WO2021046759A1 - Apparatus, system and method for quality control of polymer pellets and method for controlling production line for producing polymer pellets - Google Patents

Apparatus, system and method for quality control of polymer pellets and method for controlling production line for producing polymer pellets Download PDF

Info

Publication number
WO2021046759A1
WO2021046759A1 PCT/CN2019/105387 CN2019105387W WO2021046759A1 WO 2021046759 A1 WO2021046759 A1 WO 2021046759A1 CN 2019105387 W CN2019105387 W CN 2019105387W WO 2021046759 A1 WO2021046759 A1 WO 2021046759A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
pellets
polymer
defect
defects
Prior art date
Application number
PCT/CN2019/105387
Other languages
French (fr)
Inventor
Qiang Shen
Yangfan GUO
Yanqin PU
Aihua Chen
Tingting JIAO
Jerico Jayson UY
Hans-Bernhard Hauertmann
Original Assignee
Covestro Deutschland Ag
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Covestro Deutschland Ag filed Critical Covestro Deutschland Ag
Priority to PCT/CN2019/105387 priority Critical patent/WO2021046759A1/en
Publication of WO2021046759A1 publication Critical patent/WO2021046759A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/85—Investigating moving fluids or granular solids
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8806—Specially adapted optical and illumination features
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N2021/1765—Method using an image detector and processing of image signal
    • G01N2021/177—Detector of the video camera type
    • G01N2021/1776—Colour camera
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N2021/8411—Application to online plant, process monitoring
    • G01N2021/8416—Application to online plant, process monitoring and process controlling, not otherwise provided for
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/85—Investigating moving fluids or granular solids
    • G01N2021/8592—Grain or other flowing solid samples
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8806—Specially adapted optical and illumination features
    • G01N2021/8845—Multiple wavelengths of illumination or detection
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
    • G01N21/251—Colorimeters; Construction thereof

Definitions

  • the invention relates to an apparatus for quality control of polymer pellets, comprising at least one optical means to record information about the appearance of at least one polymer pellet, and control means.
  • the invention further relates to a system, a method for quality control of polymer pellets, and a method for controlling a production line for producing polymer pellets.
  • Polymer materials are generally produced under the form of granulates or pellets, to be then transformed into so-called finished products.
  • pellets will be used for designating polymer materials that can be employed as input material for producing finished polymer products in general, for instance as granulates.
  • pellets is meant to also include recycled polymer material.
  • pellets are aimed to be later transformed, they generally are used as input material for standardized processing steps, in order to produce finished products. Therefore, pellets are subject to quality obligations in order to sustain the reliability of the later production processes.
  • the presence of defects such as specks in pellets can lead to color inhomogeneity or holes and to increases of scrap material. Thus, unscheduled shutdowns can occur. Therefore, it is of high economical interest that pellets are defect free or at least have the less possible defects.
  • defects can hardly be completely avoided, even in an automatized pellet production.
  • defects can be encountered in raw input material, or be caused by impurities in production machines and production environment, or by uncontrolled thermal reaction or even degradation of the raw input material during the process of the pellet production.
  • BSI black speck index
  • the preparation of pellets prior to the analysis corresponds to an additional step between pellets as produced and the material that is in fact analyzed, so that the results of the analysis may be distorted without possible traceability.
  • a defect originally present in pellets as produced may for instance break down into smaller specks by the shear upon molding or transformation into sample plates or foils. Also, some specks (e.g. brown) dissolves upon molding. As an overall result, the reliability of investigation is weakened. Thus, the data obtained from the measurements do not allow a correlation between detected defects and production parameters, thus rendering adaption of production parameters difficult. In order to rectify production the primary objective, it is necessary to avoid or minimize defects in the produced pellets.
  • the object of the present invention is therefore to enable standardizing, simplifying and accelerating quality control of polymer pellets, and to improve process control in a production line for polymer pellets.
  • an apparatus for quality control of polymer pellets comprising at least one optical means to record information about the appearance of at least one polymer pellet, and control means, characterized in that the control means are configured both to detect the presence of one or more defects in the at least one polymer pellet and to identify the one or more detected defects, wherein identifying a defect is at least partially based on a comparison of the recorded information with one or more threshold values and/or reference information indicative of previous recorded information of one or more pellets having such a defect, and in that the control means are configured to output information about the quality of the one or more detected defects, wherein the output information is indicative of the identity of the one or more detected defects.
  • identifying a defect is at least partially based on a comparison of the recorded information with one or more threshold values and/or reference information indicative of previous recorded information of one or more pellets having such a defect;
  • the apparatus and the method according to the invention enable both detecting and identifying eventual defects directly in individual pellets. Accordingly, no further transformation of the material to be analyzed is necessary, thus minimizing possible distortions between the real state of produced pellets and the analysis results.
  • a further advantage offered by the present invention is, due to a direct analysis of pellets, the possibility of analyzing a great amount of pellets with reduced analysis duration in comparison to known methods, where an additional preparation step of the material to be analyzed is necessary.
  • the apparatus and method according to the invention allow for a standardization of quality control, thus allowing improved documentation and traceability regarding the quality parameters of produced pellets. In turn, this gives not only an advantage in dealing with customer complaints but also allows for deriving correlations between process parameters and pellet appearance.
  • identifying the one or more detected defects it is possible to use the information about the identity of the detected defects for controlling production devices in a control loop, thus improving the overall quality of produced pellets. Also, the automatized identification of defects allows for a faster analysis and thus an adequately reactive process control of production devices.
  • Identification based on a comparison of the recorded information with one or more threshold values and/or reference information allows for accurate differentiation between different kinds of already known defects, such as inclusions, color aberrations, foreign matter, etc.
  • taking information indicative of previous recorded information of one or more defective pellets into account permit a quality control adapted to actually encountered defects and can be executed independently from unrealistic hypothetical values, which are often employed for known quality control methods and apparatuses.
  • Information about the appearance of individual pellets can be any information that can be used to describe the appearance of individual pellets or a group of pellets, such as for instance physical or aesthetical properties that are visible under exposure to electromagnetic waves, such as visible light, ultra-violets, infrared light, and similar.
  • information can be a shape, a size, a diameter, a roundness, a roughness, a convexity, a color, a dimension, and/or a position, whereby the disclosure is not limited thereto.
  • the at least one provided optical means is preferably configured in accordance with the kind of information to record.
  • the optical means can be configured to emit infrared light and to record infrared light reflected by pellets, in order to record information about the appearance of individual pellets, which is only detectable under infrared light.
  • the optical means may for instance be configured as a CCD detector (charge-coupled device) , e.g. for recording light intensities over a surface of pixels and with a light source emitting light in the visible part of light spectrum.
  • the disclosed control means are preferably provided to perform a method according exemplary embodiment of the present invention.
  • the control means may for instance comprise at least one processor for executing computer program code for performing the required functions, at least one memory storing the program code, or both.
  • the control means may also comprise at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause an apparatus, for instance the apparatus according exemplary embodiments of the present invention, at least to perform and/or to control the method according to invention, as disclosed in this specification.
  • the control means can be part of a network, for instance a server cloud.
  • One or more pieces of information or one or more threshold values, with which the control means compare recorded information are preferably of the same kind as the recorded information.
  • the control means may for instance be configured to transform stored information or threshold values in order to enable a comparison with the recorded information.
  • control means may for instance compare recorded information with information about pellets or threshold values related to pellets presenting at least one defect.
  • the analyzed pellet will be judged (e.g. evaluated) as defective if the comparison results in a high similarity.
  • recorded information may for instance be compared with information about pellets or threshold values relating to pellets, which pellets are defect free. In this later case, the analyzed pellet will preferably be judged defective if the comparison results in a high dissemblance.
  • Output information may for instance be statistics about defect distribution over a volume or weight of pellets, and/or historic information indicative of a frequency of occurrence of one or more detected defects of a particular kind, to name but one non-limiting example.
  • the apparatus comprises output means for outputting information about the quality of one or more detected defects.
  • Output means may for instance be communication means provided to communicate with one or more electronic devices or within a digital network, or a user interface to display information.
  • Output information may for instance be expressed as machine readable information for use by further devices, such as for instance control means of devices of a production line. Alternatively or additionally, output information can be presented as to be read by a human operator.
  • a computer program is disclosed, the computer program when executed by a processor causing an apparatus, for instance the control means of such an apparatus, to perform and/or control the actions respectively steps of the method according to the present invention.
  • the computer program may be stored on computer-readable storage medium, in particular a tangible and/or non-transitory medium, as for instance in the control means.
  • control means are configured to extract at least one value for at least one characteristic parameter for the appearance of at least one polymer pellet from the recorded information, the control means are configured to compare the at least one extracted value with one or more predetermined threshold values for a characteristic parameter for the appearance of polymer pellets, wherein each respective value range of one or more value ranges is correlated with a predetermined defect category, and the control means are configured to assign a predetermined defect category to a detected defect depending on the result of the comparison.
  • the method further comprises:
  • control means are preferably configured to select value data among a data volume and/or to transform a plurality of information into representative information.
  • control means may for instance be configured to calculate a difference between intensities assigned respectively to neighbor pixels of a CCD chip, in order to determine a contrast.
  • Characteristic parameters preferably correspond to parameters, on the basis of which a defect can be detected or identified.
  • a characteristic parameter can be a hue, a contrast sharpness, a transparency, a size, a shape or similar.
  • the predetermination of defect categories is based on defects observed previously by one or more analyzed pellets, and/or in the production line. Hence, several defects can be collected, which present different features. Further, a category may for instance be determined for each kind of defect of the one or more defects to be detected. As an example, categories corresponding respectively to black specks and brown specks can be determined on the basis of different color ranges. This allows a determination of defects categories and thus an identification of defects that relies on actually encountered defects, rather than on ideal references that can eventually be far from reality.
  • the expanse of a predetermined defect category can preferably be adjusted in accordance with measurement tolerances of the optical means.
  • the definition of categories may for instance occur in accordance with variations observed for characteristic parameters related to knowns defects of the same kind.
  • the recorded information about the appearance of the at least one polymer pellet represents an image.
  • the recorded information about the appearance of the at least one polymer pellet represents an image.
  • the image preferably comprises information about e.g. one or more intensities and/or saturations, which may for instance be respectively assigned to pixels ordered into a matrix.
  • the image can comprise information distributed over two or more spatial dimensions.
  • the at least one characteristic parameter for the appearance of the at least one polymer pellet corresponds to a hue, a saturation, an intensity, a size and/or a geometrical shape of the at least one polymer pellet.
  • the at least one characteristic parameter for the appearance of the at least one polymer pellet corresponds to a hue, a saturation, an intensity, a size and/or a geometrical shape of the at least one polymer pellet.
  • anomalies such as aggregation of several pellets into doubles or triplets, or defects having a color differing from the ground color of the pellets to be produced can be identified. Also, this permits evaluating contrasts, transparency, and color changes in individual pellets. Moreover, a size distribution, and/or shape distribution over a predetermined surface on an image for instance, or a spatial position in a stream of pellets can be determined on the basis of the characteristic parameters.
  • each characteristic parameter there is preferably at least one threshold value provided, as for instance a percentage range for saturation or intensity, or a minimum value expressed in ⁇ m for a spatial dimension.
  • a hue can be defined by a vector with coordinates on axis corresponding to different colors. This can be done according to the system of the CIELAB color space (also known as CIE L*a*b*or sometimes abbreviated as simply "Lab” color space) which is a color space defined by the International Commission on Illumination (CIE) in 1976.
  • CIELAB color space also known as CIE L*a*b*or sometimes abbreviated as simply "Lab” color space
  • CIE International Commission on Illumination
  • a correspondence table can be used, in order to correlate a value to a specific geometrical shape (for instance: “1” for square, “2” for circular... ) .
  • a specific geometrical shape for instance: “1” for square, “2” for circular... .
  • control means are configured to trace a detected defect.
  • control means are configured to trace a detected defect.
  • a defect can be detected once, and followed over the duration of the analysis, in particular avoiding redundant occurrence of the same defect.
  • successively recorded information is preferably used.
  • a correlation can be established between characteristic parameters extracted from recorded information about a plurality of pellets with respective first positions and characteristic parameters extracted from recorded information about pellets moved to respective second positions. In this way, successive identical characteristic parameters can be correlated to recognize each detected defect over successive recordings.
  • two optical means are provided, wherein a first optical means is configured to record information about transparency and/or color changes within individual polymer pellets in a stream of polymer pellets, wherein a second optical means is configured to record information about the amount, the spatial distribution, the size, and/or the shape of individual polymer pellets in a stream of polymer pellets, and wherein the control means are further configured both to detect the presence of one or more defects in the at least one polymer pellets and to identify the one or more detected defects, wherein the identification is based on the recorded information as recorded by the two optical means.
  • At least one piece of information about transparency and/or color changes within individual polymer pellets in a stream of polymer pellets is recorded by first optical means, and in that at least one piece of information about the amount, the spatial distribution, the size, and/or the shape of individual polymer pellets in a stream of polymer pellets is recorded by second optical means.
  • analysis tasks can be distributed to each optical means depending on the respective configuration possibilities of each optical means.
  • using lower cost black and white camera is sufficient.
  • overall costs for the apparatus and for the method can be reduced.
  • control means are configured to control both the first optical means and the second optical means.
  • information recorded by the respective optical means are transmitted to the control means in order to achieve a conjoint analysis based on both recorded information.
  • the apparatus can be provided with a single optical means, for instance in form of a camera, which is sensitive enough to record information usable both to detect and identify defects with a reliable confidence level. In this way the apparatus can be designed in a compact and space saving way.
  • a system comprising at least two apparatuses according to any one of the claims 1 to 6, wherein the two apparatuses are configured to perform a method according to any one of claims 7 to 12 together at least partially jointly.
  • each apparatus of the system comprises control means, which may for instance be configured in a dedicated way for the respective optical means of each apparatus.
  • Each apparatus of the system may for instance be considered as an entity, which can be independently placed at a respective position on a production line.
  • a first apparatus can be arranged after a device of the production line, which is susceptible to cause inclusion defects, while a second apparatus, which is able to detect malformations of individual pellets, can be placed at the end of the production line, after pellets where subject to temperature variation susceptible to cause aggregation of several pellets together.
  • one control means among the control means provided in the system is configured to coordinate the system.
  • the control means of a first apparatus can be configured to receive analysis results from the control means of a second apparatus, and to take the received analysis results onto account to issue overall analysis results.
  • the system can be provided with higher-level control means, which allow a coordination of the respective control means for each apparatus of the system.
  • the higher-level control means can take the form of a server, for instance.
  • the apparatus, the method and the system according to the invention can be arranged or executed at the end of a production line for producing polymer pellets. This way the quality of the pellets can be controlled directly after the pellet production and the information issued from the quality control used for controlling the production line.
  • the apparatus, the method and the system according to the invention can be arranged or executed at the entrance of a production line for polymer intermediate products or finished products based on the transformation of pellets. This way, quality of pellets can be checked before a transformation processing step and the information issued from the quality check employed to adapt processing parameters on the transformation line.
  • the above named object is further achieved according to the invention with a method for controlling a production line for producing polymer pellets, wherein the production line comprises at least one apparatus, comprising:
  • Such a method allows for adapting processing parameters in a reactive way relative to defects that are eventually present in produced pellets and in accordance with parameters of apparatuses that are actually employed in the production line.
  • the achieved reactivity towards eventually upcoming defects enables counteracting unpredictable arising of events on the production line, which may cause the presence of defects in produced material.
  • the amount of produced material with off-specification characteristics can be reduced, leading to overall qualitative improvement of the production process.
  • the method allows for a fast and reliable quality control, as for instance the analysis of about 1 kg of pellets every 15 to 20 min. Therefore, a fast adaptation of process parameters is possible and less off-specification material is produced.
  • the process parameter associated to at least one apparatus of the production line is preferably representative of an aspect of the apparatus that may correlate with the arising of defects in pellets.
  • a parameter may for instance be a feeding speed for raw material at the entrance of the production line, a temperature, or a fluid that comes into contact with pellets in production or the likes, to name but a few non-limiting examples.
  • Controlling process parameters is preferably achieved by modifying a value that is used for executing a processing step in the production line.
  • a preset temperature can be stored in control means of an apparatus of the production line and used for a temperature treatment processing step. If a defect that may be caused by a too high temperature is identified, the preset temperature can be decreased and set to a lower temperature, in order to reduce probability of arising for defect of the detected kind.
  • the at least one piece of information relative to an identified defect is a possible origin for the identified defect.
  • Possible origins for defects can correspond to an excessive temperature or residence time, the used of specific metallic parts in an apparatus, which abrasion or wear can cause the presence of black specks, to name but one non-limiting example.
  • information relative to an identified defect and the process parameters are expressed under the form of values or Boolean data type, so as to facilitate comparison with each other.
  • Fig. 1 an exemplary apparatus for quality control of polymer pellets
  • Fig. 2 an exemplary flow chart for a method for quality control of polymer pellets
  • Fig. 3 an exemplary representation in table form for predetermined defect categories
  • Fig. 4 another exemplary representation in table form for predetermined defect categories.
  • Fig. 5 another exemplary representation in table form for predetermined defect categories.
  • Fig, 1 is a schematic representation of an apparatus 2 for quality control of polymer pellets,
  • the represented apparatus comprises first optical means 4, second optical means 6, first lighting means 8, second lightning means 10, control means 12 and separation means 14. Separation means 14 are arranged at the entrance of the apparatus 2, in order to separate a plurality of polymer pellets 16 into a sequence of individual pellets 18.
  • the first optical means 4 is a high-resolution 3-CCD-chip color camera, while the second optical means 6 is a CCD Line Scan sensor Camera.
  • the first lightning means 8 comprises two light emitting sources 20, 22, arranged respectively sidewise relatively to the first optical means 4, wherein the first optical means 4 and the first lightning means 8 are oriented in the same directions.
  • the second optical means 6 and the second lightning means 10 are arranged opposite to each other. Separation means 14, first optical means 4 and second optical means 6 are arranged one after the other, so that an inputted volume of pellets 16 will first be separated, then pass by the first optical means and pass by the second optical means.
  • First optical means 4 are configured to create a signal corresponding to intensities and saturations recorded for each pixel of the 3-CCD-chip and to transmit the signal to the control means 12 via a first communication connection 24.
  • the second optical means 6 are configured similarly, wherein the signal is transmitted via a second communication connection 26, and wherein the created and transmitted signal corresponds to information recorded by the CCD line scan camera.
  • the control means 12 are configured to receive and analyze signals from both optical means 4, 6. To do this, the control means 12 are provided with data storage means 28, on which a computer program for image analysis is stored, and a processor 30. The computer program allows for comparison between intensity and saturation values recorded for respective pixels, which correspond to respective positions on the 3-CCD chip. On the data storage means 28, a catalogue of threshold values corresponding to characteristic parameters representative of previous recorded information about pellets having known defects is stored.
  • the software program installed on the control means 12 is designed for executing a method according to the flow chart represented in Fig. 2.
  • Fig. 2 shows an exemplary flow chart 200 of a method for quality control of polymer pellets according to an example embodiment of the present invention.
  • a first step 201 a plurality of polymer pellets 16 is provided at the entrance of the apparatus 2, which are then separated in a second step 202 into a sequence of individual pellets 18.
  • the hence separated pellets 18 pass by the first lightning means 8 and first optical means 4, where first optical means 4 record intensity and saturation values for light reflected by light-exposed individual pellets 18, wherein the respective recorded values are associated to respective positions on a pixel matrix (third step 203) .
  • contrast changes are recorded by the second optical means 6, wherein respective contrast changes are associated to respective positions on a pixel matrix.
  • the recorded values are transmitted from the respective optical means 4, 6 to the control means 12 via the first communication connection 24 and via the second communication connection 26.
  • the control means 12 then extracts eventual color changes from the recorded intensity and saturation values and pellet shapes, pellet sizes and amount of pellets from the recorded contrast changes.
  • threshold values stored in the data storage means 28 of the control means 12 are taken into account for a comparison with the extracted color changes and pellet shapes, wherein the threshold values define ranges correlated to predetermined defect categories. Examples for threshold values and corresponding defect categories are given in Fig. 3, 4 and 5.
  • the control means 12 judges about the presence or absence of defects, and assigns in a eighth step 208 a predetermined defect category to each detected defect. Thereafter, the control means 12 takes the amount of detected defects and the assigned categories into account for determining statistics about the quality of the analyzed pellets. As a further processing step, the control means 12 outputs the assigned categories and the recorded information under the form of statistics in a ninth step 209.
  • Fig. 3 shows an exemplary representation in table form for predetermined defect categories.
  • designations for five categories are given and further, corresponding threshold values for characteristic parameters, such as hue, saturation, intensity and minimum size, are given in four juxtaposed columns.
  • the given categories correspond to different kinds of defects, such as speck inclusions, defective overall color of a pellet (third row of the table) or color anomalies.
  • the minimum size allows for differentiating between defects with reduced size in comparison to the size of a pellet and defects that are present over a whole pellet, as for instance an overall pellet color, which differs from a target transparency.
  • Fig. 4 shows another exemplary representation in table form for predetermined defect categories.
  • the given threshold values permit to categorize detected defects depending on their respective size.
  • Fig. 5 shows another exemplary representation in table form for predetermined defect categories with three columns. In the left columns category designations are given and in the right column the corresponding nature of defect. Categories are differentiated from each other by criteria given in the middle column. The criteria are expressed as result of a comparison between a shape, a sized or a number of defects extracted from the information about pellets and predetermined target size or shape stored in the control means 12. Preferably, the target values are given with respective tolerance ranges.
  • the statement of a feature comprises at least one of the subsequently enumerated features is not mandatory in the way that the feature comprises all subsequently enumerated features, or at least one feature of the plurality of the subsequently enumerated features. Also, a selection of the enumerated features in any combination or a selection of only one of the enumerated features is possible. The specific combination of all subsequently enumerated features may as well be considered. Also, a plurality of only one of the enumerated features may be possible.

Abstract

Provided is an apparatus (2) for quality control of polymer pellets，comprising at least one optical means (4,6) to record information about the appearance of at least one polymer pellet (16, 18), and control means (12), characterized in that the control means (12) are configured both to detect the presence of one or more defects in the at least one polymer pellet (16,18) and to identify the one or more detected defects， wherein identifying a defect is at least partially based on a comparison of the recorded information with one or more threshold values and/ or reference information indicative of previous recorded information of one or more pellets having such a defect, and in that the control means (12) are configured to output information about the quality of one or more detected defects, wherein the output information is indicative of the identity of the one or more detected defects．Also provided are a method for quality control of polymer pellets，a system，and a method for controlling a production line for producing polymer pellets．

Description

Apparatus, system and method for quality control of polymer pellets and method for controlling a production line for producing polymer pellets
TECHNICAL FIELD OF THE INVENTION
The invention relates to an apparatus for quality control of polymer pellets, comprising at least one optical means to record information about the appearance of at least one polymer pellet, and control means. The invention further relates to a system, a method for quality control of polymer pellets, and a method for controlling a production line for producing polymer pellets.
BACKGROUND
Polymer materials are generally produced under the form of granulates or pellets, to be then transformed into so-called finished products. In the following, the term “pellets” will be used for designating polymer materials that can be employed as input material for producing finished polymer products in general, for instance as granulates. The term pellets is meant to also include recycled polymer material.
As pellets are aimed to be later transformed, they generally are used as input material for standardized processing steps, in order to produce finished products. Therefore, pellets are subject to quality obligations in order to sustain the reliability of the later production processes. As an example, for the production of films or sheets-especially in light-colored or clear plastics-, the presence of defects such as specks in pellets can lead to color inhomogeneity or holes and to increases of scrap material. Thus, unscheduled shutdowns can occur. Therefore, it is of high economical interest that pellets are defect free or at least have the less possible defects.
However, defective pellets can hardly be completely avoided, even in an automatized pellet production. For instance, defects can be encountered in raw input material, or be caused by impurities in production machines and production environment, or by  uncontrolled thermal reaction or even degradation of the raw input material during the process of the pellet production.
Methods are known from the state of the art for controlling quality of polymer pellets. For instance, detection of defects or abnormalities in pellets is commonly carried visually by a human operator. The operator is able to detect the presence of different defects, such as so-called black and brown specks. However, this analysis is subjective, since depending on the subjective senses of the human operator and thus varying from operator to operator. Also the fatigue factor of the operator must be taken into account, and the fact that the human eye has its limits in recognizing poor color contrasts and some wavelength ranges of light spectrum. Accordingly, this method should be improved in order to achieve a more standardized and faster quality control, and thus a sustainable reliability in the quality of the controlled pellets.
Known alternatives, such as machine control, mostly require a preparation of the pellets to be analyzed, as for instance melting pellets into sample plates or foils, in order to enable a comparison with known reference values. Such known methods allow for detecting the presence of defects, and thus for counting the number of defects. It is necessary to transform the volume of pellets into a sample, and then, it is eventually enabled to determine e.g. a black speck index (BSI) , based on the class size of detected black and brown speck.
However, the preparation of pellets prior to the analysis corresponds to an additional step between pellets as produced and the material that is in fact analyzed, so that the results of the analysis may be distorted without possible traceability. A defect originally present in pellets as produced may for instance break down into smaller specks by the shear upon molding or transformation into sample plates or foils. Also, some specks (e.g. brown) dissolves upon molding. As an overall result, the reliability of investigation is weakened. Thus, the data obtained from the measurements do not allow a correlation between detected defects and production parameters, thus rendering adaption of production parameters difficult. In order to rectify production  the primary objective, it is necessary to avoid or minimize defects in the produced pellets.
SUMMARY OF SOME EXEMPLARY EMBODIMENTS
The object of the present invention is therefore to enable standardizing, simplifying and accelerating quality control of polymer pellets, and to improve process control in a production line for polymer pellets.
This object is achieved according to the invention with an apparatus for quality control of polymer pellets, comprising at least one optical means to record information about the appearance of at least one polymer pellet, and control means,, characterized in that the control means are configured both to detect the presence of one or more defects in the at least one polymer pellet and to identify the one or more detected defects, wherein identifying a defect is at least partially based on a comparison of the recorded information with one or more threshold values and/or reference information indicative of previous recorded information of one or more pellets having such a defect, and in that the control means are configured to output information about the quality of the one or more detected defects, wherein the output information is indicative of the identity of the one or more detected defects.
The above mentioned object is further achieved according to the invention with a method for quality control of polymer pellets, comprising:
- providing at least one polymer pellet;
- recording at least one piece of information about the appearance of the at least one provided polymer pellet;
- detecting the presence of one or more defects in the at least one polymer pellet and identifying the one or more detected defects, wherein identifying a defect is at least partially based on a comparison of the recorded information with one or more threshold values and/or reference information indicative of previous recorded information of one or more pellets having such a defect; and
- outputting information about the quality of one or more detected defects, wherein the output information is indicative of the identity of the one or more detected defects.
The apparatus and the method according to the invention enable both detecting and identifying eventual defects directly in individual pellets. Accordingly, no further transformation of the material to be analyzed is necessary, thus minimizing possible distortions between the real state of produced pellets and the analysis results.
A further advantage offered by the present invention is, due to a direct analysis of pellets, the possibility of analyzing a great amount of pellets with reduced analysis duration in comparison to known methods, where an additional preparation step of the material to be analyzed is necessary.
Moreover, the apparatus and method according to the invention allow for a standardization of quality control, thus allowing improved documentation and traceability regarding the quality parameters of produced pellets. In turn, this gives not only an advantage in dealing with customer complaints but also allows for deriving correlations between process parameters and pellet appearance.
By identifying the one or more detected defects, it is possible to use the information about the identity of the detected defects for controlling production devices in a control loop, thus improving the overall quality of produced pellets. Also, the automatized identification of defects allows for a faster analysis and thus an adequately reactive process control of production devices.
Identification based on a comparison of the recorded information with one or more threshold values and/or reference information allows for accurate differentiation between different kinds of already known defects, such as inclusions, color aberrations, foreign matter, etc. In particular, taking information indicative of previous recorded information of one or more defective pellets into account permit a  quality control adapted to actually encountered defects and can be executed independently from unrealistic hypothetical values, which are often employed for known quality control methods and apparatuses.
Information about the appearance of individual pellets can be any information that can be used to describe the appearance of individual pellets or a group of pellets, such as for instance physical or aesthetical properties that are visible under exposure to electromagnetic waves, such as visible light, ultra-violets, infrared light, and similar. As an example such information can be a shape, a size, a diameter, a roundness, a roughness, a convexity, a color, a dimension, and/or a position, whereby the disclosure is not limited thereto.
The at least one provided optical means is preferably configured in accordance with the kind of information to record. As an example, the optical means can be configured to emit infrared light and to record infrared light reflected by pellets, in order to record information about the appearance of individual pellets, which is only detectable under infrared light. However, for the purposes of simplicity, the optical means may for instance be configured as a CCD detector (charge-coupled device) , e.g. for recording light intensities over a surface of pixels and with a light source emitting light in the visible part of light spectrum.
The disclosed control means are preferably provided to perform a method according exemplary embodiment of the present invention. The control means may for instance comprise at least one processor for executing computer program code for performing the required functions, at least one memory storing the program code, or both. The control means may also comprise at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause an apparatus, for instance the apparatus according exemplary embodiments of the present invention, at least to perform and/or to control the method according to invention, as disclosed in this specification. The control means can be part of a network, for instance a server cloud.
One or more pieces of information or one or more threshold values, with which the control means compare recorded information, are preferably of the same kind as the recorded information. Alternatively or additionally, the control means may for instance be configured to transform stored information or threshold values in order to enable a comparison with the recorded information.
In order to detect the presence of a defect, the control means may for instance compare recorded information with information about pellets or threshold values related to pellets presenting at least one defect. In this case, the analyzed pellet will be judged (e.g. evaluated) as defective if the comparison results in a high similarity. Alternatively or additionally, recorded information may for instance be compared with information about pellets or threshold values relating to pellets, which pellets are defect free. In this later case, the analyzed pellet will preferably be judged defective if the comparison results in a high dissemblance.
Output information may for instance be statistics about defect distribution over a volume or weight of pellets, and/or historic information indicative of a frequency of occurrence of one or more detected defects of a particular kind, to name but one non-limiting example. Preferably, the apparatus comprises output means for outputting information about the quality of one or more detected defects. Output means may for instance be communication means provided to communicate with one or more electronic devices or within a digital network, or a user interface to display information. Output information may for instance be expressed as machine readable information for use by further devices, such as for instance control means of devices of a production line. Alternatively or additionally, output information can be presented as to be read by a human operator.
According to an exemplary aspect of the invention, a computer program is disclosed, the computer program when executed by a processor causing an apparatus, for instance the control means of such an apparatus, to perform and/or control the  actions respectively steps of the method according to the present invention. The computer program may be stored on computer-readable storage medium, in particular a tangible and/or non-transitory medium, as for instance in the control means.
In the following, exemplary features and exemplary embodiments of all aspects of the present invention will be described in further detail.
According to an embodiment of the apparatus, the control means are configured to extract at least one value for at least one characteristic parameter for the appearance of at least one polymer pellet from the recorded information, the control means are configured to compare the at least one extracted value with one or more predetermined threshold values for a characteristic parameter for the appearance of polymer pellets, wherein each respective value range of one or more value ranges is correlated with a predetermined defect category, and the control means are configured to assign a predetermined defect category to a detected defect depending on the result of the comparison.
According to a corresponding embodiment of the method, the method further comprises:
- extracting at least one value for at least one characteristic parameter for the appearance of the at least one polymer pellet from the recorded information;
- comparing the at least one extracted value with one or more predetermined threshold values for a characteristic parameter for the appearance of polymer pellets, wherein each respective value range of one or more value ranges is correlated with a predetermined defect category; and
- assigning a predetermined defect category to a detected defect depending on the result of the comparison.
In this way, the calculation of statistics about the quality of the analyzed pellets is facilitated. Further, a survey of quality is simplified. Also, when further information  such as a possible defect origin is correlated to an assigned defect category, adequate measures can be taken upstream in the production line in order to avoid defects of the identified category in produced pellets.
For extracting characteristic parameters from recorded information, the control means are preferably configured to select value data among a data volume and/or to transform a plurality of information into representative information. As an example, the control means may for instance be configured to calculate a difference between intensities assigned respectively to neighbor pixels of a CCD chip, in order to determine a contrast.
Characteristic parameters preferably correspond to parameters, on the basis of which a defect can be detected or identified. As an example, such a characteristic parameter can be a hue, a contrast sharpness, a transparency, a size, a shape or similar.
Preferably, the predetermination of defect categories is based on defects observed previously by one or more analyzed pellets, and/or in the production line. Hence, several defects can be collected, which present different features. Further, a category may for instance be determined for each kind of defect of the one or more defects to be detected. As an example, categories corresponding respectively to black specks and brown specks can be determined on the basis of different color ranges. This allows a determination of defects categories and thus an identification of defects that relies on actually encountered defects, rather than on ideal references that can eventually be far from reality.
The expanse of a predetermined defect category can preferably be adjusted in accordance with measurement tolerances of the optical means. Alternatively or additionally, the definition of categories may for instance occur in accordance with variations observed for characteristic parameters related to knowns defects of the same kind.
According to an embodiment of the apparatus, the recorded information about the appearance of the at least one polymer pellet represents an image.
According to a corresponding embodiment of the method, the recorded information about the appearance of the at least one polymer pellet represents an image.
In this way, extracting relevant data for the detection and identification of eventual defects from recorded information can be performed easily. In particular digital images can easily be transmitted from hardware to another, and processed for instance by a computer software stored in the control means.
The image preferably comprises information about e.g. one or more intensities and/or saturations, which may for instance be respectively assigned to pixels ordered into a matrix. The image can comprise information distributed over two or more spatial dimensions.
According to an embodiment of the apparatus, the at least one characteristic parameter for the appearance of the at least one polymer pellet corresponds to a hue, a saturation, an intensity, a size and/or a geometrical shape of the at least one polymer pellet.
According to a corresponding embodiment of the method, the at least one characteristic parameter for the appearance of the at least one polymer pellet corresponds to a hue, a saturation, an intensity, a size and/or a geometrical shape of the at least one polymer pellet.
In this way, anomalies such as aggregation of several pellets into doubles or triplets, or defects having a color differing from the ground color of the pellets to be produced can be identified. Also, this permits evaluating contrasts, transparency, and color changes in individual pellets. Moreover, a size distribution, and/or shape distribution  over a predetermined surface on an image for instance, or a spatial position in a stream of pellets can be determined on the basis of the characteristic parameters.
For each characteristic parameter, there is preferably at least one threshold value provided, as for instance a percentage range for saturation or intensity, or a minimum value expressed in μm for a spatial dimension.
A hue can be defined by a vector with coordinates on axis corresponding to different colors. This can be done according to the system of the CIELAB color space (also known as CIE L*a*b*or sometimes abbreviated as simply "Lab" color space) which is a color space defined by the International Commission on Illumination (CIE) in 1976.
For a geometrical shape a correspondence table can be used, in order to correlate a value to a specific geometrical shape (for instance: “1” for square, “2” for circular... ) . In this way, the extracted characteristic parameter can easily be compared to reference geometrical shapes, and in particular be matched by simply matching identical values.
In an embodiment of the apparatus, the control means are configured to trace a detected defect.
In a corresponding embodiment of the method, the control means are configured to trace a detected defect.
In this way, a defect can be detected once, and followed over the duration of the analysis, in particular avoiding redundant occurrence of the same defect.
For tracing a defect, successively recorded information is preferably used. As an example, a correlation can be established between characteristic parameters extracted from recorded information about a plurality of pellets with respective first positions and characteristic parameters extracted from recorded information about pellets moved to respective second positions. In this way, successive identical  characteristic parameters can be correlated to recognize each detected defect over successive recordings.
According to an embodiment of the apparatus, two optical means are provided, wherein a first optical means is configured to record information about transparency and/or color changes within individual polymer pellets in a stream of polymer pellets, wherein a second optical means is configured to record information about the amount, the spatial distribution, the size, and/or the shape of individual polymer pellets in a stream of polymer pellets, and wherein the control means are further configured both to detect the presence of one or more defects in the at least one polymer pellets and to identify the one or more detected defects, wherein the identification is based on the recorded information as recorded by the two optical means.
According to a corresponding embodiment of the method, at least one piece of information about transparency and/or color changes within individual polymer pellets in a stream of polymer pellets is recorded by first optical means, and in that at least one piece of information about the amount, the spatial distribution, the size, and/or the shape of individual polymer pellets in a stream of polymer pellets is recorded by second optical means.
In this way, analysis tasks can be distributed to each optical means depending on the respective configuration possibilities of each optical means. As an example, for counting the number of defects and recognizing geometrical features of individual pellets, using lower cost black and white camera is sufficient. As a result, overall costs for the apparatus and for the method can be reduced.
This also allows for a greater modularity and thus adaptability of the apparatus regarding the implementation of additional analysis features or method steps.
Preferably, the control means are configured to control both the first optical means and the second optical means. In this case, information recorded by the respective  optical means are transmitted to the control means in order to achieve a conjoint analysis based on both recorded information.
Alternatively, the apparatus can be provided with a single optical means, for instance in form of a camera, which is sensitive enough to record information usable both to detect and identify defects with a reliable confidence level. In this way the apparatus can be designed in a compact and space saving way.
The above named object is further achieved according to the invention by a system comprising at least two apparatuses according to any one of the claims 1 to 6, wherein the two apparatuses are configured to perform a method according to any one of claims 7 to 12 together at least partially jointly.
Accordingly, each apparatus of the system comprises control means, which may for instance be configured in a dedicated way for the respective optical means of each apparatus. Each apparatus of the system may for instance be considered as an entity, which can be independently placed at a respective position on a production line. As an example, a first apparatus can be arranged after a device of the production line, which is susceptible to cause inclusion defects, while a second apparatus, which is able to detect malformations of individual pellets, can be placed at the end of the production line, after pellets where subject to temperature variation susceptible to cause aggregation of several pellets together.
Preferably, one control means among the control means provided in the system is configured to coordinate the system. As an example, the control means of a first apparatus can be configured to receive analysis results from the control means of a second apparatus, and to take the received analysis results onto account to issue overall analysis results.
Alternatively or additionally, the system can be provided with higher-level control means, which allow a coordination of the respective control means for each apparatus  of the system. The higher-level control means can take the form of a server, for instance.
The apparatus, the method and the system according to the invention can be arranged or executed at the end of a production line for producing polymer pellets. This way the quality of the pellets can be controlled directly after the pellet production and the information issued from the quality control used for controlling the production line.
As an alternative, the apparatus, the method and the system according to the invention can be arranged or executed at the entrance of a production line for polymer intermediate products or finished products based on the transformation of pellets. This way, quality of pellets can be checked before a transformation processing step and the information issued from the quality check employed to adapt processing parameters on the transformation line.
The above named object is further achieved according to the invention with a method for controlling a production line for producing polymer pellets, wherein the production line comprises at least one apparatus, comprising:
- detecting the presence of one or more defects and identifying the one or more detected defects by applying a method according to any one of claims 7 to 12;
- matching at least one piece of information relative to an identified defect with at least one process parameter associated to at least one apparatus of the production line; and
- controlling at least one process parameter depending on the result of the matching.
Such a method allows for adapting processing parameters in a reactive way relative to defects that are eventually present in produced pellets and in accordance with parameters of apparatuses that are actually employed in the production line. The achieved reactivity towards eventually upcoming defects enables counteracting unpredictable arising of events on the production line, which may cause the presence  of defects in produced material. As a result, the amount of produced material with off-specification characteristics can be reduced, leading to overall qualitative improvement of the production process.
Also, the method allows for a fast and reliable quality control, as for instance the analysis of about 1 kg of pellets every 15 to 20 min. Therefore, a fast adaptation of process parameters is possible and less off-specification material is produced.
In general the process parameter associated to at least one apparatus of the production line is preferably representative of an aspect of the apparatus that may correlate with the arising of defects in pellets. Such a parameter may for instance be a feeding speed for raw material at the entrance of the production line, a temperature, or a fluid that comes into contact with pellets in production or the likes, to name but a few non-limiting examples.
Controlling process parameters is preferably achieved by modifying a value that is used for executing a processing step in the production line. As an example, a preset temperature can be stored in control means of an apparatus of the production line and used for a temperature treatment processing step. If a defect that may be caused by a too high temperature is identified, the preset temperature can be decreased and set to a lower temperature, in order to reduce probability of arising for defect of the detected kind.
According to an embodiment of the method for controlling a production line, the at least one piece of information relative to an identified defect is a possible origin for the identified defect.
Accordingly, possible origins for detected defects can be recognized in a fast and reliable way, thus enabling an automatized controlling of the production line with less off-specification output material.
Possible origins for defects can correspond to an excessive temperature or residence time, the used of specific metallic parts in an apparatus, which abrasion or wear can cause the presence of black specks, to name but one non-limiting example.
Preferably, information relative to an identified defect and the process parameters are expressed under the form of values or Boolean data type, so as to facilitate comparison with each other.
The features and example embodiments of the invention described above may equally pertain to the different aspects according to the present invention. It is to be understood that the presentation of the invention in this section is merely by way of examples and non-limiting. Other features of the invention will become apparent from the following detailed description considered in conjunction with the accompanying drawings. It is to be understood, however, that the drawings are designed solely for purposes of illustration and not as a definition of the limits of the invention, for which reference should be made to the appended claims. It should be further understood that the drawings are not drawn to scale and that they are merely intended to conceptually illustrate the structures and procedures described herein.
BRIEF DESCRIPTION OF THE DRAWINGS
In the figures show:
Fig. 1 an exemplary apparatus for quality control of polymer pellets;
Fig. 2 an exemplary flow chart for a method for quality control of polymer pellets;
Fig. 3 an exemplary representation in table form for predetermined defect categories; and
Fig. 4 another exemplary representation in table form for predetermined defect categories; and
Fig. 5 another exemplary representation in table form for predetermined defect categories.
DETAILED DESCRIPTION OF SOME EXEMPLARY EMBODIMENTS
The following description serves to deepen the understanding of the present invention and shall be understood to complement and be read together with the description as provided in the above summary section of this specification,
Fig, 1 is a schematic representation of an apparatus 2 for quality control of polymer pellets, The represented apparatus comprises first optical means 4, second optical means 6, first lighting means 8, second lightning means 10, control means 12 and separation means 14. Separation means 14 are arranged at the entrance of the apparatus 2, in order to separate a plurality of polymer pellets 16 into a sequence of individual pellets 18.
The first optical means 4 is a high-resolution 3-CCD-chip color camera, while the second optical means 6 is a CCD Line Scan sensor Camera.
The first lightning means 8 comprises two light emitting  sources  20, 22, arranged respectively sidewise relatively to the first optical means 4, wherein the first optical means 4 and the first lightning means 8 are oriented in the same directions. The second optical means 6 and the second lightning means 10 are arranged opposite to each other. Separation means 14, first optical means 4 and second optical means 6 are arranged one after the other, so that an inputted volume of pellets 16 will first be separated, then pass by the first optical means and pass by the second optical means.
First optical means 4 are configured to create a signal corresponding to intensities and saturations recorded for each pixel of the 3-CCD-chip and to transmit the signal to the  control means 12 via a first communication connection 24. The second optical means 6 are configured similarly, wherein the signal is transmitted via a second communication connection 26, and wherein the created and transmitted signal corresponds to information recorded by the CCD line scan camera.
The control means 12 are configured to receive and analyze signals from both  optical means  4, 6. To do this, the control means 12 are provided with data storage means 28, on which a computer program for image analysis is stored, and a processor 30. The computer program allows for comparison between intensity and saturation values recorded for respective pixels, which correspond to respective positions on the 3-CCD chip. On the data storage means 28, a catalogue of threshold values corresponding to characteristic parameters representative of previous recorded information about pellets having known defects is stored.
The software program installed on the control means 12 is designed for executing a method according to the flow chart represented in Fig. 2.
Fig. 2 shows an exemplary flow chart 200 of a method for quality control of polymer pellets according to an example embodiment of the present invention. In a first step 201, a plurality of polymer pellets 16 is provided at the entrance of the apparatus 2, which are then separated in a second step 202 into a sequence of individual pellets 18. The hence separated pellets 18 pass by the first lightning means 8 and first optical means 4, where first optical means 4 record intensity and saturation values for light reflected by light-exposed individual pellets 18, wherein the respective recorded values are associated to respective positions on a pixel matrix (third step 203) . As then passing by the second lightning means 10 and second optical means 6, in a forth step 204 contrast changes are recorded by the second optical means 6, wherein respective contrast changes are associated to respective positions on a pixel matrix.
The recorded values are transmitted from the respective  optical means  4, 6 to the control means 12 via the first communication connection 24 and via the second  communication connection 26. In a fifth step 205, the control means 12 then extracts eventual color changes from the recorded intensity and saturation values and pellet shapes, pellet sizes and amount of pellets from the recorded contrast changes. Afterwards, in an optional sixth step 206, threshold values stored in the data storage means 28 of the control means 12 are taken into account for a comparison with the extracted color changes and pellet shapes, wherein the threshold values define ranges correlated to predetermined defect categories. Examples for threshold values and corresponding defect categories are given in Fig. 3, 4 and 5.
On the basis of the result of the comparison of the seventh step 207, the control means 12 judges about the presence or absence of defects, and assigns in a eighth step 208 a predetermined defect category to each detected defect. Thereafter, the control means 12 takes the amount of detected defects and the assigned categories into account for determining statistics about the quality of the analyzed pellets. As a further processing step, the control means 12 outputs the assigned categories and the recorded information under the form of statistics in a ninth step 209.
Fig. 3 shows an exemplary representation in table form for predetermined defect categories. In a first column, designations for five categories are given and further, corresponding threshold values for characteristic parameters, such as hue, saturation, intensity and minimum size, are given in four juxtaposed columns. The given categories correspond to different kinds of defects, such as speck inclusions, defective overall color of a pellet (third row of the table) or color anomalies. The minimum size allows for differentiating between defects with reduced size in comparison to the size of a pellet and defects that are present over a whole pellet, as for instance an overall pellet color, which differs from a target transparency.
Fig. 4 shows another exemplary representation in table form for predetermined defect categories. The given threshold values permit to categorize detected defects depending on their respective size.
Fig. 5 shows another exemplary representation in table form for predetermined defect categories with three columns. In the left columns category designations are given and in the right column the corresponding nature of defect. Categories are differentiated from each other by criteria given in the middle column. The criteria are expressed as result of a comparison between a shape, a sized or a number of defects extracted from the information about pellets and predetermined target size or shape stored in the control means 12. Preferably, the target values are given with respective tolerance ranges.
It will be understood that all presented embodiments are only exemplary, and that any feature presented for a particular example embodiment may be used with any aspect of the invention on its own or in combination with any feature presented for the same or another particular example embodiment and/or in combination with any other feature not mentioned. In particular, the example embodiments presented in this specification shall also be understood to be disclosed in all possible combinations with each other, as far as it is technically reasonable and the example embodiments are not alternatives with respect to each other. It will further be understood that any feature presented for an example embodiment in a particular category (method/apparatus/computer program/system) may also be used in a corresponding manner in an example embodiment of any other category. It should also be understood that presence of a feature in the presented example embodiments shall not necessarily means that this feature forms an essential feature of the invention and cannot be omitted or substituted.
The statement of a feature comprises at least one of the subsequently enumerated features is not mandatory in the way that the feature comprises all subsequently enumerated features, or at least one feature of the plurality of the subsequently enumerated features. Also, a selection of the enumerated features in any combination or a selection of only one of the enumerated features is possible. The specific combination of all subsequently enumerated features may as well be considered. Also, a plurality of only one of the enumerated features may be possible.
The sequence of all method steps presented above is not mandatory, also alternative sequences may be possible. Nevertheless, the specific sequence of method steps exemplarily shown in the figures shall be considered as one possible sequence of method steps for the respective embodiment described by the respective figure.
The invention has been described above by means of example embodiments. It should be noted that there are alternative ways and variations which are obvious to a skilled person in the art and can be implemented without deviating from the scope of the appended claims.

Claims (15)

  1. Apparatus for quality control of polymer pellets, comprising
    - at least one optical means (4, 6) to record information about the appearance of at least one polymer pellet (16, 18) , and
    - control means (12) ,
    characterized
    - in that the control means (12) are configured both to detect the presence of one or more defects in the at least one polymer pellet (16, 18) and to identify the one or more detected defects, wherein identifying a defect is at least partially based on a comparison of the recorded information with one or more threshold values and/or reference information indicative of previous recorded information of one or more pellets having such a defect, and
    - in that the control means (12) are configured to output information about the quality of the one or more detected defects, wherein the output information is indicative of the identity of the one or more detected defects.
  2. Apparatus according to claim 1,
    characterized
    - in that the control means (12) are configured to extract at least one value for at least one characteristic parameter for the appearance of at least one polymer pellet (16, 18) from the recorded information,
    - in that the control means (12) are configured to compare the at least one extracted value with one or more predetermined threshold values for a characteristic parameter for the appearance of polymer pellets (16, 18) ,
    - wherein each respective value range of one or more value ranges is correlated with a predetermined defect category, and
    - in that the control means (12) are configured to assign a predetermined defect category to a detected defect depending on the result of the comparison.
  3. Apparatus according to claim 2,
    characterized in that the recorded information about the appearance of the at least one polymer pellet (16, 18) represents an image.
  4. Apparatus according to claim 2 or 3,
    characterized in that the at least one characteristic parameter for the appearance of the at least one polymer pellet (16, 18) corresponds to a hue, a saturation, an intensity, a size and/or a geometrical shape of the at least one polymer pellet (16, 18) .
  5. Apparatus according to any one of claims 1 to 4,
    characterized in that the control means (12) are configured to trace a detected defect.
  6. Apparatus according to any one of claims 1 to 5,
    characterized
    - in that two optical means (4, 6) are provided,
    - wherein a first optical means (4) is configured to record information about transparency and/or color changes within individual polymer pellets in a stream of polymer pellets (16, 18) ,
    - wherein a second optical means (6) is configured to record information about the amount, the spatial distribution, the size, and/or the shape of individual polymer pellets (16, 18) in a stream of polymer pellets, and
    - wherein the control means (12) are further configured both to detect the presence of one or more defects in the at least one polymer pellets (16, 18) and to identify the one or more detected defects, wherein the identification is based on the recorded information as recorded by the two optical means (4, 6) .
  7. Method for quality control of polymer pellets, comprising:
    - providing at least one polymer pellet (16, 18) ;
    - recording at least one piece of information about the appearance of the at least one provided polymer pellet (16, 18) ;
    - detecting the presence of one or more defects in the at least one polymer pellet (16, 18) and identifying the one or more detected defects, wherein identifying a defect is at least partially based on a comparison of the recorded information with one or more threshold values and/or reference information indicative of previous recorded information of one or more pellets having such a defect; and
    - outputting information about the quality of one or more detected defects, wherein the output information is indicative of the identity of the one or more detected defects.
  8. Method according to claim 7, further comprising:
    - extracting at least one value for at least one characteristic parameter for the appearance of the at least one polymer pellet (16, 18) from the recorded information;
    - comparing the at least one extracted value with one or more predetermined threshold values for a characteristic parameter for the appearance of polymer pellets (16, 18) , wherein each respective value range of one or more value ranges is correlated with a predetermined defect category; and
    - assigning a predetermined defect category to a detected defect depending on the result of the comparison.
  9. Method according to claim 7 or 8,
    characterized in that the recorded information about the appearance of the at least one polymer pellet (16, 18) represents an image.
  10. Method according to claim 8 or 9,
    characterized in that the at least one characteristic parameter for the appearance of the at least one polymer pellet (16, 18) corresponds to a hue, a saturation, an intensity, a size and/or a geometrical shape of the at least one polymer pellet (16, 18) .
  11. Method according to any one of claims 7 to 10,
    characterized in that the control means are configured to trace a detected defect.
  12. Method according to any one of claims 7 to 11,
    characterized
    - in that at least one piece of information about transparency and/or color changes within individual polymer pellets (16, 18) in a stream of polymer pellets is recorded by first optical means (4) , and
    - in that at least one piece of information about the amount, the spatial distribution, the size, and/or the shape of individual polymer pellets in a stream of polymer pellets is recorded by second optical means (6) .
  13. System comprising
    - at least two apparatuses (2) according to any one of the claims 1 to 6, wherein the two apparatuses (2) are configured to perform a method according to any one of claims 7 to 12 together at least partially jointly.
  14. Method for controlling a production line for producing polymer pellets, wherein the production line comprises at least one apparatus, comprising:
    - detecting the presence of one or more defects and identifying the one or more detected defects by applying a method according to any one of claims 7 to 12;
    - matching at least one piece of information relative to an identified defect with at least one process parameter associated to at least one apparatus of the production line; and
    - controlling at least one process parameter depending on the result of the matching.
  15. Method according to claim 14,
    characterized in that the at least one piece of information relative to an identified defect is a possible origin for the identified defect.
PCT/CN2019/105387 2019-09-11 2019-09-11 Apparatus, system and method for quality control of polymer pellets and method for controlling production line for producing polymer pellets WO2021046759A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/105387 WO2021046759A1 (en) 2019-09-11 2019-09-11 Apparatus, system and method for quality control of polymer pellets and method for controlling production line for producing polymer pellets

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/105387 WO2021046759A1 (en) 2019-09-11 2019-09-11 Apparatus, system and method for quality control of polymer pellets and method for controlling production line for producing polymer pellets

Publications (1)

Publication Number Publication Date
WO2021046759A1 true WO2021046759A1 (en) 2021-03-18

Family

ID=74867188

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/105387 WO2021046759A1 (en) 2019-09-11 2019-09-11 Apparatus, system and method for quality control of polymer pellets and method for controlling production line for producing polymer pellets

Country Status (1)

Country Link
WO (1) WO2021046759A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6635224B1 (en) * 1998-10-30 2003-10-21 General Electric Company Online monitor for polymer processes
US20100062422A1 (en) * 2004-07-07 2010-03-11 Dominique Ausserre Optical component for observing a nanometric sample, system comprising same, analysis method using same, and uses thereof
US20180128738A1 (en) * 2015-05-13 2018-05-10 Sacmi Cooperativa Meccanici Imola Societa' Cooperativa Optical inspection apparatus and method for an extruder
WO2018151843A2 (en) * 2017-02-17 2018-08-23 Life Technologies Corporation Automated quality control and spectral error correction for sample analysis instruments

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6635224B1 (en) * 1998-10-30 2003-10-21 General Electric Company Online monitor for polymer processes
US20100062422A1 (en) * 2004-07-07 2010-03-11 Dominique Ausserre Optical component for observing a nanometric sample, system comprising same, analysis method using same, and uses thereof
US20180128738A1 (en) * 2015-05-13 2018-05-10 Sacmi Cooperativa Meccanici Imola Societa' Cooperativa Optical inspection apparatus and method for an extruder
WO2018151843A2 (en) * 2017-02-17 2018-08-23 Life Technologies Corporation Automated quality control and spectral error correction for sample analysis instruments

Similar Documents

Publication Publication Date Title
EP3480735B1 (en) Inspection apparatus, data generation apparatus, data generation method, and data generation program
KR102058427B1 (en) Apparatus and method for inspection
KR102168724B1 (en) Method And Apparatus for Discriminating Normal and Abnormal by using Vision Inspection
CN110018172B (en) Product detection system and method
Eshkevari et al. Automatic dimensional defect detection for glass vials based on machine vision: A heuristic segmentation method
Birla et al. An efficient method for quality analysis of rice using machine vision system
WO2020079694A1 (en) Optimizing defect detection in an automatic visual inspection process
KR101643713B1 (en) Method for inspecting of product using learning type smart camera
JP2022133297A (en) Video system having segment of colors for enhanced visual recognition of operator
US8358831B2 (en) Probe mark inspection
JP2010008159A (en) Visual inspection processing method
CN114226262A (en) Flaw detection method, flaw classification method and flaw detection system
JPH08189904A (en) Surface defect detector
WO2021046759A1 (en) Apparatus, system and method for quality control of polymer pellets and method for controlling production line for producing polymer pellets
Jahangir Alam et al. Analysis of a printed complex image quality checking method of fabric cloth for development of an automated quality checking system
Kaushik et al. Machine vision based automated inspection approach for clutch friction disc (CFD)
CN111753599A (en) Personnel operation flow detection method and device, electronic equipment and storage medium
EP2966433B1 (en) Device for setting region of interest for analysis
JP7411155B2 (en) Color unevenness inspection device and color unevenness inspection method
Sukhia et al. Overlapping white blood cells detection based on watershed transform and circle fitting
JP4195980B2 (en) Appearance inspection method and appearance inspection apparatus using color image
US20170116746A1 (en) Method for segmenting a color image and digital microscope
JP2021006810A (en) Method and device for inspecting production quality of cable with protective sheath, especially electric cable
CN111412941A (en) Method and device for detecting mounting quality
KR100939600B1 (en) Product-quality inspection system and method thereof

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19945329

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19945329

Country of ref document: EP

Kind code of ref document: A1