METHOD FOR MONITORING THE FUNCTIONAL STATUS OF A SYSTEM FOR COMPUTER TOMOGRAPHIC INVESTIGATION OF WORKPIECES

Description

[0001] The invention relates to a method for monitoring the functional state of a system for the computed tomographic examination of workpieces according to claim 1, as well as to a corresponding device according to claim 14 and a computer program product according to claim 15.

[0002] In addition to their well-known medical applications, computed tomography (CT) examinations are increasingly being used for quality assurance and material testing in the production of workpieces. This allows for non-destructive examinations of the internal or hard-to-reach areas of a workpiece, as well as dimensional measurements on the workpiece. However, it must always be ensured that the images generated by the CT scanner actually reflect reality with sufficient accuracy. For example, wear and tear or defects in the system can lead to a gradual or temporary decline in the quality of the images. Therefore, monitoring the correct functional status of the CT scanner is necessary to ensure the meaningfulness of the collected data.

[0003] To date, various approaches exist for monitoring computed tomography examinations of workpieces. A first approach involves the use of dimensional standards. A reference value for a specific dimension is determined using a calibrated test specimen, which is then compared with the measured values ​​obtained during continuously repeated measurements of the test specimen. However, the disadvantage of this approach is that for many geometries and structures to be examined, it is not possible or cost-effective to create a metrologically appropriately calibrated dimensional standard. Therefore, the measured values ​​determined by a dimensional standard often do not correspond to the measured values ​​relevant for assessing the manufacturing quality of an object under investigation.If this initial monitoring approach is performed with a calibrated workpiece, the process is generally limited to externally accessible geometries, as only these can be calibrated externally using metrology. Capturing internal geometries by cutting and calibrating the individual parts and then using the reassembled body for monitoring leads to a variety of problems in reality. For example, deviations can arise due to deformation of the workpiece during cutting, deviations due to material loss during cutting, and deviations due to the frequently required registration of the individual parts to form an overall geometry.

[0004] Furthermore, each monitoring of the system with a measuring standard requires interruption of ongoing measurement operations, meaning that the system's utilization efficiency is suboptimal. Monitoring the system with a measuring standard typically has to be repeated at regular intervals to ensure adequate reliability of the measured values ​​determined with the system. Particularly when the permissible tolerances for measurements using the system are very narrow, very close monitoring using measuring standards may be necessary. This can lead to frequent interruptions of measurement operations, particularly for very precise measurements with narrow error tolerances, which severely limits the system's utilization efficiency.

[0005] A second approach involves monitoring the operating parameters of the system in use. Operating parameters can be, for example, electrical operating parameters (voltage or current of the X-ray tube) or mechanical parameters, such as the achievement of target positions and deviations from them. Automated checks of simple image parameters can also be performed. One disadvantage of this approach is that this monitoring is not very specific. Therefore, warnings about a system malfunction may be issued too early or too late.

[0006] For example, DE 10 2006 048 608 A1 describes a method for monitoring the performance status of an X-ray source and / or an X-ray detector, wherein a set of predeterminable test parameters is selected at least for the X-ray source, wherein X-rays emanating from the X-ray source strike at least a partial area of ​​the X-ray detector and are detected by the X-ray detector, wherein at least one characteristic variable characterizing the X-ray source and / or at least one characteristic variable characterizing the X-ray detector is recorded. The at least one recorded characteristic variable is compared with a reference characteristic variable corresponding to the recorded characteristic variable, wherein a deviation of the recorded characteristic variable from the reference characteristic variable is used as a measure of the performance status of the X-ray source and / or the X-ray detector.

[0007] Furthermore, US 2007 / 0036264 A1 discloses a method for X-ray imaging of a body using a support for receiving a body to be examined, a source emitting an X-ray beam, a detector irradiated by the X-ray beam, an image converter for converting the detected intensities into data, a means for rotating the mounted mobile support by an angle of rotation about an axis of rotation with respect to the source and the detector, and a suitably programmed computer to average the data acquired for a pair of orthogonal angles of rotation, to obtain n column and m line averages for n and m elementary segments of a band of the detector, to construct an initial image (n, m) with the n column and m line averages, to adjust the attenuation coefficient in each n × m elementary zone by a least squares method, taking into account the n column and m line averages, which are considered as boundary conditions,to repeat the previous steps for data acquired with different pairs of preferably orthogonal rotation angles, and to average the adjusted images term by term to arrive at a synthesis image expressing attenuation coefficients of the body under investigation.

[0008] Finally, implicit monitoring of images often occurs through the review and evaluation of the scan results by a human operator or expert. This often involves visually assessing the "sharpness" of cross-sectional images or the shape of surfaces. However, this monitoring only occurs selectively or locally, as it is very time-consuming to evaluate complete CT images in this way. The quality of this monitoring also depends heavily on the expert's competence and prior knowledge and is subject to natural fluctuations – e.g., due to overload, fatigue, or simply lack of time.

[0009] The present invention is therefore based on the object of providing an improved method for monitoring the functional state of a system for the computed tomographic examination of workpieces, which overcomes the aforementioned disadvantages of the prior art.

[0010] Main features of the invention are set out in claim 1, as well as in claims 14 and 15. Embodiments are the subject of claims 2 to 13.

[0011] In a first aspect, the invention relates to a computer-implemented method for monitoring the functional status of a system for the computed tomographic examination of workpieces. A computed tomographic examination of a workpiece using the system comprises performing one or more computed tomographic measurements on the workpiece, wherein the computed tomographic measurements performed by the system each yield at least one measured value for at least one measured variable. The method for monitoring the functional status initially comprises selecting measured values ​​for at least one measured variable from at least two computed tomographic measurements on one or more workpieces. Then, at least one measure of scatter for the selected measured values ​​of the at least one measured variable and at least one reference measure of scatter for measured values ​​of the at least one measured variable are determined.The functional status of the system is then determined by comparing the at least one determined degree of scatter with the at least one reference degree of scatter for the at least one measured variable. The degree of scatter provides information about the distribution of the measured values. Measurement data is obtained from at least one computed tomography image or image of a tomographed workpiece. It is therefore entirely possible for a data set obtained from a computed tomography image for an examined workpiece to be evaluated multiple times, so that one or more measured values ​​are determined for one or more measured variables from a single data set. A plurality of measurements therefore does not necessarily mean a plurality of images by the computed tomography system.

[0012] A "measured quantity" can be understood as a specific dimensionally measurable property of a workpiece or one that can be measured during non-destructive testing. For example, a measured quantity can be the position of an edge in the workpiece, the diameter of a hole, the density and / or size and / or position of pores in a workpiece, or something similar. However, a measured quantity does not necessarily have to be a dimensional property of a workpiece. Rather, a measured quantity can also be a quantity that is determined not from the image of the workpiece created by a recording, but from the image of the recording itself. For example, measured values ​​for such a measured quantity can also be determined from a projection image created during a recording or from the reconstructed volume data in the vicinity of the suspected object surface.Examples of such measured quantities would be gray value distributions or noise values ​​in defined areas of a projection image or the reconstructed volume.

[0013] The use of measured variables derived from the projection images generated during a scan has the advantage that the raw data from the scan can be used to assess the system's condition. In contrast, dimensional measured variables of the workpieces first require a reconstruction of the examined workpiece by backprojecting the projection data. Filters are typically used to obtain the sharpest possible image. However, this results in the loss of information that could be useful for assessing the functional status of the computed tomography system in use.

[0014] The measured values ​​are evaluated using a computed tomography examination of a workpiece, so that one or more numerical values ​​are obtained as a measured value. For example, a measured value can be a length, a density, or another value that characterizes the recorded image rather than the workpiece itself. However, a measured value does not necessarily have to be a directly measurable value, such as an edge length. Rather, a measured value can also be a derived value, such as an average surface quality in a specific area of ​​an examined workpiece. A measured value can also be a gray value in a specific area or a signal-to-noise ratio of an image. In general, measured values ​​​​can also be derived from the measured values ​​determined during the measurements, which allow measurements to be compared with one another.

[0015] According to the invention, at least one measure of dispersion for the measured values ​​of the measured variable under consideration is determined from the measured values ​​from at least two measurements on one or more workpieces. A measure of dispersion is a measure that contains information about the distribution of the measured values. For example, the measure of dispersion can be a variance, an interval, or a standard deviation of the measured values.

[0016] The degree of dispersion indicates a parameter that makes it possible to determine the functional status of the system.

[0017] The resulting degree of scatter is then compared with a reference degree of scatter in the subsequent process to determine the functional status of the system. The reference degree of scatter can be retrieved from a database, for example. As with the previously described degree of scatter, the reference degree of scatter can also be a measure that represents a statistic or distribution of measured values. However, the reference degree of scatter does not necessarily have to be the same type of scatter as the degree of scatter determined from the measured values.

[0018] The use of statistical variables to determine the system condition has the advantage that fluctuations due to manufacturing tolerances of the workpieces when assessing the functionality of the computer tomograph used have no or only a very small influence on the result of the assessment of the system's functional condition.

[0019] The individual procedural steps do not necessarily have to be completed in the order described above. A different order is also possible, provided it makes logical sense.

[0020] According to one embodiment, the functional status of the system can be determined by determining a difference between the at least one determined degree of variation and the at least one reference degree of variation and comparing the determined difference with at least one permissible value range for the difference. If the difference lies outside the at least one permissible value range, information is output, indicating that the determined difference lies outside at least one permissible value range.

[0021] A "difference" between a measure of scatter and a reference measure of scatter can be understood not only as a difference between the two values, but also as a percentage deviation or something similar. For example, a difference can also be determined by checking how closely an interval defined by the measure of scatter overlaps with an interval defined by the reference measure of scatter—i.e., how large the intersection is. If the determined difference lies outside at least one permissible value range, information is output. This information can be output either by the CT scanner itself, for example, in the form of a visual or acoustic warning signal, or it can be transmitted in digital form as a signal to a peripheral device and processed there.For example, if the information is passed on to a peripheral device, a log file can be created in which the detected deviations or differences are recorded. Such a log file could then be used, for example, to schedule maintenance on the CT scanner.

[0022] It should be noted that, according to the invention, several permissible value ranges can be defined for the determined difference. For example, a type of "traffic light" could be implemented for the functional status of the CT scanner. If the determined difference lies within a first value range, it is assumed that the CT scanner is functioning very well, so that no maintenance measures are necessary and the determined values ​​can be trusted without restriction. If the determined differences lie outside the first value range but still within the second value range, it is assumed that the CT scanner is no longer functioning optimally, but that no maintenance measures are currently necessary. However, it can be assumed that maintenance will be necessary in the foreseeable future, so a corresponding maintenance plan is drawn up.However, if the determined difference lies outside the first and second value ranges, it can be assumed that the CT scanner is no longer providing sufficiently accurate measurement results and that the CT scanner requires maintenance. The information output when the determined difference lies outside a value range also includes an indication of the value range(s) that were exceeded. This output can also be used to block workpieces from further production or use because the CT system did not guarantee a valid measurement.

[0023] According to a further embodiment, the output information can also be used to initiate a check of the monitored system using a measuring standard. By checking using a measuring standard, it can then be tested whether the determined difference in the scatter measurements, which led to the output of the information, was actually caused by a faulty computed tomography system. It is certainly possible that the production of the examined workpieces is subject to systematic errors, for example, caused by an excessively high production temperature or other faulty system parameters. Such sources of error can also arise gradually. This would also result in systematic deviations of the measured values ​​from the measured variables under consideration, which can lead to a deviation of the scatter measure ultimately determined from the measured values ​​from the reference scatter measure.This deviation could then be mistakenly interpreted as due to a malfunction of the measuring device. However, this incorrect interpretation can be avoided by testing using a measuring standard.

[0024] In addition to distinguishing whether a deviation in the scatter measurements is caused by faulty manufacturing or faulty recording, the described embodiment also has the advantage that the generally complex verification using a measuring standard is only carried out when a specific reason for doing so can be derived from the ongoing evaluation of measurement results. The previously described regular monitoring of the system using a measuring standard could thus be modified into needs-based monitoring. This effectively reduces the number of interruptions to measurement operations for calibration using a measuring standard compared to regular calibration and ultimately optimizes the system's utilization efficiency, effectively increasing the system's utilization efficiency.

[0025] Alternatively or in addition to the output of information as described above, according to a further embodiment, determining the functional status of the system can further comprise determining a temporal progression of the determined scatter measures for a measured variable. By comparing the determined progression with the at least one reference scatter measure, it is then estimated when a difference between the scatter measure and the reference scatter measure will lie outside at least one permissible value range for the difference. In this way, for example, a prediction can be made as to the point in time at which the CT scanner images will no longer be precise enough, due to ongoing wear, to provide representative information about the quality of the workpieces being examined.

[0026] As already explained above, the reference scatter used to determine the functional status of the system can be retrieved, for example, from a database. Alternatively, however, according to one embodiment, it can also be provided that the at least one reference scatter is determined in a training phase. The training phase initially comprises the selection of reference measured values ​​for at least one measured variable from a reference measurement series comprising a plurality of measurements on a plurality of workpieces. Subsequently, the at least one reference scatter is determined from the selected reference measured values ​​of the at least one measured variable in the reference measurement series.

[0027] The advantage of using a training phase like this to determine one or more reference scatter measurements is that workpieces that are identical in design to the workpieces that are to be analyzed and evaluated in the further course of the process can be used to determine the reference values. The process described can therefore be used to determine very meaningful reference values ​​or reference scatter measurements that are easily reproducible and comparable through measurements on the workpieces. Furthermore, measured values ​​that are already determined during workpiece testing can be used to determine the reference scatter measurements in a training phase. This means that a separate series of measurements does not necessarily have to be recorded. Instead, measured values ​​from the ongoing workpiece testing can be used, thus avoiding downtime due to monitoring the functional status of the system.

[0028] According to a further embodiment, the training phase is carried out immediately after the system for computed tomography examination of workpieces has been repaired and restarted. It is assumed that after the system has been repaired, the images from the system are accurate or contain as few errors as possible. Thus, the reference scatter values ​​determined during the training phase can be considered as accurate as possible.

[0029] The previously described training phase for determining one or more reference scatter measures can be repeated at regular or irregular intervals and / or after the inspection of a specific number of workpieces. For example, the training phase can be scheduled to be repeated after every 50, 100, or 200 workpieces have been inspected, or after a specific period of time or at defined times. Furthermore, combined profiles can also be defined that determine the execution of the training phase. For example, the training phase can be scheduled to be performed after 200 images or after 2 hours of operation, whichever occurs first.In this way, changes to the CT scanner used can be taken into account, which may lead to a systematic change in the measured values ​​but are not due to wear and tear on the CT scanner. For example, it is to be expected that over time, the operation of the CT scanner will lead to system components heating up, which in turn can result in a systematic change in the measured values. Likewise, different workloads may occur at different times during the day-to-day operation of the system, which are taken into account by repeating the training phase when determining the reference scatter measures. The result of a training phase can be a complete renewal or only a partial update of the reference scatter measures.

[0030] To account for further external influences on recorded recordings, a further embodiment provides that at least one reference scatter measure depends on an environmental parameter of the system. An "environmental parameter" can be understood, for example, as the ambient temperature, the time of day, the weather, the operating time of the system, or similar. For example, one or more reference scatter measures can be determined through repeated training phases, whereby a correlation between the reference scatter measures and the environmental parameters can be determined by simultaneously recording the environmental parameters. This correlation can then be further used to account for systematic inaccuracies in the recording of measurement data, which, however, are not due to wear and tear of the system, and to evaluate them accordingly.Accordingly, reference variance measures can be stored in a database for various sets of environmental parameters, which are used to monitor the functional status of the system depending on the prevailing environmental parameters. Furthermore, environmental parameters can also take into account other aspects such as production machine operators, metrology operators, and manufacturing tolerances.

[0031] In a further embodiment, the measured values ​​are determined from one or more projection images of the workpiece under investigation and / or from a three-dimensional reconstruction of the workpiece under investigation created using the projection images of the workpiece under investigation.

[0032] In order to permanently ensure the correct function of the computer tomography scanner used, a further embodiment provides that the selection of measured values, the determination of a scatter measure for the selected measured values, and the determination of the functional status of the system are repeated after each examination of at least one workpiece. For example, after 5, 10, 50, or 100 workpieces, a determination of the functional status of the system can be carried out again in accordance with the invention described above. When selecting the density of determinations of the functional status, i.e. the number of workpiece tests after which a determination of the functional status is carried out again based on scatter measures, the desired test density and the computing load caused by a test must be taken into account. These two factors must be weighed against each other depending on requirements and available resources.Furthermore, it can be provided that the test density is not constant over time, but varies depending on external parameters. For example, it can be provided that the test density is set very high shortly after the system is started up and initially decreases with increasing operating time. Likewise, it can also be provided that the test density is increased again after a certain operating time, since defects or wear and tear are more likely to occur after longer operation of the system. Furthermore, the test density can also depend on external parameters such as ambient temperature, or can be set at short notice by a plant manager using an appropriately configured IT system. The latter can happen, for example, if specified tolerances for variables to be measured or tested change, or if the tolerance utilization of the workpieces changes.

[0033] According to a further embodiment, the measured values ​​for determining the degree of scatter are selected from measurements on directly consecutive workpieces. This allows a good snapshot of the system's condition to be realized.

[0034] According to a further embodiment, the measured variables used to determine the scattering measures comprise workpiece-specific measured variables, whereby the workpiece-specific measured variables can only be determined for a specific workpiece type. A workpiece-specific measured variable is therefore, for example, a dimension of a specific geometry of the workpiece under investigation or an average local surface quality in a spatially precisely defined workpiece domain. The use of workpiece-specific measured variables to determine the scattering measures has the advantage that any fluctuations that occur affect precisely those measured values ​​that are also used, for example, to assess the quality of the inspected workpieces.Accordingly, with this design, it is very unlikely that a change in the functional status of the system, which would result in inaccurate or no longer representative measured values, will go undetected due to an unfavorable choice of measured values. Furthermore, these workpiece-specific measured values ​​ensure that changes in the system that are irrelevant to the measured value do not lead to a conclusion about a system fault. This makes monitoring more efficient and cost-effective.

[0035] In an analogous manner, however, embodiments can also provide for the measured variables used to determine the scatter measures to include workpiece-specific measured variables, whereby the workpiece-specific measured variables can be determined on different types of workpieces. A workpiece-specific measured variable can, for example, be the worst signal-to-noise ratio of the material gray values ​​in the relevant workpiece domains across the sample. Furthermore, the dimensions of standardized workpiece sections, such as standardized screw holes, could also be used to determine a workpiece-specific measured variable.The use of workpiece-specific measurement parameters has the advantage that different workpieces can be inspected with the same CT system, while still allowing an assessment of the system's functional status based on the measurements taken on the different workpieces. This makes the application scenarios for the monitoring more flexible.

[0036] According to a further embodiment, the measured values ​​selected for determining the at least one degree of scatter originate from at least one defined partial area of ​​a workpiece, for which a degree of scatter determined from the measured values ​​for a measured variable can be compared with the reference degree of scatter for the measured variable. In particular, a partial area of ​​a workpiece can be selected that is of particular relevance for the quality inspection of the workpiece. For example, it can be provided that the pore density in a structurally important section of an examined workpiece is used for the reference degree of scatter. In this way, it can also be ensured that a change or deterioration in the functional status of the system is reliably detected, particularly for relevant measured variables.

[0037] In a further aspect, the invention relates to a device for monitoring a functional state of a system for the computed tomographic examination of workpieces, wherein the computed tomographic examination of a workpiece comprises performing one or more computed tomographic measurements on the workpiece using a system, wherein the measurements performed by the system each yield at least one measured value for at least one measured variable. The device for monitoring the functional state is designed to: To select measured values ​​for at least one measured variable from at least two measurements on one or more workpieces, To determine at least one degree of scatter for the selected measured values ​​of the at least one measured variable, To determine at least one reference degree of scatter for measured values ​​of the at least one measured variable, and To determine the functional state of the system by comparing the at least one determined degree of scatter with the at least one reference degree of scatter for the at least one measured variable, where the measure of dispersion provides information about a distribution of the measured values.

[0038] In yet another aspect, the invention relates to a computer program product with computer-executable instructions which, when executed on a computer, cause the computer to perform the method as described above. Further features, details, and advantages of the invention emerge from the wording of the claims and from the following description of exemplary embodiments with reference to the drawings. They show: Fig. 1 is a schematic representation of an exemplary production environment for applying the method according to the invention, and Fig. 2 is a flow chart of the method according to the invention.

[0039] The Figure 1shows a schematic representation of an environment 100 in which the method according to the invention can be applied. The environment 100 is a production environment in which, by way of example, a manufacturing system 102 is arranged, which is designed to produce workpieces 104. The manufacturing system 102 can, for example, be an injection molding system for producing injection-molded parts. The workpieces 104 produced in this way are then fed to a system 106 for computed tomography examination of the workpieces 104. This can be done, for example, by means of a conveyor belt 118 shown as an example with the running direction 120. The illustrated structure could enable in-line testing of workpieces 104 produced by the manufacturing system 102.

[0040] During the computed tomography examination performed by system 106, the workpieces 104 are to be checked for their manufacturing quality. Based on this examination, a decision can be made, for example, as to whether a workpiece 104 meets a set of minimum quality requirements and can therefore be sold / distributed, or whether the workpiece 104 should be considered scrap due to insufficient manufacturing quality. For this purpose, system 106 is designed, for example, to generate radiographic projection images of the inspected workpieces 104, which can be analyzed by an IT system 108 assigned to system 106. For this purpose, IT system 108 is equipped with at least processor means 110, storage means 112, and display means 114 and is connected to system 106 via a communications link 116.The analysis by the computer system 108 can refer to both the recorded radiographic projection images and to representations of the examined workpiece 104 reconstructed from the radiographic projection images. Both three-dimensional and two-dimensional reconstructions of the workpiece 104 can be used for the analysis. Measurement data can be transmitted to the computer system 108 via the communication link 116, where it can be stored in the storage means 112 and / or processed by the processor means 110.

[0041] The processor means 110 can, for example, be one or more single- or multi-core processors. Likewise, the memory means 112 can be any desired form of data memory. The display means 114 can, for example, be implemented by one or more monitors and / or signal lights. The EDP system 108 is shown here purely as an example as a coherent unit. The EDP system 108 can, of course, also be a system of computer systems. The individual computer systems do not necessarily have to be spatially grouped. Rather, a distributed computer system can also be used as the EDP system 108, for example in the sense of cloud computing.

[0042] From the analysis of the radiographic projection images of a workpiece and / or from the analysis of a complete two- or three-dimensional reconstruction of the workpiece 104 from the projection images, the computer system 108 is configured, using appropriate software, to determine measured values ​​for various measured variables of the workpieces. For example, a measured variable can be an edge position, an average surface quality, a material density, a pore density, or similar variables. However, a measured variable can also be workpiece-specific and indicate a measure of the image quality of the recorded projection images. The measured values ​​determined for these measured variables are then checked against permissible value ranges or limit values, and based on their compliance with the value ranges and / or limit values, it is determined whether a workpiece 104 meets the quality requirements.

[0043] During operation of a computed tomography system 106, a variety of effects can occur that influence the previously described examination and evaluation of workpieces 104. For example, as the system 106 continues to operate, wear and tear on the X-ray tube used can occur, resulting in increasing noise levels in the generated projection images or reconstruction images of the workpiece 104. Furthermore, mechanical effects such as shocks and vibrations can change the setting of the computed tomography system's focusing optics, which also distorts the generated measured values. These effects can, under certain circumstances, lead to measured values ​​being determined during the analysis of a workpiece 104 that lead to an assessment of the quality of the workpiece 104 as sufficient or inadequate, even though this is not actually the case.While an erroneous assessment of the quality of a workpiece 104 as inadequate merely unnecessarily increases the scrap rate, the opposite case—i.e., an erroneous assessment of the quality of a workpiece 104 as sufficient—can quickly expose the system operator to liability if an accident or other serious malfunction of the workpiece occurs. Therefore, it must be ensured at all times that the measured values ​​determined by the system 106 are representative.

[0044] For this purpose, the computer system 108 is designed by means of the method according to the invention to derive from the measured values ​​of the workpieces 104, in addition to an assessment of their quality, also an evaluation or monitoring of the functional state of the system 106.

[0045] The method according to the invention is Figure 2 presented in the form of a flow chart 200.

[0046] In a first method step 202, measurement data for the workpiece(s) 104 are first determined by means of a computed tomography image of at least one workpiece 104. During the evaluation of the measurement data thus obtained, measurement values ​​for one or more measured variables are then determined for the at least one workpiece 104. These can be the same measured variables that are also used to evaluate the quality of the workpieces 104. However, it is also possible to define special derived variables that are used only to monitor the functional status of the system 106.

[0047] From the multitude of measured values ​​determined during the examination of workpieces 104, a set of measured values ​​is then selected in step 204 to be used to determine the functional status of system 106. For example, measured values ​​from measurements on directly consecutive workpieces 104 can be used to achieve the highest possible monitoring density. However, it is also possible to use only measurements on a single workpiece 104. Furthermore, measured values ​​with the lowest possible measurement uncertainty are preferably used to assess the functional status.

[0048] From the selected measured values, a measure of dispersion is then determined in step 206. The measure of dispersion is a statistical measure, such as the variance, a value interval, or something similar.

[0049] In order to obtain the best possible assessment of the statistics of the measured values, it is advisable to select a sufficiently large sample in step 204. However, it should also be considered that a sample that is too large will result in a high data processing effort, so a balance should be made here according to the requirements and available resources.

[0050] In addition to the degree of scatter of the measured values ​​from the measurements on the workpieces, a reference degree of scatter is also determined in step 208. The reference degree of scatter should preferably provide information on the statistics of precisely the measured variable used to determine the degree of scatter(s) in step 206. However, the type of reference degree of scatter does not necessarily have to be the same type of statistical information that was selected to determine the degree of scatter of the measured values. Provision can be made for reference degrees of scatter to be stored in the storage means 112 of the computer system 108, so that to determine the reference degrees of scatter, only the storage means 112 needs to be accessed and the required reference degrees of scatter retrieved.

[0051] However, it is also possible for the reference scatter measures to be determined in a separate training phase. In this case, measured values ​​from measurements on workpieces 104 are selected, and statistical measures are determined from the selected measured values, which are then used as the reference scatter measure. Such a training phase is preferably carried out shortly after or directly after a repair of the system 106, since at this time it can be assumed that the system is in the best possible condition and thus delivers representative results. The use of such a training phase can also have the advantage that measured values ​​from measurements on the same workpieces 104 can be used to determine the reference scatter measures, the quality of which is then to be assessed based on the images from the system 106. Thus, in particular, workpiece-specific measured variables can also be used to evaluate the functional status of the system 106.

[0052] Furthermore, it is possible to repeatedly determine reference scatter measures during ongoing operation. This allows, for example, a systematic drift in the recorded measured values, which is not attributable to wear or a fault in the system 106, to be taken into account. Accordingly, different reference scatter measures can exist for one and the same measured variable, which are applied depending on other parameters, such as the ambient temperature, the time of day, or the operating time of the system 106. Furthermore, a reference scatter measure can also consist of multiple values. For example, the reference scatter measure for a specific measured variable can include a mean value and a standard deviation.

[0053] After a reference scatter measure has been determined, a comparison of the scatter measures, i.e., the scatter measure determined from the measured values, with the reference scatter measure(s), takes place in step 210. The functional status of system 106 is then derived from this comparison. For example, a difference between the scatter measure and the reference scatter measure can be determined. The meaning of the term "difference" depends on the type of scatter measure used.

[0054] In the previously described comparison of the difference with a limit value, it is not necessarily necessary to provide for only one limit value. Rather, it can also be provided that a plurality of limit values ​​or value intervals are defined, which enable a step-by-step evaluation of the functional status of the system 106. In this way, a status traffic light for the system 106 can be implemented, which indicates, for example, whether the system 106 is operating very well, still sufficiently, or no longer sufficiently accurately. The described intermediate evaluation stage can be helpful in planning the use of the system 106. From the comparison of the scatter measure(s) with corresponding reference scatter measures, a decision is made in step 212 as to whether the condition of the system 212 is sufficient to deliver representative results or whether maintenance is necessary. If it is determined that the condition of the system 106 is OK, the inspection of workpieces 104 continues unchanged.If, however, it is determined that the condition of system 106 is no longer sufficient, a warning can be issued in step 214. In the simplest case, the warning can be a visual or acoustic signal.

[0055] At this point, it can also be provided that, after determining the condition of the system 106, information describing the condition of the system 106 is generated and stored, for example, in the storage means 112 of the IT system 108. In this case, for example, a temporal progression of the condition of the system 106 can be determined, from which, in turn, a point in time can be extrapolated at which the system 106 will no longer operate with sufficient accuracy. This allows an optimized usage and / or maintenance plan for the system 106 to be created.

[0056] The invention is not limited to one of the embodiments described above, but can be modified in many ways within the scope of the claims. List of reference symbols

[0057] 100Production environment 102Manufacturing system 104Workpiece 106Computed tomography system 108Computer system 110Processor means 112Storage means 114Display means 116Communication connection 118Travel belt 120Travel direction

Claims

1. Computer-implemented method for monitoring the functional state of an installation (106) for computed-tomography examination of workpieces (104), wherein the computed-tomography examination of a workpiece (104) comprises the performing of one or more computed-tomography measurements on the workpiece (104) by means of the installation (106), wherein the computed-tomography measurements performed by the installation (106) each result in at least one measured value for at least one measured variable, wherein the method for monitoring the functional state comprises the following steps: • selecting measured values for at least one measured variable from at least two computed-tomography measurements on one or more workpieces (104), • ascertaining at least one scatter measure for the selected measured values of the at least one measured variable, • ascertaining at least one reference scatter measure for measured values of the at least one measured variable, and • ascertaining the functional state of the installation (106) by comparing the at least one ascertained scatter measure with the at least one reference scatter measure for the at least one measured variable, wherein the scatter measure comprises information about a distribution of the measured values.

2. Method according to Claim 1, characterized in that the ascertaining of the functional state of the installation (106) comprises the following steps: • ascertaining a difference between the at least one ascertained scatter measure and the at least one reference scatter measure, • comparing the ascertained difference with at least one permissible range of values for the difference, • if the difference is outside the at least one permissible range of values, outputting information, wherein the information indicates that the difference is outside at least one permissible range of values.

3. Method according to Claim 1 or 2, characterized in that the ascertaining of the functional state of the installation (106) comprises the ascertaining of a time characteristic of the ascertained scatter measure for a measured variable, wherein comparison of the ascertained characteristic with the at least one reference scatter measure estimates when a difference between the scatter measure and the reference scatter measure will be outside at least one permissible range of values for the difference.

4. Method according to one of the preceding claims, characterized in that the at least one reference scatter measure is ascertained in a training phase, wherein the training phase comprises the following steps: • selecting reference measured values for at least one measured variable from a reference measurement series having a plurality of computed-tomography measurements on a plurality of workpieces (104), • ascertaining the at least one reference scatter measure from the selected reference measured values of the at least one measured variable of the reference measurement series.

5. Method according to Claim 4, characterized in that the training phase is performed immediately after a repair and restarting and / or an initial measurement and restarting of the installation (106) for computed-tomography examination of workpieces (104).

6. Method according to Claim 4 or 5, characterized in that the training phase is repeated at regular or irregular intervals of time and / or after the examination of a particular number of workpieces.

7. Method according to one of the preceding claims, characterized in that at least one reference scatter measure is dependent on an environmental parameter of the installation (106).

8. Method according to one of the preceding claims, characterized in that the measured values are ascertained from one or more projection images of the examined workpiece (104) and / or from a three-dimensional reconstruction of the examined workpiece (104) that is produced by means of the projection images of the examined workpiece (104).

9. Method according to one of the preceding claims, characterized in that the selecting of measured values, the ascertaining of at least one scatter measure for the selected measured values and the ascertaining of the functional state of the installation (106) are each repeated after the examination of at least one workpiece (104).

10. Method according to one of the preceding claims, characterized in that the measured values for the ascertainment of the scatter measure are selected from computed-tomography measurements on directly successive workpieces (104).

11. Method according to one of the preceding claims, characterized in that multiple measured variables are available and the measured variables comprise workpiece-specific measured variables, wherein the workpiece-specific measured variables can be ascertained only on a particular kind of workpiece.

12. Method according to one of the preceding claims, characterized in that multiple measured variables are available and the measured variables comprise workpiece-aspecific measured variables, wherein the workpiece-aspecific measured variables can be ascertained on different kinds of workpieces.

13. Method according to one of the preceding claims, characterized in that the measured values selected for the determination of the scatter measure come from at least one defined subregion of a workpiece (104), for which a scatter measure, ascertained from the measured values, for at least one measured variable can be compared with the reference scatter measure for the measured variable.

14. Apparatus (108) for monitoring a functional state of an installation (106) for computed-tomography examination of workpieces (104), wherein the computed-tomography examination of a workpiece (104) comprises the performing of one or more computed-tomography measurements on the workpiece (104) by means of an installation (106), wherein the measurements performed by the installation (106) each result in at least one measured value for at least one measured variable, characterized in that the apparatus (108) for monitoring the functional state is designed to: • select measured values for at least one measured variable from at least two measurements on one or more workpieces (104), • ascertain at least one scatter measure for the selected measured values of the at least one measured variable, • ascertain at least one reference scatter measure for measured values of the at least one measured variable, and • ascertain the functional state of the installation (106) by comparing the at least one ascertained scatter measure with the at least one reference scatter measure for the at least one measured variable, wherein the scatter measure comprises information about a distribution of the measured values.

15. Computer program product having computer-executable instructions that, when executed on a computer, cause the computer to perform the method according to one of Claims 1 to 13.