Measurement of an object at a temperature that differs from the reference temperature

Abstract

The invention relates to a method for measuring an object (2), the temperature of which differs from a reference temperature, comprising: a) in a learning step, experimentally determining a dependency of a geometry parameter (G) of the object (2) on a temperature parameter (T) of the object (2), b) in a measuring step, carrying out the following sub-steps: b1) scanning the object (2) and determining a temperature value (T w ) b2) determining a measurement value (G M ) of the geometry parameter (G) of the object (2), b3) converting the measurement value (G M ) of the geometry parameter (G) of the object (2) determined in step b2) into a reference value (G R ) of the geometry parameter (G) of the object (2) which represents a prediction of the value of the geometry parameter (G) of the object (2) at the reference temperature (T R ), wherein the temperature parameter (T) of the object (2) is a measure of the temperature of a first region (4) of the object (2) and the geometry parameter (G) of the object (2) is a measure of a geometric dimension in a second region (5) of the object (2), which differs from the first region (4).

Description

[0001] Measuring an object The invention relates to a method for measuring an object whose temperature deviates from a reference temperature. The method can be used, for example, to measure a forged component in a hot state and to obtain a prediction of the object's dimensions at room temperature. Objects are measured in many applications. For example, in production, manufactured products are measured. The results can be used to check whether the dimensions of a manufactured product correspond to the tolerance specifications. This can be done for each individual manufactured product or on a random basis, depending on the desired reliability. Such measurements of objects are performed, for example, on forged products. It is a well-known problemthat forged products usually still have a very high temperature after production. Therefore, thermal material expansion must be taken into account. The desired geometric dimensions, which may be specified, for example, in a technical drawing, are usually related to room temperature. These cold dimensions cannot be easily determined by measuring a forged product while hot. It is possible to wait until the object has cooled to room temperature, but this usually takes too long. This is particularly the case if, depending on the measurement results, an ongoing production process is to be intervened in. Methods are known from physics to mathematically describe the thermal expansion of bodies. Therefore, approaches are known to convert measured values ​​​​taken at high temperatures into cold dimensions. However, these approaches are based on a mathematical consideration,which generally only delivers usable results for simple geometries with reasonable effort. Similar problems generally arise whenever geometric parameters for an object are determined at a temperature that deviates from a reference temperature, although the value of the geometric parameters at the reference temperature is of interest. The object of the present invention is to present a method with which an object can be measured whose temperature deviates from a reference temperature, wherein a reliable forecast of the value of a geometric parameter of the object at the reference temperature can be given with little effort. This object is achieved with the method according to claim 1. Further advantageous embodiments are specified in the dependent claims. The features presented in the claims and in the description are in any,can be combined with each other in a technologically expedient manner. According to the invention, a method for measuring an object whose temperature deviates from a reference temperature is presented. The method comprises: a) in a learning step, experimentally determining a dependency of a geometric parameter of the object on a temperature parameter of the object, b) in a measuring step, performing the following substeps: b1) scanning the object with a distance sensor unit and determining a temperature value which is a measure of the value of the temperature parameter of the object during the scanning, b2) determining a measured value of the geometric parameter of the object from a signal output by the distance sensor unit in step b1),b3) Converting the measured value of the object's geometric parameter determined in step b2) based on the temperature value determined in step b1) and the dependence of the object's geometric parameter on the object's temperature parameter determined in step a) into a reference value of the object's geometric parameter, which represents a forecast for the value of the object's geometric parameter at the reference temperature, wherein the object's temperature parameter is a measure of the temperature of a first region of the object and the object's geometric parameter is a measure of a geometric quantity in a second region of the object different from the first region. The described method is used to measure an object. The object can be any physical object. For example, it can be a component,which is measured after production using the described method. This allows checking whether the component meets tolerance specifications. In particular, the described method can be used to measure a forged component as the object. This can be done immediately after the manufacturing process of the forged component, when it is still hot. In principle, the method is applicable to objects of any shape. If an object has undercuts, it may be that these cannot be fully detected. However, this can be accepted. It is preferred that the object be rotationally symmetrical. This enables particularly easy scanning. In principle, the method is applicable to objects of any size. For practical reasons, however, the method is preferably carried out with objects that can be handled by a user with their hands. This can be quantified in such a way thatthat the object preferably has an extension of a maximum of 1 m in all spatial directions, in particular a maximum of 50 cm, and / or that the object has a mass of a maximum of 50 kg, in particular a maximum of 10 kg. For example, the object can have an extension in the range of 0.5 to 30 cm in all spatial directions and / or a mass in the range of 0.1 to 5 kg. Using the described method, the object can be measured at a temperature that deviates from a reference temperature. The reference temperature is the temperature to which the measurement results should be related. Using the described method, measured values ​​can be recorded at the current temperature of the object and transferred to the reference temperature, taking into account the material expansion. The reference temperature is preferably room temperature. To avoid any possible ambiguity regarding the definition of room temperature,The reference temperature can be defined by absolute temperature specifications. It is preferred that the reference temperature be in the range of 15 to 30 °C, preferably in the range of 20 to 25 °C. This covers the range of temperatures that are usually considered room temperature. Using room temperature or a temperature within the specified range as the reference temperature has the advantage that dimensions in technical drawings are usually referenced to this temperature. However, the level of the reference temperature is not important for the functionality of the described method. The described method can therefore also be carried out with any other reference temperature. It is implicitly clear that the reference temperature must not be so high that the object to be measured would no longer have physical integrity at this temperature.for example, by melting before reaching this temperature. The object is measured at its current temperature. Preferably, the temperature of the object during measurement is higher than the reference temperature, preferably by at least 200 K, particularly preferably by at least 500 K. For example, the current temperature of the object can be in the range of 500 to 1200 °C. Here, too, it is implicitly clear that this temperature must not be so high that the object to be measured no longer has any physical integrity. The greater the difference between the current temperature of the object and the reference temperature, the greater the influence of thermal expansion and, accordingly, the more worthwhile it is toto carry out the described procedure. Using the described procedure, a prediction of the value of a geometric parameter of the object at the reference temperature can be determined. If an object is measured at a temperature that is greater than the room temperature used as the reference temperature, the calculation performed in the described procedure can also be referred to as cold dimension prediction or cold dimension determination. Any geometric dimension of the object can be considered as a geometric parameter. For example, the geometric parameter can be a length specification such as a height, a thickness, a width,a length or a diameter. The geometry parameter can also be a radius of curvature. The geometry parameter can also be the geometric position of a point of the object in space. In this case, the geometry parameter can be a coordinate. The geometry parameter can also be the distance between two points. In general, a distinction is made here between "parameter" and "value of a parameter." A parameter is an abstract physical quantity, for example, the height of an object. A value of a parameter is a numerical value specified in a suitable unit. For example, the value of the height of an object can be 10 cm. A specific parameter can take on various values,especially at different temperatures. For example, for the parameter "height" of the object, a measurement value of 10 cm can be recorded at a first temperature and a measurement value of 9.9 cm at a second temperature. The described procedure is two-step. Step a) represents a learning step. It is sufficient to perform step a) once. Step b) represents a measurement step. The object is measured using the result of step a). Step b) can be performed sequentially for a large number of similar objects. This is particularly useful in a production process,in which a large number of similar products are manufactured one after the other. Each object can be measured using the same result from the learning step. The measurement of a single object is described below. However, this can easily be transferred to the measurement of several, particularly similar, objects. In step a), a dependence of the object's geometry parameter on a temperature parameter is experimentally determined. The temperature parameter is a measure of the object's temperature. The object's temperature is generally not evenly distributed. Therefore, the focus is not on the object's temperature as such, but on the temperature parameter. The temperature parameter can, for example, be defined via a measurement specification. For example, the temperature parameter can specify the temperature,which is measured with a temperature sensor directed at a specific point on the object. It is not necessary for the temperature parameter to actually directly reflect the temperature of the object. For example, a constant offset between the values ​​of the temperature parameter and the actual temperature of the object is harmless to the functioning of the described method. The temperature parameter of the object is a measure of the temperature of a first area of ​​the object, and the geometry parameter of the object is a measure of a geometric quantity in a second area of ​​the object that is different from the first area. The temperature parameter and the geometry parameter are therefore defined for different areas of the object. If the thermal expansion of the object's material were simply determined by the thermal expansion coefficient of that material,This approach would be counterproductive. However, the described method is based on the realization that a temperature parameter can be used that does not correspond to the actual temperature of the object in the range of the geometry parameter. Rather, the temperature value is introduced as a simple numerical value, via which the measured values ​​of the object to be measured can be compared with the dependence of the geometry parameter on the temperature parameter recorded in the learning step. The dependence of the object's geometry parameter on a temperature parameter can be determined, in particular, using a reference object.by measuring corresponding pairs of values, each consisting of a value of the geometry parameter and a value of the temperature parameter. The dependency can be determined by evaluating these measurement results. The dependency can be specified, in particular, in the form of a mathematical function or value table. Ideally, the reference object corresponds exactly to the object to be measured. However, useful results can also be achieved if the reference object is only similar to the object to be measured or even differs significantly from it. Preferably, the reference object and the object to be measured are of the same type. Preferably, the reference object is made of a material,whose thermal expansion coefficient does not deviate by more than 10% from the thermal expansion coefficient of the material of the object to be measured. Particularly preferably, the reference object is made of the same material as the object to be measured. The reference object is preferably shaped such that the geometric parameter defined for the object to be measured is also defined for the reference object. Preferably, a size of the reference object deviates from the object to be measured only to the extent that the value of the geometric parameter for the reference object deviates by a maximum of 10% from the value of the geometric parameter for the object to be measured. The reference object and the object to be measured are compared at the same temperature. Particularly preferably, the reference object and the object to be measured have the same size. However, it is also possible to consider a geometric parameter for the reference object,which is only similar to the geometric parameter of the object to be measured or even deviates significantly from it. It is also possible to consider a temperature parameter for the reference object which is only similar to the temperature parameter of the object to be measured or even deviates significantly from it. In step b), the object is measured using sub-steps b1) to b3). Step b) is performed after step a). Steps b1) and b3) can be performed consecutively or overlapping in time. It is implicitly clear that a sub-step can only be started when the results of the previous sub-steps required for this sub-step are available. In step b1), the object is scanned with a distance sensor unit. In addition, in step b1), a temperature value is determined,which is a measure of the value of the temperature parameter of the object during scanning. The object is scanned in step b1) with one or more distance sensors. For the functioning of the described method, the number of distance sensors, the type of distance sensors, and the arrangement of the distance sensors are not important. Therefore, it is generally referred to as a distance sensor unit. The distance sensor unit can comprise one or more distance sensors. Preferably, the distance sensor unit comprises one, two, or four distance sensors. The distance sensor unit is suitable for determining distances between the distance sensor unit and various points on the surface of the object. The distance sensor unit is therefore configured toto capture the geometry of the object relative to the distance sensor unit. This naturally only applies within a measuring range of the distance sensor unit. The measuring range is a three-dimensional space within which the object is held during scanning. The more distance sensors the distance sensor unit comprises, the larger the measuring range can be. If the object is scanned with multiple distance sensors from different directions, the object can be captured with particularly little shadowing. This also allows the capture of features of the object that are inaccessible to a single distance sensor due to shadowing. The following describes a distance sensor.which can be used in the distance sensor unit. The distance sensor unit can comprise one or more distance sensors of this type. The distance sensor is designed as a triangulation sensor. It comprises a laser and a detector. The laser emits laser light onto the object, and the detector detects the reflected light. This allows the distance between the distance sensor and the point on the object's surface at which the laser beam was reflected from the object to be determined using triangulation. It is also possible for the distance sensor to comprise a laser and two detectors. Preferably, the two detectors are arranged symmetrically on either side of the laser. By using two detectors, the object can be detected from two perspectives. This could also be achieved with two separate distance sensors.Each of which contains a laser and a detector. However, this would require two lasers. If the distance sensor unit comprises multiple distance sensors, each of these distance sensors has its own coordinate system. For example, each distance sensor can specify the position of points in spherical coordinates, which are defined relative to the position of this distance sensor. However, the goal is naturally to specify the position of the object's points in a common coordinate system. This is preferably a Cartesian coordinate system.which is defined in relation to the arrangement used to carry out the method. The coordinates output by the distance sensors are therefore preferably subjected to a coordinate transformation and thus combined. The coordinate transformation can be determined in a referencing step. The distance sensors are calibrated relative to one another. This is preferably done with a calibration object. The calibration object can be a cuboid, for example. A rotationally symmetrical shape is less suitable. In the referencing step, the calibration object can be scanned with the distance sensor unit, i.e., with the distance sensors of the distance sensor unit. The results obtained are then superimposed in such a way thatthat for all distance sensors, the measurement points, for example, of an edge of the calibration object, are located at the same location in a Cartesian coordinate system defined relative to the arrangement. The temperature value is recorded using a temperature sensor. This is preferably done contactlessly using a temperature sensor that is directed at the object during scanning. The temperature sensor can be a pyrometer, for example. The temperature value is a numerical value. If the object has the same temperature at all points during scanning, the temperature value preferably corresponds to this temperature. However, it can generally be assumed that the temperature of the object does not have the same value at all points during scanning. Such local temperature fluctuations are not only difficult to measure; in fact, it is generally not possibleto provide a prediction for the value of the geometry parameter at the reference temperature by considering the thermal material expansion while taking into account the actual temperature distribution within the object. This can at best be achieved for simple geometries and simple temperature distributions with reasonable effort. The described method represents a considerable simplification in this respect. In any case, the temperature value is a simple numerical value. This also applies if the temperature of the object does not have the same value at all points on the object during scanning. The temperature value is therefore not defined as the temperature of the object during scanning, but as a measure of the value of the temperature parameter during scanning. This does not exclude the possibility that in the special case where the object has the same temperature at all points on the object during scanning,The temperature value corresponds to this temperature. The temperature value can be a measured value of the temperature parameter. However, this is not mandatory. The temperature value can also differ from a measured value of the temperature parameter, for example, due to an offset or a correction factor. The temperature value can be determined, in particular, as an average of several temperature measured values ​​recorded during the scanning of the object. In step b2), a measured value of the geometry parameter is determined from a signal output by the distance sensor unit in step b1). The signal from the distance sensor unit contains the information recorded by the distance sensor unit during the scanning of the object in step b1). The signal from the distance sensor unit therefore contains information from which conclusions can be drawn about the geometry of the object. This is used in step b2).by determining the measured value of the geometric parameter. The measured value of the geometric parameter is the value that the geometric parameter has during scanning, i.e., the value that the geometric parameter has at the temperature at which the object is present during scanning. The measured value of the geometric parameter can be determined, in particular, using a three-dimensional model of the object. The three-dimensional model is preferably a computer model. The three-dimensional model includes information on the position of a plurality of points of the object in space. The three-dimensional model can, in particular, be formed by a plurality of points, each of which is defined by three coordinates. It is preferredthat between steps b1) and b2), a three-dimensional model of at least a part of the object is created from the signal output by the distance sensor unit in step b1). In step b2), the measured value of the geometric parameter can then be determined from this three-dimensional model. This can be done automatically, for example, by using an appropriate algorithm to recognize the respective position of previously defined features of the object. For example, the height of the object can be determined automatically from the three-dimensional model. Alternatively, the geometric parameter can also be determined manually from the three-dimensional model. However, it is not necessary to create the three-dimensional model. For example, the three-dimensional model can be created only for those parts of the object from which the measured value of the geometric parameter can be determined. It is even possible,to forego the creation of the three-dimensional model altogether. This is possible, in particular, by processing the signal from the distance sensor unit with an algorithm that directly extracts from the signal the information relevant for determining the measured value of the geometric parameter. It is also possible for the three-dimensional model to be generated internally within the software but not output as such. In this case, the user is not aware ofthat there is a three-dimensional model. In step b3), the measured value of the geometry parameter determined in step b2) is converted into a reference value of the geometry parameter based on the temperature value determined in step b1) and the dependence of the geometry parameter on the temperature parameter determined in step a). The reference value of the geometry parameter represents a forecast for the value of the geometry parameter at the reference temperature. This means that the reference value – within the limits of unavoidable inaccuracies – corresponds to the value of the geometry parameter that would occur if the value of the temperature parameter corresponded to the reference temperature. Whether the reference temperature is actually reached after the procedure has been carried out is irrelevant. The goal of the described procedure is precisely to determine the value of the geometry parameter at the reference temperature.although the object is at a temperature different from the reference temperature. In step b3), the value that is the goal of the described method is obtained using the reference value of the geometry parameter. Ideally, the reference value of the geometry parameter corresponds to the value that the geometry parameter assumes when the measured object is at the reference temperature. Since this value cannot be measured directly when the object is at a temperature different from the reference temperature, the reference value of the geometry parameter is merely a prediction. The reference value of the geometry parameter can be obtained by converting the measured value recorded in step b2) with the dependence of the geometry parameter on the temperature parameter determined in step a). In the simplest case, this is done by multiplying by a factor,which can be determined based on the dependence of the geometry parameter on the temperature parameter determined in step a). Alternatively or additionally, the dependence of the geometry parameter on the temperature parameter can also comprise an addition or subtraction. The described method is preferably carried out with an arrangement having a housing. The distance sensor unit and a temperature sensor for detecting the temperature value are preferably arranged in the housing. The object can be placed in the housing for measurement. The housing can preferably be closed, for example, via a door, a folding flap, or a simple flap. This allows the object to be measured with the housing closed. This reduces environmental influences on the measurement and achieves particularly high reproducibility. If the distance sensor unit comprises a laser,The lockable housing also ensures laser safety. Furthermore, the lockable housing prevents a user from coming into contact with the object during measurement. This is particularly advantageous when the object is being measured while hot. The described method determines a reference value for a geometry parameter. This means that a respective reference value is determined for at least one geometry parameter. It is possible and even preferred that the described method determines a respective reference value for several geometry parameters. For each of the geometry parameters, what is described here for one geometry parameter applies accordingly. For the sake of clarity, however, the focus here is primarily on a single geometry parameter. If several geometry parameters are considered,The method comprises the following steps: a) for several geometric parameters of the object, experimentally determining a respective dependency of the geometric parameter of the object on a temperature parameter of the object in a learning step, b) in a measuring step, performing the following sub-steps: b1) scanning the object with a distance sensor unit and determining a temperature value which is a measure of the value of the temperature parameter of the object during the scanning, b2) for each of the geometric parameters of the object, determining a respective measured value of the geometric parameter from a signal output by the distance sensor unit in step b1),b3) for each of the geometry parameters of the object, converting the respective measured value of the geometry parameter determined in step b2) based on the temperature value determined in step b1) and the respective dependency of the geometry parameter on the temperature parameter of the object determined in step a) into a respective reference value of the geometry parameter, which represents a forecast for the value of this geometry parameter at the reference temperature, wherein the temperature parameter of the object is a measure of the temperature of a first region of the object and at least one of the geometry parameters of the object is a measure of a geometric quantity in a respective second region of the object different from the first region. In this case, the learning step a) is preferably carried out jointly for all geometry parameters. However, it is also conceivable,that a separate learning step is performed only for some of the geometry parameters or even only for a single one of the geometry parameters. In addition to the at least one geometry parameter of the object, which is a measure of a geometric quantity in a respective second region of the object different from the first region, there may be one or more geometry parameters that are a measure of a geometric quantity in the first region of the object. There may be one or more second regions. For each of the second regions, there may be one or more geometry parameters that are a measure of a geometric quantity in this second region of the object. In a preferred embodiment of the method, the dependence of the geometry parameter of the object on the temperature parameter of the object is determined in step a),by determining, for a reference object at multiple temperatures, a value of a geometry parameter of the reference object and a value of a temperature parameter of the reference object. In this embodiment, the learning step with a reference object is carried out analogously to the measurement step with the object to be measured. For example, a lengthy cooling-down of the reference object is waited for once, in order to be able to dispense with this for all subsequent objects to be measured. A geometry parameter and a temperature parameter are defined for the reference object. These are preferably defined analogously to the geometry parameter and temperature parameter of the object to be measured, respectively. However, this is not absolutely necessary. The dependence of the object's geometry parameter on the object's temperature parameter can also be determined bythat the reference object is used to determine the dependence of a differently defined geometric parameter of the reference object on a differently defined temperature parameter of the reference object, and this is transferred to the object to be measured, accepting a corresponding inaccuracy. In the simplest case, the transfer can be achieved by directly equating the dependence of the geometric parameter of the reference object on the temperature parameter of the reference object, determined for the reference object, with the dependence of the geometric parameter of the object to be measured on the temperature parameter of the object to be measured. However, there is also nothing to prevent, for example, a correction factor from being taken into account during the transfer.which was obtained from empirical values. In one example, a measured value of 10 cm for a height is determined for the reference object at a temperature of 900 °C, and then a measured value of 9.9 cm for this same height is determined at room temperature. If a measured value of 10 cm for this height is determined for an object to be measured that is identical to the reference object at a temperature of 900 °C, there is no need to wait for the object to cool down. Instead, it can be immediately predicted that this height will also be 9.9 cm for the object to be measured at room temperature. The best results can be obtained when the reference object exactly matches the object to be measured. However, this is not absolutely necessary. In general, the more closely the reference object matches the object to be measured, the better the results. For example, if a large number of similar objects are produced,One of these objects can be used as the reference object. However, it is also possible to apply the result of a learning step performed for a reference object of a first type to an object of a second type and to accept the inaccuracy resulting from the differences between the types. However, it is preferred that a separate learning step is performed for each type of object. The best results can be obtained if, in learning step a), a measured value of the geometry parameter was recorded at exactly the temperature at which a measured value of the geometry parameter for the object to be measured is also recorded. However, it is not necessary to measure in step b) at a temperature that was also used in step a). For intermediate values ​​between the values ​​used in step a), interpolation can be performed. Extrapolation even makes it possiblein step b) at a temperature that lies outside a temperature range covered in step a). In general, the closer the temperature of the object in step b) is to a value used in step a), the better the results. Preferably, in step a), measured values ​​are recorded at temperatures that cover a temperature range in which the temperature of the object is expected to lie in step b1). Particularly preferably, in step a) measured values ​​are recorded at temperatures that are evenly distributed over this temperature range. The temperature range preferably extends around a temperature expected for step b1). If, for example, in step b) a large number of similar forged objects are measured immediately after their production, it can be assumed that these objects have a similar temperature. It is therefore sufficient,for a range around this temperature, measure values ​​in step a). In particular, it is not necessary to cover the entire temperature range between the expected temperature of the object in step b1) and the reference temperature in step a). Rather, it is sufficient to cover the temperature range around this expected temperature and the reference temperature in step a). The best results can be obtained if, in the learning step, a measurement value of the geometry parameter was recorded at exactly the reference temperature. However, this is not absolutely necessary. By interpolation or extrapolation, a reference temperature can also be determined that was not taken into account in learning step a). In general, the closer the reference temperature is to a value used in step a), the better the results. Preferably, in step a), measure values ​​are recorded at temperatures,which cover a temperature range extending over at least 50 K and which deviates by at least 100 K, in particular by at least 300 K, from the reference temperature, and on the other hand, in step a), a measured value is recorded which deviates by no more than 20 K from the reference temperature. Preferably, no further values ​​are recorded between this temperature range and the individual temperature value. In general, the best results can be obtained if the measurement of the object in step b) is carried out exactly analogously to the measurement of the reference object in step a). The preferred features of measurement step b) described herein are therefore also preferred analogously for measurement step a). However, it is not absolutely necessary that steps a) and b) be carried out exactly analogously to one another. The results are better,the smaller the differences between the procedure in steps a) and b). However, there is nothing to be said against accepting disadvantages in this regard. For example, to save costs, the implementation of a precisely matching learning step a) can be dispensed with if the results of a previously implemented learning step a) can be used, which is at least similar to the measurement step to be performed. In a further preferred embodiment of the method, the dependence of the geometric parameter of the object on the temperature parameter of the object is determined in step a),by determining a mathematical function from pairs of values, each consisting of a value of the reference object's geometry parameter and a value of the reference object's temperature parameter. In the simplest case, the mathematical function can express the dependence of thermal expansion on a thermal expansion coefficient ^ of the object's material. However, the described method also provides usable results when the object's geometry is more complex and / or when the temperature distribution within the object is not homogeneous. In this case, the described calculation could not simply be performed using the thermal expansion coefficients. In particular, it would be unclear which value should be used for the temperatures. Because the dependence of the geometry parameter on the temperature parameter is determined experimentally in a learning step in step a),These difficulties can be avoided with the described method. In this respect, it should also be noted that the temperature parameter used does not necessarily have to correspond exactly to the temperature of the object. Such a parameter would not be clearly defined if the object had an inhomogeneous temperature distribution. It is not mandatory to use the function specified above. Alternatively, non-linear functions can also be used. In general, any mathematical function can be used that can be reasonably adapted to the measurement data determined in the learning step. Depending on the location of the measurement points, the appropriate function for adaptation can be selected. In a further preferred embodiment of the method, the learning step a) comprises the following substeps:which are each carried out once at several different temperatures of the reference object: a1) Scanning the reference object with the distance sensor unit and determining a temperature value, which is a measure of the value of the temperature parameter of the reference object during the scanning, a2) Determining a measured value of the geometric parameter of the reference object from a signal output by the distance sensor unit in step a1). In this embodiment, a temperature value is determined in each case in step a1). The best results can be obtained if this is carried out exactly analogously to the determination of the temperature value in step b1). Preferably, therefore, the same measuring specification is used in steps a1) and b1). In particular, it is preferred that the same measuring device is used for steps a1) and b1). In this case, it is irrelevantwhether the signal output by this measuring device correctly indicates the temperature of the reference object or the object. However, it is not mandatory that the temperature value in step a1) be determined exactly as the temperature value in step b1). It is also possible that an inaccuracy resulting from a deviation in the determination of the temperature value is accepted. Regarding step b), it was previously stated thatthat the measured value of the geometry parameter can be determined, in particular, based on a three-dimensional model of the object. This applies analogously to step a). In particular, before step a2), a three-dimensional model of at least a portion of the reference object can be created from a signal output by the distance sensor unit in step a1). In step a2), the measured value of the geometry parameter can then be determined from this three-dimensional model of the reference object. In a further preferred embodiment of the method, steps a1) and a2) are performed cyclically while the reference object cools down. The simplest way to perform learning step a) isby cooling the object in the environment and continuously determining value pairs, each consisting of a temperature value and a measured value of the geometry parameter. In this way, a particularly large amount of data is obtained. With regard to the functionality of the described method, however, there is nothing to prevent measurements in the learning step only at certain temperatures. In a further preferred embodiment of the method, the object is scanned with the distance sensor unit in step b1) by rotating the object about a rotation axis relative to the distance sensor unit. In this embodiment, the object can be scanned, for example, by recording measured values ​​with the distance sensor unit at constant angular intervals, for example every 0,5°. The object is preferably held on a turntable for measurement. The turntable is preferably driven by a motor. The motor is preferably a servomotor. This allows the rotation of the object to be controlled particularly precisely. This is advantageous because the angular position of the object is relevant for scanning. If the method is carried out using an arrangement with a lockable housing, the housing can also prevent a user from being injured by the rotational movement of the object or, for example, a turntable used for this purpose. If the learning step is carried out with sub-steps a1) and a2), the reference object is preferably scanned with the distance sensor unit in step a1) by rotating the reference object relative to the distance sensor unit about a rotation axis. In this respect, the learning step a) is also carried out analogously to the measurement step b). If the reference object is scanned,By rotating the reference object relative to the distance sensor unit about a rotation axis, steps a1) and a2) can each be performed once per revolution. This has proven particularly practical because a particularly large amount of data is obtained in learning step a). The rotation axis preferably coincides with an axis of the object to be measured. However, this is not necessarily the case. In a further preferred embodiment of the method, the temperature value of the object is determined in step b1) as an average of measured values ​​recorded by a temperature sensor during scanning. By averaging the measured values, a simple numerical value can be obtained, which is a measure of the temperature of the object. It is not necessarythat this mean value actually corresponds to the average temperature of the object. In a further preferred embodiment of the method, the orientation of the temperature sensor relative to the object remains unchanged during the scanning in step b1). In particular, in this embodiment, it is preferred that the temperature sensor is configured for contactless measurement of the temperature of the object. This can be achieved, for example, using a pyrometer as the temperature sensor. The fact that the orientation of the temperature sensor relative to the object is unchanging during the scanning in step b1) means that the temperature sensor is always aligned along a fixed direction. The fact that the temperature sensor is used to measure the temperature of the object impliesthat this fixed direction is directed toward the object. The object's temperature is measured at the intersection of this direction with the object's surface. If the object is rotated relative to the temperature sensor during scanning, the temperature is still not measured at a fixed point on the object. Rather, the temperature is measured along a line that runs along the object's surface. If the temperature value is determined by averaging the temperature measurements during scanning, the temperature value is the average temperature of the object at the described line on the object's surface. It has been found that the temperature value can be determined particularly easily in this way, and that satisfactory results are also achieved. The invention is explained in more detail below with reference to the figures. The figures show a particularly preferred embodiment.to which the invention is not limited. The figures and the proportions depicted therein are merely schematic. They show: Fig. 1: a side view of an arrangement for carrying out a method according to the invention for measuring an object, Fig. 2: a top view of the arrangement from Fig. 1, Fig. 3: the object shown in Figs. 1 and 2 in an enlarged side view, Fig. 4: a simplified visualization of the measuring range of the distance sensor unit of the arrangement from Figs. 1 and 2, Fig. 5: an illustration of the experimental determination of the dependence of a geometric parameter of a reference object on a temperature parameter of a reference object, which can be used with the arrangement from Figs. 1 and 2, Fig. 6: an illustration of the conversion of a measured value of the geometric parameter of the object based on the dependence shown in Fig. 5. Fig. 1 shows an arrangement 1,with which an object 2 can be measured. The object 2 is also shown, but is not part of the arrangement 1. The arrangement 1 has a housing 12. Within the housing 12, a turntable 10 is arranged, on which the object 2 is held. The object 2 can be rotated together with the turntable 10 about a rotation axis 6. In this case, the object 2 can be scanned with a distance sensor unit 3. The distance sensor unit 3 comprises two pairs, each consisting of a laser 8 and a detector 9. Furthermore, the arrangement 1 comprises a temperature sensor 7, with which a temperature value of the object 2 can be recorded. In Fig. 2, the arrangement 1 from Fig. 1 is shown in a plan view. The description of Fig. 1 applies analogously to Fig. 2. With the arrangement 1 from Figs. 1 and 2, a method for measuring the object 2 can be carried out. The object 2 has a temperature,which deviates from a reference temperature. The method comprises: a) in a learning step, experimentally determining a dependence (visualized in Fig. 5) of a geometric parameter ^ of the object 2 on a temperature parameter ^ of the object 2, b) in a measuring step, performing the following substeps: b1) scanning the object 2 with a distance sensor unit 3 and determining a temperature value ^, ^ , which is a measure of the value of the temperature parameter ^ of the object 2 during scanning, b2) Determining a measured value ^ ^ of the geometry parameter ^ of the object 2 from a signal output by the distance sensor unit 3 in step b1), b3) converting the measured value G determined in step b2) M of the geometry parameter ^ of object 2 based on the temperature value determined in step b1) ^ ^and the dependence of the geometry parameter ^ of the object 2 on the temperature parameter ^ of the object 2 determined in step a) into a reference value ^ ^ of the geometry parameter ^ of object 2, which is a forecast for the value of the geometry parameter ^ of object 2 at the reference temperature ^ ^ represents. In Fig. 3, the object 2 shown in Figs. 1 and 2 is shown enlarged. The object 2 is rotationally symmetrical. Two geometric parameters ^ are shown as examples. These are the height of the upper section of the object 2 and the diameter of the upper section of the object 2. Also shown is a first region 4, which is formed as a line running along the circumference of the object 2. On this line, the temperature value ^ is measured in step b1) during scanning. ^recorded with the temperature sensor 7 by keeping the orientation of the temperature sensor 7 relative to the object 2 unchanged during scanning and rotating the object 2 relative to the temperature sensor 7. Furthermore, a second region 5 is shown, which represents the upper section of the object 2. It is irrelevant how far the second region 5 extends downwards. It is sufficient that the second region 5 does not extend into the first region 4. The temperature parameter ^ is thus a measure of the temperature of the first region 4 of the object 2 and the geometry parameter ^ is a measure of a geometric quantity in the second region 5 of the object 2, which is different from the first region 4. Fig. 4 shows a measuring region 11 of the distance sensor unit 3. The measuring range results from the intersection region of the light beams emitted by the two lasers 8, which are indicated by dashed lines.The measuring range is symmetrical to the rotation axis 6, which is also shown. However, this is not mandatory. The rotation axis 6 can even lie outside the measuring range 11. In this case, the rotation of the object 2 would cause the part of the object 2 to be scanned to enter the measuring range 11. Fig. 5 shows how the dependence of the geometry parameter ^ of the object 2 on the temperature parameter ^ of the object 2 can be determined by taking a measured value ^ for a reference object at several temperatures. ^ ^ a geometry parameter ^ ^ of the reference object and a temperature value of the temperature parameter ^ ^ of the reference object. A pair of measured values ​​^ ^ ^ of the geometry parameter ^ ^ of the reference object and temperature value of the temperature parameter ^ ^ of the reference object is used for the reference temperature ^ ^and is therefore considered a reference value ^ ^ ^ of the geometry parameter ^ of the reference object at the reference temperature ^ ^ By adapting a linear mathematical function, a dependence of the geometry parameter ^ ^ of the reference object from the temperature parameter ^ ^ of the reference object. This is indicated by a dashed line. This dependency can be transferred from the reference object to the object 2 to be measured. Fig.6 shows a plot of the geometry parameter ^ of object 2 against the temperature parameter ^ of object 2. The mathematical function determined from Fig.5 is also shown as a dashed line. If a measured value ^ ^ of the geometry parameter ^ of object 2 at a temperature value ^ ^ recorded, this can be converted into a reference value ^ taking into account the mathematical function ^of the geometry parameter ^ of object 2 at the reference temperature ^ ^ This is indicated by a dotted arrow, which runs parallel to the dashed line. The measured value ^ ^ of the geometry parameter ^ of object 2 is thus multiplied by a factor in the example shown to the reference value ^ ^ of the geometry parameter ^ of the object 2, where the factor of the gradient corresponds to the linear function determined in Fig.5.

[0002] List of reference symbols 1 Arrangement 2 Object 3 Distance sensor unit 4 First area 5 Second area 6 Rotation axis 7 Temperature sensor 8 Laser 9 Detector 10 Turntable 11 Measuring area 12 Housing ^ Geometric parameters of the object ^ ^ Measured value of the geometry parameter of the object ^ ^ Reference value of the object's geometry parameter ^ Temperature parameter of the object ^ ^ Temperature value of the object ^ ^Geometry parameters of the reference object ^ ^ ^ Measured value of the geometry parameter of the reference object ^ ^ ^ Reference value of the geometry parameter of the reference object ^ ^ Temperature parameters of the reference object ^ ^ ^ Temperature value of the reference object ^ ^ Reference temperature

Claims

Claims 1. Method for measuring an object (2) whose temperature deviates from a reference temperature, comprising: a) in a learning step, experimentally determining a dependency of a geometric parameter (^) of the object (2) on a temperature parameter (^) of the object (2), b) in a measuring step, carrying out the following sub-steps: b1) scanning the object (2) with a distance sensor unit (3) and determining a temperature value (^ ^ ), which is a measure of the value of the temperature parameter (^) of the object (2) during scanning, b2) determining a measured value (^ ^ ) of the geometry parameter (^) of the object (2) from a signal output by the distance sensor unit (3) in step b1), b3) converting the measured value (G) determined in step b2) M ) of the geometry parameter (^) of the object (2) based on the temperature value (^) determined in step b1). ^) and the dependence of the geometry parameter (^) of the object (2) on the temperature parameter (^) of the object (2) determined in step a) into a reference value (^ ^ ) of the geometry parameter (^) of the object (2), which provides a forecast for the value of the geometry parameter (^) of the object (2) at the reference temperature (^ ^ ), wherein the temperature parameter (^) of the object (2) is a measure of the temperature of a first region (4) of the object (2) and the geometry parameter (^) of the object (2) is a measure of a geometric size in a second region (5) of the object (2) different from the first region (4).

2. Method according to claim 1, wherein the dependence of the geometry parameter (^) of the object (2) on the temperature parameter (^) of the object (2) is determined in step a) by determining for a reference object at several temperatures in each case a Value of a geometry parameter (^ ^) of the reference object and one value of a temperature parameter (^ ^ ) of the reference object are determined.

3. The method according to claim 2, wherein the dependence of the geometry parameter (^) of the object (2) on the temperature parameter (^) of the object (2) is determined in step a) by selecting from value pairs each consisting of a value of the geometry parameter (^ ^ ) of the reference object and a value of the temperature parameter (^ ^ ) of the reference object, a mathematical function is determined.

4. The method according to claim 2 or 3, wherein the learning step a) comprises the following sub-steps, which are each carried out once at several different temperatures of the reference object: a1) scanning the reference object with the distance sensor unit (3) and determining a temperature value (^ ^ ^ ), which is a measure of the value of the temperature parameter (^ ^) of the reference object during scanning, a2) Determining a measured value (^ ^ ^ ) of the geometry parameter (^ ^ ) of the reference object from a signal output by the distance sensor unit (3) in step a1).

5. The method according to claim 4, wherein steps a1) and a2) are carried out cyclically while the reference object cools down.

6. The method according to any one of the preceding claims, wherein the object (2) is scanned in step b1) with the distance sensor unit (3) by rotating the object (2) relative to the distance sensor unit (3) about a rotation axis (6).

7. The method according to any one of the preceding claims, wherein the temperature value (^ ^) is determined in step b1) as an average value of measured values ​​of a temperature sensor (7) recorded during the scanning.

8. The method according to claim 7, wherein an orientation of the temperature sensor (7) relative to the object (2) remains unchanged during the scanning in step b1).

Images