Generating a simulation model with respect to an actually produced object

Abstract

The invention relates to a method and a corresponding device for generating a modified simulation model, wherein the method comprises the following steps: A) providing a surface data set (22) of an actually produced object 2, wherein the actually produced object has been produced on the basis of a target shape (14), and wherein the surface data set (22) comprises data which are based on a measurement of at least one part of the surface of the actually produced object (2); B) providing a simulation model, wherein an initial shape state (23) of the simulation model represents the target shape (14) of the object (2) and wherein the initial shape state (23) is initially defined as the current shape state of the simulation model, wherein the simulation model is designed to simulate mechanical stresses within the object (2) which occur when the object (2) undergoes a given deformation; C) determining a further shape state (24) of the simulation model, in which, among all possible shape states in which, starting from the current shape state of the simulation model, at least one local region B1 of the simulation model approximates the surface data set (22), the simulation model has minimal mechanical stress energy; D) taking the further shape state of the simulation model determined in step C) as the new current shape state of the simulation model; E) defining the modified simulation model with the new current shape state from step D) as the modified simulation model of the actually produced object (2). The invention further relates to a method and a corresponding arrangement for the further use of the modified simulation model in subsequent simulations. The invention also relates to a corresponding computer programme and to a corresponding computer-readable storage medium.

Description

[0001] Creation of a simulation model in relation to an actually manufactured object

[0002] The invention relates to a method for generating a simulation model of an actually manufactured object. The invention further relates to a device for generating such a simulation model, wherein the device is specifically configured for carrying out the method. The invention also relates to an arrangement with a corresponding device for further use of the generated simulation model. Finally, the invention relates to a computer program and a computer-readable storage medium, each comprising instructions that, when the computer program is executed by a computer or an arrangement of computers, cause the computer(s) to execute the method.

[0003] Many industrial simulation applications require realistic simulation models to describe or model the current state of an actual manufactured object. These realistic simulation models do not correspond to the target data of an object, such as CAD data, and therefore do not represent the target state of the object.

[0004] In the prior art, various methods are proposed for generating such a realistic simulation model, which best represents the actual state of a manufactured object. EP 3382581 B1 describes a best-fit, for example using least squares, between the simulation model derived from target data and the surface data set based on a measurement. In EP 3382581 B1, the simulation model data used in the best-fit are assigned local mathematical weighting factors for the best fit. These weighting factors depend on the locally present stress tensors and are, for example, inversely proportional to the magnitudes of the locally present stress tensors.

[0005] In other words, in EP 3382581 B1, all available data are subjected to a global best-fit with the aim of minimizing the discrepancies between the surface dataset and the simulation model using least squares for all available data. In this process, local elements of the simulation model with high local stresses are given the opportunity, through weighting, to adapt less closely to the surface dataset than local elements of the simulation model with low local stresses.

[0006] The method proposed in EP 3382581 B1 has several disadvantages. Firstly, it includes all data from the surface dataset. This means that measurement artifacts, areas with poor data quality or density, and areas with strong curvatures, edges, or holes cannot be excluded from the global best-fit. Furthermore, the global best-fit itself is problematic because its sole objective is to minimize all distances in the Gaussian mean for all elements of the simulation model. This process deforms the shape of the simulation model into the surface dataset in a single step, causing the elements or areas of the simulation model to be automatically and uncontrollably compressed, stretched, or curved by the best-fit.Secondly, the best-fit method in EP 3382581 B1 does not globally consider the mechanical state within the simulation model; it merely indirectly reduces the need for adjustments for local elements with high local stresses through appropriate local weighting factors. However, even local elements with high local stresses in the simulation model are automatically included in the best-fit, albeit with lower weights for the fit.

[0007] The object of the present invention is therefore to overcome the disadvantages of the prior art and to provide an improved method for generating a simulation model. In particular, a suitable device, a suitable arrangement for using the simulation model, a computer program for executing the method, and a computer-readable storage medium containing instructions for executing the method are also to be provided.

[0008] To solve the aforementioned problem, a method, particularly computer-implemented, for generating a modified simulation model is now proposed, which comprises the following steps:

[0009] A) Receiving a surface data set of an actually manufactured object, wherein the actually manufactured object was manufactured based on a target shape, and wherein the surface data set includes data based on a measurement of at least a part of the surface of the actually manufactured object; B) Receiving a simulation model, wherein an initial shape state of the simulation model represents the target shape of the object, and wherein the initial shape state is initially set as the current shape state of the simulation model, and wherein the simulation model is designed to simulate mechanical stresses within the object that occur during a presumed deformation of the object.

[0010] C) Determining a further shape state of the simulation model in which the simulation model exhibits a minimum mechanical strain energy among many or all possible shape states in which, starting from the current shape state of the simulation model, at least a local area of ​​the simulation model approximates the surface data set, D) Adopting the further shape state of the simulation model determined in step C) as the new current shape state of the simulation model, E) Defining the modified simulation model with the new current shape state from step D) as the modified simulation model of the actually manufactured object.

[0011] Furthermore, to solve the aforementioned problem, a device for generating a modified simulation model is proposed, wherein the device is designed in particular for carrying out a method according to one of the embodiments of the method according to the invention described below and wherein the device comprises:

[0012] - an interface for receiving a surface data set of an actually manufactured object, wherein the actually manufactured object was manufactured based on a target shape, and wherein the surface data set includes data based on a measurement of at least a part of the surface of the actually manufactured object,

[0013] - an interface for receiving a simulation model, wherein an initial shape state of the simulation model represents the desired shape of the object, and wherein the initial shape state is initially defined as the current shape state of the simulation model, wherein the simulation model is designed to simulate mechanical stresses within the object that occur during an assumed deformation of the object,

[0014] a modification device designed to modify the simulation model, taking into account the surface data set, into a modified simulation model, and performing the following steps: • Determining a shape state of the simulation model in which the simulation model exhibits a minimum mechanical strain energy among many or all possible shape states where, starting from the current shape state of the simulation model, at least a local area of ​​the simulation model approximates the surface data set; and

[0015] • Adopting the simulation model in the specific form state of the minimum mechanical strain energy as the modified simulation model,

[0016] a determination device designed to determine, after completion of the modification of the simulation model, the then existing modified simulation model as the modified simulation model of the actually manufactured object.

[0017] Further embodiments of the device result from the corresponding embodiments of the method set out below.

[0018] Within the scope of the present invention, a surface dataset is understood to be a dataset of surface data of an actually manufactured object, processed from measurement data. This dataset is thus based on a measurement of at least a portion of the surface of an actually manufactured object and, based on its surface data, represents a mathematical representation of the actual physical surface coordinates of at least a portion of the measured surface of the actually manufactured object. This can, for example, be a triangular mesh of mathematical points that are fitted into the point clouds of the manufactured object, acquired by an optical sensor, using a best-fit technique. Furthermore, it can also be the mathematical description of substitute elements that are fitted into the corresponding point clouds using a best-fit technique.The fitting of corresponding surface, shape, or geometric elements, such as circles, slots, cylinders, etc., into measured point clouds is well known. Furthermore, hybrid methods for the mathematical representation of surface data from triangular meshes and substitute elements for the mathematical representation of the measured physical surface as a surface dataset are also conceivable. Therefore, within the scope of this application, a surface dataset is to be understood as any possibility of the mathematical representation of measured physical surface coordinates. This definition of a surface dataset is not limited to the aforementioned representational methods but is intended to encompass all possibilities by which measured physical surface coordinates can be mathematically made available for further data processing in a computer or network.

[0019] The measurement data for the actual manufactured object can include not only sensor data for capturing the object's surfaces, but also volume data from a CT scanner, from which the surface data of the object's interfaces within and at its boundaries can be calculated. Furthermore, all optical or tactile sensors known in coordinate measuring technology for capturing surfaces and contours can be used to measure the actual manufactured object, regardless of whether these sensors are mounted on a coordinate measuring machine, a robot, or individually in a production line.

[0020] Within the scope of the present invention, a simulation model is understood to be a model for modeling an object that can not only adequately describe the object's surface data based on its external shape state as a mathematical representation of the surface data of the simulation model, but can also adequately simulate the mechanical properties within the object. Therefore, a simulation model, metaphorically speaking, consists of an external part for describing the shape state and an internal part for describing the mechanical properties within the object. Typically, such simulation models are therefore volume models for describing an object. A finite element method (FEM) model serves as an example.

[0021] Within the scope of the present invention, a form state of the simulation model comprises a mathematical representation for at least the surface coordinates of the simulation model. Therefore, the form state describes at least the external form state of the simulation model. Specifically, the form state, as a mathematical representation, describes the surface data of the simulation model in its respective state. The form state, like the surface data set or the desired form, can be constructed from various mathematical representations. Therefore, the definition of the term "form state" is not limited to the aforementioned representational possibilities, but rather encompasses all methods by which the surface coordinates of a simulation model can be mathematically provided for further data processing in a computer or network.Within the scope of the present invention, the initial form state of the simulation model represents the desired form and can therefore, for example, correspond to the desired form, be derived from it, or approximate it. The form state determined by the method according to the invention, compared to the initial form state, already originates from a modified simulation model that has already completed at least one step of the approximation process to the surface data set according to the method according to the invention.

[0022] Within the scope of the present invention, a target form can be understood, for example, as a target form derived from CAD data of the actually manufactured object, which, as a mathematical representation, describes the desired target surface data of the actually manufactured object. The target form, like the surface data set, can be constructed from various possibilities of mathematical representation. Therefore, the definition of the term target form is not limited to the aforementioned possibilities of representation, but is intended to encompass all possibilities by which the desired target surface coordinates of an actually manufactured object can be mathematically provided for further data processing in a computer or a network.

[0023] In the context of the present invention, a local area of ​​the simulation model is understood to be a spatially contiguous but limited area of ​​surface, volume, and / or elements within the simulation model. This local area may also contain holes, recesses, or non-elements. CAD patches of the object's CAD model, which can be automatically defined by a test plan, a historical lock-up table, or user selection, can also qualify as such a local area. Depending on the structure of the simulation model, elements of the local area, as well as elements of sub-areas, can include local surface elements, local triangular meshes of nodes, local volume elements, or surface or volume geometry elements used as mathematical substitutes for holes, slots, protrusions, or recesses.According to the invention, it was first recognized that, in order to obtain a realistic simulation model of an actually manufactured object, not all areas of the simulation model may generally be included when approximating the surface dataset. Areas whose shape suggests that their inclusion in the approximation would lead to inaccurate simulation models, or areas for which corresponding empirical data already exists, should be disregarded when generating a realistic simulation model or a modified simulation model of an actually manufactured object. Therefore, in the present invention, the approximation of the simulation model to the surface dataset is performed only for at least a local area of ​​the simulation model.

[0024] Furthermore, it was recognized according to the invention that considering the global strain energy of the simulation model is useful in order to decide which shape state of the simulation model, and thus which modified simulation model, most realistically represents the actually manufactured object. The simulation model with the lowest strain energy among the many or all possible simulation models whose shape states allow an approximation of the surface data set has proven in practice to be the most suitable simulation model of an actually manufactured object for subsequent uses. In such subsequent uses of simulation models, the internal stress states of the simulation model are generally removed before its use, so that the simulation model can be used for the intended application or the specific simulation of the object in a stress-free state.The invention involves simulation. Therefore, for subsequent uses or simulations, it is helpful or advantageous if the simulation model generated according to the invention, i.e., the modified simulation model, is already in a state of minimum strain energy. Consequently, in the present invention, the global strain energy of the simulation model is always considered across all areas of the simulation model in order to determine which simulation model, as the modified simulation model, is most suitable for representing an actually manufactured object. Thus, while the approximation to the surface data set according to the invention only occurs in at least one local area of ​​the simulation model, the strain energy of the simulation model is considered globally across all areas of the simulation model. According to one embodiment of the method according to the invention,In the device according to the invention, in step C) the approximation of at least one local area of ​​the simulation model is carried out by:.

[0025] Subdividing at least one local area of ​​the simulation model into sub-areas of the simulation model or adopting a corresponding, existing subdivision;

[0026] For each sub-area of ​​the simulation model, determine according to a predefined assignment rule whether an assignable sub-area of ​​the surface data set exists, and if so, assign the existing assignable sub-area of ​​the surface data set to the sub-area of ​​the simulation model;

[0027] from the sub-areas of the simulation model for which an associated sub-area of ​​the surface dataset exists, determine active sub-areas of the simulation model that meet at least one first predefined criterion;

[0028] The beginning of a process of relocating or shifting the active sub-areas of the simulation model, taking into account a constraint that prescribes either an active approach to the assigned sub-area of ​​the surface data set or at least a partial superimposition of the assigned sub-area of ​​the surface data set, but in each case offers or enables at least one degree of freedom for relocating or shifting the active sub-area of ​​the simulation model.

[0029] The predefined assignment rule could, for example, consist of assigning each sub-region of the simulation model to the nearest sub-region (e.g., of the same size) of the surface dataset (e.g., according to the Euclidean distance). Alternatively, the sub-region of the surface dataset that is intersected by a normal vector to the surface of the simulation model sub-region (or by a straight line obtained by extending the normal vector) could be assigned.

[0030] By restricting the active approximation to the active sub-regions, problematic sub-regions can be excluded from the active approximation. These are areas where, for example, the target shape already exhibits significant surface curvature and / or where, for example, significant surface curvature would result from the adaptation to the surface dataset. If a change to the simulation model were actively mandated in such problematic areas, this would often lead to very high local stresses, which would therefore also contribute significantly to the overall mechanical strain energy. This would be detrimental to finding a simulation model that generally approximates the surface dataset well.

[0031] The first predefined criterion used to determine the active sub-areas of the simulation model can, in particular, consider one or more of the following sub-criteria. If multiple sub-criteria are considered, then all of these sub-criteria must be fulfilled within the scope of the first predefined criterion:

[0032] ■ The surface normal of the sub-area of ​​the simulation model does not deviate too much (for example, not more than a specified limit for the magnitude of the angle of deviation) from the surface normal of the associated sub-area of ​​the surface data set.

[0033] ■ The surface curvature (for example, represented by the radius of curvature) of the sub-area of ​​the simulation model does not deviate too much (for example, not more than a specified limit for the amount of the difference in curvature, for example, the maximum curvature in the respective sub-area) from the surface curvature of the associated sub-area of ​​the surface data set.

[0034] ■ The (for example, Euclidean or directed) distance of the sub-area of ​​the simulation model to the associated sub-area of ​​the surface dataset is not too large (for example, not larger than a predefined limit for the magnitude of the distance).

[0035] In particular, the process of relocating or shifting the active sub-areas can be initiated by determining a relocation or shifting area for each active sub-area, taking into account its applicable constraint. This area allows for the relocation or shifting of the active sub-area into a plurality and, in particular, an infinite number of different positions and / or orientations of the relocated or shifted active sub-area. Specifically, a constraint can be defined for various (for example, infinitesimally small) active sub-areas of the simulation model, each requiring the active sub-area to approach the surface dataset or at least partially overlap an associated sub-area of ​​the surface dataset, while still allowing at least one degree of freedom for the relocation or shifting.Shifting the active sub-area of ​​the simulation model offers or enables this. Furthermore, different constraints for approximation can be defined for different active sub-areas, especially when the active sub-areas consist of different surface, volume, node, or substitute elements.

[0036] The degrees of freedom for relocation or displacement can be restricted, for example, by constraints as follows: The sub-areas of the simulation model remain movable along the relocation or displacement area according to their respective constraints. This movable area extends perpendicular to a connecting line between the unmoved or undisplaced sub-area of ​​the simulation model and an associated sub-area of ​​the surface dataset. This area can be a planar surface or, for example, a curved surface that extends at a constant distance from the unmoved or undisplaced sub-area, or whose shape is determined from the simulation model and / or the surface dataset.

[0037] Of the various possible form states, all of which satisfy the displacement or shift ranges specified by the constraints for the active sub-areas of the simulation model, a state must now be determined in which the simulation model exhibits the lowest strain energy and is thus adopted as the modified simulation model. This can optionally be an intermediate result, i.e., the intermediate result forms the basis for at least one further step of the modification, in particular by repeating previously executed steps on the new basis. It should be noted that the approximation of the simulation model according to the invention generally does not lead to a complete alignment with the surface data set. Rather, the alignment can only occur in at least one local area, and within that local area, only in individual sub-areas.In a further embodiment of the inventive method or device, after determining the form state of minimum mechanical strain energy for the relocated active sub-areas of the simulation model, it is checked whether a second predetermined criterion is met, wherein if the second criterion is not met, the affected active sub-area is deactivated and the process of relocating or shifting the active sub-areas of the simulation model under step C) is carried out again without taking the constraint into account for the deactivated sub-area or sub-areas.

[0038] The second predefined criterion, the non-fulfillment of which deactivates a sub-area and renders it inactive, can consider one or more of the following sub-criteria. If several of the following sub-criteria are considered, then fulfilling just one criterion is sufficient to deactivate the affected sub-area:

[0039] ■ The local mechanical stress or the mechanical strain energy in the sub-area of ​​the simulation model is not too large (for example, not larger than a specified limit value of stress or strain energy or area-specific or volume-specific stress or strain energy) when the local area of ​​the simulation model is approximated to the surface data set.

[0040] ■ The surface normal of the active sub-area of ​​the simulation model before the relocation or displacement does not deviate too much (for example, not more than a specified limit value for the magnitude of the angle of deviation) from the surface normal of the relocated or displaced active sub-area.

[0041] Even the target shape represented by the initial form state of the simulation model can deviate significantly from the surface data set.

[0042] In particular, deviations from the target shape can occur during the manufacturing of the actual object, which are not suitable for meaningful adjustments to the target shape. For example, a curved area of ​​the actual object may differ significantly in its radius of curvature from the radius of curvature in the corresponding area of ​​the target shape. Furthermore, a highly curved area of ​​the actual object may exist where the corresponding area of ​​the target shape is not curved or only slightly curved, or vice versa. Additionally, it is possible that a recess or protrusion exists only in one corresponding area of ​​the target shape and the surface data set.In the case of the embodiments described above with active sub-areas, it is therefore possible, in particular by means of the second specified criterion, to exclude the possibility that such local areas contain active sub-areas.

[0043] In one embodiment of the inventive method or device, steps C) and D) are repeated so that the shape state of the simulation model is gradually approximated to the surface data set, starting from the desired shape of the object, at least for the at least one local area of ​​the simulation model as a whole.

[0044] As mentioned above, the simulation model can be modified iteratively in several successive steps. The step of modifying the simulation model can therefore be repeated at least once, with the modified simulation model obtained from the previous step being used as the model to be modified in the current step. This allows the simulation model to be gradually approximated to the surface data set, starting from the desired shape of the object. The device's modification mechanism can be designed accordingly. This incremental modification enables a stable process with a good approximation to the surface data set. Otherwise, the approximation could, in some cases, lead to unrealistic results in certain areas.

[0045] In a further embodiment of the inventive method or device, sub-areas of the surface data set are determined based on a global analysis of the surface data set, which are excluded for assignment to the sub-areas of the simulation model.

[0046] This allows, in particular, the exclusion of areas of the surface dataset whose data quality is insufficient for relocating active sub-areas due to various reasons. For example, the corresponding point cloud density in the original measurement data may not have been sufficient to create a sufficiently dense triangular network of nodes for the surface dataset, or the original measurement data may exhibit excessive fluctuation or nonsensical measurements in the relevant areas.

[0047] In one embodiment of the inventive method or device, the simulation model is a finite element method (FEM) model. Such a model has the advantage that it can now be derived very easily and automatically from a CAD model of the object, and that the mechanical properties of the object are implemented in the model from the outset.

[0048] In a further embodiment of the method according to the invention or in an arrangement with a device according to the invention, the modified simulation model is used in a further step in:

[0049] a simulation of a deformation of the object by forces acting on the object from the outside;

[0050] a simulation of the object in a mechanically stressed state; a determination of a deformation of the object by simulating forces acting on the object from the outside, saving a determination result and using the determination result for at least one other actually manufactured object of the same type; a creation of at least one further simulation model with a further shape state for the same object, so that the different simulation models describe the object with two different shape states, and determination of differences in the shape of the object represented by the at least two simulation models;

[0051] Validation of a process of manufacturing the object and / or determining a deformation field between the modified simulation model and another simulation model of the same or an identical actually manufactured object.

[0052] In these applications of the modified simulation model, for example, the influence of gravity on the desired installation position of the object is simulated as forces acting on the object from the outside. Furthermore, the installation of a door or fender into a car body can be simulated by simulating the object in a mechanically stressed state. In addition, such a simulation can also be applied to identical doors or fenders to detect deformation. Therefore, the manufacturing process of the object can also be validated or optimized using the simulation model.Furthermore, deformation fields can also be determined as the difference between two simulation models, wherein one simulation model was obtained according to the invention as a modified simulation model of an actually manufactured component and the other simulation model for the difference formation can be another simulation model of the same component, for example the simulation model whose initial shape state corresponds to the target shape of the component, or another simulation model of an identical component.

[0053] In a further embodiment of the method according to the invention, or in an arrangement with a device according to the invention, the method is carried out in a first step for a first actually manufactured object in a first state after the object has been manufactured, and in a further step the method is carried out for the first or a second actually manufactured object in a second state, wherein the second state has different externally acting forces compared to the first state. Subsequently, a deformation field is determined between the modified simulation models of the first and the further step.

[0054] In this use of the simulation model modified according to the invention, for example, the modified simulation model of an actually manufactured car door and the modified simulation model of this car door in an installation position with externally imposed constraint forces can be used to generate a deformation field for these two states by taking the difference between the two simulation models, which can then be transferred to the modified simulation model of another actually manufactured car door in order to simulate its subsequent installation situation.

[0055] The device or arrangement mentioned above according to the present invention can, in particular, be a computer or a computer arrangement. One of the embodiments of the arrangement described in this description can also be implemented, in particular, by a computer, a computer arrangement, or a networked computer arrangement. This means, in particular, that the various components of the device or arrangement can be implemented by a single computer or by an arrangement of computers. With regard to the measuring device or coordinate measuring machine mentioned below, this means that it can be additionally present and / or that a computer or computer arrangement of the measuring device can be part of the computer or computer arrangement.

[0056] The invention further relates to a computer program comprising instructions which, when the computer program is executed by a computer or by an arrangement of computers, cause the computer(s) to execute the method in one of its embodiments or to execute the computer-implemented part of the method in one of its embodiments.

[0057] Furthermore, the invention relates to a computer-readable storage medium comprising instructions which, when executed by a computer or by an arrangement of computers, cause the computer(s) to execute the method in one of its embodiments or to execute the computer-implemented part of the method in one of its embodiments.

[0058] The device, arrangement, or computer or computer arrangement mentioned above may, for example, consist of or include a single computer or a computer network. The computer, or at least one of the computers, may, in particular, be an analog computer, a digital computer, and / or a hybrid computer with regard to its mode of operation.

[0059] In terms of its size and design, it can be a smartphone, a personal digital assistant (PDA), a tablet computer, an embedded system (e.g., embedded in the control computer of a coordinate measuring machine), a single- or multi-board computer, a personal computer (PC), a desktop computer, a workstation, a host computer or server integrated into a computer network, a thin client computer, a netbook, a notebook, a laptop, a mainframe computer, or a supercomputer, with some of the aforementioned types also being implemented by a single computer, such as a PC with multiple circuit boards. Furthermore, the computer, or at least one of the computers, can have one or more central processing units (CPUs) and / or one or more processing cores per CPU.Graphics cards or other dedicated cards with processing units that are part of a computer can, alone or in combination with other computers or processing units, constitute the means for executing the procedure. The computer network or networks mentioned may be a local or non-local network, or a combination thereof. In particular, a local network may be a Body Area Network (BAN), a Wireless Body Area Network (WBAN), a Personal Area Network (PAN), a Wireless Personal Area Network (WPAN), a Local Area Network (LAN), or a Wireless LAN (WLAN). A non-local network may, in particular, be a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a Global Area Network (GAN), a Virtual Private Network (VPN), or a Storage Area Network (SAN).

[0060] Furthermore, it should be noted that although the computer or computers are preferably prompted to execute the method by a computer program, the means for executing the method may at least also include a hardware-implemented, preferably programmable arrangement (for example, an arrangement of logic gates), such as an ASIC (Application-Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0061] The computer-readable storage medium is, for example, a digital medium such as a Compact Disc (CD), a floppy disk, a Digital Versatile Disc (DVD), a hard drive, a memory card or a mass storage device, or it can also be an analog computer-readable medium such as text (which is, for example, a printout of program code), an image, an arrangement of images or an analog flat or disc-shaped data carrier.

[0062] The term coordinate measuring machine (CMM) encompasses all types of devices used to determine the coordinates of workpieces. One class of CMMs determines surface coordinates, meaning they measure the coordinates of points on the workpiece's surface. Another class, either alternatively or additionally, can determine internal coordinates within workpieces. These include CMMs that utilize invasive radiation penetrating the workpiece material, specifically measuring the intensity of the radiation passing through the workpiece. Typically, the workpiece is irradiated from various directions, and the results are used to perform a computer-aided reconstruction of the scanned workpiece. Such methods are also known as computed tomography (CT).

[0063] The term coordinate measuring machine (CMM) encompasses, in particular, devices with an optical sensor or sensor system. This will be discussed in more detail below. Specifically, the term also includes classic coordinate measuring machines, such as portal-type or gantry-type machines, and CMMs with a movable bridge, on which, in particular, a quill with a sensor movable relative to the bridge is mounted, as well as articulated arm machines and machines with a hexapod mechanism. Furthermore, it includes devices in which at least one sensor is fixed with respect to at least one degree of freedom of the relative movement between the sensor and the workpiece, and in which the workpiece to be measured is movable relative to the at least one sensor, as, for example, in a machine with a movable measuring table.The term coordinate measuring machine also encompasses machines that, while not primarily designed as coordinate measuring machines, are configured to operate like one. In particular, these machines have at least one measuring sensor used to determine the coordinates. Examples include robots, such as robot arms with rotary joints, to which a sensor for detecting the workpiece surface (e.g., a fringe projection sensor) is attached instead of or in addition to a tool, or machine tools to which a measuring sensor (e.g., a tactile sensor) is attached instead of or in addition to a machining tool. Hexapod mechanisms are also known, to which a sensor for detecting the workpiece surface (e.g., a tactile sensor) is attached instead of a machining tool.

[0064] Person-operated (e.g., handheld) sensors for detecting a workpiece are also known, where at least one sensor is optionally mounted on a movable mechanism. In this case, there is no motor drive to generate the sensor's movement. Instead, the movement is effected by a person. Both scanning and scanning of the object are possible, with the sensor being positioned at a fixed angle. The triangulation principle is frequently used in person-operated sensors: for example, a known pattern is projected onto the object, and at least one image of the object's surface is captured from a different angle. Laser radiation can also be used instead of a pattern projection when applying the triangulation principle.In particular, the location where the reflected laser radiation is received contains information about the surface coordinates of the object being measured. However, devices with a person-operated sensor are also known, which have at least one motor to assist in the movement and / or positioning of the sensor.

[0065] The invention is not limited with regard to the types of sensors used by a coordinate measuring machine to determine the coordinates. Tactile sensors have already been mentioned as an example; these can be, for instance, of the switching or measuring type. In particular, the tactile sensors can be passive or active. Active sensors can be configured to generate a probing force with which a tactile probe contacts the surface of a workpiece to be measured. Optical sensors are frequently used alternatively or additionally, e.g., a projection sensor, in particular a fringe projection sensor, a laser triangulation sensor, a line scan camera, a camera for capturing two-dimensional images, a camera for capturing three-dimensional images, an arrangement with at least two cameras, or a confocal chromatic sensor.Furthermore, there are, for example, capacitive sensors and inductive sensors.

[0066] The term coordinate measuring machine therefore also includes 3D scanners. As mentioned above, but not limited to handheld sensors, triangulation-based systems such as laser scanners and projection sensors are particularly advantageous because they can generate many measurement points with their 3D coordinates in a short time. Projection sensors project a planar pattern, e.g., a striped pattern, onto the object being measured and capture images of the object along with the projected pattern using at least one image acquisition unit (especially a camera). In a specific configuration, two image acquisition units may be present, capturing the object from different viewing angles. By evaluating the image captures, the 3D coordinates of surface points on the object can be determined. To measure an object, or...To fully capture the surface of the object being measured, a single measurement position is often insufficient, so the relative position of the 3D scanner to the object is usually changed multiple times. This positioning can be done manually (handheld), or semi-automatically or fully automatically, for example, by guiding the 3D scanner with a robot. The position can also be changed by moving the object relative to the 3D scanner, for example, using a rotary table.

[0067] Exemplary embodiments of the invention will now be described with reference to the accompanying drawing. The individual figures in the drawing show:

[0068] Fig. 1 shows an arrangement with a measuring device, a measured object and a data processing device,

[0069] Fig. 2 shows a cross-section through the target shape of an object, in particular the object from Fig. 1, wherein the shape is defined by target data; Fig. 3 shows a cross-section through the upper surface of the object from Fig. 2, wherein, in addition to the target shape of the object and an initial shape state of the simulation model approximating the target shape, the surface data set of an actually manufactured object is also shown, which was manufactured according to the target shape, but where deviations have occurred, which are exaggerated in Fig. 3.

[0070] Fig. 4 shows a representation as in Fig. 3 with the surface data set and a shape state of the simulation model approximating the surface data set, which can be obtained by adapting an initial shape state of the simulation model under modification of the simulation model according to the invention.

[0071] Fig. 5 shows a sub-area of ​​an initial shape state of the simulation model, for example the initial shape state from Fig. 3, and a sub-area of ​​a surface data set, for example from Fig. 3 and Fig. 4, also for the purpose of illustrating an assignment between corresponding sub-areas.

[0072] Fig. 6 shows a representation similar to Fig. 5, but with a different way of assigning corresponding sub-areas.

[0073] Fig. 7, similar to Fig. 5 and Fig. 6, shows a sub-area of ​​the initial shape state and a sub-area of ​​a surface data set, schematically depicting a plurality of mutually associated sub-areas; Fig. 8 illustrates part of a first step in the approximation of the shape state of the simulation model to the surface data set; Fig. 9 shows nodes of a simulation model to illustrate an intrinsic coupling of the simulation model between neighboring nodes.

[0074] Fig. 10 is a flowchart illustrating the steps of an embodiment of the method according to the invention and

[0075] Fig. 11. schematically shows a data processing device with a provisioning device, a modification device and a specification device.

[0076] Fig. 1 shows a sketch of a possible device 1 for the three-dimensional optical measurement of objects (e.g., the illustrated object 2 with a recess 15) with a sensor 3 suitable for acquiring surface data of the object 2. The sensor 3 comprises a projection unit 4 and at least one image acquisition unit 5 and can be guided by a robot, a coordinate measuring machine, or manually. The projection unit 4 has a light source 6, a pattern generator 7, and projection optics 8. The sensor 3 is connected to an evaluation unit 9. The evaluation unit 9 analyzes the measurement images acquired by the at least one image acquisition unit 5 and provides a surface data set representing the 3D component geometry of the object 2.In an embodiment described below with reference to further figures, the surface data set represents, for example, only the actual physical surface data on one side of the measured object 2, such as the surface oriented towards the upper right in Fig. 1, since, for example, corresponding measurement images were only taken there. The evaluation unit 9 is preferably suitable for performing the necessary data processing steps with a computer program. The evaluation unit 9 is connected to a data processing device 10 via a data interface 11. The data interface 11 is designed to transmit data, such as the surface data set from the evaluation unit 9 and / or data such as the target shape of the object 2 and / or data from a virtual clamping device, to the data processing device 10.In a preferred embodiment, the data processing unit 10 is configured to perform a method for generating a simulation model of the measured object 2 according to one of the embodiments of the method described in this description. Furthermore, a display unit 12 is provided, which is connected to the data processing unit 10 and displays to the user, among other things, the surface data set or the generated simulation model, as well as optionally further information such as the results of a target / actual comparison. The data processing unit 10 is preferably suitable for performing the necessary data processing steps with a computer program. In a particularly preferred embodiment, the evaluation unit 9 and the data processing unit 10 are combined in a single data processing unit.

[0077] Fig. 2 sketches a desired shape 14 of an object, in particular object 2 from Fig. 1. As already indicated in Fig. 1, the object has a recess 15. The desired shape 14 can, for example, be represented by or derived from CAD data.

[0078] Figure 3 shows a section through the surface data set 22 and through the target shape 14 of object 2 from Figure 2, focusing on the side or surface shown at the top of Figure 2. The deviations between the surface data set 22 and the target shape 14, also indicated by arrows, are exaggerated compared to actual practice.

[0079] The desired shape 14 from Fig. 2 can initially be modeled by an initial shape state 23 of the simulation model. Here, the initial shape state 23 of the simulation model represents the desired shape 14 and is therefore, for example, derived from the desired shape 14, approximated to the desired shape 14, or even corresponds to the mathematical or software-based description of the desired shape 14, or is a copy of this mathematical / software-based description of the desired shape 14. According to the invention, such an initial shape state 23 of the simulation model is subsequently approximated to the surface data 22, which is based on a measurement of at least a part of the surface of the actually manufactured object 2; that is, the simulation model involved is thereby modified or altered.This method according to the invention yields a simulation model that is more suitable for practical application, since – as will be explained in more detail below – the areas of the surface data set 22 with large deviations, large fluctuations or measurement artifacts, see the area of ​​the surface data set 22 marked by the dashed circle in Figure 3, are not actively taken into account when changing or modifying the simulation model.

[0080] Fig. 4 schematically shows, for the case already depicted in Fig. 3, the result of approximating a shape state 24 of the simulation model to the surface data set 22. The modified shape state of the simulation model, which results from the initial shape state 23 of the simulation model through the approximation to the surface data set, is now labeled with reference numeral 24 in Fig. 4. It can be seen that no agreement was achieved in the area outlined by a dashed circle in Fig. 3. Very good agreement was achieved in the central area of ​​the representation. Good agreement was achieved in the left area of ​​the representation, although there are somewhat larger deviations in the shapes in the sub-region of the greatest curvature of the surface. For better visibility, the deviations have again been exaggerated.

[0081] In Fig. 4, three local sub-areas Us1, Us2, and Us3 of the simulation model are shown for a local area B1 of the simulation model. Each sub-area extends between two vertically oriented dashed lines and is marked by a double arrow. Within the scope of this application, sub-areas of the simulation model are designated by the singular and plural Us, unless specific sub-areas Us1, Us2, or Us3 are explicitly referenced in the figures or text. The same applies to the sub-areas Uo of the surface data set. The active approximation of the shape state of the simulation model to the surface data set 22 was performed in these three local sub-areas Us1, Us2, and Us3 of the simulation model shown in Fig. 4.However, the approximation did not only take place in the local sub-areas Us1, Us2 and Us3, because the mechanical properties of the object taken into account in the simulation model and the requirement of the lowest strain energy for the form state 24 lead to the areas outside the sub-areas Us1, Us2 and Us3 as well as the areas outside the local area B1 being passively traced or approximated by the simulation model and thus an overall approximation of the form state 24 of the simulation model to the surface data set 22 has taken place.

[0082] The procedure outlined in Fig. 4 uses the inventive method for generating a modified simulation model, which comprises the following steps:

[0083] A) Receiving a surface data set 22 of an actually manufactured object 2, wherein the actually manufactured object was manufactured based on a target shape 14, and wherein the surface data set 22 includes data based on a measurement of at least a part of the surface of the actually manufactured object 2,

[0084] B) Receiving a simulation model, wherein an initial shape state 23 of the simulation model represents the desired shape 14 of the object 2, and wherein the initial shape state 23 is initially defined as the current shape state of the simulation model, wherein the simulation model is designed to simulate mechanical stresses within the object 2 that occur during an assumed deformation of the object 2.

[0085] C) Determining a further shape state 24 of the simulation model in which the simulation model exhibits a minimum mechanical strain energy among many or all possible shape states in which, starting from the current shape state of the simulation model, at least one local area B1 of the simulation model approximates the surface data set 22,

[0086] D) Adopting the further shape state of the simulation model determined in step C) as the new current shape state of the simulation model, E) Defining the modified simulation model with the new current shape state from step D) as the modified simulation model of the actually manufactured object 2.

[0087] Thus, the form state 24 shown in Fig. 4 exhibits the lowest mechanical strain energy compared to all possible form states of the simulation model in which an approximation of the depicted local region B1 of the simulation model to the surface data set 22 is possible. In the method according to the invention, a local approximation of at least one local region B1 of a form state 24 of the simulation model to the surface data set 22 is therefore carried out, but under a global consideration of the mechanical strain energy of the form state 24 of the simulation model as a whole, which is not limited to the local region B1, but includes all regions of the simulation model.

[0088] An approximation of at least one local area B1 of the simulation model in step C) of the method according to the invention can now be carried out in an embodiment of the method according to the invention by the following further steps:

[0089] Subdividing at least one local area B1 of the simulation model into sub-areas of the simulation model Us or adopting a corresponding, existing subdivision;

[0090] For each sub-area of ​​the simulation model Us, determine according to a predefined assignment rule whether an assignable sub-area of ​​the surface data set Uo exists, and if so, assign the existing assignable sub-area of ​​the surface data set Uo to the sub-area of ​​the simulation model Us;

[0091] from the sub-areas of the simulation model Us for which an associated sub-area of ​​the surface data set Uo exists, determine active sub-areas Us1, Us2, Us3 of the simulation model that satisfy at least one first predefined criterion;

[0092] Initiation of a process of relocating or shifting the active sub-areas of the simulation model Us1, Us2, Us3, taking into account a constraint that prescribes either an active approach to the assigned sub-area of ​​the surface data set or at least a partial superimposition of the assigned sub-area of ​​the surface data set, but in each case offers or enables at least one degree of freedom for relocating or shifting the active sub-area of ​​the simulation model Us1, Us2, Us3.

[0093] Sub-areas Us of the simulation model can include individual nodes, multiple nodes, surface elements or even geometric elements, depending on which components the simulation model is locally constructed from in the affected sub-areas Us.

[0094] Such an approximation under step C) of the method according to the invention means, with reference to Fig. 4, that an active approximation to the surface data set 22 takes place only for the active sub-areas Us1, Us2, and Us3 of the local area B1 of the simulation model, since it was only possible to assign sub-areas of the surface data set to these sub-areas Us1, Us2, and Us3, and these sub-areas Us1, Us2, and Us3 satisfy at least a first predefined criterion. The other sub-areas Us of the simulation model, on the other hand, remain inactive or passive during the active approximation and are merely pulled along by the intrinsic couplings within the simulation model during the active approximation of the active sub-areas Us1, Us2, and Us3 by the requirement of the lowest strain energy and are thus "passively" relocated or shifted.

[0095] This means, for example, that other or inactive sub-areas Us of the simulation model, where the associated sub-areas Uo of the surface data set 22 exhibit excessively large deviations, fluctuations, or even measurement artifacts (see the area of ​​the surface data set 22 marked by the dashed circle in Figure 3), can be disregarded when the simulation model is modified or altered. With regard to Figure 4, this means that in the active sub-areas Us1, Us2, and Us3, an active approximation of the form state 24 of the simulation model to the surface data set 22 has taken place, which is also reflected in the good agreement between the form state 24 and the surface data set 22 in the active sub-areas Us1, Us2, and Us3 of Figure 4.Figure 4 shows that in the other inactive sub-areas Us of the local area B1 only a passive approximation to the surface data set 22 has taken place, since the requirement of the lowest strain energy for the form state 24 also forces the other or other inactive sub-areas Us of the simulation model to make a meaningful passive approximation to the surface data set 22 due to the mechanical properties of the object 2 stored in the simulation model, without having to imitate or replicate the deviations, fluctuations or measurement artifacts present there in the surface data set 22.

[0096] The first predefined criterion used to determine the active sub-areas of the simulation model can, in particular, consider one or more of the following sub-criteria. If multiple sub-criteria are considered, then all of these sub-criteria must be fulfilled within the scope of the first predefined criterion:

[0097] ■ The surface normal of the sub-area Us of the simulation model does not deviate too much (for example, not more than a specified limit value for the magnitude of the angle of deviation) from the surface normal of the associated sub-area Uo of the surface data set 22.

[0098] ■ The surface curvature (for example, represented by the radius of curvature) of the sub-area Us of the simulation model does not deviate too much (for example, not more than a specified limit value for the amount of the difference in curvature, for example, the maximum curvature in the respective sub-area) from the surface curvature of the associated sub-area Uo of the surface data set 22.

[0099] ■ The (for example, Euclidean or directed) distance of the sub-area Us of the simulation model to the assigned sub-area Uo of the surface dataset 22 is not too large (for example, not larger than a predefined limit for the magnitude of the distance). The predefined assignment rule for the sub-areas Us and Uo can, for example, consist of assigning to each sub-area of ​​the simulation model Us the sub-area (for example, of the same size) of the surface dataset Uo that is closest to it (for example, according to the Euclidean distance).

[0100] Alternatively, the subregion of the surface dataset Uo can be assigned to the subregion that is intersected by a normal vector of a node K of the subregion of the simulation model Us (or by a straight line obtained by extending the normal vector). As mentioned, the subregions can be chosen to be infinitesimally small or at least very small, such that a normal vector is uniquely defined. Even for subregions that are not infinitesimally small, a normal vector can be uniquely defined, for example, by referring to the surface of the subregion at its surface centroid. When dividing the data into subregions and assigning them to subregions, it should be noted that the subregions under consideration do not necessarily have to form a continuous surface or volume.The process of changing the simulation model also works without continuous areas, since in the respective shape state of the simulation model a passive approximation of neighboring areas also takes place, namely due to the minimization of the mechanical strain energy over the entire simulation model.

[0101] The process of relocating or shifting the active sub-areas Us can be initiated, in particular, by determining a relocation or shift area for each active sub-area Us, taking into account the constraints applicable to it. This area allows the active sub-area Us to be relocated or shifted to approximate its assigned sub-area Uo of the surface data set 22. The relocation or shift area allows...

[0102] The displacement area is a plurality and, in particular, an infinite number of different positions and / or orientations of the relocated or displaced active sub-area Us, resulting in an infinite number of possible form states 24 in which at least the local area B1 of the simulation model approximates the surface data set 22. A unique position or orientation for the respective sub-area Us then results from minimizing the mechanical strain energy across the entire simulation model. A constraint can be understood, for example, as equations or inequalities for at least one coordinate of the sub-area Us of the simulation model, by means of which the approximation of the sub-area Us of the simulation model to the associated sub-area Uo of the surface data set 22 is achieved.

[0103] Figure 5 shows a sub-region of the initial form state 23 of the simulation model, for example, the initial form state 23 from Figure 3, and a sub-region of the surface data set 22, for example, the surface data set from Figures 3 and 4. The initial form state 23 and the surface data set 22 are separated by a distance, and the sub-region of the initial form state 23 belongs to a first local region of the simulation model, in which the approximation of the form states of the simulation model to the surface data set 22 takes place. For this purpose, sub-regions of the simulation model and the surface data set 22 are assigned to each other. In Figure 5, only one sub-region is shown for each of the surface data set 22 and the initial form state 23, namely a sub-region Us of the simulation model and a sub-region Uo of the surface data set 22.Starting from the sub-area Us, the associated sub-area Uo is found, for example, as the nearest sub-area of ​​the surface data set 22.

[0104] Figure 6 shows an alternative method of assignment. The assigned sub-area Uo of the surface dataset 22 is the area whose territory is intersected by a normal vector of a node K of the sub-area Us of the simulation model. Such a normal vector of a node K can be calculated by considering the neighboring nodes and the surface elements spanned by these neighboring nodes. In contrast to Figure 5, the normal vector of node K is therefore represented by a vertical arrow in Figure 6.

[0105] Figure 7, similar to Figures 5 and 6, shows a sub-area of ​​the initial form state 23 of the simulation model and a sub-area of ​​the surface data set 22, schematically depicting a plurality of mutually associated sub-areas of the simulation model and the surface data set 22. Each sub-area extends between two transverse dashed lines. For two of the sub-areas Us1 and Us2 of the initial form state 23 of the simulation model, an arrow indicates which sub-area of ​​the surface data set 22 is assigned. When determining the active sub-areas in which the approximation of the form states of the simulation model to the surface data set 22 is to be actively carried out, different results are obtained for sub-areas Us1 and Us2 of the simulation model. While sub-area Us1 is identified as the active sub-area, this is not the case for sub-area Us2.The reason is that the criterion for selecting a sub-area as the active sub-area of ​​these two sub-areas Us1 and Us2 is only met for sub-area Us1. In this embodiment, the criterion is that the mean surface curvature of the corresponding sub-areas does not differ too much. The mean surface curvature of the corresponding sub-area Uo2 is much greater than that of sub-area Us2, so sub-area Us2 is not a suitable active area for an active approximation.

[0106] Fig. 8 illustrates part of a first step in the approximation of shape states of the simulation model starting from the initial shape state 23 to the surface data set 22. For the active sub-area Us1, a dotted line running between it and the associated sub-area Uo1 is shown.

[0107] Simplified with respect to the representation in Fig. 8, the constraint for relocating or shifting this active sub-area Us1 stipulates that, for example, the surface center (e.g., surface centroid) of sub-area Us1 is relocated or shifted to this line, which is located at a distance of 80 percent of the distance to the associated sub-area Uo1. The dotted line extends along the surface data set 22 and is therefore generally curved. However, there are other ways to prescribe a stepwise approximation through the respective constraint. In particular, it could be a different percentage of the distance, or the approximation could depend on the shape of the two forms in the sub-area where the associated sub-areas are located.

[0108] Figures 3 to 8 only show the simplified two-dimensional case, as they are cross-sectional representations. In practice, the sub-areas are, for example, surfaces, triangular networks of nodes, individual nodes, or geometric elements, such that the constraint for a surface corresponding to the dotted line in Figure 8 applies as the target surface of the considered sub-areas. For example, the center point of a surface as a sub-area can therefore also be moved or shifted in directions perpendicular to the plane of Figure 8, as long as it is moved or shifted onto the surface indicated by the dotted line in Figure 8. Due to the corresponding degrees of freedom for moving or shifting the center point of the surface, there is an infinite number of modified form states of the simulation model. Additional degrees of freedom for moving or shifting further active sub-areas of the simulation model are also available.The displacement of their surface centers is a key factor. However, the possible shape states of the modified simulation model, which satisfy the constraint for all active sub-areas, are also defined by the fact that in each state the entire shape must conform to the modeling rules of the simulation model used. Since the simulation model simulates mechanical stresses, unrealistically large stresses do not occur between adjacent sub-areas after their relocation or displacement. This also applies to sub-areas or parts of the simulation model that are not active sub-areas. The implementation of such rules in simulation models is well-known in the field, so it will not be discussed further here.

[0109] The present invention is based on the idea of ​​identifying, from the possible states of the simulation model, the state exhibiting a minimum of mechanical strain energy. As mentioned, only states that comply with the rules of the modeling by the simulation model used are considered. In particular, an optimization algorithm known from numerical methods, such as a downhill simplex algorithm, can be applied.

[0110] After identifying the state with the minimum strain energy, the correspondingly modified simulation model can be adopted as an intermediate result and can form the starting point for repeating the steps of approximating the surface data set. Thus, steps C) and D) of the method according to the invention can be carried out repeatedly, so that the shape state 24 of the simulation model in Fig. 4 is gradually approximated to the surface data set 22, starting from the initial shape state 23 of the simulation model in Fig. 3, which approximates the target shape 14 of the object 2, or which represents this target shape 14 of the object 2, or which even corresponds to this target shape 14 of the object 2.

[0111] The geometric configuration of the form state of the simulation model and the surface data set, schematically depicted in Figures 3 to 8, is merely one of many possible embodiments. For example, sub-areas along edges of the two data sets can be assigned to each other, active sub-areas can be determined from these, and the approximation of the form states can be carried out in these active sub-areas. In this case, inactive sub-areas of the simulation model are located, for example, in an edge segment that exhibits excessive deviations in its orientation from the assigned edge segment of the surface data set.

[0112] In one embodiment of the method according to the invention, a further control step is performed for the determined form state 24 of minimum mechanical strain energy. For each of the relocated or displaced active sub-areas of the simulation model Us1, after determining the form state 24 of minimum mechanical strain energy, it is checked whether a second predefined criterion is met. If the second criterion is not met, the affected active sub-area is deactivated. Subsequently, the process of relocating or displacing the active sub-areas of the simulation model Us1 is repeated in step C), without considering the constraint for the deactivated sub-area or sub-areas. This can then result in a new form state 24 of minimum mechanical strain energy.

[0113] The second predefined criterion, the non-fulfillment of which deactivates a sub-area and renders it inactive, can consider one or more of the following sub-criteria. If multiple sub-criteria are considered as the second predefined criterion, then fulfilling just one sub-criterion is sufficient to deactivate the affected sub-area.

[0114] ■ The local mechanical stress or the mechanical strain energy in the sub-area Us of the simulation model is not too large (for example, not larger than a specified limit value of stress or strain energy or area-specific or volume-specific stress or strain energy) when the local area B1 of the simulation model is approximated to the surface data set 22.

[0115] ■ The surface normal of the active sub-area Us of the simulation model before the relocation or displacement does not deviate too much (for example, not more than a predefined limit for the magnitude of the angle of deviation) from the surface normal of the active sub-area Us of the simulation model to be relocated or displaced before the relocation or displacement. This control or verification step ensures that no modified simulation models with form states 24 are output as purported real models of the actually manufactured object 2 for further use, which are based on unrealistic stress states within the model or on unrealistic form states of the model.

[0116] In a further embodiment of the method according to the invention, sub-areas Uo of the surface data set 22 can be determined based on a global analysis of the surface data set 22, which are excluded from the assignment to the sub-areas Us of the simulation model. This is particularly useful if no, insufficient, nonsensical, unrealistic or unexpected surface data is available for the affected sub-areas Uo of the surface data set.

[0117] Figure 9 shows three selected nodes K1, K2, K3 of a simulation model. The initial form state 23 and the approximate form state 24 of the simulation model can be implemented using such nodes. The intrinsic couplings kp of the simulation model for modeling the mechanical connections between any two adjacent nodes are indicated by straight lines. Only the intrinsic coupling kp12 between nodes K1, K2, the intrinsic coupling kp13 between nodes K1, K3, and the intrinsic coupling kp23 between nodes K2, K3 are explicitly shown. However, further straight lines radiating from nodes K1, K2, K3 indicate that nodes K1, K2, K3 are intrinsically coupled or mechanically connected to other nodes not shown.

[0118] Each of the intrinsic couplings kp models a mechanical stress for the current node positions or for the node positions of the respective shape state of the simulation model, which can also assume the value zero. In particular, the value of the mechanical stress depends on the relative position of the two nodes coupled to each other via the intrinsic coupling kp. For example, for a specific zero-relative position of the nodes, the mechanical stress is equal to zero. Taking into account the spring constant D derived from the elasticity and / or stiffness of the material of the modeled object, the mechanical force F arising when AL deviates from the zero-relative position can be calculated in the embodiment described here as follows: F = D * ΔL (1)

[0119] The mechanical potential energy E can be calculated as follows in the case of the spring constant D:

[0120] E = D / 2 * (ΔL) 2(2)

[0121] The mechanical strain energy can be obtained by summing the contributions of all intrinsic couplings for at least one local area and, depending on the design and application, optionally also beyond. Since the invention aims to find a state of minimum strain energy, the strain energy does not need to be calculated with absolute precision. The calculated value of the strain energy can, for example, deviate from the correct value by a constant factor.

[0122] Numerous variations of this specific embodiment are possible. It has already been discussed that the force between two adjacent model elements is not proportional to the displacement from a rest position.

[0123] The zero position must be considered. In particular, quadratic or higher-order terms can be taken into account. Alternatively or additionally, the intrinsic couplings can be defined not only for two adjacent nodes, but for a group of adjacent nodes. Furthermore, alternatively or additionally, the intrinsic couplings can be related to adjacent finite elements. These are just a few examples of possible variations. Accordingly, the simulation model can also be based on a so-called spring model, a so-called spring-mass model, or on a planar or volumetric model according to the finite element method (FEM), the latter already including an intrinsic model of the mechanical properties of objects.

[0124] An embodiment of the method according to the invention will now be described with reference to Fig. 10. In a first step S1 or A) of the method according to the invention, a surface data set 22 of an actually manufactured object 2 is provided or received from a data storage device 30 or a network. The actually manufactured object was produced based on a target shape 14, and the surface data set 22 comprises data based on a measurement of at least a part of the surface of the actually manufactured object 2.

[0125] In a subsequent step S2 or B) of the method according to the invention, a simulation model is provided or received from a data storage device 30 or network, the initial form state 23 of which represents the desired form 14 of the object 2. Furthermore, this initial form state 23 is defined as the current form state of the simulation model. This simulation model is designed to describe or simulate mechanical stresses within the simulated object that occur during an assumed deformation of the object 2. This can be achieved, for example, by describing mechanical forces, such as compressive and tensile forces, or by describing at least an equivalent quantity, such as compressive or tensile stress.Furthermore, it is possible, for example, for the simulation model to describe the contribution to the mechanical strain energy resulting from the positions of the support points of the model elements for each adjacent model element (such as nodes).

[0126] In a subsequent step S4 or C) of the method according to the invention, a further shape state 24 of the simulation model is determined in which the simulation model exhibits a minimum mechanical strain energy among many or all possible shape states in which, starting from the current shape state, at least one local area B1 of the simulation model approaches the surface data set 22. In step S3, the specifications (for example, regarding the at least one local area B1 and the constraints) are read in, for example, by accessing the data memory 30. In particular, alternatively or additionally, user specifications can be taken into account, for example, regarding a change to the at least one local area B1. In this way, for example, areas with measurement gaps and / or areas with implausible measured values ​​can be taken into account.In particular, these areas can be excluded from local area B1. Returning to step S4, and taking into account the specifications of step S3 for at least one local area B1, the process of approximating the shape state of the simulation model to the surface data set 22, considering the simulated mechanical stresses, is initiated, and the shape state 24 with the lowest mechanical strain energy is determined. In the following step S5 or D) of the method according to the invention, the further shape state of the simulation model with the minimum mechanical strain energy, determined in step C), is adopted as the new current shape state of the simulation model.

[0127] Optionally, as indicated by dashed arrows in Fig. 10, steps S4 and S5 can be repeated, starting with the current form state 24 of the modified simulation model and determining at least one local area B1 accordingly in a modified form from the modified simulation model of the last current form state. Once the iterations have been completed, for example, because a predefined limit for the deviation of the new current form state 24 of the modified simulation model from the surface data set 22 with respect to all active sub-areas is undercut, the resulting modified simulation model with the last current form state can be output as the final or modified simulation model.

[0128] In the following step S6, the final, modified simulation model output in step S5 is used at:

[0129] - a simulation of a deformation of object 2 by forces acting on the object from the outside;

[0130] a simulation of object 2 in a mechanically stressed state; - a determination of a deformation of object 2 by simulating forces acting on the object from the outside, saving a determination result and using the determination result for at least one other actually manufactured object of the same type; a creation of at least one further simulation model for the same object 2 with a further shape state, so that the different simulation models describe the object with two different shape states, and determination of differences in the shape of the object represented by the at least two simulation models;

[0131] - Validation of a process for the manufacture of object 2 and / or the determination of a deformation field between the modified simulation model and another simulation model of the same or an identical actually manufactured object.Furthermore, the inventive method for generating a modified simulation model can also be used to determine a deformation field, wherein in a first step the inventive method is carried out for a first actually manufactured object in a first state after the object has been manufactured, and wherein in a further step the inventive method is carried out for the first or a second actually manufactured object in a second state, wherein the second state has different externally acting forces compared to the first state, and wherein a deformation field is determined between the modified simulation models of the first and the further step.

[0132] Fig. 11 shows a data processing device 10, for example the data processing device 10 from Fig. 1, with a provision or

[0133] The system comprises a receiving device 35, a modifying device 36, and a fixing device 37. The providing / receiving device 35 is configured to provide the surface data set 22 and the simulation model of object 2, derived, for example, from CAD and / or FEM data of object 2, if already available, or to receive the surface data set 22 and the simulation model of object 2, derived, for example, from CAD and / or FEM data of object 2, from the data storage device 30 or a network. The modifying device 36 generates the modified simulation model from this data, the current form state 24 of which is the lowest strain energy, and which has been approximated to the surface data set 22 of the actually manufactured object, at least for a local area (B1) of the simulation model.The fixing device 37 fixes the modified simulation model with the current form state of lowest strain energy as the modified simulation model based on surface data of an actually manufactured object. In particular, the aforementioned devices of the data processing device 10 can be implemented by a common computer program, whereby alternatively the provision or reception device 35 can optionally be implemented separately from the modification device 36 and the fixing device 37.

Claims

Patent claims:

1. A method for generating a modified simulation model, wherein the method comprises the following steps: A) Receiving a surface data set (22) of an actually manufactured object (2), wherein the actually manufactured object was manufactured based on a target shape (14), and wherein the surface data set (22) includes data based on a measurement of at least a part of the surface of the actually manufactured object (2), B) Receiving a simulation model, wherein an initial shape state (23) of the simulation model represents the desired shape (14) of the object (2) and wherein the initial shape state 23 is initially defined as the current shape state of the simulation model, wherein the simulation model is designed to simulate mechanical stresses within the object (2) that occur during an assumed deformation of the object (2), C) Determining a further shape state (24) of the simulation model in which the simulation model exhibits a minimum mechanical strain energy among many or all possible shape states in which, starting from the current shape state of the simulation model, at least one local area (B1) of the simulation model approximates the surface data set (22), D) Adopting the further shape state of the simulation model determined in step C) as the new current shape state of the simulation model, E) Defining the modified simulation model with the new current shape state from step D) as the modified simulation model of the actually manufactured object (2).

2. The method of claim 1, wherein in step C) the approximation of the at least one local area (B1) of the simulation model is carried out by: Subdividing at least one local area (B1) of the simulation model into sub-areas of the simulation model (Us) or adopting a corresponding existing subdivision; - Determine for each sub-area of ​​the simulation model (Us) according to a predefined assignment rule whether an assignable sub-area of ​​the surface data set (Uo) exists, and if so, assign the existing assignable sub-area of ​​the surface data set (Uo) to the sub-area of ​​the simulation model (Us); - from the sub-areas of the simulation model (Us) for which an associated sub-area of ​​the surface data set (Uo) exists, determine active sub-areas (Us1, Us2, Us3) of the simulation model that satisfy at least one first predefined criterion; Initiation of a process of relocating or shifting the active sub-areas of the simulation model (Us1, Us2, Us3) taking into account a constraint that prescribes either an active approach to the associated sub-area of ​​the surface data set or at least a partial superimposition of the associated sub-area of ​​the surface data set, but in each case offers or enables at least one degree of freedom of relocating or shifting the active sub-area of ​​the simulation model (Us1, Us2, Us3).

3. Method according to claim 2, wherein, for the relocated or displaced active sub-areas of the simulation model (Us1, Us2, Us3), after determining the form state (24) of minimum mechanical strain energy, it is checked whether a second predetermined criterion is met, wherein if the second criterion is not met, the affected active sub-area is deactivated, and wherein the process of relocating or displacing the active sub-areas of the simulation model (Us1, Us2, Us3) is repeated under step C) without considering the constraint for the deactivated sub-area or sub-areas.

4. Method according to one of claims 1-3, wherein steps C) and D) are repeated such that the shape state (23; 24) of the simulation model is gradually approximated to the surface data set (22) at least for the at least one local area (B1) of the simulation model as a whole, starting from the desired shape (14) of the object.

5. Method according to claim 2, wherein sub-areas of the surface data set (Uo) are determined based on a global analysis of the surface data set (22) which are excluded for assignment to the sub-areas (Us) of the simulation model.

6. Method according to any one of claims 1 to 5, wherein the simulation model is a model according to the finite element method (FEM).

7. Method according to any one of claims 1-6, wherein the modified simulation model is used in a further step in: - a simulation of a deformation of the object (2) by forces acting on the object from the outside; a simulation of the object (2) in a mechanically stressed state; a determination of a deformation of the object (2) by simulating forces acting on the object from the outside, saving a determination result and using the determination result for at least one other actually manufactured object of the same type; a creation of at least one further simulation model with a further shape state for the same object (2), such that the different simulation models describe the object with two different shape states, and determination of differences in the shape of the object represented by the at least two simulation models; Validation of a process of manufacturing the object (2) and / or - of determining a deformation field between the modified simulation model and another simulation model of the same or an identical actually manufactured object.

8. Method according to claim 7 for determining a deformation field, wherein in a first step the procedure is carried out for a first actually manufactured object in a first state after the object has been manufactured, wherein in a further step the procedure is carried out for the first or a second actually manufactured object in a second state, wherein the second state has different externally acting forces compared to the first state, and wherein a deformation field is determined between the modified simulation models of the first and the further step.

9. Device for generating a modified simulation model, wherein the device is configured in particular for carrying out a method according to one of the preceding method claims and wherein the device comprises: - an interface for receiving a surface data set (22) of an actually manufactured object (2), wherein the actually manufactured object was manufactured based on a target shape (14), and wherein the surface data set (22) comprises data based on a measurement of at least a part of the surface of the actually manufactured object (2), - an interface for receiving a simulation model, wherein an initial shape state (23) of the simulation model represents the target shape (14) of the object (2), and wherein the initial shape state (23) is initially defined as the current shape state of the simulation model, wherein the simulation model is designed to simulate mechanical stresses within the object (2) that occur during an assumed deformation of the object (2), a modification device (36) designed to modify the simulation model to a modified simulation model, taking into account the surface data set (22), and to perform the following steps: • Determining a shape state (24) of the simulation model in which the simulation model exhibits a minimum mechanical strain energy among many or all possible shape states in which, starting from the current shape state of the simulation model, at least one local area (B1) of the simulation model approximates the surface data set (22); and • Adopting the simulation model in the specified form state (24) of the minimum mechanical strain energy as the modified simulation model, a determination device (37) which is designed to determine, after completion of the modification of the simulation model, the then existing modified simulation model as the modified simulation model of the actually manufactured object (2).

10. Device according to the preceding claim, wherein the changing device (36) is configured to perform the following steps when approaching the at least one local area (B1) of the simulation model in the step to determine a shape state in which the simulation model has a minimum mechanical strain energy. - Subdividing at least one local area (B1) of the simulation model into sub-areas of the simulation model (Us) or adopting a corresponding existing subdivision; For each sub-area of ​​the simulation model (Us), determine according to a predefined assignment rule whether an assignable sub-area of ​​the surface data set (Uo) exists, and if so, assign the existing assignable sub-area of ​​the surface data set (Uo) to the sub-area of ​​the simulation model (Us); - from the sub-areas of the simulation model (Us) for which an associated sub-area of ​​the surface data set (Uo) exists, determine active sub-areas (Us1, Us2, Us3) of the simulation model that satisfy at least one first predefined criterion; Initiation of a process of relocating or shifting the active sub-areas of the simulation model (Us1, Us2, Us3) taking into account a constraint that prescribes either an active approach to the associated sub-area of ​​the surface data set or at least a partial superimposition of the associated sub-area of ​​the surface data set, but in each case offers or enables at least one degree of freedom of relocating or shifting the active sub-area of ​​the simulation model (Us1, Us2, Us3).

11. Device according to claim 10, wherein the modification device (36) is configured to check, for the relocated or displaced active sub-areas of the simulation model (Us1, Us2, Us3), after determining the form state (24) of minimum mechanical strain energy, whether a second predetermined criterion is met, wherein if the second criterion is not met, the affected active sub-area is deactivated, and wherein the process of relocating or displacing the active sub-areas of the simulation model (Us1, Us2, Us3) is repeated under step C) without considering the constraint for the deactivated sub-area or sub-areas.

12. Device according to one of claims 9 to 11, wherein the changing device (36) is configured to perform the step for determining a form state in which the simulation model exhibits a minimal mechanical exhibits strain energy, and to repeat the step of adopting the simulation model in the determined form state of minimum mechanical strain energy as the modified simulation model at least once, so that the form state (23; 24) of the simulation model is gradually approximated to the surface data set (22) at least for the at least one local area (B1) of the simulation model as a whole, starting from the target form (14) of the object.

13. Device according to one of claims 9 to 12, wherein the simulation model is a model according to the finite element method (FEM).

14. Arrangement with a device according to one of claims 9 to 13, wherein the arrangement further comprises a further device configured to use the modified simulation model in a further step for - simulating a deformation of the object (2) by forces acting on the object from the outside; Simulation of the object (2) in a mechanically stressed state; - Determining a deformation of the object (2) by simulating forces acting on the object from the outside, saving a determination result and using the determination result for at least one other actually manufactured object of the same type; creating at least one further simulation model with a further shape state for the same object (2), so that the different simulation models describe the object (2) with two different shape states, and determining differences in the shape of the object represented by the at least two simulation models; Validation of a process of manufacturing the object (2) and / or determination of a deformation field between the modified simulation model and another simulation model of the same or an identical actually manufactured object.

15. Arrangement according to claim 14 for determining a deformation field, wherein the arrangement is configured, In a first step, to create a first adapted simulation model for a first actually manufactured object in a first state after the object's manufacture, - in a further step to create another adapted simulation model for the first or a second actually manufactured object in a second state, where the second state exhibits different external forces compared to the first state, and the arrangement is designed to determine a deformation field between the adapted simulation models of the first and the subsequent step.

16. Computer program comprising instructions which, when the computer program is executed by a computer or by an arrangement of computers, cause it / them to execute the method according to any one of claims 1 to 8.

17. Computer-readable storage medium comprising instructions which, when executed by a computer or by an arrangement of computers, cause it / them to execute the method according to any one of claims 1 to 8.

Images