CN111443661A - Method and device for automatically machining workpieces by means of a machine tool - Google Patents
Method and device for automatically machining workpieces by means of a machine tool Download PDFInfo
- Publication number
- CN111443661A CN111443661A CN202010042329.XA CN202010042329A CN111443661A CN 111443661 A CN111443661 A CN 111443661A CN 202010042329 A CN202010042329 A CN 202010042329A CN 111443661 A CN111443661 A CN 111443661A
- Authority
- CN
- China
- Prior art keywords
- machining
- workpiece
- variable
- parameters
- modeled
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003754 machining Methods 0.000 title claims abstract description 206
- 238000000034 method Methods 0.000 title claims abstract description 136
- 230000008569 process Effects 0.000 claims abstract description 90
- 238000012545 processing Methods 0.000 claims abstract description 37
- 238000005457 optimization Methods 0.000 claims abstract description 32
- 230000001419 dependent effect Effects 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 21
- 238000013531 bayesian neural network Methods 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims 3
- 230000008859 change Effects 0.000 description 8
- 238000005520 cutting process Methods 0.000 description 5
- 238000005219 brazing Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 238000003466 welding Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005553 drilling Methods 0.000 description 2
- 238000011478 gradient descent method Methods 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 238000003801 milling Methods 0.000 description 2
- 238000005245 sintering Methods 0.000 description 2
- 238000004026 adhesive bonding Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000005304 joining Methods 0.000 description 1
- 238000003698 laser cutting Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000010399 physical interaction Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005476 soldering Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/4097—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by using design data to control NC machines, e.g. CAD/CAM
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q15/00—Automatic control or regulation of feed movement, cutting velocity or position of tool or work
- B23Q15/007—Automatic control or regulation of feed movement, cutting velocity or position of tool or work while the tool acts upon the workpiece
- B23Q15/12—Adaptive control, i.e. adjusting itself to have a performance which is optimum according to a preassigned criterion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32153—Exchange data between user, cad, caq, nc, capp
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32187—Correlation between controlling parameters for influence on quality parameters
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/36—Nc in input of data, input key till input tape
- G05B2219/36289—Cutting, machining conditions by optimisation of time, cost, accuracy
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Human Computer Interaction (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Manufacturing & Machinery (AREA)
- Probability & Statistics with Applications (AREA)
- Mechanical Engineering (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Numerical Control (AREA)
- Automatic Control Of Machine Tools (AREA)
Abstract
The invention relates to a method for machining one or more workpieces with an automated machine tool, wherein a machining head and the workpiece are moved relative to each other along a machining path, having the following steps each being carried out for machining one of the one or more workpieces: processing the workpiece along the processing track according to preset processing parameters; measuring one or more process parameters during processing of the workpiece along the process path; the process parameters are optimized by means of an optimization method based on a cost function and at least one boundary condition, wherein, by means of a process model, a modeled process variable and an uncertainty variable associated with the modeled process variable are determined as a function of the process parameter considered during the optimization, wherein the boundary condition is associated with a value range, which surrounds the modeled process variable and is dependent on a predefined probability, wherein the modeled process variable, the uncertainty variable and the predefined probability describe the value range surrounding the modeled process variable.
Description
Technical Field
The invention relates to a method for automatically controlling a machine tool for machining workpieces. The method relates in particular to the optimization of a machining path, according to which a workpiece is machined by a machine tool.
Background
In the cutting of workpieces by automated machine tools, various factors can influence the machining time, the machining quality and the quality of the machined workpiece. In general, for machining, a machining path is predefined with which the machining head of the machine tool is moved relative to the workpiece to be machined. In this case, the drive-off (Abfahren) on the machining trajectory is achieved on the basis of a plurality of machining parameters which are decisive for the machining time, the machining quality and the quality of the machined workpiece.
Machining methods can include cutting techniques, such as, for example, drilling, milling, sawing, welding, etc., cutting techniques, such as, for example, laser cutting, heat treatment, such as, for example, soldering, sintering, etc., and joining techniques, such as, for example, gluing, bonding, etc., and generally other methods in which the tool is moved relative to the workpiece and material-related (stofflich) measures, i.e., measures for material change and/or measures for shape change, are carried out.
Disclosure of Invention
According to the invention, a method for machining a workpiece with a machine tool according to claim 1, and a corresponding device and machine tool according to the parallel claims are proposed.
Further embodiments are specified in the dependent claims.
According to a first aspect, a method for machining a workpiece with an automated machine tool is proposed, wherein a machining head and the workpiece are moved relative to each other along a machining path, wherein the machining path is determined by machining parameters, having the following steps:
-machining the workpiece along the machining trajectory according to predetermined machining parameters;
-measuring one or more process variables, in particular one or more profiles of the process variables, during the processing of the workpiece along the processing path;
optimizing the machining parameters by means of an optimization method based on a cost function and at least one boundary condition,
wherein a modeled process variable and an uncertainty variable associated with the modeled process variable are determined from the process variable considered during the optimization by means of a process model, which is provided as a regression model, wherein the boundary condition is associated with a value range for the modeled process variable that depends on a predefined probability, wherein the modeled process variable, the uncertainty variable and the predefined probability specification (angelen) surround the value range of the modeled process variable.
In automated machining methods, the machining of the workpiece is usually carried out by a movement of the machining head of the tool relative to the workpiece, so that a material change or a shape change of the workpiece is achieved. The movement of the machining head follows a machining path and can be characterized by machining parameters. The machining path can be defined as a path parameterized by the machining parameters or can be divided into path segments defined, for example, by linear segment curves of the respective individual machining parameters. Depending on the type of machining method, the machining parameters can be the feed speed, the feed force of the machining tool onto the workpiece, the tool rotational speed (in a rotating machining head, such as, for example, in drilling), the machining temperature of the machining head, etc.
The trajectory section of the processing trajectory is defined by a section time (Abschnittszeit) or a section distance. The segment times or segment distances may have the same or different durations or lengths. The machining method carries out the machining in each track section according to machining parameters predefined for this purpose. The parametric transitions may be smoothed by interpolation according to their physical meaning. Thus, for example, the feed rate or the tool temperature does not change stepwise.
Depending on the choice of the machining parameters of the machining in the individual machining paths, the speed, duration and/or quality of the machining are better or worse. It can thus be provided that the cost function considered for the optimization is dependent on the total duration for the machining and/or one or more quality parameters which characterize the quality of the machining and which depend on the modeled machining variables, wherein the one or more quality parameters are determined in particular by the profile of the modeled machining variables, and in particular the mean, maximum, minimum of the profile as modeled machining variables are evaluated for the entire machining path or individually for one, some or all path segments.
The optimization of the machining method is carried out according to a cost function and taking into account boundary conditions. The boundary conditions may be considered during the optimization process alone or as part of a cost function. The optimization can be carried out automatically by an optimization process, wherein the quality of the machining is ensured by one or more corresponding boundary conditions with respect to the quality parameters with optimized machining parameters.
The boundary conditions are defined with respect to the process parameters. Since the process variables cannot be measured during the optimized calculation, these are provided as modeled process variables according to the process model. Since the process model is often defective, for the above-described method a regression model is used as the process model, which, in addition to the model values, also provides uncertainty parameters which describe the reliability of the model values. By defining a boundary condition with a value range in which the true values of the process variables under consideration are within a predetermined probability by the boundary condition, the quality of the process can be avoided by avoiding process variables outside the value range for too long a duration. Although it is possible that the true values of the process variables are sometimes outside the value range defined by the boundary conditions, this is the case only in a relatively short time period due to the predetermined probability.
According to the method described above, the machining target specified by the minimization of the total cost is now optimized while probabilistically complying with the boundary conditions. For this purpose, a gradient of the machining model of the target variable is calculated and is followed to such an extent that at least with a certain predetermined probability a boundary condition is complied with, i.e. a boundary value of the relevant machining variable lies with a predetermined probability within a value range of the relevant machining variable.
For example, the specific feed speed of the tool determines the specific feed force (machining variable) acting on the workpiece. When machining a workpiece, this feed force can be recorded with respect to the feed speed (machining parameters) and a corresponding machining model learned. As a boundary condition in the optimization, the feed force can be limited for reasons of tool protection or wear reduction. Through the training of the machining model, a regression with uncertain limits is generated between the feed force and the feed speed. The boundary conditions suitable for the optimization can now be predefined such that the predefined boundary values for the feed force (process variable) are not exceeded with a specific probability. This method enables a better trajectory determination of the machining trajectory for the machine tool.
The optimization of the machining method can be carried out automatically by an optimization process, wherein a cost function is also defined as a function of the quality parameters and the machining parameters of each trajectory section are optimized on the basis of the cost function.
The cost function describes cost values that depend on quality parameters that respectively represent an evaluation of the machining method or a part thereof. For example, the quality parameters can specify the time taken for machining the workpiece, the occurrence of vibrations during the machining of the workpiece, material wear of the machining head and the machining accuracy with respect to the machined workpiece, the temperature of the machining head and/or the feed force acting on the machining head.
The method described above uses a trainable machining model via physical interaction relationships in machining a workpiece in order to improve the optimization of the machining path. This is achieved by selecting a model as the processing model which provides, on the one hand, a regression curve of the model values and, on the other hand, the respective uncertainties of the individual model values of the regression curve. Thus, the trainable model enables a reliable optimization of the cost function. When using, for example, a gradient descent method, the gradient used for the optimization can be calculated by the model, but also the spacing of its reliable boundaries, in which the modeled process parameters are specified, can be determined.
In principle, the optimization method is carried out by optimizing the machining parameters according to an optimization objective, which can be predefined, for example, as a minimization of the machining time or generally as a minimization of the cost according to a cost function, and which can take into account boundary conditions, such as, for example, quality and machining, and wear of the tool. In this case, the machining parameters are optimized by minimizing the costs, while at the same time complying with the boundary conditions.
It may be provided that the machining model is selected as a trainable machining model, wherein the learning process is carried out on the basis of one or more machining variables measured with respect to the machining with predefined machining parameters. In this way, it can be provided that the trainable processing model is updated after each or a corresponding number of processing steps, wherein the processing result is taken into account during the learning process depending on the processing parameters used.
The trainable processing model may be any regression model that is also capable of providing uncertainty in model values relative to each of the processing parameters. Furthermore, a regression model is advantageous which also accounts for gradients with respect to each of the process parameters. In particular, gaussian process models, bayesian linear regression models, regular kernel ridge regression models, bayesian neural networks, and/or the like may be used.
Drawings
The embodiments are explained in detail below with the aid of the figures. In which is shown:
fig. 1 shows a schematic view of an automated machine tool configured for carrying out a method for machining a workpiece,
fig. 2 shows a schematic diagram for explaining a flow chart of a method for machining a workpiece by means of the machine tool of fig. 1; and is
Fig. 3 shows a diagram for illustrating the consideration of the uncertainty values of the modeled process variables.
Detailed Description
Fig. 1 shows a machine tool 1 with a machining head 6, which can be driven in rotation by a spindle 7. The machining head 6 can be designed in particular as a cutting tool, with which the workpiece 2 can be machined. The machining head 6 is operated by a drive unit 3 which machines the workpiece 2 in a material or shape-changing manner by means of a suitable movement of the machining head 6 of the tool. For example, the processing head 6 may be a cutting tool such as a drill, a milling head, a laser, a heat treatment tool such as a brazing pin, a sintering tool, a welding head (resistance welding), etc., an ultrasonic head (ultrasonic bonding), a bonding nozzle, etc.
The drive unit 3 is also capable of causing a feed (vortirib) of the processing head 6 in order to process the workpiece 2 along a distance. The processing head 6 can also be connected to a sensor system 9 in order to measure the forces acting between the processing head 6 and the workpiece 2, the temperature of the processing head 6 and/or the workpiece 2 or other resulting output variables. Furthermore, the workpiece 2 may be provided with a sensing system 8, which may measure vibrations, temperature and/or other characteristics of the workpiece during processing.
By means of the control unit 10, which actuates the drive unit 3 in such a way that the workpiece 2 is machined according to the machining path, the sensor values of the sensor systems 8, 9 can be evaluated as described below.
The machining is carried out in a controlled manner by the control unit 10 according to a machining trajectory which is determined by the machining parameters. The processing track may comprise a plurality of track sections. The machining parameters describe preset machining values for machining the workpiece. The machining parameters can specify, for example, a thrust quantity (zustelung), a rotational speed, a depth of cut, a feed speed and/or the like for each path section of the machining path. If different parameter values of the processing parameter are specified for different track sections, the transitions between the parameter values of the processing parameter can be smoothed in order to avoid a step-like change of the parameter values. In particular, in the case of transitions between processing track sections, the parameter values can be interpolated, in particular linearly. Parameter steps which are technically not feasible can be avoided. Alternatively, the machining path can be defined without sections, wherein the profile of the machining path is defined by the profile of one or more machining parameters.
During the machining according to the machining trajectory, the control unit 10 can detect, take into account and temporarily store sensor values relating to the machining head 6 (via the sensor system 9) and to the workpiece (via the sensor system 8).
The method for operating the machine tool 1 is described in more detail below with the aid of the flow chart of fig. 2.
In step S1, the workpiece 2 is machined according to the corresponding machining parameters initially or by optimizing a predefined machining path. For this purpose, for example, the feed rate, the advance, the rotational speed and/or the depth of cut can be predefined in order to carry out an exemplary sawing process with a circular saw. For the brazing process, for example, the feed rate, the brazing material input and the brazing temperature can be predefined.
While the machining head 6 is moved away from the machining path for machining, sensor values of the sensor systems 8, 9 are detected in step S2, which sensor values relate to the generated machining variables, such as, for example, the feed force acting on the workpiece 2 counter to the feed direction, the rotational torque for driving the circular saw at a predetermined rotational speed, the tool temperature, the workpiece temperature during machining, the workpiece vibration during machining, etc. The change curve of the corresponding processing parameter is stored according to the processing track.
In step S3, a machining model that maps the machining process in a model manner is trained on the basis of the used machining parameters and the measured machining parameters. The process model may be, for example, a gaussian process, bayesian linear regression, regular kernel ridge regression, bayesian neural network, or the like. In general, the process model is selected such that it is capable of providing a regression of at least one of the process parameters based on the process parameters. Furthermore, the machining model is selected in such a way that, in addition to the estimated model values of the machining variables, it can also output in each case an uncertainty value for the model value of the relevant machining variable. The uncertainty value is defined, for example, by a factor describing the standard deviation, an uncertainty range around the model value of the process model, within which the true value of the relevant process variable lies with a predefined uncertainty.
According to the approximation method, the machining parameters for the path segments are then optimized in step S4. For this purpose, the optimization of the machining parameters for the travel away from the machining trajectory is carried out by means of a suitable optimization method (for example, a gradient descent method, etc.) as a function of a target preset value (which may correspond, for example, to a minimization of a cost function).
The cost function may define a cost in terms of a processing duration, one or more quality parameters of the processing, and the like. For example, the quality parameters for one, several or all track sections can specify a corresponding duration for the machining, a corresponding measure of the profile of the vibrations occurring during the machining of the workpiece (measured as vibrations by the sensor system 8), a corresponding measure of the material wear of the machining head (determined by the profile of the feed force, which corresponds to the force acting on the machining head by the feed of the machining head and is measured by the sensor system 9), and a corresponding measure of the machining accuracy of the machined workpiece, which specifies how the machined workpiece deviates from the theoretical state of the workpiece. The profile of the machining variable, such as, for example, the profile of the feed force, the profile of the occurrence of vibrations, etc., can be specified as an average, maximum, minimum of the relevant machining variables for the entire machining path or individually for one, some or all path segments.
The new optimized processing parameters are now provided for subsequent processing of another workpiece.
The optimization method can also take into account, in addition to the predefined costs or objective function, one or more boundary conditions, which can be predefined with respect to one or more process variables. In this case, the boundary conditions can be predetermined for the processing variables that result from the processing parameters that are taken into account during the execution of the optimization method. An optimization solution in which the limit value is exceeded or undershot by the expected value of the process variable is correspondingly not permissible. For this purpose, the modeled machining variables are determined as model values from the trainable, in particular self-learned machining model as a function of the respectively considered machining parameters and are accordingly taken into account during the minimization of the target variables/costs as a function of the cost.
In this case, the boundary condition is not only defined as a fixed boundary value of the process variable, but also the uncertainty of the corresponding model value of the process variable, which is described by the corresponding uncertainty value, can be taken into account in a suitable manner during the optimization. In this case, one or more boundary values or ranges of the permissible values of the process variable depend on the modeled values of the process variable and their associated uncertainty values for the particular combination of parameter values.
In the case of a probability predefined by the boundary conditions, a value range of the model value surrounding the machining variable can be determined from the model values of the machining variable for a given predefined value of a particular machining parameter or for a given predefined value of a particular combination of machining parameters under consideration, which value range indicates that the actual value of the machining variable lies within the value range with a predefined probability when machining with the corresponding machining parameter value. In which the uncertainty of modeling by means of a trainable machining model is reflected.
By means of this machining model, a point-to-point (post) distribution is obtained for each individual point of the modeled machining trajectory, that is to say, for the case of a gaussian process as a regression model, this corresponds to a gaussian distribution.
Numerical range of the process variables
The actual value of the process variable lies within this value range with a predetermined probability, as determined by the following equation:
wherein μ corresponds to a model value of a process variable corresponding to the process model, p corresponds to a predefined probability, and the actual value of the process variable lies within a predefined distribution interval defined by z (predefined as the quantile z of the standard deviation σ). The upper and lower boundary values x1, x2 for the scatter interval are derived from the solution of the above equation. The scatter interval describes a value range of the machining variable within which the actual value of the machining variable lies with a predetermined probability.
As is shown by way of example in fig. 3, a model curve, which is specified in particular by the machining model, for the modeled machining variable x can describe a value range profile as a function of an exemplary machining parameter z. The model curve is used as a curve of the change of the modeled process variable x with respect to the exemplary process variable z. The model curve describes the course of the lower and upper range limits GU, GO of the value range of the respective values of the modeled process variable, which should not be undershot or undershot with a probability predetermined by the limit conditions. The lower/upper range boundaries GU, GO of the relevant modeled process variables can then be used as boundary values for the boundary conditions, which indicate the particular measurement reliability. For example, the range boundary is generated by a predetermined probability.
In this way, the adherence to the boundary conditions for the currently considered values of the machining parameters is checked during the optimization. In order to obtain the corresponding range boundary by predetermining the probability, the equation (distribution function) for the cumulative density is considered as a starting point, the desired probability is brought in and solved according to the standard deviation. This value is used with the post standard deviation of the gaussian process (e.g. at 99% a width of z =2.576 σ is obtained). (i.e., upper border = μ + z σ).
For example, in fig. 3, the range of values surrounding the modeled generated force is illustrated based on a modeled profile of the generated force (machining variable x) acting on the tool at a particular feed speed. The boundary conditions are predefined as conditions for the process variables to be within the range of two standard deviations (i.e., with a probability of approximately 95%) in this case.
If the machining parameter "feed speed" is to be selected such that the force acting on the tool should not exceed 120N with a probability of 95%, a feed speed of 320 mm/min is thereby obtained in the case of the upper limit value of the numerical range 120N.
The method described above can be carried out cyclically during the machining of a workpiece by a machine tool. The method can also be used in advance in the initial setting of the machine tool in order to determine a set of initial machining parameters for machining the workpiece. When used during the optimization method, the feed rate is limited as a boundary condition to a value at which the forces acting on the tool are not exceeded with a specific probability.
Claims (12)
1. Method for machining one or more workpieces (2) with an automated machine tool (1), wherein a machining head (6) and the workpiece (2) are moved relative to one another along a machining path, wherein the machining path is determined by machining parameters, having the following steps, each of which is carried out for machining one of the one or more workpieces (2):
-machining (S1) the workpiece (2) along the machining trajectory according to pre-given machining parameters;
-measuring (S2) one or more process variables, in particular one or more profiles of process variables, during the processing of the workpiece (2) along the processing trajectory;
-optimizing (S4) the machining parameters by means of an optimization method based on a cost function and at least one boundary condition,
wherein the modeled process variable and the uncertainty variable associated with the modeled process variable are determined from the process variable considered during the optimization by means of a process model, which is provided as a regression model,
wherein the boundary condition is associated with a value range surrounding the modeled process variable that is dependent on a predefined probability,
wherein the modeled machining variable, the uncertainty variable and the predefined probability account for a range of values surrounding the modeled machining variable.
2. The method according to claim 1, wherein the boundary condition is satisfied when the predefined boundary value lies within a value range for the process variable.
3. The method according to claim 1 or 2, wherein the machining trajectory is divided into trajectory sections in which the workpiece (2) is machined on the basis of one or more machining parameters.
4. A method according to claim 3, wherein the cost function considered for the optimization is dependent on one or more quality parameters dependent on the modeled process variable, wherein the one or more quality parameters are determined in particular by or dependent on a profile of the modeled process variable, and in particular as an average, maximum, minimum of the profile of the modeled process variable are specified for the entire process trajectory or individually for one, some or all trajectory sections.
5. The method of any of claims 1-4, wherein the process parameters include a measure of one or more of:
-vibrations occur during machining of the workpiece (2),
-material wear of the machining head (6);
-the temperature of the workpiece (2);
-the temperature of the machining head (6);
-a feed force acting on the machining head (6) and the workpiece (2); and
-a description of the machining accuracy of the machined workpiece (2).
6. Method according to any one of claims 1 to 5, wherein the machining parameters for the machine tool (1) comprise one or more of the following parameters:
-a feed speed of the machining head (6),
-a force of feed,
-the temperature of the machining head (6),
-the rotational speed of the machining head (6); and
-the power of the machining head (6).
7. Method according to one of claims 1 to 6, wherein the machining model is selected as a trainable machining model, wherein a learning process is carried out on the basis of one or more machining parameters measured with respect to machining the workpiece (2) with the predefined machining parameters.
8. The method of any one of claims 1 to 7, wherein the process model comprises a Gaussian process model, a Bayesian linear regression model, a regular kernel ridge regression model, and a Bayesian neural network.
9. Device, in particular control unit, for machining a workpiece with an automated machine tool (1), wherein a machining head (6) and the workpiece (2) are moved relative to each other along a machining path, wherein the machining path is determined by machining parameters, wherein the device is designed for machining one of one or more workpieces (2) in each case:
-machining the workpiece (2) along the machining trajectory according to predefined machining parameters;
-measuring one or more process variables, in particular one or more profiles of process variables, during the processing of the workpiece (2) along the processing path;
optimizing the machining parameters by means of an optimization method based on a cost function and at least one boundary condition,
wherein the modeled process variable and the uncertainty variable associated with the modeled process variable are determined from the process variable considered during the optimization by means of a process model, which is provided as a regression model,
wherein the boundary condition is associated with a value range surrounding the modeled process variable that is dependent on a predefined probability,
wherein the modeled machining variable, the uncertainty variable and the predefined probability account for a range of values surrounding the modeled machining variable.
10. A machine tool (1) having a drive unit (3) for moving a workpiece (2) and a machining head (6) relative to each other and an arrangement according to claim 9 for controlling the drive unit (3) along a machining trajectory as a function of machining parameters.
11. A computer program for performing all the steps of the method according to any one of claims 1 to 8 when the computer program is implemented on a data processing apparatus.
12. A storage medium readable by machine, on which a computer program according to claim 11 is stored.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE102019200482.2A DE102019200482A1 (en) | 2019-01-16 | 2019-01-16 | Method and device for the automated machining of a workpiece with a machine tool |
| DE102019200482.2 | 2019-01-16 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111443661A true CN111443661A (en) | 2020-07-24 |
Family
ID=71131771
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010042329.XA Pending CN111443661A (en) | 2019-01-16 | 2020-01-15 | Method and device for automatically machining workpieces by means of a machine tool |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111443661A (en) |
| DE (1) | DE102019200482A1 (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP4019177A1 (en) | 2020-12-22 | 2022-06-29 | FRONIUS INTERNATIONAL GmbH | Method and device for assessing the quality of a processing process |
| CN113059184B (en) * | 2021-03-30 | 2022-08-05 | 南京航空航天大学 | Optimization method for parameters of ingot blank spray forming process |
| DE102021204215A1 (en) * | 2021-04-28 | 2022-11-03 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung eingetragener Verein | Process for the automated processing of workpiece surfaces by polishing, grinding or painting |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040093107A1 (en) * | 2002-11-08 | 2004-05-13 | Good Richard P. | Probability constrained optimization for electrical fabrication control |
| CN101493686A (en) * | 2009-02-26 | 2009-07-29 | 上海交通大学 | Cutting tool mode parameter uncertain curve five-shaft numerical control process parameter optimizing method |
| US20110301723A1 (en) * | 2010-06-02 | 2011-12-08 | Honeywell International Inc. | Using model predictive control to optimize variable trajectories and system control |
| CN102609591A (en) * | 2012-02-16 | 2012-07-25 | 华中科技大学 | Optimization method of cutting parameters of heavy machine tool |
| CN104965483A (en) * | 2015-07-16 | 2015-10-07 | 北京工业大学 | Multi-axis numerical-control machine tool processing precision retaining ability optimization method based on robust design |
| US20160041536A1 (en) * | 2014-08-05 | 2016-02-11 | Mitsubishi Electric Research Laboratories, Inc. | Model Predictive Control with Uncertainties |
| US20170297198A1 (en) * | 2013-03-15 | 2017-10-19 | Carnegie Mellon University | Supervised Autonomous Robotic System for Complex Surface Inspection and Processing |
| CN107480354A (en) * | 2017-07-28 | 2017-12-15 | 东北大学 | A kind of turnery processing parameter optimization method for considering uncertain parameter |
| CN107976956A (en) * | 2017-11-22 | 2018-05-01 | 沈阳机床股份有限公司 | The multi objective cutting data generating algorithm and cutting parameter optimization method of numerically-controlled machine tool |
-
2019
- 2019-01-16 DE DE102019200482.2A patent/DE102019200482A1/en active Pending
-
2020
- 2020-01-15 CN CN202010042329.XA patent/CN111443661A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040093107A1 (en) * | 2002-11-08 | 2004-05-13 | Good Richard P. | Probability constrained optimization for electrical fabrication control |
| CN101493686A (en) * | 2009-02-26 | 2009-07-29 | 上海交通大学 | Cutting tool mode parameter uncertain curve five-shaft numerical control process parameter optimizing method |
| US20110301723A1 (en) * | 2010-06-02 | 2011-12-08 | Honeywell International Inc. | Using model predictive control to optimize variable trajectories and system control |
| CN102609591A (en) * | 2012-02-16 | 2012-07-25 | 华中科技大学 | Optimization method of cutting parameters of heavy machine tool |
| US20170297198A1 (en) * | 2013-03-15 | 2017-10-19 | Carnegie Mellon University | Supervised Autonomous Robotic System for Complex Surface Inspection and Processing |
| US20160041536A1 (en) * | 2014-08-05 | 2016-02-11 | Mitsubishi Electric Research Laboratories, Inc. | Model Predictive Control with Uncertainties |
| CN104965483A (en) * | 2015-07-16 | 2015-10-07 | 北京工业大学 | Multi-axis numerical-control machine tool processing precision retaining ability optimization method based on robust design |
| CN107480354A (en) * | 2017-07-28 | 2017-12-15 | 东北大学 | A kind of turnery processing parameter optimization method for considering uncertain parameter |
| CN107976956A (en) * | 2017-11-22 | 2018-05-01 | 沈阳机床股份有限公司 | The multi objective cutting data generating algorithm and cutting parameter optimization method of numerically-controlled machine tool |
Non-Patent Citations (1)
| Title |
|---|
| 陈志同 等: ""面向单元切削过程的切削参数优化模型"", 《机械工程学报》, vol. 45, no. 5, 15 May 2009 (2009-05-15), pages 230 - 236 * |
Also Published As
| Publication number | Publication date |
|---|---|
| DE102019200482A1 (en) | 2020-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111443661A (en) | Method and device for automatically machining workpieces by means of a machine tool | |
| KR100428413B1 (en) | Apparatus and Method for Processing Simulation in Annecy Machining | |
| JP5710391B2 (en) | Processing abnormality detection device and processing abnormality detection method for machine tools | |
| CN111687652B (en) | Grip force adjusting device and grip force adjusting system | |
| JP2019082836A (en) | Tool life prediction device | |
| CN106584462A (en) | Method for adjusting running speed of robot in real time | |
| JP7101131B2 (en) | Numerical control system | |
| US9459166B2 (en) | Cutting resistance analysis device, cutting and machining device equipped with same, and cutting resistance analysis program | |
| CN105382412B (en) | Laser processing device with high speed positioning function | |
| JP6209392B2 (en) | Interference confirmation device | |
| JP7053518B2 (en) | Cutting fluid amount adjustment device and cutting fluid amount adjustment system | |
| JP6896737B2 (en) | How to monitor machine tools and controllers | |
| JP2005301440A (en) | Machining time calculating device | |
| EP3293593B1 (en) | Method for estimating error propagation | |
| Pereverzev et al. | Optimization of control programs for numerically controlled machine tools by dynamic programming | |
| JP2009518705A (en) | Method and control device for determining the period until essential maintenance of machine elements | |
| KR101896291B1 (en) | Tool path correction method of machining tools | |
| KR100372366B1 (en) | Numerical controlling unit using machining information | |
| EP3939744A1 (en) | Chip processing device for machine tool and chip processing method | |
| KR20080105691A (en) | Step length optimization method for multi axis milling process | |
| JP2016134078A (en) | Numerical control device for control over machine tool based on skiving processing command | |
| JP7177669B2 (en) | Analysis device, analysis method and processing system | |
| CN111103844A (en) | Program correcting device | |
| JP2020160830A (en) | Numerical control device, machine tool, control program, and storage medium | |
| US20240231312A1 (en) | Numerical control device and numerical control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |