WO1996026511A1 - Method of detecting systematic defects in quality control processes - Google Patents
Method of detecting systematic defects in quality control processes Download PDFInfo
- Publication number
- WO1996026511A1 WO1996026511A1 PCT/EP1996/000744 EP9600744W WO9626511A1 WO 1996026511 A1 WO1996026511 A1 WO 1996026511A1 EP 9600744 W EP9600744 W EP 9600744W WO 9626511 A1 WO9626511 A1 WO 9626511A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- sensors
- measuring
- deviations
- mean value
- value
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000009897 systematic effect Effects 0.000 title claims abstract description 22
- 238000003908 quality control method Methods 0.000 title claims abstract description 7
- 230000007547 defect Effects 0.000 title abstract description 7
- 238000005259 measurement Methods 0.000 claims abstract description 37
- 230000000052 comparative effect Effects 0.000 claims abstract 7
- 238000001514 detection method Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000012937 correction Methods 0.000 claims description 2
- 230000007257 malfunction Effects 0.000 abstract 1
- 238000012360 testing method Methods 0.000 description 21
- 238000004519 manufacturing process Methods 0.000 description 9
- 238000011156 evaluation Methods 0.000 description 5
- 238000003070 Statistical process control Methods 0.000 description 2
- 230000002950 deficient Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000015654 memory Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000002872 Statistical quality control Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 229930014626 natural product Natural products 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000010972 statistical evaluation Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B29/00—Checking or monitoring of signalling or alarm systems; Prevention or correction of operating errors, e.g. preventing unauthorised operation
- G08B29/16—Security signalling or alarm systems, e.g. redundant systems
Definitions
- the invention relates to a method for detecting systematic errors, in particular for automatic fault detection in quality controls, by means of measuring devices equipped with measuring sensors with a combination of three individual sensors of the same type for parallel measurement on the same object, the measured values of the individual sensors being recorded, the mean value and / or a comparison value of the measured values of the individual sensors is created.
- Such a method is known from DE 40 22 954 AI, which describes an error detector for determining significant differences in the outputs of redundantly provided signal transmitters, which is acted upon by differences in the output signals of the signal transmitters and which delivers an error message if the outputs of the signal transmitters correspond to a significant difference by a criterion characterizing a threshold value.
- Fail-safe techniques for industrial controls are known from industrie-elektrik-elektronik, 24th year 1979, No. 9/10, S 224-228.
- B a redundant system with fail-safe subsystems for error detection, systems with two identical and independent channels with bit processor, program memory, process bus, timer and input-output modules are recommended.
- the two processors operate in isochronous mode so that it is possible to monitor each step of the processing by comparing all bit lines of the buses.
- a device for determining surface defects of high quality surfaces with a scanner that can be positioned over the surface and illuminates the measuring points of the surface with a light beam of suitable wavelength at a constant angle of incidence and at different heights and circumferential angles around the reflected one Beam arranged light sensors which are connected to an electronic evaluation device for quality evaluation of the surface.
- reduction devices are provided which contain addressable memories with previously calculated and stored values and also mean value devices in which mean values of the intensities are determined in parallel from separately supplied signals.
- a large number of measuring and testing devices are known which use sensors - the sensors being tactile or optical or any other sensors - to test or measure a wide variety of substances and materials and products and to display or record the results.
- Another area of measurement technology concerns quality control. For any measurement, different factors can lead to incorrect measurement results, which in turn can lead to considerable consequential damage.
- One problem is to recognize systematic measurement errors that z. B. caused by incorrectly set, defective, aged, misaligned or soiled measuring sensors. These are extremely difficult to recognize if the measurement scatter is within the normal range and thus pretends that the measurement is correct. Several measurements in a row do not reveal such systematic errors.
- the object of the invention is to provide a method for recognizing systematic errors by means of measuring devices equipped with sensors, which makes it possible to control the measuring system, that is to say to estimate the measured values obtained and to systematically deviate the measured results promptly during the actual measuring or testing process recognize without additional control measures having to be carried out, and at the same time also carry out a quality control of the test objects.
- object is also understood here to mean three or more individual objects which are taken from an overall object, i. H. they can also be individual objects in a batch.
- the method according to the invention makes it possible to identify systematic measurement errors of test or measuring devices with a high degree of probability and thus to significantly increase the quality of the check, since this also includes statements the results are secured in the interval between two prescribed calibrations.
- Advantageous embodiments of the invention make it particularly easy to identify systematic measurement errors and to assign them to individual measurement sensors, which can then be exchanged. Furthermore, it is possible, if a certain measured value or a size calculated by a device with a processor is exceeded, an acoustic or optical signal, for. B. also to switch off or interrupt further measurement processes, which then allows a check. Furthermore, during measurements it is possible to identify systematic errors of each individual measuring sensor and to eliminate them during an evaluation or to save a determined constant so that the correct values are automatically given despite the deviation. Another advantage is that a statistical evaluation of entire series of measurements can be carried out, which can then be used to control and regulate and control the quality.
- the method for recognizing systematic measurement errors on measuring devices equipped with measuring sensors is that at least three mutually independent measuring sensors measure the same object in parallel, preferably simultaneously.
- the method can be used with any measuring sensors as well as with measuring and testing devices.
- the individual sensors are preferably arranged within a measuring head, without they being able to influence one another.
- the individual measured values of the at least three sensors are determined in parallel - as soon as possible or at the same time - to save time, possibly converted into signals and fed to an evaluation unit that can display the individual measured values of each sensor separately, but also via a processor the mean value, the median value or a
- the comparison value of the three measurements and the difference between the individual measurement values from the mean value, median value or the comparison value are determined, displayed and, if appropriate, also stored.
- a signal can be triggered so that, if necessary, a further measurement can be carried out on the same object in a changed measuring position, in which case the last measurement can also be deleted if necessary, since it should not be used for statistics.
- a systematic measurement value deviation of a measuring circuit of the test device can be reliably detected by means of a statistical mean value and / or comparison value deviation control of a series of measurements.
- the deviating sensor can be recognized and the deviation variable can be determined, with which it is then possible to correctly calculate the measured values. With this method, it is possible to reliably detect deviations that are significantly smaller than the natural product measurement scatter.
- the evaluation can be done via an integrated machine control or a HOST control.
- the trend calculation can be a special solution for individual devices represent logarithmic correction of past mean and / or comparison value deviations.
- An advantage of this method is that no buffer battery is required as a buffer in the device in the event of a power failure or when the device is switched off. Only two values per measurement sensor and one value for the number of measurements are to be saved as statistical values. This can be done by taking suitable measures without a battery in an EEPROM.
- a prerequisite for the method to function properly is that a safe parallel measurement is possible.
- a quantity that can be controlled in parallelized test sensors - according to the laws of statistical process control - represents the mean value and / or comparison value deviation for at least three or even more sensors used at the same time. If one sensor has a permanent mean value and / or or comparison value deviation, the statistical process control methods can be applied to the measuring device, as are known from production control. This means that the known control methods can also be used for statistical processes, the use of which is aimed at the mean and / or comparison value deviation.
- a great further advantage of the method is that everything can run automatically and only when certain deviations a signal is sent which interrupts the measuring process and enables a check.
- the defective location of the manufacturing tool can be easily determined.
- the deviations, which are determined by the measuring sensor or the measuring sensors, then do not indicate any systematic measuring errors which have their cause in the measuring device, but in the faulty production, e.g. B. by worn punch, by which the properties of the test specimens are significantly affected.
- This method can thus be used to reliably identify and specifically identify manufacturing faults of the measurement objects that have not previously been testable as product-independent systematic faults.
- Critical manufacturing processes can be monitored for systematic errors by means of parallelization.
- test sensors measuring in parallel are arranged at defined measuring points in the parallel process branches.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention concerns a method of detecting systematic defects, in particular for automatically detecting malfunctions in quality control processes, by means of measuring apparatus equipped with measuring sensors and with a combination of three individual sensors of the same type for parallel measurements on the same object. The measured values of the individual sensors are detected and the mean value and/or a comparative value of the measured values of the individual sensors is calculated. The deviations of the at least three individual sensors from the common mean value and/or the deviations in the comparative values of the individual sensors are determined. The mean value and/or comparative value and/or the mean value deviations and/or comparative value deviations are compared with one another during a sequence of measurements carried out at various measuring locations on a given object among an amount of identical objects or on an amount of identical objects, so that defects in the measuring apparatus can simultaneously be detected and the quality of the objects can be controlled.
Description
VERFAHREN ZUM ERKENNEN VON SYSTEMATISCHEN FEHLERN BEI QUALITÄTSKONTROLLENMETHOD FOR DETECTING SYSTEMATIC ERRORS IN QUALITY CONTROLS
Die Erfindung betrifft ein Verfahren zum Erkennen von systematischen Fehlern, insbesondere zur automatischen Störungserkennung bei Qualitätskontrollen, mittels mit Meßsensoren ausgestatteten Meßgeräten mit einer Kombination aus drei einzelnen Sensoren desselben Typs zur Parallelmessung am selben Objekt, wobei die Meßwerte der einzelnen Sensoren erfaßt, der Mittelwert und/oder ein Vergleichswert der Meßwerte der einzelnen Sensoren erstellt wird.The invention relates to a method for detecting systematic errors, in particular for automatic fault detection in quality controls, by means of measuring devices equipped with measuring sensors with a combination of three individual sensors of the same type for parallel measurement on the same object, the measured values of the individual sensors being recorded, the mean value and / or a comparison value of the measured values of the individual sensors is created.
Ein derartiges Verfahren ist aus der DE 40 22 954 AI bekannt Darin ist ein Fehlerdetektor zur Feststellung signifikanter Differenzen der Ausgänge von redundant vorgesehenen Signalgebern beschrieben, der von Differenzen der Ausgangssignale der Signalgeber beaufschlagt ist und eine Fehlermeldung liefert, wenn die Ausgänge der Signalgeber einen vorgegebenen, eine signifikante Differenz durch einen Schwellwert charakterisierenden Kriterium entsprechen.Such a method is known from DE 40 22 954 AI, which describes an error detector for determining significant differences in the outputs of redundantly provided signal transmitters, which is acted upon by differences in the output signals of the signal transmitters and which delivers an error message if the outputs of the signal transmitters correspond to a significant difference by a criterion characterizing a threshold value.
Aus industrie-elektrik-elektronik, 24. Jg. 1979, Nr. 9/10, S 224-228 sind Fail-safe-Techniken für Industrie-Steuerungen bekannt, so z. B. ein redundantes System mit Fail-safe- Teilsystemen zur Fehlererkennung, wobei Systeme mit zwei identischen und voneinander unabhängigen Kanälen mit Bitprozessor, Programmspeicher, Prozeßbus, Zeitgeber und Eingabe-Ausgabe-Bausteinen empfohlen werden. Die beiden Prozessoren arbeiten taktsynchron, so daß es möglich ist, jeden Schritt der Verarbeitung durch Vergleich aller Bitleitungen der Busse zu überwachen.
Aus der DE 43 24 800 AI ist eine Vorrichtung zur Bestimmung von Oberflächenfehlern von Oberflächen hoher Güte mit einem scannend über die Oberfläche positionierbaren Abtaster bekannt, der Meßpunkte der Oberfläche mit einem Lichtstrahl geeigneter Wellenlänge in konstantem Einfallswinkel beleuchtet und in unterschiedlichen Höhenwinkeln und Umfangswinkeln um den reflektierten Strahl angeordnete Lichtsensoren aufweist, die an eine elektronische Auswerteinrichtung zur Qualitätsbewertung der Oberfläche angeschlossen sind. In der Auswerteinrichtung sind Reduktionseinrichtungen vorgesehen, die adressierbare Speicher mit zuvor berechneten und abgespeicherten Werten und auch Mittelwerteinrichtungen enthalten, in denen parallel aus getrennt zugeführten Signalen Mittelwerte der Intensitäten bestimmt werden.Fail-safe techniques for industrial controls are known from industrie-elektrik-elektronik, 24th year 1979, No. 9/10, S 224-228. B. a redundant system with fail-safe subsystems for error detection, systems with two identical and independent channels with bit processor, program memory, process bus, timer and input-output modules are recommended. The two processors operate in isochronous mode so that it is possible to monitor each step of the processing by comparing all bit lines of the buses. From DE 43 24 800 AI a device for determining surface defects of high quality surfaces is known with a scanner that can be positioned over the surface and illuminates the measuring points of the surface with a light beam of suitable wavelength at a constant angle of incidence and at different heights and circumferential angles around the reflected one Beam arranged light sensors which are connected to an electronic evaluation device for quality evaluation of the surface. In the evaluation device, reduction devices are provided which contain addressable memories with previously calculated and stored values and also mean value devices in which mean values of the intensities are determined in parallel from separately supplied signals.
Es ist eine Vielzahl von Meß- und Prüfgeräten bekannt, die mittels Sensoren - wobei die Sensoren Tast- oder optische oder auch beliebige andere Sensoren sein können - unterschiedlichste Stoffe und Materialien und Produkte prüfen bzw. messen und die Ergebnisse anzeigen oder auch aufzeichnen. Ein weiteres Gebiet der Meßtechnik betrifft die Qualitätskontrolle. Bei jeglichen Messungen können unterschiedliche Faktoren zu falschen Meßergebnissen führen, die wiederum zu erheblichen Folgeschäden führen können. Ein Problem besteht darin, systematische Meßfehler zu erkennen, die z. B. durch falsch eingestellte, defekte, gealterte, verstellte oder verschmutzte Meßsensoren verursacht werden. Diese sind außerordentlich schwer zu erkennen, wenn die Meßwertstreuung im Rahmen des üblichen liegt und somit eine ordentliche Messung vortäuscht . Mehrere Messungen in Folge lassen solche systematischen Fehler nicht erkennen. Es wird geschätzt, daß bei Meßergebnissen von Prüf- und Meßgeräten die unerkannten oder vermeintlichen systematischen Meßfehler in Abhängigkeit vom Meßsystem und Meßprinzip bis zu 30 oder 40 % betragen. Wenn man vermutet, daß die ermittelten Werte nicht korrekt sind, werden derzeit mit weiteren
Kontrollgeräten die Meßwerte überprüft und/oder die Meßsensoren mit Kontroll- oder Kalibriereinheiten kontrolliert. Systematische Fehler sind bisher nur mit rückführbaren Meßnormalen (Masse, Länge, Temperatur) als Kontrolleinheit sicher feststellbar. Wegen des Aufwands erfolgt in der Regel eine solche Überprüfung auf systematische Meßfehler nur in größeren Zeitabständen. Derzeit ist lediglich festgelegt, in definierten Zeitabständen die Meßsicherheit der Prüf- oder Meßgeräte durch Kalibrieren oder Eichen sicherzustellen, vgl. auch die DIN-ISO 9000 ff. Ein Kalibrieren der Meßgeräte bzw. Sensoren in kurzen Zeitabständen erhöht zwar die Meßsicherheit, der Aufwand ist aber entsprechend hoch.A large number of measuring and testing devices are known which use sensors - the sensors being tactile or optical or any other sensors - to test or measure a wide variety of substances and materials and products and to display or record the results. Another area of measurement technology concerns quality control. For any measurement, different factors can lead to incorrect measurement results, which in turn can lead to considerable consequential damage. One problem is to recognize systematic measurement errors that z. B. caused by incorrectly set, defective, aged, misaligned or soiled measuring sensors. These are extremely difficult to recognize if the measurement scatter is within the normal range and thus pretends that the measurement is correct. Several measurements in a row do not reveal such systematic errors. It is estimated that in the case of measurement results from test and measuring devices, the undetected or supposed systematic measurement errors are up to 30 or 40% depending on the measuring system and measuring principle. If one suspects that the determined values are not correct, currently with further Control devices the measured values checked and / or the measuring sensors checked with control or calibration units. Systematic errors have so far only been reliably ascertainable with traceable measurement standards (mass, length, temperature) as a control unit. Because of the effort involved, such a check for systematic measurement errors is usually only carried out at longer intervals. At the moment it is only stipulated to ensure the measuring reliability of the test or measuring devices by calibrating or calibrating at defined time intervals, cf. DIN-ISO 9000 ff. Calibrating the measuring devices or sensors at short intervals increases the measurement reliability, but the effort is correspondingly high.
Aufgabe der Erfindung ist es, ein Verfahren zum Erkennen von systematischen Fehlern mittels mit Sensoren ausgestatteten Meßgeräten anzugeben, das es ermöglicht, das Meßsystem zu kontrollieren, also die gewonnenen Meßwerte einzuschätzen und systematische Abweichungen der Meßergebnisse bereits zeitnah während des eigentlichen Meß- bzw. Prüfverfahrens zu erkennen, ohne daß zusätzliche Kontrollmaßnahmen durchgeführt werden müssen, und dabei gleichzeitig auch eine Qualitätskontrolle der Meßobjekte durchzuführen.The object of the invention is to provide a method for recognizing systematic errors by means of measuring devices equipped with sensors, which makes it possible to control the measuring system, that is to say to estimate the measured values obtained and to systematically deviate the measured results promptly during the actual measuring or testing process recognize without additional control measures having to be carried out, and at the same time also carry out a quality control of the test objects.
Diese Aufgabe wird durch ein Verfahren mit den Merkmalen des Anspruchs 1 gelöst. Unter dem Begriff "Objekt" werden hier auch drei oder mehr Einzelobjekte verstanden, die einem Gesamtobjekt entnommen sind, d. h. sie können auch Einzelobjekte einer Charge sein.This object is achieved by a method with the features of claim 1. The term “object” is also understood here to mean three or more individual objects which are taken from an overall object, i. H. they can also be individual objects in a batch.
Vorteilhafte Ausführungen der Erfindung sind den Unteransprüchen zu entnehmen.Advantageous embodiments of the invention can be found in the subclaims.
Durch das erfindungsgemäße Verfahren ist es möglich, systematische Meßfehler von Prüf- oder Meßgeräten mit hoher Wahrscheinlichkeit zu erkennen und damit die Qualität der Überprüfung wesentlich zu erhöhen, da damit auch Aussagen
über die Ergebnisse im Intervall zwischen zwei vorgeschriebenen Kalibrierungen gesichert sind.The method according to the invention makes it possible to identify systematic measurement errors of test or measuring devices with a high degree of probability and thus to significantly increase the quality of the check, since this also includes statements the results are secured in the interval between two prescribed calibrations.
Durch vorteilhafte Ausgestaltungen der Erfindung lassen sich systematische Meßfehler besonders leicht erkennen und einzelnen Meßsensoren zuordnen, die dann ausgetauscht werden können. Weiterhin ist es möglich, bei Überschreiten eines bestimmten Meßwerts oder einer durch ein Gerät mit Prozessor errechneten Größe, ein akustisches oder optisches Signal z. B. auch zur Abschaltung oder zur Unterbrechung weiterer Meßvorgänge, auszusenden, das dann eine Überprüfung erlaubt. Weiterhin ist es bei Messungen möglich, systematische Fehler jedes einzelnen Meßsensors zu erkennen und diesen bei einer Auswertung zu eliminieren bzw. eine ermittelte Konstante zu speichern, so daß automatisch die richtigen Werte trotz der Abweichung angegeben werden. Ein weiterer Vorteil besteht darin, daß auch eine statistische Auswertung ganzer Meßreihen erfolgen kann, die dann zur Steuerung und Regelung und Kontrolle der Qualität eingesetzt werden kann.Advantageous embodiments of the invention make it particularly easy to identify systematic measurement errors and to assign them to individual measurement sensors, which can then be exchanged. Furthermore, it is possible, if a certain measured value or a size calculated by a device with a processor is exceeded, an acoustic or optical signal, for. B. also to switch off or interrupt further measurement processes, which then allows a check. Furthermore, during measurements it is possible to identify systematic errors of each individual measuring sensor and to eliminate them during an evaluation or to save a determined constant so that the correct values are automatically given despite the deviation. Another advantage is that a statistical evaluation of entire series of measurements can be carried out, which can then be used to control and regulate and control the quality.
Verfahren zur statistischen Qualitätsregelung bei Stückgut und Fließprozessen sind bekannt, z. B. aus atp - Automatisierungstechnische Praxis 35 (1993) 10,Methods for statistical quality control in general cargo and flow processes are known, for. B. from atp - automation technology practice 35 (1993) 10,
S. 553 ff.P. 553 ff.
Nachfolgend wird die Erfindung erläutert.The invention is explained below.
Das Verfahren zum Erkennen von systematischen Meßfehlern an mit Meßsensoren ausgestatteten Meßgeräten besteht darin, daß mindestens drei voneinander unabhängige Meßsensoren parallel - vorzugsweise gleichzeitig - dasselbe Objekt messen. Das Verfahren ist unter Verwendung beliebiger Meßsensoren sowohl bei Meß- als auch bei Prüfgeräten anwendbar. Die einzelnen Sensoren werden vorzugsweise innerhalb eines Meßkopfes angeordnet, ohne daß sie sich gegenseitig beeinflussen können, was z. B. bei elektrisch oder magnetisch arbeitenden Meßsensoren zu beachten ist . Hier wäre für die notwendige
Abschirmung zu sorgen. Die einzelnen Meßwerte der mindestens drei Sensoren werden parallel - aus Zeitersparnis möglichst zeitnah oder zeitgleich - ermittelt, gegebenenfalls in Signale umgeformt und einer Auswerteinheit zugeführt, die die einzelnen Meßwerte jedes Sensor getrennt anzeigen kann, aber auch über einen Prozessor den Mittelwert, den Medianwert oder einen Vergleichswert der drei Messungen und die Differenz der einzelnen Meßwerte vom Mittelwert, Medianwert oder dem Vergleichswert ermittelt, anzeigt und gegebenenfalls auch speichert.The method for recognizing systematic measurement errors on measuring devices equipped with measuring sensors is that at least three mutually independent measuring sensors measure the same object in parallel, preferably simultaneously. The method can be used with any measuring sensors as well as with measuring and testing devices. The individual sensors are preferably arranged within a measuring head, without they being able to influence one another. B. with electrical or magnetic measuring sensors. Here would be for the necessary To provide shielding. The individual measured values of the at least three sensors are determined in parallel - as soon as possible or at the same time - to save time, possibly converted into signals and fed to an evaluation unit that can display the individual measured values of each sensor separately, but also via a processor the mean value, the median value or a The comparison value of the three measurements and the difference between the individual measurement values from the mean value, median value or the comparison value are determined, displayed and, if appropriate, also stored.
Bei Überschreiten einer bestimmten Größe einer Abweichung, sei es bei den ermittelten Werten der einzelnen Sensoren oder vom Mittelwert, Medianwert oder vom Vergleichswert kann ein Signal ausgelöst werden, damit gegebenenfalls am selben Objekt in einer veränderten Meßstellung eine weitere Messung vorgenommen werden kann, wobei dann die letzte Messung gegebenenfalls auch gelöscht werden kann, da sie für die Statistik nicht verwendet werden sollte. Durch die Veränderung der Meßstellung (Ortslage) soll ausgeschlossen werden, daß ein technischer oder sonstiger Defekt am Meßobjekt Ursache der abweichenden Werte ist.If a certain size of a deviation is exceeded, be it with the determined values of the individual sensors or from the mean value, median value or from the comparison value, a signal can be triggered so that, if necessary, a further measurement can be carried out on the same object in a changed measuring position, in which case the last measurement can also be deleted if necessary, since it should not be used for statistics. By changing the measuring position (location) it should be excluded that a technical or other defect in the measuring object is the cause of the deviating values.
Wenn mehrere parallele Messungen am gleichen Objekt erfolgen, kann über eine statistische Mittelwert- und/oder Vergleichswertabweichungskontrolle einer Meßreihe eine systematische Meßwertabweichung eines Meßkreises des Prüfgerätes sicher erkannt werden. Hierbei kann sowohl der abweichende Sensor erkannt, als auch die Abweichungsgröße ermittelt werden, mit der es dann möglich ist, die gemessenen Werte richtigzurechnen. Mit dieser Methode ist es möglich, Abweichungen sicher zu erkennen, die wesentlich kleiner als die natürliche Produktmeßwertstreuung sind.If several parallel measurements are carried out on the same object, a systematic measurement value deviation of a measuring circuit of the test device can be reliably detected by means of a statistical mean value and / or comparison value deviation control of a series of measurements. In this case, the deviating sensor can be recognized and the deviation variable can be determined, with which it is then possible to correctly calculate the measured values. With this method, it is possible to reliably detect deviations that are significantly smaller than the natural product measurement scatter.
Die Auswertung kann über eine integrierte Maschinensteuerung oder auch eine HOST-Steuerung erfolgen. Eine spezielle Lösung für einzelne Geräte kann die Trendberechnung mit
logarithmischer Korrektur zurückliegender Mittelwer - und/oder Vergleichswertabweichungen darstellen. Ein Vorteil dieser Methode ist, daß als Zwischenspeicher im Gerät bei Stromausfall oder beim Ausschalten des Geräts keine Pufferbatterie erforderlich ist. Es sind als Statistikwerte lediglich zwei Werte pro Meßsensor und ein Wert für die Anzahl der Messungen zu speichern. Dies kann durch geeignete Maßnahmen ohne Batterie in einem EEPROM erfolgen.The evaluation can be done via an integrated machine control or a HOST control. The trend calculation can be a special solution for individual devices represent logarithmic correction of past mean and / or comparison value deviations. An advantage of this method is that no buffer battery is required as a buffer in the device in the event of a power failure or when the device is switched off. Only two values per measurement sensor and one value for the number of measurements are to be saved as statistical values. This can be done by taking suitable measures without a battery in an EEPROM.
Eine Voraussetzung für eine einwandfreie Funktion des Verfahrens besteht darin, daß eine sichere Parallelmessung möglich ist. In diesem Zusammenhang wäre es auch möglich, den Mikroprozessor mit unterschiedlichen Berechnungsalgorithmen zu parallelisieren, was im vorliegenden Falle heißt, mindestens dreifach auszuführen.A prerequisite for the method to function properly is that a safe parallel measurement is possible. In this context, it would also be possible to parallelize the microprocessor with different calculation algorithms, which in the present case means to execute at least three times.
Eine - nach den Gesetzen der statistischen Prozeßkontrolle - kontrollierbare Größe bei parallelisierten Prüfsensoren stellt die Mittelwert- und/oder Vergleichswertabweichung bei mindestens drei oder auch mehr gleichzeitig im Einsatz befindlichen Sensoren dar. Hat ein Sensor im Verhältnis zu den beiden anderen eine bleibende Mittelwert- und/oder Vergleichswertabweichung, können auf das Meßgerät die statistischen Prozeßkontrollmethoden angewendet werden, wie sie aus der Produktionskontrolle bekannt sind. Damit sind auch die bekannten Regelverfahren für statistische Prozesse anwendbar, wobei deren Einsatz auf die Mittelwert- und/oder Vergleichswertabweichung gerichtet ist.A quantity that can be controlled in parallelized test sensors - according to the laws of statistical process control - represents the mean value and / or comparison value deviation for at least three or even more sensors used at the same time. If one sensor has a permanent mean value and / or or comparison value deviation, the statistical process control methods can be applied to the measuring device, as are known from production control. This means that the known control methods can also be used for statistical processes, the use of which is aimed at the mean and / or comparison value deviation.
Ein großer weiterer Vorteil des Verfahrens ist, daß alles automatisch ablaufen kann und lediglich bei bestimmten Abweichungen ein Signal ausgesandt wird, das den Meßprozeß unterbricht und eine Überprüfung ermöglicht.A great further advantage of the method is that everything can run automatically and only when certain deviations a signal is sent which interrupts the measuring process and enables a check.
Erkennbar sind generell auch systematische Fehler, von parallel hergestellten Probe- bzw. Prüfkörpern, die mit verschiedenen Werkzeugen hergestellt sind, wenn die Probe-
bzw. Prüfkörper jeweils eines von mehreren gleichartigen Herstellungswerkzeugen immer demselben Meßsensor zugeführt werden. Hierdurch ist es möglich, herstellungsbedingte Fehler der Probe- bzw. Prüfkörper durch eines der Werkzeuge zu erkennen. Es können die systematischen Abweichungen gegenüber den Parallelmessungen an den weiteren Meßsensoren die Meßwertabweichungen erkennen lassen oder diese Abweichungen werden dadurch erkannt, daß jeder der parallel hergestellten Probe- bzw. Prüfkörper nacheinander einem zweiten, dritten etc. Meßsensor zugeführt wird. Bei jeweils gleicher Ausrichtung aller Probe- bzw. Prüfkörper in bezug auf die Meßposition und gegebenenfalls auch die Meßrichtung, können auch lokale Defekte detektiert werden, die durch kleine Abweichungen eines der Herstellungswerkzeuge verursacht sind. In diesem Falle kann die defekte Stelle des Herstellungswerkzeugs problemlos ermittelt werden. Die Abweichungen, die durch den Meßsensor bzw. die Meßsensoren ermittelt werden, zeigen dann keine systematischen Meßfehler an, die ihre Ursache im Meßgerät haben, sondern in der fehlerhaften Herstellung, z. B. durch abgenutzte Stanzmesser, durch die die Eigenschaf en der Prüfkörper wesentlich beeinflußt werden.In general, systematic errors can also be identified, from test specimens or test specimens produced in parallel, which are produced with different tools, if the test specimens or test specimens of one of several identical production tools are always fed to the same measuring sensor. This makes it possible to identify manufacturing-related defects in the test specimens or test specimens using one of the tools. The systematic deviations from the parallel measurements on the further measuring sensors can reveal the measured value deviations or these deviations are recognized by the fact that each of the test specimens or test specimens produced in parallel is successively fed to a second, third etc. measuring sensor. With the same orientation of all specimens or test specimens with respect to the measuring position and possibly also the measuring direction, local defects can also be detected which are caused by small deviations in one of the production tools. In this case, the defective location of the manufacturing tool can be easily determined. The deviations, which are determined by the measuring sensor or the measuring sensors, then do not indicate any systematic measuring errors which have their cause in the measuring device, but in the faulty production, e.g. B. by worn punch, by which the properties of the test specimens are significantly affected.
Hierdurch lassen sich auch parallele Produktionsverfahren auf systematische Anlagenfehler überwachen.In this way, parallel production processes can also be monitored for systematic system errors.
Damit können durch dieses Verfahren bisher nicht prüfbare Herstellungsfehler der Meßobjekte als produktunabhängige systematische Fehler sicher erkannt und gezielt identifiziert werden. Kritische Herstellungsprozesse können durch entsprechende Parallelisierung auf systematische Fehler überwacht werden. Hierzu werden an definierten Meßstellen in den parallelen Prozesszweigen parallel messende Prüfsensoren angeordnet .
This method can thus be used to reliably identify and specifically identify manufacturing faults of the measurement objects that have not previously been testable as product-independent systematic faults. Critical manufacturing processes can be monitored for systematic errors by means of parallelization. For this purpose, test sensors measuring in parallel are arranged at defined measuring points in the parallel process branches.
Claims
1. Verfahren zum Erkennen von systematischen Fehlern, insbesondere zur automatischen Störungserkennung bei Qualitätskontrollen, mittels mit Meßsensoren ausgestatteten Meßgeräten mit einer Kombination aus drei einzelnen Sensoren desselben Typs zur Parallelmessung am selben Objekt, wobei die Meßwerte der einzelnen Sensoren erfaßt, der Mittelwert und/oder ein Vergleichswert der Meßwerte der einzelnen Sensoren erstellt wird, dadurch gekennzeichnet, daß die Mittelwertabweichungen der mindestens drei einzelnen Sensoren vom gemeinsamen Mittelwert und/oder die Vergleichswertabweichungen der einzelnen Sensoren ermittelt werden, wobei der Mittelwert und/oder Vergleichswert und/oder die Mittelwert- und/oder Vergleichswertabweichungen während einer Folge von Messungen an verschiedenen Meßorten desselben Objekts einer Menge gleicher Objekte oder an einer Menge gleicher Objekte miteinander verglichen werden, so daß gleichzeitig Fehler des Meßgeräts erkennbar und eine Qualitätskontrolle der Objekte durchgeführt wird.1. Method for detecting systematic errors, in particular for automatic fault detection in quality controls, using measuring devices equipped with measuring sensors with a combination of three individual sensors of the same type for parallel measurement on the same object, the measured values of the individual sensors being recorded, the mean value and / or a Comparative value of the measured values of the individual sensors is created, characterized in that the mean value deviations of the at least three individual sensors from the common mean value and / or the comparative value deviations of the individual sensors are determined, the mean value and / or comparative value and / or the mean value and / or Deviations in comparison values during a sequence of measurements at different measuring locations of the same object, a set of the same objects or on a set of the same objects are compared with one another, so that errors in the measuring device can be recognized at the same time and a quality control e of the objects is carried out.
2. Verfahren nach Anspruch 1, dadurch gekennzeichnet, daß die von den Sensoren erfaßten Meßwerte in Signale umgewandelt werden, die einem Prozessor des Meßgerätes zugeführt, dort verarbeitet, gespeichert und angezeigt, mit im Prozessor eingespeicherten vorgegebenen Werten verglichen werden und bei Überschreiten einer vorgegebenen Abweichungsgröße ein Signal ausgegeben wird. 2. The method according to claim 1, characterized in that the measured values detected by the sensors are converted into signals which are fed to a processor of the measuring device, processed, stored and displayed there, compared with predetermined values stored in the processor and when a predetermined deviation variable is exceeded a signal is output.
3. Verfahren nach Anspruch 2, dadurch gekennzeichnet, daß die von den Sensoren erfaßten Meßwerte im Prozessor einer Trendberechnung der Mittelwert- und/oder Vergleichswertabweichungen unterzogen werden und darauf beruhend eine logarithmische Korrektur früherer Mittelwert- und/oder Vergleichsabweichungen durchgeführt wird.3. The method according to claim 2, characterized in that the measured values detected by the sensors in the processor are subjected to a trend calculation of the mean and / or comparison value deviations and, based on this, a logarithmic correction of previous mean value and / or comparison deviations is carried out.
4. Verfahren nach Anspruch 2 oder 3, dadurch gekennzeichnet, daß die gemessenen und verarbeiteten Werte jedes Sensors im Prozessor statistisch ausgewertet und zusätzlich zur Steuerung der Qualität der Objekte verwendet werden. 4. The method according to claim 2 or 3, characterized in that the measured and processed values of each sensor in the processor are statistically evaluated and used in addition to control the quality of the objects.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE19506388.0 | 1995-02-23 | ||
| DE1995106388 DE19506388C1 (en) | 1995-02-23 | 1995-02-23 | Method for recognizing systematic errors, in particular for the automatic detection of malfunctions in quality controls, by means of measuring devices equipped with measuring sensors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO1996026511A1 true WO1996026511A1 (en) | 1996-08-29 |
Family
ID=7754880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/EP1996/000744 WO1996026511A1 (en) | 1995-02-23 | 1996-02-22 | Method of detecting systematic defects in quality control processes |
Country Status (2)
| Country | Link |
|---|---|
| DE (1) | DE19506388C1 (en) |
| WO (1) | WO1996026511A1 (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE19754847A1 (en) * | 1997-12-10 | 1999-06-17 | Buna Sow Leuna Olefinverb Gmbh | Uni and multivariate method of data analysis in online process control using fuzzy logic |
| DE10340793B4 (en) * | 2003-09-02 | 2006-07-27 | Deutsches Zentrum für Luft- und Raumfahrt e.V. | Method for the computational determination of systematic errors of processed time series measurement data |
| US8428909B2 (en) | 2007-09-20 | 2013-04-23 | Siemens Industry, Inc. | Use of statistics to determine calibration of instruments |
| WO2009067051A1 (en) * | 2007-11-21 | 2009-05-28 | Volvo Construction Equipment Ab | Method for calibrating sensors |
| RU2466369C2 (en) * | 2007-11-21 | 2012-11-10 | Вольво Констракшн Эквипмент Аб | Method of calibrating sensors |
| US8655624B2 (en) * | 2008-06-11 | 2014-02-18 | Fluke Corporation | System and method for objective self-diagnosis of measurement device calibration condition |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0349788A1 (en) * | 1988-06-09 | 1990-01-10 | IVECO FIAT S.p.A. | Method and apparatus for controlling the regeneration of at least one particulate filter installed in a diesel engine |
| US4926364A (en) * | 1988-07-25 | 1990-05-15 | Westinghouse Electric Corp. | Method and apparatus for determining weighted average of process variable |
| DE3842032C1 (en) * | 1988-12-14 | 1990-05-31 | Juergen Dr. 8028 Taufkirchen De Zorn | Ball-controlled 3-coordinate probe |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4580234A (en) * | 1982-08-02 | 1986-04-01 | Nippon Business Consultant Co., Ltd. | Measurement system of enhanced reliability |
| DD262778A3 (en) * | 1987-04-21 | 1988-12-14 | Inst Stahlbeton | CIRCUIT ARRANGEMENT FOR LINEARIZING AND NORMING THE NOMINAL FIELD OF A CHANGE VOLTAGE MEASUREMENT SIGNAL GENERATED IN A MEASUREMENT TRANSMITTER |
| DE4022954A1 (en) * | 1990-07-19 | 1992-01-23 | Bodenseewerk Geraetetech | Fault detector, using redundant signal emitter outputs - user to determines fault probabilities to eliminate stochastic faults |
| DE4039647A1 (en) * | 1990-12-12 | 1992-06-17 | Rolf Wendler | Measurement value processing system for rolling mill - uses multiple channel measurement value acquisition system between processing system and rolling mill |
| DE4244014A1 (en) * | 1992-12-24 | 1994-07-07 | Bodenseewerk Geraetetech | Detection and identification of sensor errors e.g. for aircraft controller |
| DE4324800C2 (en) * | 1993-07-23 | 1997-05-22 | Olaf Dr Ing Schnabel | Device for determining defects of high quality surfaces |
-
1995
- 1995-02-23 DE DE1995106388 patent/DE19506388C1/en not_active Expired - Fee Related
-
1996
- 1996-02-22 WO PCT/EP1996/000744 patent/WO1996026511A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0349788A1 (en) * | 1988-06-09 | 1990-01-10 | IVECO FIAT S.p.A. | Method and apparatus for controlling the regeneration of at least one particulate filter installed in a diesel engine |
| US4926364A (en) * | 1988-07-25 | 1990-05-15 | Westinghouse Electric Corp. | Method and apparatus for determining weighted average of process variable |
| DE3842032C1 (en) * | 1988-12-14 | 1990-05-31 | Juergen Dr. 8028 Taufkirchen De Zorn | Ball-controlled 3-coordinate probe |
Also Published As
| Publication number | Publication date |
|---|---|
| DE19506388C1 (en) | 1996-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DE2843810C3 (en) | Method and device for tightening screw connections and for quality control | |
| DE102017211737B4 (en) | Monitoring device and method for monitoring a system | |
| EP1269762B2 (en) | Protective device for safeguarding a dangerous area and method for verifying the functional reliability of such a device | |
| DE68903400T2 (en) | DEVICE AND METHOD FOR INSPECTING THE SIDE OF A TIRE. | |
| EP1472890B1 (en) | Method for verifying the functional reliability of an image sensor and device comprising an image sensor | |
| DE102016112650A1 (en) | AUTOMATED ROBOT MOUNTING SYSTEM | |
| EP0613732B1 (en) | Method and device for testing the working capacity of inspection units in bottle inspecting machines | |
| EP0416370B1 (en) | Method and device for detecting and indentifying sensor errors | |
| DE102020128435A1 (en) | Diagnostic device | |
| DE2653590C3 (en) | Device for determining defects in two-dimensional patterns, in particular in photo masks | |
| DE19506388C1 (en) | Method for recognizing systematic errors, in particular for the automatic detection of malfunctions in quality controls, by means of measuring devices equipped with measuring sensors | |
| DE2704983A1 (en) | METHOD FOR AUTOMATICALLY DETECTING DEFECTS IN THE SURFACE OR DIMENSIONS OF AN OBJECT AND DEVICE FOR EXECUTING THE METHOD | |
| EP2815230B1 (en) | Inspection method | |
| DE102015108851B4 (en) | Method for operating a plurality of measuring machines and overall apparatus comprising at least two measuring machines | |
| DE102009012887B4 (en) | Method for checking incorrect installation of vehicle sensors | |
| EP4130898B1 (en) | Safety system and method with a safety system | |
| EP4295202B1 (en) | Method and system for measuring components and program | |
| DE102016121831B3 (en) | Method for checking the dimensional accuracy of a part gripper | |
| DE102019114537A1 (en) | OPTOELECTRONIC SENSOR COMPONENT FOR LIGHT MEASUREMENT WITH BUILT-IN REDUNDANCY | |
| DE102019205812A1 (en) | Defect identification in a manufacturing or machining process for a component, in particular for a control board, with a sensor carrier part | |
| DE102019209293A1 (en) | Generation of an output signal value from sensor signal values from at least three redundant sensors | |
| DE102024200772B3 (en) | Method for determining a state of a test bench, computer program, control device and test bench | |
| DE19819264A1 (en) | Method for testing an integrated circuit arrangement and integrated circuit arrangement therefor | |
| EP0425897B1 (en) | Method for operating a control system | |
| DE102023112283A1 (en) | Method for determining state information of a solar radiation sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| AK | Designated states |
Kind code of ref document: A1 Designated state(s): JP US |
|
| AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE |
|
| DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
| 122 | Ep: pct application non-entry in european phase |