Machining processor

[0002] The present invention relates to a control apparatus for a machine tool and a machining system comprising the control apparatus and a machine tool wherein, by supplying a raw workpiece and inputting data regarding a machining profile of a final product (hereinafter referred to as machining profile data), the workpiece to be machined is machined according to the machining profile data so that a final product can be fabricated.

[0003] In the conventional method of machining a workpiece by a NC machine tool, the first step is to prepare a drawing representing the profile of a product to be machined. A programmer determines the machining steps from the drawing and creates a NC program manually or by an automatic programming unit. An operator inputs the NC program into the NC machine tool while, at the same time, setting up the workpiece on the NC machine tool manually or by using an automatic workpiece changer. Then, the cutting tool to be used is preset, and the amount of tool offset is defined. The cutting tool is then mounted in the tool magazine of the NC machine tool. After that, the NC program is executed thereby to machine the workpiece and fabricate a product. Various inventions have hitherto been developed with the aim of automating these steps as far as possible and reflecting the knowhow accumulated by programmers and operators on the machining steps.

[0004] First to be cited is an automatic programming system disclosed in Japanese Unexamined Patent Publication (Kokai) No. 4-315550. This system comprises profile recognition means for extracting a machining profile from the data representing the machining profile of a workpiece, a machining conditions memory for storing optimum machining conditions, and automatic machining conditions setting means for automatically setting optimum machining conditions stored in the machining conditions memory. As a result, machining conditions can be set automatically without the manual operation of setting the machining conditions by the operator, thus eliminating the human errors which often accompany manual operation and making a superior machining operation possible. Further, the load on the operator and the time required for the operator to tend to the machine are reduced.

[0005] A second conventional machining system is disclosed in Japanese Unexamined Patent Publication (Kokai) No. 4-138504. In this system, data on a workpiece including the material, surface roughness and the dimensional accuracy are stored beforehand and machining conditions are determined by a first neural network. The machining conditions can be corrected by the operator. This system further comprises learning means which, after an actual machining operation, corrects the machining conditions based on the machining result to produce corrected machining conditions while at the same time correcting the weight of the first neural network. The system furthermore comprises adaptive control means including a sensor for detecting the sparks, sound and the force generated during the machining process, and a second neural network supplied with the data from the sensor as temporal data in which the data is averaged with a predetermined time margin to detect the machining conditions at the moment thereby to correct the machining conditions dynamically. Even without a skilled operator, therefore, the workpiece can be machined under optimum machining conditions.

[0006] A third conventional technique concerns a method using the numerical control disclosed in Japanese Unexamined Patent Publication (Kokai) No. 9-26811. According to this method, a machining process and a machine tool are optimally selected in accordance with simplified input databased on the registration of various information files, the input of machining pattern data, the processing of a finish pattern, pattern recognition and the determination of the machining process. In this way, a machining area and machining steps high in production efficiency are selected, and tools, machining conditions and a tool path most suitable for the input patterns are determined. Both the production efficiency and the machining accuracy are improved further by post-machining measurement and correction.

[0007] Techniques are well known for automatically creating a tool path from the profile data of a product to be machined. An NC program can be automatically prepared by adding various machining conditions to the profile data. According to the first prior art described above, machining conditions are selected from a database in accordance with a predetermined algorithm based on the profile data of the product to be machined. These machining conditions can be considered static ones. In contrast, the second prior art is such that machining conditions which undergo a constant change are detected by a sensor, and the machining conditions set based on the detection result are adoptively controlled using the learning function of a neural network, so that dynamic machining conditions are determined in accordance with ever-changing machining requirements. The first and second prior art emphasize the automatic determination of machining conditions.

[0008] In the third conventional method, the operator inputs data, and machining conditions are automatically determined using a technique similar to the first and second prior art. In addition, automatic determination of a cutting tool and a tool path combined with a technique of measurement and correction after machining to finish an intended product without human labor.

[0009] These conventional techniques are based on the architecture of securing a high accuracy and a high production efficiency by feedback correction of the machining conditions, but not intended to realize a high-accuracy, high-efficiency machining process by predicting machining requirements and determining a tool path and machining conditions based on the prediction.

[0010] An object of the present invention is to provide a machine tool control apparatus and a machining system including the control apparatus and a machine tool, in which an intended product can be automatically machined at high efficiency while meeting the precision requirements in response to only profile data on the product to be finished and data on the workpiece to be machined.

[0011] Another object of the invention is to provide a machine tool control apparatus and a machining system including the control apparatus and a machine tool, in which machining requirements are predicted and a tool path and machining conditions are determined automatically in conformance with the predicted machining requirements, thus making possible a high-precision, high-speed machining process.

[0012] According to a first aspect of the invention, there is provided a machine tool control apparatus supplied with machining profile data for machining a workpiece, comprising input means for inputting workpiece data including machining profile data on a final workpiece profile and the material and the profile of the workpiece to be machined, data storage means for storing at least a selected one of machine data representing the specifications of the machine tool used for machining the workpiece and tool data representing the specifications of the tools held in the machine tool, and tool path determining means for generating a tool path for machining the workpiece and determining the conditions for machining the workpiece such as the rotational speed of the main spindle and the feed rate of the machine tool based on the data stored in the data storage means.

[0013] According to another aspect of the invention, there is provided a machining system supplied with the machining profile data for machining a workpiece, comprising: a machine tool for machining the workpiece; input means for inputting the machining profile data on the final profile of the workpiece and the material and profile of the workpiece to be machined; data storage means for storing at least a selected one of the machine data representing the specifications of the machine tool for machining the workpiece and the tool data representing the specifications of the tools held in the machine tool; predictive calculation means for predicting at least a machining load or interference between a tool and the workpiece based on the data input from the input means and the data stored in the data storage means; machining process determining means for selecting a tool and a machining pattern and determining a machining process based on the data input from the input means and the data stored in the data storage means; tool path determining means for generating a tool path for machining the workpiece and determining the conditions for machining the workpiece including the rotational speed of the main spindle and the feed rate of the machine tool based on the tool, the machining pattern and the machining process selected and determined by the machining process determining means, the result of predictive calculation made by the predictive calculation means, the data input by the input means and the data stored in the data storage means; and numerical control means for controlling the operation of the machine tool according to the tool path and the machining conditions generated and determined by the tool path determining means.

[0014] FIG. 1 is a block diagram showing a machine tool control apparatus according to an embodiment of the invention.

[0015] FIG. 2 is a block diagram showing a database.

[0016] FIG. 3 is a block diagram showing a tool path determining unit.

[0017] FIG. 4 is a flowchart for a machining process determining unit.

[0018] FIGS. 5A to 5H are diagrams schematically showing the types of machining patterns.

[0019] FIG. 6 is a block diagram showing a predictive calculation unit.

[0020] FIGS. 7A and 7B are schematic diagrams for explaining a method of determining a machining area.

[0021] FIGS. 8A and 8B are schematic diagrams for explaining the effects of the invention, in which FIG. 8A shows a profile drawn based on the machining profile data input, and FIG. 8B is a diagram for comparing the results of machining by use of a control apparatus according to the invention and the prior art.

[0022] Embodiments of the invention will be described with reference to the accompanying drawings.

[0023] FIG. 1 is a block diagram showing a control apparatus 100 for a machine tool according to an embodiment of the present invention.

[0024] The control apparatus 100 includes, as main components thereof, an input unit 1, a database 3, a tool path determining unit 5 and a predictive calculation unit 7. Though not shown in detail, the control apparatus 100 can be configured of a CPU, a RAM, a ROM, an input/output interface, a data memory and a bidirectional bus for connecting these component elements.

[0025] The operator inputs the machining profile data 1aof a product to be machined from an input unit 1. This machining profile data 1acan be, for example, electronic pattern information such as CAD data. Also, the machining profile data 1aincludes data on the machining accuracy and the surface roughness.

[0026] The input unit 1 is further supplied with workpiece data 1bproviding data on the profile and the material of the workpiece to be machined for making the product. The workpiece data 1binclude data on the dimensions and profiles of the fittings, jigs such as pallets, etc. for mounting and fixing the work on the machine tool 11, the mounting position on the machine tool 11 and the mounting position of the workpiece to be machined on a jig. The input unit 1 includes not only a keyboard but also such information media as a floppy disk and a magneto-optical disk, driving units thereof and an interface to a network which stores the data.

[0027] The data input by way of the input unit 1 are stored in the database 3 functioning as a data storage means. The database 3 includes a machine database 3a, a tool/holder database 3b, a machining conditions database 3c, a material database 3d, a NC/servo database 3eand an input database 3f. The database 3 can be configured of a data storage unit such as a hard disk drive or an optical disk device. The machine database 3a, the tool/holder database 3b, the machining conditions database 3c, the material database 3d, the NC/servo database 3eand the input database 3fin the database 3 can be configured either as different data storage units or can be incorporated in a single data storage unit partitioned into a plurality of areas assigned the databases 3ato 3f, respectively.

[0028] With reference to FIG. 2, the data stored in the machine database 3ainclude those on the stroke of each feed shaft, the maximum rotational speed of the main spindle, and the maximum feed rate of the machine tool 11, and the deformation characteristic of the machine tool 11 with respect to temperature and the deformation characteristic of the machine tool 11 due to the weight of the work. The data stored in the tool/holder database 3bincludes serial numbers of the tools, dimensions and the shape of the tool holder, material and life of the tools, the trip characteristic and the run-out characteristic of the tools under load, and the dimensions and shape of the forward end of the main spindle. The data stored in the machining conditions database 3cinclude the amount of feed and the amount of cut per blade, the pick feed amount, whether coolant is used or not, the machining pattern, the partitioning data for dividing the machining surface into a plurality of machining areas, and fundamental data for selecting an optimum tool for machining a given machining surface. The data stored in the material database 3dinclude the type, hardness, tensile strength, modulus of elasticity, etc. of a material. The data stored in the NC/servo database 3einclude specifications, setup parameters, time constant of the servo, and the gain of the numerical control unit. The input database 3fhas stored therein the machining profile data 1aand the workpiece data 1binput from the input unit 1.

[0029] The data stored in the database 3 are at least one of the data input arbitrarily by the operator, the data registered in the machine tool 11 and the data stored in a predetermined memory.

[0030] The tool path determining unit 5 will be described with reference to FIGS. 3 and 4.

[0031] The tool path determining unit 5 includes a machining process determining unit 5a, a tool path generating unit 5b, a machining conditions determining unit 5cand a correction unit 5ddescribed in detail below.

[0032] First, the machining profile data 1aand the workpiece data 1bstored in the input database 3fare sent to the machining process determining unit 5a(step S11). Based on these data, the machining process determining unit 5arecognizes the profile of the machining surface of the workpiece to be machined and the work to be finished. Then, the machining process determining unit 5apartitions the machining surface into a plurality of machining areas with the curvature, the inclination and depth of the surface to be machined as surface parameters based on the surface partitioning data stored in the machining conditions database 3 and the recognized profile of the machining surface (step S13). Then, from among the fundamental data and the machining patterns stored in the machining conditions database 3cfor selecting an optimum tool for machining surfaces, the machining process determining unit 5aselects an optimum tool and an optimum machining pattern for machining each machining area in accordance with each surface parameter thereof (step S15). In the case, where a steep slope is machined with a scan path for example, the tool would be overloaded or develop a shimmy. In order to prevent this, a machining path along a contour is selected for a steep slope. At the same time, the data on whether a cooling is required to be supplied or not is introduced from the machining conditions database 3c. Then, the order of machining the respective machining areas is determined (step S17).

[0033] Examples of machining patterns are shown in FIGS. 5A to 5H. FIG. 5A shows a scan machining path, and FIG. 5B a contour machining path. FIG. 5C shows a scan machining path, i.e. what is called the character line machining path, intended to improve the cutting efficiency by setting only the portion having a work as a machining area, and FIG. 5D is a radiation machining path centered at about a point O. FIG. 5E shows a contour machining path for a work having a substantially predetermined stock allowance from a final profile of a material such as cast iron which is cutting a predetermined amount each time along the direction of the normal to the final profile by progressively offsetting the work, and FIG. 5F shows a scan machining path for a work having a substantially predetermined stock allowance from a final profile of a material such as cast iron which is cut a predetermined amount each time progressively along the direction of the normal to the final profile thereof.

[0034] FIG. 5G shows a contour machining path intended to improve the cutting efficiency by defining a machining area only at the portion covered by a work, and starting the machining operation with the portion covered by a work or decreasing the pick feed rate when the work approaches a final profile, for example. FIG. 5H shows a machining path, in which the boundary of a machining area is overlapped automatically with an adjacent machining area, and the tool is retracted over the overlapped portion smoothly thereby to prevent a misalignment at the boundary. For example, the tool begins to be moved along the normal from a relief point a predetermined distance off the tool path associated with the size of the particular tool toward the machining area, within which the tool is moved along the tool path corresponding to the size of the particular tool to cut the machining area. These machining patterns are stored in the database 3 in a form reflecting the accumulated know-how in such a manner that a desired machining pattern can be selected for a corresponding surface parameter such as the curvature, inclination and depth of each machining area. The shown machining patterns represent only several examples to which the present invention is not limited.

[0035] The data determined in the machining process determining unit 5aon the machining area, the tool, the machining pattern and the machining sequence are sent to the tool path generating unit 5b. The tool path generating unit 5bdetermines optimum values of the feed rate per blade, the depth of cut and the pick feed length for machining a particular work from various amounts of these parameters stored in the machining conditions database 3cin correspondence with the data on the machining area, the tool, the machining pattern, the workpiece data 1b, in particular, the data on the material of the workpiece to be machined, stored in the input database 3f, and the data on the type, tensile strength and modulus of elasticity of the material stored in the material database 3d, while at the same time generating a tool path and a pick feed path. The tool path and the pick feed path are generated according to the formula stored in the machining conditions database 3c.

[0036] In place of the machining process determining unit 5aselecting and determining the tool, the machining pattern and the machining process, the operator can arbitrarily input the data on the tool, the machining pattern and the machining process.

[0037] Then, the machining conditions determining unit 5cdetermines the rotational speed of the main spindle and the feed rate of the machine tool 11 from the feed rate, the cut rate and the pick feed rate per blade based on the result of the calculation input from the tool path generating unit 5b, while at the same time applying a tool change command for each step of the process and a command for inserting a step of measurement by a sensor unit 13 described later to a numerical control unit 9.

[0038] The correction unit 5dissues a correction command to the tool path generating unit 5bor the machining process determining unit 5abased on the result of the predictive calculation input from the predictive calculation unit 7 described later.

[0039] Instead of determining the machining conditions in the machining conditions determining unit 5c, the machining conditions arbitrarily input by the operator or the machining conditions stored in a predetermined memory can be used with equal effect.

[0040] The numerical control unit 9 generally includes a well-known NC unit 9aand a servo system 9bhaving a servo motor and an amplifier for controlling the revolutions of the main spindle of the machine tool 11. Specifically, the NC unit 9agenerates a command for moving each of the X, Y and Z feed shafts based on the tool path data and the machining conditions data, as is well known, while at the same time issuing commands for starting/stopping the main spindle, turning on/off the coolant, automatically changing the tool and automatically changing the pallet, etc. of the machine tool 11. The servo system 9b, in response to a move command from the NC unit 41, controls the servo motor of the feed shaft. In this way, the machine tool 11 is controlled by the numerical control unit 9.

[0041] The machine tool 11 is well-known equipment such as a machining center having an automatic tool changer, for example. The machine tool 11 comprises a sensor unit 13 including machine sensors having a temperature sensor such as thermistor for measuring the temperature of various parts of the machine such as the column and the bed, the coolant and the environment in which the machine tool 11 is installed, and an ampere meter for measuring the current value supplied to the motor of the machine tool 11, a tool sensor for measuring the length, diameter and the profile of the forward end of the tool mounted on the machine tool 11, and a work sensor for measuring the profile of the work being machined. The values detected by the sensor unit 13 are sent to the predictive calculation unit 7 described later. Further, the tool length, the tool diameter, the wobble of the tool in rotation, the off-center, etc. measured by the tool sensor of the sensor unit 13 are sent through the data correction unit 15 to the tool/holder database 3bof the database 3. Thus, the contents of the database 3, especially, the friction characteristic, the service life and the wobble characteristic of the tool are corrected and updated.

[0042] Also, the temperature of various parts of the machine body, the coolant temperature, the temperature of the environment, etc. measured by the temperature sensor included in the mechanical sensor of the sensor unit 13 are applied through the data correction unit 15 to the machine database 3aof the database 3, so that the related portion of the machine database 3ais corrected and updated. As an alternative, the positioning error of the machine tool 11 in operation is calculated from the machining error measured by the work sensor of the sensor unit 13, and applied through the data correction unit 15 to the machine database 3aof the database 3, so that the corresponding portion of the machine database 3ais corrected and updated. It is also possible to employ a method in which the machining error measured by the work sensor of the sensor unit 13 is sent to the correction unit 5dof the tool path determining unit 5 for generating correction data from the machining error data, and a corrected tool path is generated by the tool path generating unit 5bfor re-machining.

[0043] Now, the predictive calculation unit 7 will be described with reference to FIG. 6.

[0044] The predictive calculation unit includes, as main component elements thereof, a machining state simulation unit 7a, a machine behavior simulation unit 7b, a workpiece simulation unit 7cand a tool behavior simulation unit 7d.

[0045] The machining state simulation unit 7areceives the machining profile data 1aand the workpiece data 1bfrom the input database 3f, and the dimensions and the profile of the tool, the tool holder and the forward end of the main spindle determined by the machining process determining means 5afrom the tool/holder database 3b. Further, the machining state simulation unit 7areceives the feed rate, the cut rate, the pick feed rate, the tool path and the pick feed path per blade determined and calculated by the tool path generating unit 5band also receives the rotational speed of the main spindle and the feed rate determined by the machining conditions determining unit 5c.

[0046] The machining state simulation unit 7a, before starting the machining process, i.e. when the tool path determining unit 5 first generates a tool path or the like, predicts whether the tool, the tool holder or the forward end of the main spindle interferes with the work during the machining operation based on the machining profile data 1a, the workpiece data 1b, and the profile and dimensions of the tool, the tool holder and the forward end of the main spindle, the tool path and the pick feed path. In the case where interference is predicted during the machining operation in a given machining area, the machining state simulation unit 7ainstructs the machining process determining unit 5ato define the same machining area as a no-machining area. In the case where the particular no-machining area can be machined by another tool or in the case where the machining pattern of the no-machining area can be changed, then the machining process determining unit 5achanges the tool or the machining pattern, as the case may be, determines the machining area, the tool, the machining pattern and the machining sequence afresh, and sends the result to the tool path determining unit 5. Then, this process is executed again. In the case where the tool or the machining pattern cannot be changed, on the other hand, the machining process determining unit 5asends the machining sequence with the no-machining area removed from the machining process to the tool path determining unit 5.

[0047] An example of a method for determining a machining area will be explained with reference to FIGS. 7A, 7B. Assume that an area 21afree of interference with a given tool is predicted by calculations in the machining state simulation unit 7aand that an area that can be machined by the particular tool can be defined as an area 21bhaving a surface inclination of 66 degrees or less (FIG. 7A). The portion overlapped by the two areas is determined as a machining area 21c(FIG. 7B).

[0048] The machining process determining unit 5acan alternatively define a range to be machined as a portion overlapped between the machining area determined by checking for interference between the tool and the work as in the method shown in FIGS. 7A, 7B, the machining area arbitrarily determined by checking the machining profile data 1ainput from the input unit 1 for any interference between the work and the tool and the machining area determined according to the surface inclination, the surface curvature and the depth included in the machining profile data 1ainput from the input unit 1.

[0049] The machining state simulation unit 7afurther checks, in real time, for a similar interference also after the machining process has started. In the case where interference between the tool and the work is predicted as a result of the check, a data indicating it is applied to the tool path generating unit 5b. The tool path generating unit 5bcalculates and generates a machining route avoiding the interference such as along the Z-axis.

[0050] Also, the machining state simulation unit 7apredicts an internal-angle portion of the work being machined, based on the machining profile data 1, the workpiece data 1b, the machining pattern and the tool path while the actual machining work is going on. The result of prediction is applied to the machining conditions determining unit 5c, which generates machining conditions for decelerating the tool feed rate at the internal-angle portion.

[0051] Further, the machining state simulation unit 7apredicts by calculation a machining load from the workpiece data 1b, the feed rate, the cut rate, the pick feed rate and the tool path per bite determined and calculated by the tool path generating unit 5b, and the rotational speed of the main spindle and the feed rate determined by the machining conditions determining unit 5c. Furthermore, the machining state simulation unit 7apredicts the current profile or the work being machined, the contact point between the tool and the work and the weight of the work, from the machining profile data 1a, the workpiece data 1b, the profile and dimensions of the tool, the tool holder and the forward end of the main spindle determined by the machining process determining unit 5a, and the pick feed rate and the tool path determined and calculated by the tool path generating unit 5b.

[0052] Also, the machining state simulation unit 7apredicts the load inertia of the work based on the predicted work weight change and the machining conditions determined by the machining conditions determining unit 5c. The predicted load inertia is applied to the numerical control unit 9 for correcting the parameters of the servo system 9bof the numerical control unit 9.

[0053] In the case where what is called an “air cut” is predicted in which the work fails to be cut by the tool moving along a given tool path, based on the predicted contact point between the tool and the work, then the machining state simulation unit 7asends data informing the machining conditions determining unit 5cof the fact. In response, the machining conditions determining unit 5ccan give a command to the numerical control unit 9 to move the tool along the tool path at a maximum feed rate through the air-cut predicted area.

[0054] The machine behavior simulation unit 7bpredicts the thermal deformation of the machine tool 11 based on the data on the thermal deformation characteristic of the machine against the temperature stored in the machine database 3aand the temperature data supplied from the temperature sensor of the sensor unit 13. Also, the machine behavior simulation unit 7bpredicts the deformation of the machine tool 11 due to the work weight based on the work weight data predicted by calculation in the machining state simulation unit 7aand the deformation characteristic data of the machine under the weight of the work stored in the machine database 3a.

[0055] The workpiece simulation unit 7creceives the machining profile data 1aand the workpiece data 1bfrom the input database 3f, and receives, from the tool/holder database 3b, the dimensions and profile of the tool, the tool holder, and the forward end of the main spindle determined by the machining process determining unit 5a. Further, the workpiece simulation unit 7creceives the feed rate, the cut rate, the pick feed rate, the tool path and the pick feed path per blade determined and calculated by the tool path determining unit 5b, and also receives the rotational speed of the main spindle and the feed rate determined by the machining conditions determining unit 5c. The workpiece simulation unit 7cpredicts by calculation an intermediate profile of the work being machined at each moment based on the data received from the database, and applies the result of predictive calculation to the machining state simulation unit 7aand the machine behavior simulation unit 7b.

[0056] The tool behavior simulation unit 7dpredicts the trip and the run-out of the tool from the data on the machining load predicted by the machining state simulation unit 7aand the data on the trip and the runout characteristics with respect to the tool load stored in the tool/holder database 3b. Also, the tool behavior simulation unit 7dpredicts by calculation the wear and the wear distribution of the tool from the data on the tool life stored in the tool/holder database 3band the machining load, the machining contact point, the cut rate and the machining time predicted by the machining state simulation unit 7a. The wear distribution is defined as a prediction, for a ball end mill, for example, as to whether the forward end of the mill wears, and at which position the forward end wears, the straight portion or other portions. Based on this predictive calculation of the wear, the expiry of the service life of the tool is judged from the predictive calculation of the wear. Further, the tool sensor of the sensor unit 13 corrects the prediction value based on the change in the tool profile (diameter, length, center position, orientation, etc. of the tool) while turning.

[0057] The trip predictively calculated in the tool behavior simulation unit 7dis sent to the correction unit 5dof the tool path determining unit 5. In the case where the predicted tool trip is larger than a predetermined value, the tool generating unit 5bgenerates a tool path taking the tool trip into account. In the case where a large machining load is imposed for relatively moving the work and the tool to each other in the horizontal direction, for example, the tool is deformed in such a way as to trip forward in the direction of movement. Therefore, the movement of the forward end of the tool lags behind the coordinate of the main spindle. As a result, the feed rate is required to be corrected in such a manner that the forward end of the tool on the main spindle moves to a predetermined machining position in order to permit the required machining operation. A high-precision machining operation thus becomes possible.

[0058] Now, a display unit 17 will be explained. The display unit 17 includes an image processing unit 17aand a monitor 17bsuch as a CRT. The display unit 17 can display the machining area, the machining pattern, the profile and the serial number of the tool, etc. determined in the machining process determining unit 5aof the tool path determining unit 5. The image processing unit 17areceives the data on the curvature, the inclination and the depth of the machining area from the machining process determining unit 5aand can generate the data for displaying each area three-dimensionally in different colors. Further, though not shown in detail in FIG. 1, the image processing unit 17ais connected to the database 3, the predictive calculation unit 7, the numerical control unit 9 and the sensor unit 13 and thus can display the information from each of these component parts graphically or as a text. The display unit 17 can graphically display, for example, the result of the check for the work-tool interference based on the data from the predictive calculation unit 7. Further, the data on the machining accuracy included in the machining profile data 1ainput to the input database 3fand the work profile measurement data obtained in the sensor unit 13 can be displayed in comparison with each other. Also, as described already, it is possible to graphically display the portion of the work remaining uncut by a machining path avoiding the work-tool interference.

[0059] The operation according to an embodiment of the invention will be described.

[0060] First, the operator inputs the machining profile data 1aand the workpiece data 1bfrom the input unit 1 into the database 3. According to the machining profile data 1aand the workpiece data 1bthus input and the data stored in the database 3, the tool path determining unit 5 selects and determines a machining area, a tool, a machining pattern and a machining process. Then, a tool path is determined by the tool path generating unit 5bof the tool path determining unit 5, and machining conditions by the machining conditions determining unit 5c. The machine tool 11 is driven under the control of the numerical control unit 9 according to the machine drive data including the tool path and the machining conditions generated and determined. In this way, the work is machined and a product is finished.

[0061] In the process, the predictive calculation unit 7 makes various calculations for predicting the machining load and the work-tool interference based on the machining profile data 1aand the workpiece data 1binput by the operator from the input unit 1 and the various data stored in the database 3. The result of predictive calculation is applied to the machining process determining unit 5aof the tool path determining unit 5. A tool, a machining pattern and a machining process are thus determined in the predictive calculation unit 7 according to the result of predictive calculation. In similar fashion, based on the result of predictive calculation in the predictive calculation unit 7, a tool path is determined by the tool path generating unit 5bof the tool path determining unit 5 and machining conditions by the machining conditions determining unit 5c.

[0062] As described above, according to this invention, various calculations are carried out in advance by the predictive calculation unit 7 for predicting the machining load, the interference between the work and the tool, etc. based on the machining profile data 1aand the workpiece data 1binput from the input unit 1 and the data stored in the database 3. In other words, unlike in the prior art in which a work is first machined followed by detection and feedback of the machining requirements in an attempt to control the operation optimally, the method according to the present invention starts with predictive calculations, and the result of-the predictive calculations is fed forward and used for machining the work. Also, the detected machining requirement is fed back for further improving the accuracy of the feedforward control. Consequently, the operation is optimized and the work can be machined with high accuracy and move quickly than in the prior art.

[0063] As can be seen from the foregoing description, according to this invention, the operator inputs only the profile data on the product to be machined and the data on the profile and material of the work to be loaded and machined. Thus, the operation of machining a work meeting the precision requirement can be automatically performed within a shorter length of time as a process meeting the required accuracy is executed automatically with a proper tool, along a proper tool path and under proper machining conditions.

[0064] An example comparison between the prior art and the present invention is shown in FIGS. 8A, 8B. In the case where the machining profile databased on the drawing as shown in FIG. 8 is input, according to the prior art, assume that the feed rate of the tool is kept so high that the work is machined at 6 m/min, for example, in performing the machining operation. Corners are often undesirably dulled as shown by one-dot chain in FIG. 8B (the portions designated by reference numerals 23a, 23b) and the radius of curvature is shortened (the portion designated by numeral 23c). Further, according to the prior art, in the case where the tool is fed along a circumference at high speed, the true roundness is decreased as shown by numeral 23dor the thickness is increased as shown by numeral 23E. If the tool feed rate is reduced to, say, 200 mm/min in the prior art, on the other hand, a high-precision machining operation as shown by numerals 23f, 23gbecomes possible. According to this invention, in contrast, as shown by solid line in FIG. 8B, the above-mentioned inconvenience is not caused even when the machining operation is performed with the tool kept fed at high speed. This can be achieved only by reason of the fact that the predictive calculation unit 9 calculates a predicted value of the machining load and checks for interference between the tool and the work in advance, and that various data stored in the database 3 are appropriately utilized.