Patent Application
20160202691
The present invention pertains to the field of quality assurance for production processes. More specifically, the present invention relates to a system and a method for controlling a production facility based on measurements of samples of the output of a production process, the output being defined by nominal data, for instant being provided as a mechanical drawing and/or a CAD model, and for compensating systematic errors occurring in the production process.
It is common practice during the industrial production of goods such as a car to measure features and properties of its different components. These measurements can be carried out in special measurement cells by means of either contact or non contact measuring gauges, for example based on laser or photogrammetric principles. Such a procedure, for instance, is disclosed in DE 195 44 240 A1.
U.S. Pat. No. 7,672,500 discloses a method for monitoring and visualizing the output of a production process, whose output materials or items are inspected by one or more inspection units. The inspection units scan or otherwise inspect each of a series of items or material being produced by a production process, and an image is generated representing each of the inspected items, wherein differences between the items can be visually coded.
The purpose of such methods is to determine possible errors of the measured object during product development, launch or during production. Disadvantageously though, in the measurement process, there can occur various additional errors, that prevent or complicate the determination of the errors of the object. This is especially the case if a high precision detection of errors is needed.
The quantities produced in the controlled production process can lie in a range between one piece production and mass production. The parts are produced in a production facility which can incorporate a broad scale of different manufacturing techniques and technologies. Depending on the specific manufacturing technique the installation set-up of the facility can vary. For example, the production facility can comprise a CNC machine—including programming means as well as electronic controller means—if the parts are to be produced by milling or turning.
The parts to be produced are specified by nominal data in a drawing and/or a CAD model that defines theoretical dimensions of the part in combination with appropriate tolerances. The tolerances define the accepted deviations between the specified theoretical dimensions of the nominal data and the real dimensions of a produced part.
The manufacturing processes referred to also include a quality control step wherein measures are taken to ensure the desired quality of the produced parts, i.e. to ensure that the percentage of “good parts” does not fall below a defined minimum. The quality control step consists of two sub-steps:
Today, in manufacturing processes of the kind described above, the measurement facility that detects the quality of the produced parts can be e. g. a coordinate measurement machine or an articulated arm (e. g. a measuring roboter arm). With this measurement equipment the “good parts” are detected by measuring one or more defined part dimensions (measurement step).
If the measurements show that the deviations between the measured part dimensions and the theoretical values defined in the nominal data exceed the accepted tolerances, appropriate parameter values of the production facility are amended to compensate these production errors. In the above example of a CNC machine this could be the case e. g. if a milling tool changes its characteristics due to wear.
This parameter value amendment, however, has the strong disadvantage that it needs good knowledge about the general installation set-up of the facility, the current constitution of the facility and about the cause of the production errors. This requires highly skilled personal, and moreover—since the production error causes often are not exactly known—in many cases a try-and-error approach has to be performed which is time consuming and costly.
Therefore, a possibility to compensate the production errors without changing the parameter values of the production facility—i.e. wherein the production facility can be treated as a “black box”—would be highly advantageous.
Some embodiments of the present invention provide an improved method and an improved system for controlling a production process of an object in a production facility.
Some embodiments provide such a method and such a system wherein no amendments in the production facility are necessary.
Some embodiments provide such a method and such a system wherein an intensity of the quality control is adaptable to an output of the production process.
Some embodiments provide an error compensation unit as a part of a system for controlling a production process for compensating errors in a production process without having to perform amendments in the production facility.
According to some embodiments, a first aspect of the invention relates to a method for controlling a production process of an object in a production assembly and for compensating errors occurring in the production process, the method comprising generating actual property data comprising obtained values of properties of at least one sample object produced in the production assembly according to a production model, performing a nominal-actual value comparison with the obtained values of properties of the actual property data and set values of corresponding properties of nominal property data of the object, thereby generating deviation data, and automatically creating an adapted production model based on the nominal property data and on the deviation data. The adapted production model is usable in an adapted production process for producing an adapted object in the production assembly and differs from the nominal property data so that the errors occurring in the production process are at least partially compensated in the adapted production process.
In one embodiment of the method according to the invention, generating the actual property data comprises measuring properties of at least one sample object in a measurement facility, wherein the object has been produced in the production assembly according to the production model.
In another embodiment of the method according to the invention, generating the actual property data comprises predictably calculating properties of an object virtually produced in the production assembly according to the production model and according to a provided production facility model.
In a further embodiment of the method according to the invention, creating the adapted production model is based on information about machine parameters of the production assembly, particularly on available processing means, such as tools, of the production assembly and/or adjustability of the processing means.
Particularly, the deviation data describes deviations between the obtained values and the corresponding set values and/or between the obtained values and corresponding production values of the production model.
In one embodiment of this method, creating the adapted production model comprises adapting at least one set or production value based on nominal property data and deviation data of the respective property of the object, so that the adapted production model comprises at least one adapted value of at least one property of the object, wherein the adapted values are amended with respect to the set or production values in such a way that an adapted object produced in the production assembly according to the adapted production model has smaller deviations with respect to the nominal property data than the at least one sample object.
In one embodiment, the nominal property data is provided as a model of the object, particularly a computer-aided design model or a mechanical drawing.
In another embodiment, the production model is based on the nominal property data.
In another embodiment, the production model is amended with respect to the nominal property data, and/or the method comprises amending the production model, wherein the amendment is based on information about the production assembly, particularly on information about known systematic errors.
In yet another embodiment, creating the adapted production model is also based on the production model.
In yet another embodiment, the production model comprises control commands for controlling the production process, and the adapted production model comprises adapted control commands for controlling the adapted production process.
In a further embodiment, the nominal property data comprises threshold values for the set values, each threshold value describing a maximum allowable deviation from the respective set value, wherein the adapted production model comprises adapted control commands that are adapted for producing in the production assembly an adapted object having at least one adapted value of a property, the respective obtained value of which exceeds a respective threshold value, and wherein the at least one adapted value deviates from the respective set value in such a way that a deviation between the set value and the corresponding obtained value is at least partially compensated when producing the adapted object.
In one embodiment, the method according to the invention comprises
In one embodiment, this method comprises
wherein the selecting and/or measuring are subject to a statistical process control step comprising monitoring the measurement results, and comprising
Particularly, the method comprises adapting a measurement program of the measurement facility based on the monitored measurement results, on the currently monitored ambient and/or production assembly parameters and/or on the identified correlations, and/or analyzing the statistical distribution comprises analyzing mean value, standard deviation, kind of statistical distribution function, number and distribution of outlier values, stability and/or trends.
In particular, the ambient and/or production assembly parameters comprise room temperature, machine temperatures at relevant positions, air humidity and barometric pressure, vibrations of the basement and/or the machines, noise level, brightness of illumination, time of day, day of week, calendar date, number and/or identity of persons currently working at the production assembly, machine hours since last maintenance or tool change, kind of last maintenance, storing time and conditions of the at least one sample object, current batch number and total number of the production lot, and/or processed materials.
A second aspect of the invention relates to a self-compensating manufacturing system adapted to produce at least one object in a production assembly, the production assembly comprising a production facility having at least one processing means and being adapted to produce the object, and a production control unit having means for storing and/or obtaining a production model and being adapted to control a production process of the production facility based on the production model. According to the invention, the manufacturing system comprises an error compensation unit for compensating errors in the production process of an object, the error compensation unit being adapted
In one embodiment, the manufacturing system according to the invention comprises a measurement facility that is adapted to generate the actual property data by measuring values of properties of at least one sample object that has been produced in the production assembly according to the production model, and a quality assurance facility that is adapted to perform the nominal-actual value comparison with the measured values and corresponding set values and to generate deviation data describing deviations between the measured values and the corresponding set values.
In one embodiment, the measurement facility comprises a coordinate measuring machine, an articulated arm, a laser scanner, a structured light measurement device, a coating or lamination thickness measurement device, a weighing device, a hardness measuring device, a temperature measuring device, and/or a device for measuring voltage, electric current, electric resistance and/or dielectric strength.
In another embodiment, the production facility comprises an additive manufacturing machine, a CNC machine, a pressing machine, a rolling machine, a wire and/or plate bending machine, a grinding, sanding and/or polishing machine, and/or a welding machine.
In a further embodiment, the processing means comprises at least one tool providing a 3D printing, drilling, turning, milling, cutting, honing, sanding, grinding, polishing, pressing, rolling, bending and/or welding functionality.
In another embodiment, the manufacturing system comprises a production error predicting means adapted to generate the actual property data by predictably calculating values of properties of an object virtually produced according to the production model and according to a provided production facility model of the production facility, to perform the nominal-actual value comparison with the calculated values and set values of corresponding properties of the nominal property data, and to generate deviation data describing deviations between the calculated values and the corresponding set values.
In particular, the provided production facility model comprises a geometry and/or stiffness model of at least one part of the production facility comprising geometric and/or stiffness data applicable for predicting production errors of an object virtually produced in the production facility, and/or the production facility comprises internal measurement means adapted for measuring properties of at least a part of the production facility, wherein the provided production facility model is based on the measured properties of the production facility.
A third aspect of the invention relates to a measurement facility for measuring values of properties of at least one object that is produced in a production facility according to a production model, the measurement facility being adapted to perform a nominal-actual value comparison with the measured values and set values of corresponding properties of nominal property data of the object, thereby generating deviation data. According to the invention, the measurement facility comprises an error compensation unit for compensating errors occurring in the production process of the object, the error compensation unit being adapted to automatically create an adapted production model based on the nominal property data and on the deviation data, wherein the adapted production model is usable in an adapted production process for producing an adapted object in the production assembly, and differs from the nominal property data so that errors occurring in the production process are at least partially compensated in the adapted production process.
In one embodiment, the measurement facility comprises a coordinate measuring machine, an articulated arm, a laser scanner, a structured light measurement device, a coating or lamination thickness measurement device, a weighing device, a hardness measuring device, a temperature measuring device, and/or a device for measuring voltage, electric current, electric resistance and/or dielectric strength.
In another embodiment of the measurement facility according to the invention, the measured properties comprise linear or angular dimensions, lengths, distances, radii, diameter, position and/or orientation, curvature, thickness, volume, weight, roughness, hardness, surface quality parameters, optical, thermal and/or electrical properties.
In yet another embodiment, the measurement facility is adapted for use with a manufacturing system according to the invention.
In a further embodiment, the measurement facility according to the invention comprises a statistical process control unit having a computing means for performing a statistical process control step that comprises monitoring measurement results and analyzing the measurement results with respect to changes over time, analyzing a statistical distribution of the measurement results, monitoring ambient and production assembly parameters and determining correlations between these, and/or monitoring ambient and/or production assembly parameters and determining correlations between the measurement results and at least one of these parameters. The statistical process control step further comprises controlling the measuring of properties and/or a selecting of sample objects from a plurality of objects that have been produced in the production assembly, the controlling being based on the analyzed measurement results.
Particularly, the statistical process control unit is adapted to adapt a measurement program of the measurement facility based on the monitored measurement results, on monitored ambient and/or production assembly parameters and/or on identified correlations, and/or analyzing the statistical distribution comprises analyzing mean value, standard deviation, kind of statistical distribution function, number and distribution of outlier values, stability and/or trends.
In particular, the ambient and/or production assembly parameters comprise room temperature, machine temperatures at relevant positions, air humidity and barometric pressure, vibrations of the basement and/or the machines, noise level, brightness of illumination, time of day, day of week, calendar date, number and/or identity of persons currently working at the production assembly, machine hours since last maintenance or tool change, kind of last maintenance, storage time and conditions of the at least one sample object, current batch number and total number of the production lot, and/or processed materials.
A fourth aspect of the invention relates to an error compensation unit for compensating errors in a production process of an object in a production assembly, particularly being adapted for use in a manufacturing system according to the invention and/or being adapted for use with a measurement facility according to the invention. According to the invention, the error compensation unit comprises data obtaining means adapted to obtain nominal property data of the object and deviation data that has been generated in the course of a nominal-actual value comparison with obtained values of properties of actual property data of at least one sample object and set values of corresponding properties of the nominal property data, a data storage device adapted to store the nominal property data and the deviation data, a calculation device adapted to automatically create an adapted production model based on the nominal property data and on the deviation data, and data provision means adapted to provide the adapted production model to the production assembly for an adapted production process of an adapted object, wherein the adapted production model differs from the nominal property data so that errors occurring in the production process are at least partially compensated in the adapted production process.
In one embodiment of the error compensation unit, the deviation data describes deviations between obtained values (e. g. measured or calculated values) of properties of actual property data of at least one sample object and set values of corresponding properties of the nominal property data and/or between the obtained values and corresponding production values of a production model.
In another embodiment, the data obtaining means is adapted to obtain information about machine parameters of the production assembly for performing the production process, particularly information about available processing means of the production assembly, and/or adjustability of the processing means, wherein the calculation device is adapted to create the adapted data set also based on the information about machine parameters of the production assembly.
In another embodiment, the error compensation unit comprises data output means, particularly a graphical display, adapted to provide the nominal property data, the deviation data, the adapted production model and/or graphical representations thereof to a user.
In yet another embodiment of the error compensation unit, the data obtaining means and the data provision means comprise a cable, means for allowing a wireless data transfer, or means for accepting a storage medium, in particular a compact disc or flash memory device.
Another aspect of the invention relates to a computer program product comprising program code which is stored on a machine-readable medium, or being embodied by an electromagnetic wave comprising a program code segment, and having computer-executable instructions for performing, particularly when executed on a computing unit of an error compensation unit according to the invention, the step of automatically creating an adapted production model based on the nominal property data and on the deviation data of the method according to the invention.
The invention in the following will be described in detail by referring to example embodiments that are accompanied by figures, in which:
In
If these deviations exceed a pre-defined threshold value (e. g. defined in the CAD data 30) actions are required to alter the production process in order to produce objects that meet the pre-defined threshold values. To achieve this aim, it is established practice to adjust the production assembly accordingly. For instance, a production program of the production control unit 11 can be amended or tools in the production facility 1 can be exchanged or altered. In the example of
Properties of sample objects are measured in the measurement facility 4, and deviation data 34 is generated based on the determined deviations from the nominal data 30. This deviation data 34 is then provided to an error compensation unit 6. Based on the deviation data 34 and on the nominal data 30—and optionally on information about the properties of the production facility 1—an adapted production model 36 is generated by an algorithm of the error compensation unit 6 based on the production model. This adapted production model 36, which is adapted to compensate the measured deviations, is provided to the production control unit 11.
In the example of
In one example, the diameter of the drilled hole might be too big compared to the nominal data 30 and the original production model 31. Thus, in the adapted production model 36 the size of the drilled hole might be reduced with respect to the nominal data 30. The object is then produced according to the adapted production model 36, i.e. with a smaller hole than before. As there are no further adaptations in the production facility 1, one can expect that an object is produced having a drilled hole that is too big with respect to the adapted production model 36, but nevertheless meets the threshold values with respect to the original nominal data 30.
In another example, it is the position of the drilled hole that is not correct; e. g. the measured value of the X coordinate is too small compared to the nominal data 30. Thus, in the adapted production model 36 the position of the drilled hole might be amended with respect to the nominal data 30, e. g. increasing the value of the X coordinate, in order to obtain a drilled hole at the desired position when producing an adapted object according to the adapted production model 36.
Adaptation of the model takes place with the production phase already in progress. The adapted objects are intended for sale or further processing. The process can be an at least partially automated loop with the steps
These steps, and particularly the amendment of the CAD-data, can be controlled by predefined algorithms, so that no user interaction is necessary.
The adapted production model 36 is not intended to represent the part how it should be produced (i.e. is no nominal data). Instead, it represents an object that intentionally deviates from the nominal data 30 in order to compensate opposed deviations that occur during the production. As it is solely intended for this purpose, the adapted production model 36 need not comprise any thresholds or tolerance values. Further, it may be a volatile data set and need not be archived.
Optionally, values provided in the initial production model 31 can already deviate from the nominal data 30. For instance, if systematic errors of the production facility 1 are known beforehand, the affected values can already be amended, e. g. manually by a user, before producing the first part 3. In this case, the measured values optionally might as well be compared with the adapted values of the initial production model 31 instead of the original values of the nominal data 30.
The method 100 starts with the provision 110 of model or nominal data of the objects that are to be produced, e. g. a CAD model of the object. Then, according to the provided model or nominal data, a number of objects (i.e. at least one object) are produced 120 in a production facility. A number of samples are selected 130 from the number of produced objects. Certain properties of the selected samples are measured 140 in a measurement facility, and the measured values are compared 150 with set values of the provided model or nominal data. Then, it is checked 160 whether the measured properties deviate from those of the nominal data more than a pre-defined threshold. If the samples are OK, production is continued without any changes. If the samples are not OK, i.e. if thresholds are exceeded for at least one measured value, according to the invention, for compensating the production errors, an adapted production model is created 170 and provided to the production assembly in order to produce adapted objects according to this adapted model. The adapted model does not provide nominal values of the properties that did not meet the defined thresholds. Instead, it is the purpose of the adapted model to provide false values for these properties in order to compensate errors of the production facility. Optionally, also the future selection 130 of samples can be influenced by the adapted model, as well as the measurement 140 of the samples (illustrated by the dashed arrows).
In
The production facility 1 and production control unit 11 here are shown as parts of a production assembly 10, which can be regarded as a “black box”, as according to the invention no amendments with respect to the production facility 1 are necessary. The production control unit 11 interprets the production model 31 into steering commands (e. g. CAM data) for the production facility 1. According to these steering commands, the production facility 1 produces a number of objects 3 as output. Before shipping or further processing of the output, samples of the output (i.e. at least one sample object) are measured in a measurement facility 4.
In the measurement facility 4 parameters of the sample objects are measured that are required to detect the quality of the produced output 3. These parameters particularly comprise geometrical part dimensions, such as lengths, distances, diameters, position coordinates, or thicknesses. For measuring these, the measurement facility 4 can comprise coordinate measuring machines (CMM), articulated arms, distance meters, optical measurement devices (e. g. transmitted light), image processing devices with cameras, devices for measuring thickness via electrical or ultrasonic parameters, mechanical gauges or standards, callipers etc. Furthermore, the parameters can comprise surface quality parameters, such as linear or angular dimensions, lengths, distances, radii, diameter, position and/or orientation, curvature, thickness, volume, weight, roughness, hardness, surface quality parameters, optical, thermal and/or electrical properties. Optionally, the measurement facility 4 can be integrated into the production assembly 10 (not shown here). For instance, CMM functionality can be physically integrated into a CNC production machine. This is advantageous because both types of machine have the need for positioning in three dimensions with high accuracy, so that synergies can be employed by fusing these two machines into one.
The results of the measurements are provided to a quality control unit 5, where the compliance of pre-defined thresholds provided by the nominal data 30 is monitored.
If the measured values are within the thresholds, i.e. if the measured sample objects are OK, the production process can continue without any amendments. If at least one threshold is exceeded, according to the invention, the data is provided to an error compensation unit 6, which creates an adapted model 36. The geometry and the dimensions of the adapted model 36 are based on the nominal data 30, and/or on the preceding production model 31, but its dimensions are changeable to a certain extent based on special error compensation values that are defined by the error compensation unit 6. Then, instead of the preceding production model 31, the model data of this new virtual part provided by the adapted production model 36 is used to program and/or control the production facility 1.
The error compensation values are defined in dependence of the production errors, i.e. the deviations 34 between the measured values of the properties (e. g. dimensions) of the produced parts 3 and the theoretical values of the nominal data 30.
The compensation values are defined in such a way that by applying them to the dimensions of the virtual part, the measured production errors are compensated without having to alter any of the parameters of the production facility 1 itself. If, for example, a measured length of a produced part is too big compared to the nominal CAD model 30, the corresponding length of the virtual part is decreased, so that this production error will be compensated without changing anything in the production assembly 10—which thus can be treated as a black box.
The definition of the error compensation values further depends on the definition of the tolerances of the original part as defined in the nominal data 30. This is important because it must be ensured that changing a dimension at one point of the part, i.e. where the deviations have been measured, does not lead to an unwanted dimensional change at another location of the part which could lead to exceeding tolerances there.
Additionally or alternatively, the error compensation values can be used to compensate the production errors on the level of the tool kinematics instead of on the model description level. Consequently, the error compensation values directly influence the production control unit 11 which control the production facility 1 (e. g. by a CAM program, “G-code”). For the example of a CNC machine that would mean that the error compensation values could alter the trace route of a milling tool.
Optionally, the error compensation values are defined not only in dependence of the measured deviations but additionally in dependence of relevant information about the production facility 1 (information flow represented by the dashed arrow). As described above, the error compensation values defined by the error compensation unit to adapt the geometry and dimensions of the adapted production model 36 are based on the deviations between the measured dimensions of produced parts 3 and the theoretical values of the nominal data 30.
However, in some situations the permitted changes of the dimensions of the adapted model 36 are not unlimited but restricted due to specific machine parameters of the production facility 1. An example for such a situation would be the diameter of a drilled hole which cannot vary continuously but might be limited to a discrete number of values, namely the diameters of the drill tools available in the production facility 1. If, for example, the theoretical diameter of the hole would be 10 mm and the produced parts show an actual diameter of 10.17 mm, then it will not be possible to reduce the theoretical value of the virtual part down to 9.83 mm, since there is no drill available with such a diameter. However, if there is a drill with a diameter of 9.8 mm, by using that drill, one could expect the resulting diameter to be about 9.97 mm, which then could meet a predefined threshold value for the 10 mm hole. Therefore, preferably, the error compensation unit 6 has access to all needed information about the production facility 1 which is relevant for choosing valid values for all changing dimensions of the virtual part—e. g. including information about all possible tool choices inside the production facility 1.
According to this embodiment, the manufacturing process comprises a statistical process control step to optimize the efficiency of the quality control step by minimizing the time and effort needed to perform the measurement step. This comprises monitoring the measurement results and analysing them with respect to changes over time. Particularly, the statistical distribution of the results can be analyzed, e. g. mean value, standard deviation, kind of statistical distribution function (e. g. Gaussian distribution), number and distribution of outlier values, stability, trends etc.
The statistical process control step may furthermore comprise monitoring ambient and production assembly parameters and trying to identify correlations. The monitoring can comprise measuring and/or identifying room temperature, machine temperatures at relevant positions, air humidity and barometric pressure, vibrations of the basement and/or the machines, noise level, brightness of illumination, time of day, day of week, calendar date, number and/or identity of persons currently working at the production assembly, machine hours since last maintenance or tool change, kind of last maintenance, storage time and conditions of the at least one sample object, current batch number and/or total number of the production lot, processed materials (e. g. what kind of steel or aluminum is currently used, which alloy, heat treatment, supplier) etc. Then, possible correlations between changes in the behaviour of the measurement results and the monitored ambient and/or machine parameters can be determined.
The statistical process control step may furthermore comprise adapting or optimizing the measurement program, i.e. trying to optimize the efficiency of the measurement program based on the identified behaviour of the current measurement results, on the currently monitored ambient and/or machine parameters and on the identified correlations. For example, if a measurement value stays very stable over time and has been found not to be affected from any fluctuating ambient parameters, then the measurement frequency (how many parts of a mass production need to be measured) can possibly be reduced.
Optionally, the measurement program of the measurement facility 4 can be adapted based on the changes between the nominal data 30 or the initial production data 31 and the adapted model 36. As described above, the measurement facility 4 is used to measure all the parameters (geometrical and others) of a produced part which are critical to identify the part as a “good part” or not. All these parameters are defined in a measurement program, which furthermore defines the frequency of each measurement, i.e. how many parts out of a production lot have to be measured and which selection strategy is used (every part, random samples or every 10th part etc.), the accuracy of each measurement, the measurement procedure (e. g. taking a mean value out of x single measurements) etc. For the definition of the measurement program, information from the nominal data is used about which dimensions and/or other properties are critical for the quality of the produced parts. The defined measurement program can then be adapted and thus optimized according to the adapted production model during the production process.
Optionally, a further quality control step can be performed using the SPC unit 8. According to the defined measurement program, samples of the produced parts 3 are measured in the measurement facility 4. The measured actual values of the critical part dimensions and/or other measured properties are then compared to the theoretical dimensions defined in the nominal data 30.
Normally, the actual measured values and the theoretical values will be not exactly the same. Nevertheless, the produced part can be a “good part” as long as the deviations between these values stay below defined thresholds. The definition of these thresholds is also taken from the nominal data 30. Based on all determined deviations of all measured dimensions and other part parameters in comparison to the accepted values the produced part is treated as a “good part” or a “bad part”. Furthermore, based on these results, the error compensation unit 6 defines error compensation values as described above. However, not only the part currently measured is taken into account but also formerly measured parts. The historical development of all the measured values is used to define the compensation values. For example, if only one part is identified as “bad” whereas all prior parts have been identified as “good”, then the error compensation unit will probably not immediately change the parameters of the virtual part since this bad part is probably only a single outlier. Instead of immediately changing the virtual part, the measurement program will be amended via the SPC unit 8 by increasing the measurement frequency.
In the embodiment shown in
For this part production simulation process, the manufacturing system is equipped with a production error predicting means 15 which is capable to predict reaction forces and displacements of the tools (and/or tool holders) of the production facility 1′ in the simulated part production process.
At the initial state, a geometrical calibration model and/or a stiffness calibration model of the production facility 1′ are provided to the error predicting means 15. This means that the errors of the production facility 1′ are known, e. g. the pitch, yaw and roll errors of a tool of the production facility 1′, and can thus directly be compensated by an adapted production model 36 even before producing the first sample part. The same applies to the position-dependent stiffness of a tool. As shown here, the production facility 1′ optionally comprises an internal measurement means 14 which is adapted for measuring properties of at least a part of the production facility 1′, e. g. certain tools. The provided production facility model is then based on values of the properties measured by the internal measurement means 14.
The production error predicting means 15 is adapted to predictably calculate properties of an object, as if this object would be produced in the production facility 1 according to the nominal data based production model 31 and according to a provided production facility model. This production facility model comprises a geometry model and/or a stiffness model of the production facility or parts thereof, particularly tools of the production facility. The geometry or stiffness model comprises geometric or stiffness data, respectively, applicable for predicting production errors of an object virtually produced in the production facility.
Furthermore, the production error predicting means 15 is adapted to determine deviations between the properties of the virtually produced object and the nominal data 30, to generate deviation data 34 describing these deviations, and to provide this deviation data 34 to the error compensation unit 6.
With the forces and displacements determined in the production facility 1′ of
As no real production takes place in the part production simulation process, the measurement facility 4 and the quality control 5 are not needed, until real objects 3′ are produced according to the adapted model 36.
Although the invention is illustrated above, partly with reference to some preferred embodiments, it must be understood that numerous modifications and combinations of different features of the embodiments can be made. All of these modifications lie within the scope of the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 15151145.8 | Jan 2015 | EP | regional |
