining path based on the NC program stored in the storage unit 4, derives a reference machine axis path, conversion processing of the reference machine axis path, acceleration / deceleration processing of the converted machine axis path, and special command input The corresponding machine axis direction (X-axis direction, Y-axis direction, W) of each support 102b, 106a, 108a, 110a per set cycle time according to the presence / absence of a special command input to the device 122 and the type of the special command. (Axis direction, Z-axis direction) and the like are calculated. The computing unit 12 includes a post-processing mechanical axis path deriving unit 20 and a transfer amount deriving unit 22 as functional blocks.

  The processed machine axis path deriving unit 20 obtains a processed machine axis path by performing various arithmetic processes based on the NC program stored in the storage unit 4. As illustrated in FIG. 3, the post-processing machine axis path deriving unit 20 includes a reference machine axis path deriving unit 24, a path converting unit 26, and an acceleration / deceleration filter 28 as functional blocks.

  Based on the NC program, the reference machine axis path deriving unit 24 derives a reference machine axis path that represents the position in each machine axis direction to which the supports 102b, 106a, 108a, 110a are to be transferred as a function of the set time t. Is. The reference mechanical axis path deriving unit 24 includes a program reading unit 30, a curved surface interpolation unit 32, a command path calculation unit 34, an acceleration / deceleration calculation unit 36, and a time adjustment unit 38 as functional blocks. .

  The program reading unit 30 reads a machining path (tool path) from the NC program stored in the storage unit 4. The machining path represents the trajectory that the tool 105 moves when machining the workpiece 100 and the speed of the tool 105. The machining path read by the program reading unit 30 is stored in the memory 5.

  The curved surface interpolation unit 32 performs an interpolation operation so that the machining path becomes a smooth path as necessary. When the interpolation calculation of the machining path is performed by the curved surface interpolation unit 32, the machining path after the interpolation calculation is stored in the memory 5.

  The command path calculation unit 34 calculates a command path that is a moving component for each machine axis direction from the machining path stored in the memory 5. The calculated command path for each machine axis direction is stored in the memory 5.

  The acceleration / deceleration calculation unit 36 calculates acceleration / deceleration according to acceleration / deceleration conditions for each machine axis direction with respect to the command path for each machine axis direction stored in the memory 5, that is, allowable acceleration or allowable jerk for each machine axis direction. And a machine axis path for each machine axis direction is calculated as a function of the set time t. The calculated machine axis path for each machine axis direction is stored in the memory 5 as a machine axis path before adjustment.

  The time adjustment unit 38 performs an operation for time adjustment on the pre-adjustment mechanical axis path stored in the memory 5 and calculates a reference mechanical axis path. The calculated reference machine axis path is stored in the memory 5.

  The path conversion unit 26 performs a specific conversion process on the reference machine axis path for each machine axis direction calculated by the acceleration / deceleration calculation unit 36 of the reference machine axis path deriving unit 24, thereby performing post-conversion for each machine axis direction. Deriving the machine axis path. The specific conversion process performed by the path conversion unit 26 includes a calculation process for correcting in advance an error caused by an acceleration / deceleration process described later by the acceleration / deceleration filter 28. As calculation processing for this error correction, the reference machine point regarding the point where the error with the converted machine axis path is equal to or less than the correction boundary value among the end points of the later-described block of the reference machine axis path for each machine axis direction A correction amount corresponding to an error between the mechanical axis path and the reference mechanical axis path obtained when the conversion unit acceleration / deceleration process, which is the same process as the later-described acceleration / deceleration process by the acceleration / deceleration filter 28, is performed on the mechanical axis path. Calculation processing is performed for each machine axis direction, and the reference machine axis path for each machine axis direction is corrected in a direction (cancellation direction) that cancels the error with the calculated correction amount for each machine axis direction. . As the calculation process for error correction, the specific conversion process performed by the path conversion unit 26 includes a post-conversion machine axis path among end points of blocks to be described later of a reference machine axis path for each machine axis direction. For the point where the error exceeds the correction boundary value, the correction amount for the point is calculated by proportional distribution from the correction amount for the two nearest command points before and after that point, and the calculated correction amount Similar to the above, a calculation process for correcting the reference machine axis path is performed.

  For example, when the trajectory of the tool 105 obtained from the reference machine axis path indicates a circle on the XY plane as shown in FIG. 4, the same conversion unit acceleration / deceleration processing as that described later is performed. When the process is performed on the reference machine axis path, the trajectory of the tool 105 indicated by the processed machine axis path is shifted radially inward of the path indicated by the reference machine axis path as shown in FIG. The path conversion unit 26 calculates a correction amount corresponding to an error between the machine axis path after processing that generates such a deviation and the reference machine axis path for each machine axis direction (X axis, Y axis) and The reference machine axis path for each machine axis direction (X axis, Y axis) is corrected in a direction that cancels the error with the calculated correction amount for each machine axis direction. As a result, the path conversion unit 26 calculates a post-conversion mechanical axis path that provides a trajectory of the tool 105 as shown in FIG. 5, for example. That is, in the form shown in FIG. 5, the trajectory of the tool 105 obtained from the post-conversion mechanical axis path results in a trajectory obtained by correcting the trajectory of the tool 105 obtained from the reference mechanical axis path radially outward.

  The acceleration / deceleration filter 28 transfers the support bodies 102b, 106a, 108a, 110a obtained from the post-conversion machine axis path to the post-conversion machine axis path for each machine axis direction derived by the path conversion unit 26. Acceleration / deceleration processing is performed to mitigate changes in speed. Details of the acceleration / deceleration processing by the acceleration / deceleration filter 28 will be described later.

  The transport amount deriving unit 22 supports each support body per predetermined set cycle time in each machine axis direction from the machine axis path after acceleration / deceleration processing by the acceleration / deceleration filter 28 (hereinafter, referred to as post-process machine axis path in some cases). The transfer amounts of 102b, 106a, 108a, and 110a are derived. The set cycle time is set as a length of time that the set time advances while the reference time advances by the reference cycle time. The reference cycle time is a time length corresponding to the interpolation calculation period when the transfer amount deriving unit 22 derives the transfer amount. When calculating the transfer amount, the transfer amount deriving unit 22 supports each of the supports 102b and 106a per set cycle time according to whether or not a special command is input to the special command input device 122 and the type of the special command. , 108a, 110a are calculated from the machine axis path after processing. Specifically, when there is no special command input to the special command input device 122, the transfer amount deriving unit 22 determines each support 102b, 106a, 108a, per set cycle time from the post-processing machine axis path. While the transfer amount in the corresponding machine axis direction of 110a is calculated, when a special command is input to the special command input device 122, the length of the set cycle time is accordingly sent to the special command input device 122. The length of the support 102b, 106a, per set cycle time after the change is changed from the length in the state immediately before the input of the special command to the length corresponding to the speed change of the object indicated by the special command. The transfer amounts in the machine axis direction corresponding to 108a and 110a are calculated from the post-process machine axis path.

  The transfer control unit 14 supports the support bodies 102b, 106a, 102b corresponding to the transfer parts 102c, 106b, 108b, 110b according to the transfer amounts of the support bodies 102b, 106a, 108a, 110a derived by the transfer amount deriving part 22. 108a and 110a are moved in the corresponding machine axis direction. Specifically, the transfer control unit 14 creates a servo command pulse for instructing the transfer amount for each machine axis direction derived by the transfer amount deriving unit 22 and creates the servo command pulse for each machine axis direction. Is transferred to the corresponding transfer units 102c, 106b, 108b, 110b, the amount of transfer per set cycle time indicated by the servo command pulse sent to each transfer unit 102c, 106b, 108b, 110b, per reference cycle time. Control for transferring the supports 102b, 106a, 108a, 110a corresponding to the above is performed.

  Next, a numerical control process performed by the numerical control apparatus 2 according to the present embodiment will be described. In order to simplify the following description, it is assumed that one unit of the set time and the set cycle time is equal to the length of the reference cycle time.

  First, the processed machine axis path deriving unit 20 of the calculation unit 12 derives a processed machine axis path based on the NC program stored in the storage unit 4 (step S2 in FIG. 6). The specific process of this derivation is shown in the flowchart of FIG.

  When deriving the machine axis path after this processing, first, the program reading unit 30 reads the NC program stored in the storage unit 4, and derives the machining path from the read NC program (step S22 in FIG. 7). The derived machining path is stored in the memory 5.

  Next, when the curved surface interpolation unit 32 needs to smooth the locus indicated by the machining path stored in the memory 5, the curved surface interpolation of the machining path is performed (step S24). The curved surface interpolation unit 32 stores the processing path after the curved surface interpolation in the memory 5.

  Next, the command path calculation unit 34 converts the machining path stored in the memory 5 into a movement component for each machine axis direction using an inverse kinematic relational expression, thereby moving the machine path for each machine axis direction. A command path representing the component is calculated (step S26). The command path calculation unit 34 stores the calculated command path for each machine axis direction in the memory 5.

  Next, the acceleration / deceleration calculation unit 36 performs acceleration / deceleration calculation according to the allowable acceleration and allowable jerk for each machine axis direction for the command path stored in the memory 5, thereby adjusting each machine axis direction. A front mechanical axis path is calculated (step S28). The acceleration / deceleration calculation unit 36 causes the memory 5 to store the pre-adjustment machine axis path for each calculated machine axis direction.

  Thereafter, the time adjustment unit 38 adjusts the time of the pre-adjustment machine axis path for each machine axis direction stored in the memory 5 to calculate a reference machine axis path for each machine axis direction (step S29). Specifically, the conversion unit acceleration / deceleration processing described later is directly performed on the part corresponding to the acceleration period immediately after the start of the object transfer and the part corresponding to the deceleration period immediately before the stop of the object transfer in the pre-adjustment machine axis path. When performing the above, there is a difference in the set times in those portions. For this reason, the time adjustment unit 38 performs time adjustment for eliminating the deviation of the set time in advance before the conversion unit acceleration / deceleration processing described later to derive the reference mechanical axis path. In this embodiment, since the linear distribution function f (T) is used in the conversion unit acceleration / deceleration processing as will be described later, the time adjustment unit 38 obtains the following equation according to the linear distribution function f (T). Used to adjust the time of the machine axis path before adjustment. The following expression represents a general expression of a function used for time adjustment in a form corresponding to a linear distribution form.

  In this equation, t is the set time before time adjustment, tj (t) is the time after time adjustment, and Tj is the machine axis direction corresponding to the pre-adjustment machine axis path to be adjusted. Is the time constant of. Further, t = 0 is a start time when the object starts to be transferred, and t = te is an end time when the object is stopped.

Specifically, the time adjustment unit 38, when performing the time adjustment, sets the pre-adjustment mechanical axis path from the set time t = 0 to t = te to a time unit tt [ i] ≦ t ≦ tt [i + 1] (i = 0, 1, 2,..., n−1, tt [0] = 0, tt [n] = te), and each time tt The machine axis position P _A i at [i] is obtained, and the time array is adjusted according to the time adjustment function t [i] = tj (tt [i]). Then, the time adjustment unit 38 has a plurality of linear blocks P _A i (t) (t [i] in which the position of the reference machine axis path at each time t [i] after the time adjustment is the machine axis position P _A i. ] ≦ t ≦ t [i + 1]) (i = 0, 1, 2,..., N−1, t [0] = Tj / 2, t [n] = te−Tj / 2). The plurality of linear blocks P _A i (t) constitute a reference machine axis path P _A (t). That is, in the time adjustment process, the time adjustment unit 38 adjusts the time of the pre-adjustment mechanical axis path and obtains a plurality of linear blocks that are subdivided, thereby obtaining a reference mechanical axis path including the plurality of linear blocks. The time adjustment unit 38 causes the memory 5 to store a reference machine axis path P _A (t) composed of a plurality of blocks P _A i (t) for each machine axis direction obtained as described above.

  Next, the path conversion unit 26 performs conversion processing on the reference machine axis path for each machine axis direction stored in the memory 5 to derive a converted machine axis path for each machine axis direction (step S30). . Specifically, the path conversion unit 26 performs a calculation process according to the flowchart shown in FIG. 8 as the conversion process. Hereinafter, this calculation process will be described in detail.

  First, the path conversion unit 26 performs the following conversion unit acceleration / deceleration processing with respect to the reference machine axis path for each machine axis direction derived as described above and stored in the memory 5 (step S42). ). In order to simplify the explanation, the following description will focus on the acceleration / deceleration processing of the conversion part of the reference machine axis path in one machine axis direction, but the reference machine axis paths in all other machine axis directions are also described. The same conversion unit acceleration / deceleration processing is performed for each machine axis direction.

The path conversion unit 26 performs acceleration / deceleration processing for each block P _A i (t) of the reference machine axis path. Specifically, the path converting unit 26, a setting time which is in the reference machine axis predetermined block P _{A i} (t) in the path and t, the mechanical axis direction of the reference machine axis path subjected to the conversion unit deceleration processing the half of the constant Tj when set as a, a distribution function which satisfies the following equation (4) and f (T), the converted portion deceleration process corresponding to the predetermined block P _{a i} (t) The block of the machine axis path is Q _A i (t), and the speed function of the predetermined block P _A i (t) at each set time T in the distribution target section from t−a to t + a is represented by P _A i ′ ( T), a speed function Q _A i ′ (t) that is a first-order differential function at a set time t of the block Q _A i (t) after the conversion unit acceleration / deceleration processing is calculated based on the following equation (5). To do.

In this conversion unit acceleration / deceleration processing, as shown in the above formulas (4) and (5), the speed at each set time T within the distribution target section [ta, t + a] before and after the set time t. The distribution value P _A i ′ (T) · f (t−T when the function P _A i ′ (T) is distributed according to the distribution function f (T) in the interval from T−a to T + a ) Is integrated over the interval from t−a to t + a.

  The distribution function f (T) used by the path conversion unit 26 in the above arithmetic processing is the same as the distribution function used when the acceleration / deceleration filter 28 performs the acceleration / deceleration processing. In this embodiment, the path conversion unit 26 and the acceleration / deceleration filter 28 both perform distribution in a linear distribution format (a format in which the speed function is evenly distributed over the entire distribution interval), and the distribution function f (T) corresponding to this distribution format is It is a linear partition function expressed by the following equation.

The speed function Q _A i ′ (t) obtained by the above equation (5) is obtained by calculating the time constant Tj = 2a from the section width t [i + 1] −t [i] of the block P _A i (t) corresponding to the speed function. Larger and smaller cases are represented differently. The speed function Q _A i ′ (t) in each case is as follows.

When the section width t [i + 1] −t [i] of the block P _A i (t) is larger than the time constant Tj = 2a, t [i] −a <t [i] <t [i] + a < The relationship t [i + 1] −a <t [i + 1] <t [i + 1] + a holds (see FIG. 9). The speed function P _A i ′ (T) of the block P _A i (t) in this case is represented by a thick line in FIG. In this case, the calculated speed function Q _A i ′ (t) is equal to the speed function q _A i11 ′ (t) in the section from the following t [i] −a to t [i] + a and t [i ] + 'and (t), t [i + 1] t from -a [i + 1] + speed function of the section up to a _q a i13' a from t [i + 1] speed function of the section up -a _q a i12 and (t) It is subdivided into.

_A speed function q _A i11 ′ (t) in a section from t [i] −a to t [i] + a is expressed by the following equation (7).

_A speed function q _A i12 ′ (t) in a section from t [i] + a to t [i + 1] −a is expressed by the following equation (8).

_A function q _A i13 ′ (t) in a section from t [i + 1] −a to t [i + 1] + a is expressed by the following equation (9).

Next, when the section width t [i + 1] −t [i] of the block P _A i (t) is smaller than the time constant Tj = 2a, the section width t [i + 1] −t [i] is the time constant. It is classified into a case where it is smaller than Tj = 2a and larger than a value a half of the time constant Tj, and a case where the section width t [i + 1] -t [i] is smaller than a value a half of the time constant Tj. . FIG. 10 shows a case where the section width t [i + 1] −t [i] of the block P _A i (t) is smaller than the time constant Tj = 2a and larger than the half value a of the time constant Tj. In this case, the relationship t [i] −a <t [i] <t [i + 1] −a <t [i] + a <t [i + 1] <t [i + 1] + a holds. FIG. 11 shows a case where the section width t [i + 1] −t [i] of the block P _A i (t) is smaller than the half value a of the time constant Tj = 2a. The relationship t [i] −a <t [i + 1] −a <t [i] <t [i + 1] <t [i] + a <t [i + 1] + a holds. The speed function P _A i ′ (T) of the block P _A i (t) in these cases is represented by a thick line in FIGS. In these cases, the calculated speed function Q _A i ′ (t) is the following speed function q _A i21 ′ (t) in a section from t [i] -a to t [i + 1] -a. , T [i + 1] -a to t [i] + a interval speed function q _A i22 ′ (t) and t [i] + a to t [i + 1] + a interval velocity function q _A i23 ′ ( t).

_A function q _A i21 ′ (t) in a section from t [i] -a to t [i + 1] -a is expressed by the following equation (10).

_A function q _A i22 ′ (t) in a section from t [i + 1] −a to t [i] + a is expressed by the following equation (11).

_A function q _A i23 ′ (t) in a section from t [i] + a to t [i + 1] + a is expressed by the following equation (12).

Then, the path conversion unit 26 calculates the overall speed function Q _A ′ (t) by accumulating the speed functions Q _A i ′ (t) for each block calculated as described above, and the calculated overall The mechanical axis path Q _A (t) after the conversion unit acceleration / deceleration processing is calculated by integrating the speed function Q _A ′ (t).

Thereafter, the path conversion unit 26 subtracts the reference machine axis path P _A (t) for the corresponding machine axis direction from the calculated machine axis path Q _A (t) after the conversion unit acceleration / deceleration processing for each machine axis direction. Thus, a reference error, which is an error between the machine axis path Q _A (t) after the conversion unit acceleration / deceleration processing and the reference machine axis path after the time adjustment in each machine axis direction, is calculated (step S44). At this time, the path conversion unit 26 sets the reference machine after the time adjustment corresponding to the end point value of each block Q _A i (t) of the machine axis path Q _A (t) after the conversion unit acceleration / deceleration processing. The reference error is calculated by subtracting the end point value of the axis path block P _A i (t). Specifically, the path conversion unit 26 calculates a reference error e [i] for each end point of each block Q _A i (t) by the following equation (13).

e [i] = Q _A (tt [i]) − P _A (t [i]) (13)
The path conversion unit 26 calculates the reference error e [i] for each end point of each block Q _A i (t) as described above, and then uses each end point of each block Q _A i (t) as a correction boundary value. Based on this, the first correction target point and the second correction target point are classified (step S46). The second correction target point includes a corner portion (for example, the vertex of the polygon when the trajectory indicated by the mechanical axis path indicates a polygonal trajectory), and the specification immediately after the first start point of the mechanical axis path. The end point of the block included in the section (constant section immediately after the start of transfer), the end point of the block included in the specific section (predetermined section immediately before the stop of transfer) immediately before the last end point of the machine axis path, and the speed change need to be suddenly relieved There is an end point of a block included in a specific section. The first correction target point is an end point other than the second correction target point, and is an end point of a block included in a section instructing smooth transfer in the mechanical axis path. The correction boundary value is stored as a set value in the memory 5, and the path conversion unit 26 has a point that the reference error e [i] is less than or equal to the correction boundary value among the end points of each block Q _A i (t). Is a first correction target point, and a point where the reference error e [i] exceeds the correction boundary value is a second correction target point.

Then, the path conversion unit 26 calculates a correction amount for each of the first correction target point and the second correction target point (step S48). Specifically, for the first correction target point, the path conversion unit 26 sets a value equal to the reference error e [i] of the point as the correction amount h1 [i] for the first correction target point. The path conversion unit 26 for the second correction target point, and its second correction target point and the distance L _b between the most recent first correction target point before the second correction target point, the first The second correction target point is obtained by obtaining _a distance La between the two correction target points and the first correction target point immediately after the second correction target point, and proportionally allocating by the following equation (14). A correction amount h2 [i] is obtained. In the following equation (14), e _b [i] is a reference error for the first correction target point immediately before the second correction target point, and e _a [i] is the second correction target. This is a reference error for the first correction target point immediately after the target point.

h2 [i] = (e _b [i] × L _a + e _a [i] × L _b ) / (L _b + L _a ) (14)
Then, the path conversion unit 26 stores in the memory 5 the correction amounts h1 [i] and h2 [i] for each end point of each block Q _A i (t) obtained as described above.

Thereafter, the path conversion unit 26 corrects the reference machine axis path P _A (t) with the correction amounts h1 [i] and h2 [i] for each end point of each block Q _A i (t) stored in the memory 5. The post-conversion mechanical axis path P _B (t) is calculated. Specifically, the path conversion unit 26 corrects the first correction target point among the end points of each block P _A i (t) of the reference machine axis path P _A (t) with the correction amount h1 [i], Of the end points of each block P _A i (t), the second correction target point is corrected with the correction amount h2 [i] (step S50). Specifically, for the first correction target point, the path conversion unit 26 subtracts the correction amount h1 [i] set for the point from the end point value of the block P _A i (t) of the corresponding reference machine axis path. Then, the corrected value of the first correction target point is obtained, and for the second correction target point, the correction amount h2 set for the point from the end point value of the block P _A i (t) of the corresponding reference machine axis path The corrected value of the second correction target point is obtained by subtracting [i]. The path conversion unit 26 stores the mechanical axis path composed of the blocks obtained by the correction in this way in the memory 5 as a converted mechanical axis path. Specifically, the path conversion unit 26 corrects each block P _B i (t) (t [i] ≦ t ≦ t [i + 1]) (i = 0, 1, 2,... The converted machine axis path P _B (t) consisting of n−1) is stored in the memory 5.

Next, the acceleration / deceleration filter 28 derives a processed mechanical axis path by performing acceleration / deceleration processing on the converted mechanical axis path stored in the memory 5 (step S32 in FIG. 7). The acceleration / deceleration process by the acceleration / deceleration filter 28 is exactly the same as the acceleration / deceleration process by the path conversion unit 26. Namely, acceleration and deceleration filter 28, speed function _{P B} in the formula (5) speed function _{P A} i of the 'blocks _P B i after the conversion of the (T) mechanical axis path _P B (t) (t) The velocity function Q _B i ′ (t) of each block Q _B i (t) of the post-processing mechanical axis path Q _B (t) is calculated using the expression replaced with i ′ (t), and each of the calculated blocks The speed function Q _B i ′ (t) is integrated in the same manner as described above to calculate the overall speed function Q _B ′ (t), and the calculated overall speed function Q _B ′ (t) is integrated. A post-processing machine axis path Q _B (t) is obtained. Specifically, the acceleration / deceleration filter 28 uses a distribution function f (T) that satisfies the above equation (4), and calculates the speed function Q _B i ′ (t) based on the following equation (15). To do.

  By performing the acceleration / deceleration process by the acceleration / deceleration filter 28, a post-process mechanical axis path that reduces the change in the transfer speed of each of the supports 102b, 106a, 108a, 110a indicated by the post-conversion mechanical axis path is derived. The By the way, with this acceleration / deceleration processing, the processed machine axis path changes from the converted machine axis path and causes an error with respect to the converted machine axis path. Since the rear machine axis path is corrected in a direction to cancel out the error caused by the acceleration / deceleration process in advance, the post-process machine axis path does not cause an error with respect to the reference machine axis path. The processed machine axis path calculated by the acceleration / deceleration filter 28 is stored in the memory 5.

  The acceleration / deceleration processing by the acceleration / deceleration filter 28 is a conventional acceleration / deceleration processing after pulse interpolation (after interpolation) that can correct the portion of abnormal data that causes mechanical shock so that the change in the transfer speed indicated by the portion is reduced. A process substantially equivalent to the acceleration / deceleration process) is performed on the converted machine axis path.

  Specifically, in the conventional acceleration / deceleration processing after pulse interpolation, the mechanical axis path is set to p (t), the cycle time is set to Δt seconds, the time constant is set to τ = n · Δt (seconds), Cycle (0), and the cycle from time i · Δt to time (i + 1) · Δt is cycle (i). Then, the transfer pulse Δp (i) indicating the transfer amount in the cycle (i) can be expressed by the following equation (16).

Δp (i) = p ((i + 1) · Δt) −p (i · Δt) (16)
In post-interpolation acceleration / deceleration processing using a linear distribution function, the transfer pulse Δp (i) of cycle (i) is set to a length equal to the time constant τ between cycle (i) and cycle (i + n−1). The transfer pulses of all cycles are distributed according to a distribution method in which (p ((i + 1) · Δt) −p (i · Δt)) / n is evenly distributed to each cycle in the distribution interval, and then each cycle The transfer pulse after the acceleration / deceleration process after interpolation is calculated by integrating the distribution value of the transfer pulse every time. From this, the transfer pulse Δq [j] of the j-th cycle (j) after post-interpolation acceleration / deceleration processing is the transfer pulse Δp [i] of each cycle from cycle (j−n + 1) to cycle (j). (I = j−n + 1, j−n + 2,..., J−2, j−1, j) is obtained by accumulating each distribution value multiplied by 1 / n. Specifically, the transfer pulse Δq [j] after post-interpolation acceleration / deceleration processing is calculated by the following equation (17).

  By the way, since the transfer pulse Δq [j] after the post-interpolation acceleration / deceleration process corresponds to the transfer amount during the cycle time Δt seconds in the cycle (j), both sides of the above equation (17) are divided by Δt. To obtain the average speed in cycle (j). That is, the average speed in the cycle (j) is obtained by the following equation (18).

  Here, since the time constant τ = n · Δt, it can be expressed as 1 / n = 1 / (τ · Δt), where j · Δt is time t and i · Δt is time T, 18) can be transformed into the following equation (19).

  This mathematical formula (19) shows the distribution value when the average speed (p (T + Δt) −p (T)) / Δt in the cycle (i) is evenly distributed in the distribution section having the section width of τ · Δt. This means that integration is performed for each cycle from the cycle of i = j−n + 1 to the cycle of i = j. Since the average speed in the cycle (j) is equal to the average speed Δq (t) / Δt at time t, the following equation (20) is established.

  Here, when n is infinite, Δt can be expressed as dt, Δq (t) / Δt can be expressed as q ′ (t), and (p (T + Δt) −p (T)) / Δt. Can be expressed as p ′ (T). As described above, j · Δt = t, and (j−n + 1) · Δt = j · Δt−n · Δt + Δt = t−τ + Δt. When n is infinite, Δt is approximated to 0, so t−τ + Δt becomes t−τ. From these, the following equation (21) can be derived based on the above equation (20).

  In the above embodiment, since the half value of the time constant is defined as a, the time constant τ in the above equation (20) is replaced with 2a and the linear distribution function f (T) = 1 / 2a. Therefore, the above equation (21) is expressed as the following equation (22).

From this equation (22), if an acceleration / deceleration process is to be performed so that the converted machine axis path directly corresponds to the conventional post-interpolation acceleration / deceleration process, a predetermined block Q _A i (t) after the acceleration / deceleration process is obtained. It is conceivable that the velocity function Q _A i ′ (t) is determined according to the following equation (23).

  Further, in the conventional post-interpolation acceleration / deceleration processing, the transfer pulse of each cycle is distributed in a distribution section having a section width equal to the time constant after the time corresponding to the end point of the cycle. Accordingly, the corresponding distribution function f (T) satisfies the following expression (24).

  The contents of the acceleration / deceleration processing expressed by these equations (24) and (23) are as follows. The distribution range and the integration range of the distribution value of the speed function are compared with the distribution interval and the integration range in the acceleration / deceleration processing by the acceleration / deceleration filter 28. The content of the acceleration / deceleration process by the acceleration / deceleration filter 28 is substantially the same, only by shifting by a value a which is half the time constant. Accordingly, the acceleration / deceleration process by the acceleration / deceleration filter 28 has an effect substantially equivalent to the conventional post-interpolation acceleration / deceleration process, that is, an effect of reducing the change in the transfer speed of the abnormal data portion.

  Next, the transfer amount deriving unit 22 initially sets a predetermined set time Ts to 0 and also sets a predetermined set cycle time ΔT to 1 (step S6 in FIG. 6).

  Next, the transfer amount deriving unit 22 further calculates a value g (t) obtained by differentiating the set time function T (t) representing the set time Ts as a function with respect to the reference time t with the reference time t at the reference time t. The differentiated value dg is initially set to 0 (step S8).

Next, the transfer amount deriving unit 22 determines each machine axis direction per set cycle time ΔT based on the data of the processed machine axis path Q _B (t) for each machine axis direction stored in the memory 5. Each transfer amount dP [axis] is calculated by the following equation (25) (step S10).

dP [axis] = Q _B (T + ΔT) [axis] −Q _B (T) [axis] (25)
Next, the transfer amount deriving unit 22 uses the transfer amount dP [axis] for each machine axis direction calculated as described above to determine the current transfer speed V of the object in the synthesis direction in each machine axis direction (hereinafter referred to as the synthesis rate). V) is calculated by the following equation (26) (step S11).

V = | dP [] | (26)
In this equation (26), | dP [] | is a combined transfer amount obtained by synthesizing the transfer amount for each machine axis direction per set cycle time ΔT, and the transfer amount deriving unit 22 determines the direction of each machine axis. It is obtained by synthesizing each transfer amount dP [axis].

  Next, the transfer control unit 14 drives each transfer unit 102c, 106b, 108b, 110b according to the transfer amount dP [axis] for each machine axis direction derived by the transfer amount deriving unit 22 (step S12). . At this time, the transfer control unit 14 instructs a servo command to instruct to transfer each support 102b, 106a, 108a, 110a by the transfer amount dP [axis] per reference cycle time (1 mmsec in the present embodiment). A pulse is created for each machine axis direction, and the created servo command pulse for each machine axis direction is output to the corresponding one of the transfer units 102c, 106b, 108b, 110b. As a result, the servo motors of the respective transfer units 102c, 106b, 108b, 110b, according to the servo command pulse from the transfer control unit 14, support members 102b, 106a, 108a, 108b, 106a, 108a, The corresponding one of 110a is transferred in the corresponding machine axis direction.

Next, the transfer amount deriving unit 22 determines whether or not the processing for the entire period of the processed mechanical axis path Q _B (t) stored in the memory 5 has been completed (step S14). Specifically, the data of the post-processing mechanical axis path Q _B (t) stored in the memory 5 is subjected to arithmetic processing for obtaining the transfer amount dP [axis] in order for each set cycle time ΔT as described above. Therefore, in step S14, it is determined whether or not the calculation process has been completed for the entire period of the processed mechanical axis path Q _B (t) stored in the memory 5. Here, when the transfer amount deriving unit 22 determines that the processing for the entire period of the processed mechanical axis path Q _B (t) stored in the memory 5 has been completed, the numerical control process by the numerical controller 2 is performed. finish.

  In parallel with the arithmetic processing as described above, the acceleration / deceleration request monitoring unit 10 monitors whether or not a special command is input to the special command input device 122 in a process as shown in FIG.

  Specifically, the acceleration / deceleration request monitoring unit 10 first determines whether or not the machine tool is in continuous operation (step S122). Here, if the acceleration / deceleration request monitoring unit 10 determines that the machine tool is in continuous operation, it next determines whether or not the emergency stop button 124a has been pressed (step S124).

  When the acceleration / deceleration request monitoring unit 10 detects an emergency stop signal transmitted from the stop signal transmission unit 124b and determines that the emergency stop button 124a has been pressed, it issues an emergency stop request (step S126). Thereafter, the process of step S122 is performed again. On the other hand, when determining that the emergency stop button 124a has not been pressed, the acceleration / deceleration request monitoring unit 10 next determines whether or not the override dial 128a has been turned (step S128).

  Here, when the acceleration / deceleration request monitoring unit 10 detects the speed change signal transmitted from the speed change signal transmission unit 128b and determines that the override dial 128a is rotated, the rotation direction of the override dial 128a is determined. Then, a speed change request including information on the override coefficient k corresponding to the rotation amount is issued (step S130), and then the process of step S122 is performed again. If the acceleration / deceleration request monitoring unit 10 determines that the override dial 128a is not turned, the acceleration / deceleration request monitoring unit 10 performs the process of step S122 again without issuing a speed change request.

  Further, when the acceleration / deceleration request monitoring unit 10 determines that the machine tool is not in continuous operation in the determination in step S122, each support body by each transfer unit 102c, 106b, 108b, 110b according to the emergency stop request. It is determined whether or not the transfer of 102b, 106a, 108a, 110a is stopped (step S132).

  If the acceleration / deceleration request monitoring unit 10 determines that the transfer of each of the supports 102b, 106a, 108a, 110a is stopped, it next determines whether the restart button 126a has been pressed (step). S134). At this time, if the acceleration / deceleration request monitoring unit 10 detects the restart signal transmitted from the restart signal transmitting unit 126b and determines that the restart button 126a has been pressed, it issues a restart request (step S1). After that, the process of step S122 is performed again. If the acceleration / deceleration request monitoring unit 10 determines that the restart button 126a has not been pressed, the acceleration / deceleration request monitoring unit 10 performs the process of step S122 again without issuing a restart request.

On the other hand, when the transfer amount deriving unit 22 determines in the determination in step S14 that the processing for the entire period of the processed mechanical axis path Q _B (t) stored in the memory 5 has not been completed, Then, it is determined whether or not an emergency stop request is issued from the acceleration / deceleration request monitoring unit 10 (step S16 in FIG. 6). Here, when it is determined that the emergency stop request has been issued, the transfer amount deriving unit 22 executes the emergency stop process shown in FIG.

  Specifically, the transfer amount deriving unit 22 first temporarily stores the current value of the synthesis speed V of the object in the memory 5 (step S52), and then the deceleration stop period time and the stop cycle time. A variation function g (t) is calculated (step S54). Here, the transfer amount deriving unit 22 determines that the value gs of the stop-time cycle time fluctuation function g (t) at the beginning of the deceleration stop period time is equal to the set cycle time ΔT, and the stop-time cycle time fluctuation at the end of the deceleration stop period time. The deceleration stop period time and stop cycle time variation function g (t) are calculated such that the value ge of the function g (t) is 0 and the combined speed at the start of the deceleration stop period time is V. To do. The calculation process of the deceleration stop period time and the stop cycle time variation function g (t) is shown in FIG. FIG. 16 shows the calculation of the restart acceleration period time and restart cycle time fluctuation function g (t) in the restart process described later, and the speed change period time and speed change cycle time fluctuation function g in the speed change process. A process that is commonly applicable to the calculation of (t) is shown. That is, the transfer amount deriving unit 22 is a period time required for the operation with the speed change in the common calculation process in the operation with the speed change of the object performed in accordance with the input of the special command to the special command input device 122. And a cycle time variation function g (t) representing the variation of the cycle time within the period time.

  The stop cycle time variation function g (t) is, for example, when the object is moving at a constant speed at the beginning of the deceleration stop period time (when the transfer amount deriving unit 22 receives an emergency stop request). 17, when the object is accelerating at the beginning of the deceleration stop period time, the object is represented by, for example, a curve as shown in FIG. 18, and the object is at the beginning of the deceleration stop period time. When the vehicle is decelerating, for example, it is represented by a curve as shown in FIG.

  In calculating the deceleration stop period time and the stop-time cycle time fluctuation function g (t), the transfer amount deriving unit 22 first determines the second-order differential value j of the stop-time cycle time fluctuation function g (t) and the stop-time cycle time. The slope α of the first derivative of the variation function g (t) is obtained (step S142 in FIG. 16). At this time, the secondary differential value j is obtained by the following equation (27), and the slope α of the primary differentiation is obtained by the following equation (28).

j = J / V (27)
α = A / V (28)
J is an allowable jerk in the combined movement direction of the object, and A is an allowable acceleration in the combined movement direction of the object. These values of J and A are parameters set for the calculation process of time and g (t) of the present embodiment, and the allowable jerk as an acceleration / deceleration condition defined based on the machine characteristics of the machine tool and The value is set to about half of the allowable acceleration.

  Next, the transfer amount deriving unit 22 determines whether or not the value at the point of t = 0 of the first-order differential value dg at the reference time t of the stop cycle time variation function g (t) is 0 or more ( Step S144). Here, when the transfer amount deriving unit 22 determines that the value of this dg is 0 or more, the second derivative of the first half of the quadratic curve represented by the stop-time cycle time variation function g (t). The value j1 is set to -j (step S146). On the other hand, when determining that the value of dg is smaller than 0, the transfer amount deriving unit 22 sets the secondary differential value j1 to j (step S148).

  Next, the transfer amount deriving unit 22 compares the vertex of the first half of the quadratic curve of the stop cycle time variation function g (t) with respect to the start point of the stop cycle time variation function g (t) in the deceleration stop period time. The position (t0, E0) is temporarily calculated by the following equations (29) and (30), and the cycle time at the time of stop in the range from the top of the first half of the quadratic curve to the end of the second half of the quadratic curve A change amount E of the value of the variation function g (t) is provisionally calculated by the following equation (31) (step S150).

t0 = −dg / j1 (29)
E0 = (dg / 2) × t0 (30)
E = ge-gs-E0 (31)
Next, the transfer amount deriving unit 22 determines whether or not the change amount E of the value of the stop-time cycle time variation function g (t) is 0 or more (step S152). Here, if the transfer amount deriving unit 22 determines that the change amount E is equal to or greater than 0, the transfer amount deriving unit 22 obtains the second-order differential value j1 of the first half of the quadratic curve of the stop cycle time variation function g (t). j is set, and the second-order differential value j2 of the second half of the quadratic curve of the stop-time cycle time variation function g (t) is set to −j (step S154). On the other hand, when determining that the change amount E is smaller than 0, the transfer amount deriving unit 22 sets the secondary differential value j1 of the first half of the quadratic curve to −j and the quadratic curve. Is set to j, and the gradient α of the primary derivative of g (t) is set to be positive or negative (step S156).

  Next, the transfer amount deriving unit 22 calculates the relative position (t0, E0) of the vertex of the first half of the quadratic curve with respect to the start point of the quadratic curve of the cycle time variation function g (t) at the time of stop. , (30), and the amount of change E of g (t) from the top of the first half to the end of the second half of the quadratic curve is recalculated using the above equation (22) (step S158).

  Next, the transfer amount deriving unit 22 determines the value gu of the apex of the first half of the quadratic curve of the stop cycle time variation function g (t) and the starting point of the quadratic curve to the inflection point. , Time t1 from the inflection point of the quadratic curve to the end point, change amount G1 of g (t) in the range from the start point to the inflection point of the quadratic curve, and the quadratic curve A change amount G2 of g (t) in the range from the inflection point to the end point of the curve and a change amount G of g (t) in the range from the start point to the end point of the quadratic curve are calculated (step S160). . At this time, the transfer amount deriving unit 22 calculates the gu by the following equation (32), calculates the t1 by the following equation (33), and calculates the t2 by the following equation (34). Further, the transfer amount deriving unit 22 calculates G1 by the following equation (35), calculates G2 by the following equation (36), and calculates G by the following equation (37).

gu = gs + E0 (32)
t1 = α / j1 (33)
t2 = −α / j2 (34)
G1 = (α / 2) × t1 (35)
G2 = (α / 2) × t2 (36)
G = G1 + G2 (37)
Next, the transfer amount deriving unit 22 determines that the absolute value | E | of the change amount of g (t) in the range from the first half of the quadratic curve to the end of the second half is from the start point to the end point of the quadratic curve. It is determined whether or not the absolute value | G | of the change amount of g (t) in the range up to is greater than or equal to (step S162). Here, when | E | is greater than or equal to | G |, a straight line portion is interposed between the first and second curve portions of the quadratic curve as shown in FIG. Therefore, the transfer amount deriving unit 22 has three sections: a first half curve section 0 ≦ t ≦ T1, a straight section T1 <t <T2, and a second curve section T2 ≦ t ≦ time. The cycle time fluctuation function g (t) at the time of stopping in each of the above is individually obtained (step S164).

  Specifically, the transfer amount deriving unit 22 obtains a stop-time cycle time variation function g (t) in a section of 0 ≦ t ≦ T1 by the following equation (38).

g (t) = gu + (j1 / 2) × (t−t0) ² (38)
Further, the transfer amount deriving unit 22 obtains the stop cycle time variation function g (t) in the section of T1 <t <T2 by the following equation (39).

g (t) = gu + G1 + α × (t−T1) (39)
Further, the transfer amount deriving unit 22 obtains a stop-time cycle time variation function g (t) in a section of T2 ≦ t ≦ time by the following equation (40).

g (t) = ge + (j2 / 2) × (t-time) ² (40)
In this case, the T1 is the time taken from the start point of the quadratic curve to the first inflection point (end point of the first half curve portion), and is obtained by the following equation (41), and the T2 Is the time taken from the start point of the quadratic curve to the second inflection point (end point of the straight line portion), and is obtained by the following equation (42). Further, the deceleration stop period time is a time taken from the start point to the end point of the quadratic curve, and is obtained by the following equation (43).

T1 = t0 + t1 (41)
T2 = T1 + (EG) / α (42)
time = T2 + t2 (43)
On the other hand, when the | E | is smaller than the | G |, there is no straight line portion between the first half curve portion and the second half curve portion of the quadratic curve, and these two curve portions are continuous. Therefore, the transfer amount deriving unit 22 changes the cycle time at the time of stop in each of the two sections of the first half curve section 0 ≦ t ≦ T1 and the second half curve section T1 <t ≦ time. The function g (t) is obtained individually (step S165).

  Specifically, the transfer amount deriving unit 22 obtains a stop-time cycle time variation function g (t) in a section of 0 ≦ t ≦ T1 by the following equation (44).

g (t) = gu + (j1 / 2) × (t−t0) ² (44)
Further, the transfer amount deriving unit 22 obtains a stop cycle time variation function g (t) in a section of T1 <t ≦ time by the following expression (45).

g (t) = ge + (j2 / 2) × (t-time) ² (45)
In this case, the deceleration stop period time is obtained by the following equation (46).

time = T1 + t2 (46)
Here, T1 is the time taken from the start point of the quadratic curve to the inflection point (end point of the first half curve portion), and is obtained by the following equation (47).

T1 = t0 + t1 (47)
Moreover, t1 is calculated | required by the following formula | equation (48), and t2 is calculated | required by the following formula | equation (49).

t1 = [2 × E × j2 / {j1 × (j2−j1)}] ^1/2 (48)
t2 = −j1 / j2 × t1 (49)
As described above, the stop cycle time fluctuation function g (t) and the deceleration stop period time are obtained.

  Next, the transfer amount deriving unit 22 initially sets the reference time t to 0 (step S56 in FIG. 13).

  Thereafter, the transfer amount deriving unit 22 counts up the reference time t by 1 (step S58), and subsequently, the set cycle time during deceleration that is the cycle time of the set time Ts in this emergency stop process (deceleration stop period time). dT2 is calculated based on the stop-time cycle time variation function g (t) obtained as described above (step S60). At this time, the length of the set cycle time dT2 during deceleration calculated is the state immediately before the emergency stop request is issued from the acceleration / deceleration request monitoring unit 10, that is, immediately before the emergency stop button 124a of the stop command input device 124 is pressed. The length is reduced from the length of the set cycle time ΔT in the state. The transfer amount deriving unit 22 calculates the deceleration set cycle time dT2 by substituting the reference time t counted up in step S58 into the stop cycle time variation function g (t).

  Next, the transfer amount deriving unit 22 transfers the transfer amount dP [axis] for each machine axis direction per set cycle time dT2 during deceleration, that is, the transfer amount dP [axis] for each machine axis direction from the set time Ts to T + dT2. ] Is calculated by the following equation (50) (step S62).

_{dP [axis] = Q B (} T + dT2) [axis] -Q B (T) [axis] ··· (50)
The transfer amount dP [axis] for each machine axis direction per deceleration set cycle time dT2 thus determined is per set cycle time ΔT in a state immediately before an emergency stop request is issued from the acceleration / deceleration request monitoring unit 10. This is smaller than the transfer amount for each machine axis direction.

  Thereafter, similarly to step S12, the transfer control unit 14 determines each transfer unit according to the transfer amount dP [axis] for each machine axis direction per deceleration set cycle time dT2 calculated by the transfer amount deriving unit 22. 102c, 106b, 108b, 110b is driven, and the supports 102b, 106a, 108a, 110a corresponding to the transfer units 102c, 106b, 108b, 110b are transferred by dP [ Axis] is transferred (step S64).

  Next, the transfer amount deriving unit 22 updates the set time Ts by adding the set cycle time dT2 during deceleration calculated in Step S60 to the set time Ts (Step S66).

  Thereafter, the transfer amount deriving unit 22 determines whether or not the reference time t is equal to or greater than the deceleration stop period time (step S68). Here, when the transfer amount deriving unit 22 determines that the reference time t is equal to or longer than the deceleration stop period time, the transfer of the object is stopped, and the transfer amount deriving unit 22 Then, the value of the synthesis speed V stored in the memory 5 is read (step S70). Thereafter, when a restart command is input to the restart command input device 126, a restart process (see FIG. 14) for restarting the transfer of the object is performed. On the other hand, when it is determined in step S68 that the reference time t has not passed the deceleration stop period time, the transfer amount deriving unit 22 performs the processing from step S58 onward.

  Next, the restart process of the object will be described. In this restart process, the transfer amount deriving unit 22 determines whether or not a restart request is output from the acceleration / deceleration request monitoring unit 10 (step S72). Here, when it is determined that the restart request is not issued from the acceleration / deceleration request monitoring unit 10, the transfer amount deriving unit 22 repeatedly performs the determination in step S72. On the other hand, when the transfer amount deriving unit 22 determines that the restart request is issued from the acceleration / deceleration request monitoring unit 10, the transfer acceleration period time and the restart cycle time variation function g (t ) Is calculated (step S74). At this time, the transfer amount deriving unit 22 has the value gs of the restart cycle time variation function g (t) at the start of the restart acceleration period time being 0, and the restart cycle time at the end of the restart acceleration period time. The restart acceleration period time and the cycle time at restart satisfying the condition that the value ge of the variation function g (t) is equal to the current override coefficient k and the combined speed at the end of the restart acceleration period time is V. A variation function g (t) is calculated. The calculation process of the restart acceleration period time and the restart cycle time variation function g (t) is the above-described calculation process of the deceleration stop period time and the stop cycle time variation function g (t) (steps S142 to S142 in FIG. 16). The same as S165). In this calculation process, both t0 and E0 are 0, and dg is 0.

  The restart cycle time variation function g (t) is the absolute value of the change amount of g (t) in the section from the apex of the first-order quadratic curve to the end point of the second-order quadratic curve represented by the function | E Is equal to or greater than the absolute value | G | of the amount of change in g (t) during the restart acceleration period time, a straight line portion is present between the first half curve portion and the second half curve portion as shown in FIG. If the | E | is smaller than the | G |, the first-half curve portion and the second-half curve portion shown in FIG. 22 are continuous without a straight line portion. It is represented by a curved line.

  Then, after calculating the restart acceleration period time and the restart cycle time variation function g (t), the transfer amount deriving unit 22 performs the processes in steps S76 to S88 in FIG. 14 in steps S56 to S68 in FIG. Do the same as the process. At this time, the transfer amount deriving unit 22 calculates the acceleration set cycle time dT2 in step S80 based on the restart cycle time variation function g (t) calculated in step S74, and in step S82, the calculation is performed. A transfer amount dP [axis] for each machine axis direction per acceleration set cycle time dT2 is calculated. By performing the restart process as described above, the stopped object resumes moving and accelerates to a speed according to the current override coefficient k.

  When the transfer amount deriving unit 22 determines in step S88 that the reference time t is equal to or greater than the restart acceleration period time, the transfer amount deriving unit 22 performs the processing from step S8 onward in FIG. Try again.

  By the way, if the transfer amount deriving unit 22 determines in step S16 that an emergency stop request has not been issued from the acceleration / deceleration request monitoring unit 10, then a speed change request is issued from the acceleration / deceleration request monitoring unit 10. It is determined whether or not (step S18).

  Here, when it is determined that the speed change request is issued from the acceleration / deceleration request monitoring unit 10, the transfer amount deriving unit 22 performs the speed change process of the object shown in FIG.

  Specifically, the transfer amount deriving unit 22 first calculates the speed change period time and the speed change cycle time variation function g (t) (step S90). At this time, the transfer amount deriving unit 22 determines that the value gs of the speed change cycle time variation function g (t) at the beginning of the speed change period time becomes equal to the set cycle time ΔT, and the speed change time at the end of the speed change period time. The speed change period time and the speed change cycle satisfying the condition that the value ge of the cycle time variation function g (t) is equal to the override coefficient k after the speed change and the combined speed at the beginning of the speed change period time is V. A time variation function g (t) is calculated.

  The speed change cycle time variation function g (t) when the acceleration operation of the object is performed is the constant speed at the start of the speed change period time (when the transfer amount deriving unit 22 receives the speed change request). When moving, it is represented by a curve as shown in FIG. 23, for example. On the other hand, when the object is accelerating at the beginning of the speed change period time, the speed change cycle time variation function g (t) when the acceleration operation of the object is further executed is as shown in FIG. The cycle time variation function g (t) at the time of speed change when the object is decelerating when the object is decelerating at the beginning of the speed change period time is shown in FIG. It is represented by a curve as shown.

  In addition, the speed change cycle time variation function g (t) when the object is decelerated is constant at the start of the speed change period time (when the transfer amount deriving unit 22 receives the speed change request). When moving at a speed, it is represented by a curve as shown in FIG. 26, for example. On the other hand, when the object is accelerating at the beginning of the speed change period time, the speed change cycle time variation function g (t) when the object is decelerating is a curve as shown in FIG. The speed change cycle time variation function g (t) in the case where the object is further decelerating when the object is decelerating at the beginning of the speed change period time is shown in FIG. It is represented by such a curve.

  23 to 28 are all g (t) in the section from the vertex of the first-order quadratic curve to the end point of the second-order quadratic curve represented by the speed change cycle time variation function g (t). ) Change absolute value | E | is smaller than the absolute value | G | of the change amount of g (t) in the speed change period time. The variation function g (t) is represented by a curve having a shape in which the first-half curve portion and the second-half curve portion are continuous without passing through the straight-line portion. On the other hand, when | E | is greater than or equal to the above | G |, the cycle time at the time of speed change is determined by a curve having a shape in which the first half curve portion and the second half curve portion of each figure are connected via a straight line portion. A variation function g (t) is represented.

  Then, after calculating the speed change period time and the speed change cycle time variation function g (t), the transfer amount deriving unit 22 sets the reference time t to 0 (step S92), and then sets the reference time t to 1. Counts up only (step S94).

  Next, the transfer amount deriving unit 22 calculates a set cycle time ΔT based on the speed change cycle time variation function g (t) calculated in step S90 (step S96). This set cycle time ΔT is obtained by substituting the reference time t counted up in step S94 into the speed change cycle time variation function g (t).

  Next, the transfer amount deriving unit 22 calculates the transfer amount dP [axis] for each machine axis direction per the calculated set cycle time ΔT in the same manner as in step S10 (step S98).

  Next, the transfer amount deriving unit 22 calculates the current synthesis speed V of the object from the transfer amount dP [axis] for each machine axis direction calculated in Step S98 (Step S100). The method for calculating the composite speed V is the same as the method for calculating the composite speed V in step S11.

  Thereafter, the transfer control unit 14 drives each of the transfer units 102c, 106b, 108b, and 110b in the same manner as in Step S12 according to the transfer amount dP [axis] calculated in Step S98 (Step S102). Thereafter, the transfer amount deriving unit 22 adds the set cycle time ΔT calculated in step S96 to the set time Ts to update the set time Ts (step S104).

  Next, the transfer amount deriving unit 22 calculates the dg by differentiating the speed change cycle time variation function g (t) obtained in step S90 with respect to the reference time t (step S106).

  Thereafter, the transfer amount deriving unit 22 determines whether or not the reference time t is equal to or greater than the speed change period time (step S108), and if it is determined that the reference time t is equal to or greater than the speed change period time, If the process after step S8 is performed again and it is determined that the reference time t has not passed the speed change period time, then whether or not an emergency stop request has been issued from the acceleration / deceleration request monitoring unit 10 is determined. Is determined in the same manner as in step S16 (step S110).

  Here, when the transfer amount deriving unit 22 determines that an emergency stop request is issued from the acceleration / deceleration request monitoring unit 10, the emergency stop process of steps S52 to S68 is performed, and the acceleration / deceleration request monitoring unit 10 is performed. If it is determined that an emergency stop request has not been issued, the transfer amount deriving unit 22 next determines whether or not a speed change request has been issued from the acceleration / deceleration request monitoring unit 10 as in step S18. Judgment is made (step S112). When the transfer amount deriving unit 22 determines that a speed change request is issued from the acceleration / deceleration request monitoring unit 10, the speed change process after step S90 is performed again. On the other hand, when the transfer amount deriving unit 22 determines that the speed change request is not issued from the acceleration / deceleration request monitoring unit 10, the processes after step S94 are performed again.

  By the way, when the transfer amount deriving unit 22 determines that the speed change request is not issued from the acceleration / deceleration request monitoring unit 10 in the determination of step S18, the transfer amount deriving unit 22 next sets the set time Ts. The set cycle time ΔT is added to update the set time Ts (step S20 in FIG. 6). Thereafter, the processes after step S10 are performed again.

  As described above, the numerical control process by the numerical controller 2 of the present embodiment is performed.

  In the present embodiment, the acceleration / deceleration filter 28 uses the respective support bodies 102b, 106a, and 108a obtained from the post-conversion machine axis path with respect to the post-conversion machine axis path for each machine axis direction calculated by the path conversion unit 26. , 110a, the acceleration / deceleration process is performed to alleviate the change in the transfer speed. Even if the reference machine axis path includes abnormal data that causes a mechanical shock, the abnormal data is converted into a mechanical shock by the acceleration / deceleration process. Can be corrected to show a gradual speed change that does not cause

  Further, in the present embodiment, the mechanical axis path obtained when the path conversion unit 26 performs the conversion unit acceleration / deceleration process, which is the same process as the acceleration / deceleration process by the acceleration / deceleration filter 28, with respect to the reference machine axis path. A correction amount corresponding to an error with the machine axis path is calculated for each machine axis direction, and a reference machine axis path for each machine axis direction in a direction to cancel the error with the calculated correction amount for each machine axis direction. Therefore, the post-conversion machine axis path is corrected in such a way that an error in each machine axis direction caused by the acceleration / deceleration processing by the acceleration / deceleration filter 28 is estimated in advance and the error is offset. Machine axis path. For this reason, when the acceleration / deceleration filter 28 performs acceleration / deceleration processing on the converted mechanical axis path, errors in all machine axis directions caused by the acceleration / deceleration processing can be suppressed. An error in the work shape of the workpiece caused by the deceleration process can be suppressed.

Further, in the present embodiment, when the acceleration / deceleration filter 28 performs acceleration / deceleration processing, the speed function P _B ′ (T) at each set time T in the section from t−a to t + a is represented by each set time T. Unlike the case where the speed function at each set time is distributed to the distribution sections after each set time, since the distribution is performed in the distribution sections from T−a to T + a that are spread evenly in the front and rear as the center, etc. There is no set time delay due to the distribution form of the speed function in the section instructing the transfer of the object at high speed. Therefore, the speed function Q _B ′ (t) is obtained by integrating the distribution value P _B ′ (T) · f (t−T) of the speed function P _B ′ (T) distributed by the acceleration / deceleration filter 28 in the distribution section. ), And in the post-process mechanical axis path Q _B (t) calculated by further integrating the calculated speed function Q _B ′ (t), the speed in the section instructing the transfer of the object at a constant speed It is possible to prevent a time delay caused by the function distribution form.

In this embodiment, when the path conversion unit 26 calculates a correction amount corresponding to an error between the mechanical axis path after the conversion unit acceleration / deceleration processing and the reference mechanical axis path, the machine axis after the conversion unit acceleration / deceleration processing is calculated. Reduces the phase error factor caused by the speed function distribution format in the conversion unit acceleration / deceleration process from the correction amount while simplifying the time correspondence between the points to be subtracted between the path and the reference machine axis path. can do. Specifically, suppose that the path conversion unit 26 has a distribution section having a section width corresponding to a time constant after the speed at each set time in a section of the reference mechanical axis path in the conversion section acceleration / deceleration processing. When calculating the machine axis path after the processing according to a calculation formula that integrates each distributed value, calculates the speed function, and further integrates the calculated speed function. Since the set time in the machine axis path after the processing is delayed with respect to the set time in the reference machine axis path, the reference machine axis path is subtracted from the processed machine axis path without considering the delay in the set time. When a correction amount corresponding to an error between the two machine axis paths is obtained, a phase error is included in the correction amount. In such a case, if the phase error element is to be excluded from the correction amount, the set time between the machine axis path after the processing and the reference machine axis path generated due to the distribution form of the speed function described above It is necessary to obtain an error between the machine axis path after the process and the reference machine axis path after performing association (time adjustment) for eliminating the delay, and the process of the association is complicated. On the other hand, in the present embodiment, the path conversion unit 26 centers the speed function P _A i ′ (T) at each set time T of the reference machine axis path around each set time T in the conversion unit acceleration / deceleration processing. As a result of the distribution in the distribution section from T-a to T + a that spreads evenly in the forward and backward directions, the distribution of the velocity function P _A i ′ (T) is caused in the section instructing the transfer of the object at a constant speed. It is possible to prevent the set time from being delayed. Therefore, in the machine axis path Q _A (t) after the conversion unit acceleration / deceleration processing, the coordinate values at the same time t coincide with the reference machine axis path in the section instructing the transfer of the object at a constant speed. To do. For this reason, when calculating the correction amount corresponding to the error between the machine axis path Q _A (t) after the conversion unit acceleration / deceleration processing and the reference machine axis path, While simplifying the association of the set times, it is possible to reduce the phase error element resulting from the distribution format of the speed function of the reference mechanical axis path from the correction amount.

In the present embodiment, when the acceleration / deceleration filter 28 performs acceleration / deceleration processing, the speed function P _B i ′ (T) at each time T in the section from t−a to t + a is centered on each time T. Unlike the case where the speed function at each time is distributed in the distribution section after each time, the distribution of the speed function is as follows. There is no time delay due to the distribution format. Therefore, in the mechanical axis path Q _B (t) after the acceleration / deceleration processing by the acceleration / deceleration filter 28, it is possible to prevent a time delay due to the speed function distribution form.

  Further, in the present embodiment, the time adjustment unit 38 transfers the part corresponding to the acceleration period (0 ≦ t <Tj) immediately after the start of object transfer and the object transfer for the machine axis path before adjustment in each machine axis direction. In order to adjust the time for the portion corresponding to the deceleration period immediately before the stop (te−Tj <t ≦ te), before the path conversion unit 26 performs the conversion unit acceleration / deceleration process, the conversion unit acceleration / deceleration process is performed in advance. It is possible to obtain a reference machine axis path that is corrected in a direction that cancels out the set time difference between the portions corresponding to the acceleration period and the deceleration period. For this reason, it is possible to prevent a set time shift from occurring in each part corresponding to the acceleration period and the deceleration period of the mechanical axis path after the path conversion unit 26 performs the conversion unit acceleration / deceleration processing. It is possible to calculate an appropriate correction amount for each part.

  Further, in the present embodiment, when a special command is input to the special command input device 122 to cause the machine tool to perform an irregular operation accompanied by a change in the speed of the object, the length of the set cycle time common to each machine axis direction is increased. Is changed from the length in the state immediately before the input of the special command to the length corresponding to the speed change instructed by the special command, and at the set cycle time after the change calculated from the machine axis path after processing. Each support 102b, 106a, 108a, 110a is transferred according to the transfer amount of each support 102b, 106a, 108a, 110a in the corresponding machine axis direction. For this reason, when an operation involving irregular speed changes of the object is executed separately from the normal transfer of the object at the time of machining the workpiece, the object deviates from the trajectory indicated by the machine axis path after processing. Can be prevented. Specifically, when the length of the set cycle time common to each machine axis direction is changed according to a special command as in this embodiment, it is calculated per set cycle time from the processed machine axis path. The relative positional relationship between the machine axis directions defined by the post-processing machine axis path for each machine axis direction is maintained for the transfer amount of the supports 102b, 106a, 108a, 110a in each machine axis direction. It is calculated in the state. For this reason, if each support body 102b, 106a, 108a, 110a is transferred by each transfer part 102c, 106b, 108b, 110b according to the calculated transfer amount in each machine axis direction per set cycle time, An operation involving a change in the speed of the object is performed in a state in which the relative positional relationship between the machine axis directions defined by the post-processing machine axis path in each machine axis direction is maintained. As a result, even if an irregular operation accompanied by a change in the speed of the object is performed, the object can be prevented from deviating from the trajectory indicated by the post-processing machine axis path in each machine axis direction.

  Further, in the present embodiment, the trajectory indicated by the original machining path when the object performs an operation accompanied by a speed change according to the special command while obtaining the effect of suppressing the mechanical shock caused by the abnormal data by the acceleration / deceleration processing. Can be prevented from coming off. Specifically, if acceleration / deceleration processing after pulse interpolation is performed in order to suppress mechanical shock caused by abnormal data, the transfer pulse after the post-interpolation acceleration / deceleration processing is indicated. The movement trajectory of the object is deviated from the movement trajectory indicated by the original machining path due to an error caused by the post-interpolation acceleration / deceleration processing. In this case, when an operation involving a change in the speed of the object according to the special command is performed according to the transfer pulse after the acceleration / deceleration processing after interpolation, the object moves out of the locus indicated by the original machining path. It will be. In contrast, in the present embodiment, the acceleration / deceleration process by the acceleration / deceleration filter 28 as described above obtains the effect of suppressing the mechanical shock caused by the abnormal data, and the acceleration / deceleration is performed in the machine axis path after the acceleration / deceleration process. Due to the effect of error correction by the conversion processing of the path conversion unit 26 before processing, it is possible to suppress the locus indicated by the machine axis path after the acceleration / deceleration processing from deviating from the locus indicated by the machining path. Accordingly, it is possible to prevent the object from deviating from the trajectory indicated by the original machining path when performing an operation accompanied by a speed change according to the mechanical axis path after the acceleration / deceleration processing.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description of the embodiments but by the scope of claims for patent, and further includes meanings equivalent to the scope of claims for patent and all modifications within the scope.

  The numerical controller according to the present invention is not limited to application to a machine tool as shown in FIG. For example, the numerical control device according to the present invention can be applied to a machine tool having a so-called portal frame, a machine tool having a parallel mechanism, a lathe, and other various machine tools.

  In the above embodiment, the linear distribution function is used in the conversion unit acceleration / deceleration process by the path conversion unit 26 and the acceleration / deceleration process by the acceleration / deceleration filter 28. However, the distribution function used for these acceleration / deceleration processes is as described above. It is also possible to use a distribution function of a type other than a straight linear distribution function. For example, a so-called bell-shaped distribution function may be used for these acceleration / deceleration processes. Such a bell-shaped distribution function f (t) is expressed by the following equation, for example.

  Note that f (t) = 0 outside the distribution interval −a ≦ t ≦ a.

  Further, when the conversion unit acceleration / deceleration process is performed using a bell-shaped distribution function, in step S41 of FIG. 8, the portion of the acceleration period immediately after the start of transfer in the reference machine axis path and the transfer using the following equation: It is preferable to perform time adjustment for eliminating the time lag after the conversion unit acceleration / deceleration processing in the portion of the deceleration period immediately before the stop.

  In addition, the reference machine axis path deriving unit locally smoothes only a portion instructing rapid acceleration / deceleration that may cause a machine shock to the machine tool in the reference machine axis path derived by the acceleration / deceleration calculation unit. A local acceleration / deceleration filter for interpolation may be provided. In this case, the path conversion unit may perform conversion processing on the reference mechanical axis path after being locally interpolated by the local acceleration / deceleration filter.

  In the above embodiment, the machine tool is provided with a special command input device, and when a special command is input to the special command input device, the length of the set cycle time is changed and the machine corresponding to each support body. Although the transfer amount in the axial direction is calculated, the numerical control device according to the present invention may be applied to a machine tool that does not include such a special command input device. Also, even if the machine tool is equipped with a special command input device, when a special command is input to the special command input device, the transfer corresponding to the set cycle time in which the length is changed according to the content of the special command. The amount is derived from the machine axis path in each machine axis direction, and the operation of the object is controlled according to the special command by a method other than the method of changing the transfer speed of each support according to the derived transfer amount. The numerical control device according to the present invention may be applied to a simple machine tool. In these cases, in the numerical control process shown by the flowchart of FIG. 6, the processing of steps S16 and S18 is omitted, and the entire period of the processed mechanical axis path in which the transfer amount deriving unit is stored in the memory in step S14. If it is determined that the process has not been completed yet, the process of step S20 may be performed next.

  Further, the reference machine axis path deriving unit does not necessarily include the curved surface interpolation unit and the acceleration / deceleration calculation unit. That is, the program reading unit of the reference machine axis path deriving unit derives a machining path from the NC program, and converts the machining path into a moving component in each machine axis direction without performing interpolation calculation or acceleration / deceleration calculation. A reference machine axis path for each machine axis direction may be derived.

  Further, the control of the transfer device by the numerical control device of the present invention is not necessarily applied to all of the deceleration at the time of emergency stop of the object, the acceleration at the restart, and the acceleration / deceleration by the override device. For example, the control of the transfer device by the numerical control device of the present invention may be applied only to the deceleration at the time of emergency stop of the object, and the transfer device by the numerical control device of the present invention is only applied to the acceleration at the time of restart of the object. Control may be applied, and control of the transfer device by the numerical control device of the present invention may be applied only to acceleration / deceleration of the object by the override device. Further, the control of the transfer device by the numerical control device of the present invention may be applied to any two of the deceleration at the time of emergency stop of the object, the acceleration at the restart, and the acceleration / deceleration by the override device.

Further, in the conversion unit acceleration / deceleration process by the path conversion unit, the speed function Q _A i ′ (t) of the predetermined block Q _A i (t) after the conversion unit acceleration / deceleration process according to the following equations (51) and (52): ).

  However, in this case, since the time in the machine axis path after the conversion unit acceleration / deceleration processing is delayed from the time in the reference machine axis path, the path conversion unit When calculating the correction amount corresponding to the error of, after associating the times (time adjustment) of the points to be subtracted from each other between the mechanical axis path after the conversion unit acceleration / deceleration processing and the reference mechanical axis path, It is necessary to calculate the error by subtracting points.

  Further, the processing by the acceleration / deceleration filter may be repeatedly performed. In this case, it is possible to obtain a processed machine axis path with less error with respect to the reference machine axis path.

  In the above embodiment, the time adjustment unit 38 performs time adjustment for the acceleration period immediately after the start of the transfer of the object and the deceleration period immediately after the stop of the transfer of the object. However, such time adjustment is not necessarily required. Absent. That is, the reference machine axis path deriving unit does not include a time adjusting unit, and the machine axis path calculated by the acceleration / deceleration calculating unit is set as the reference machine axis path, and the reference converting machine axis whose time is not adjusted by the path converting unit. The conversion unit acceleration / deceleration processing may be performed on the path.

2 Numerical Control Device 14 Transfer Control Unit 22 Transfer Amount Deriving Unit 24 Reference Machine Axis Path Deriving Unit 26 Path Conversion Unit 28 Acceleration / Deceleration Filter (Acceleration / Deceleration Processing Unit)
100 Workpiece 102 Workpiece transfer device (transfer device)
102b Work support (support)
102c Workpiece transfer unit (transfer unit)
105 Tool 106 Vertical transfer device (transfer device)
106a Vertical support (support)
106b Vertical transfer part (transfer part)
108 Tool first horizontal transfer device (transfer device)
108a First horizontal support (support)
108b 1st horizontal transfer part (transfer part)
110 Tool second horizontal transfer device (transfer device)
110a Second horizontal support (support)
110b Second horizontal transfer section (transfer section)
122 Special command input device

Claims (5)

A plurality of transfer devices that transfer a workpiece or a tool that processes the workpiece as a target to transfer the target in order to process the workpiece, and each of the transfer devices includes a support for supporting the target; A numerical control device that is provided in each machine tool having a transfer unit that transfers the object by transferring the support in a specific machine axis direction, and performs numerical control of each of the transfer devices,
Based on a machining command that represents a movement trajectory of the tool during machining of the workpiece, a reference machine axis path that represents a position in the machine axis direction to which the support is to be transferred as a function of a predetermined set time is derived. A reference machine axis path deriving unit;
A path conversion unit that derives a converted machine axis path for each machine axis direction by performing a specific conversion process on the reference machine axis path for each machine axis direction derived by the reference machine axis path deriving unit. When,
Acceleration / deceleration processing for mitigating changes in the transfer speed of each support obtained from the post-conversion machine axis path with respect to the post-conversion machine axis path in the respective machine axis directions derived by the path conversion unit An acceleration / deceleration processing unit for performing
A transfer amount for deriving a transfer amount of each of the supports per predetermined set cycle time in each of the machine axis directions from a processed machine axis path that is a machine axis path that has been subjected to acceleration / deceleration processing by the acceleration / deceleration processing unit. A derivation unit;
A transfer control unit that transfers the support corresponding to each transfer unit according to the transfer amount derived by the transfer amount deriving unit;
The specific conversion process performed by the path conversion unit includes a machine axis path and a reference machine axis obtained when the same process as the acceleration / deceleration process by the acceleration / deceleration process unit is performed on the reference machine axis path. A correction amount corresponding to an error with respect to the path is calculated for each of the machine axis directions, and the reference machine for each of the machine axis directions is in a direction that cancels the error with the calculated correction amount for each of the machine axis directions. A numerical control apparatus including an arithmetic processing for calculating the converted mechanical axis path corrected for an axis path. The acceleration / deceleration processing unit sets the set time as t as the acceleration / deceleration process, and sets a half value of a time constant set in the machine axis direction of the converted machine axis path to be subjected to the acceleration / deceleration process. a, a speed function that is a primary differential function of the post-conversion mechanical axis path at each set time T in a section extending from t−a to t + a is P _B ′ (T), and satisfies the following equation (1), When the distribution function representing a specific distribution format for distributing the speed function P _B ′ (T) is f (T), the speed function Q _B ′ (t ) Is calculated based on the following equation (2), and the calculated speed function Q _B ′ (t) is integrated to calculate the post-process mechanical axis path Q _B (t). Item 2. The numerical control device according to Item 1.

As the same process as the acceleration / deceleration process, the path conversion unit calculates a speed function, which is a first-order differential function of the reference mechanical axis path at each set time T in the section from t−a to t + a, as P _A ′ ( T), a speed function Q _A ′ (t), which is a linear differential function of the mechanical axis path after performing the same processing as the acceleration / deceleration processing, is calculated based on the following equation (3), and the calculation 3. The numerical value according to claim 2, wherein an arithmetic process for calculating a mechanical axis path Q _A (t) after performing the same process as the acceleration / deceleration process is performed by integrating the speed function Q _A ′ (t). Control device.

The reference machine axis path deriving unit includes a time adjusting unit that derives the reference machine axis path after adjusting the set time, and the time adjusting unit sets the set time before time adjustment to t, The set time after the time adjustment is set to tj (t), the time constant in the machine axis direction of the target for time adjustment is set to Tj, the time of the start point of the transfer start of the target is set to t = 0, and the target is transferred When the stop end time is t = te , the equation (1) is satisfied, and the distribution function representing the specific distribution format is f (T), the time adjustment is performed based on the following equation: Item 4. The numerical control device according to Item 2 or 3 .

The machine tool includes a special command input device for inputting a special command for instructing an operation accompanied by a speed change of the target object separately from a normal transfer of the target object at the time of machining the workpiece. And
In response to the special command being input to the special command input device, the transfer amount deriving unit instructs the length of the set cycle time from the length immediately before the input of the special command. The length of the object is changed according to the change in speed of the object, and the amount of transfer in the machine axis direction corresponding to each of the supports per set cycle time after the length is changed is the post-processing machine. The numerical control device according to claim 1, wherein the numerical control device is derived from an axis path.

Images