CN114076765A - Method and device for on-line monitoring of foam quality during foam production - Google Patents
Method and device for on-line monitoring of foam quality during foam production Download PDFInfo
- Publication number
- CN114076765A CN114076765A CN202010801935.5A CN202010801935A CN114076765A CN 114076765 A CN114076765 A CN 114076765A CN 202010801935 A CN202010801935 A CN 202010801935A CN 114076765 A CN114076765 A CN 114076765A
- Authority
- CN
- China
- Prior art keywords
- foam
- profile
- point cloud
- defect
- conveyor
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
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
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
- G01N21/8914—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the material examined
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B17/00—Measuring arrangements characterised by the use of infrasonic, sonic or ultrasonic vibrations
- G01B17/06—Measuring arrangements characterised by the use of infrasonic, sonic or ultrasonic vibrations for measuring contours or curvatures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/02—Analysing fluids
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biochemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Acoustics & Sound (AREA)
- Engineering & Computer Science (AREA)
- Textile Engineering (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a process and a device for producing foam by a continuous foam block foaming method, in particular to a method and a device for monitoring the foam quality on line in the foam production process. In the method for on-line monitoring of foam quality in a foam production process according to the present invention, raw materials for producing foam are injected onto a conveyor floor of a conveyor through a mixing nozzle provided at one end of the conveyor, comprising the steps of: continuously acquiring section profile data of the foam at a preset position to generate a topography map of the surface of the foam, wherein the preset position is positioned on a traveling path of the conveying bottom plate; and detecting the defects of the foam based on the topography of the foam surface.
Description
Technical Field
The invention relates to a process and a device for producing foam by a continuous foam block foaming method, in particular to a method and a device for monitoring the foam quality on line in the foam production process.
Background
The continuous foam slabstock process is a common production process for polyurethane foam, and is well suited for large-scale industrial production. In this process, the polyol, isocyanate, water and auxiliary blowing agents, catalysts, stabilizers and other additives as raw materials are pumped into a mixing head in precise metered amounts. FIG. 1 is a view of a typical continuous foam slabstock process production apparatus. Referring to fig. 1, the materials are mechanically agitated or otherwise mixed and spread on a base plate 110, which transports the mixture within a discharge air channel 120. During transport, the mixture chemically reacts to foam into blocks 130 of a certain size (e.g., up to 220cm in width, 120cm in height, and unlimited in length). The block (polyurethane foam) is formed and then cut to the desired size and shape.
In the continuous foam block foaming process, foam defects are important quality control indicators, which are directly related to the physical properties of the foam and also determine the material cost of subsequent products. In intelligent manufacturing, monitoring of quality performance indicators is the basis for data analysis and quality optimization. However, there is no method and device for monitoring the quality of foam on line accurately and efficiently.
Disclosure of Invention
It is an object of the present invention to provide a method and apparatus for on-line monitoring of foam quality during foam production which has the advantages of simplicity, efficiency, etc.
In a method for on-line monitoring of foam quality in a foam production process according to one aspect of the present invention, raw materials for producing foam are injected onto a conveyor floor of a conveyor through a mixing nozzle provided at one end of the conveyor, the method comprising the steps of:
continuously acquiring top profile data of a section of the foam at a preset position to generate a topographic map of the surface of the foam, wherein the preset position is located on a traveling path of the conveying bottom plate; and
and detecting the defects of the foam based on the topography of the surface of the foam.
Alternatively, in the above method, the top profile data is acquired in a line scan manner using a laser sensor or an ultrasonic sensor disposed above the foam at the cross section.
Optionally, in the above method, the top contour data includes spatial position coordinates of points of a top contour of the foam, the topographic map is formed by stitching a plurality of top contours, and the defect of the foam is detected in the following manner:
comparing a profile taken on the topographical map in a first direction with a first standard profile template, and comparing a profile taken on the topographical map in a second direction with a second standard profile template, wherein the first and second directions are perpendicular to a depth direction of the foam and to each other; and
and judging the existence and the type of the defect according to the comparison result.
Optionally, in the above method, the first standard contour template and the second standard contour template are a characteristic threshold or a standard contour curve determined based on a production quality specification.
Optionally, in the above method, the top contour data includes spatial position coordinates of points of a top contour of the foam, the topographic map is a point cloud map including the spatial position coordinates of the points of the top contour of the foam, and the defect of the foam is detected in the following manner:
searching a target area with a sudden depth change in the point cloud image; and
and identifying the target area based on a neural network model to judge whether the defect exists and the type of the defect.
Optionally, in the above method, the types of defects include surface cracks, surface blisters, and surface bulges.
Optionally, in the above method, the following steps are further included: and optimizing the process parameters and/or the formula in the product manufacturing scheme by comparing the defect detection results corresponding to a plurality of product manufacturing schemes.
In an apparatus for on-line monitoring of foam quality in a foam production process according to another aspect of the present invention, raw materials for producing foam are injected onto a conveying floor of a conveyor through a mixing nozzle provided at one end of the conveyor, the apparatus comprising:
a top profile measurement unit configured to continuously acquire top profile data of a cross section of the foam at a predetermined position on a travel path of the conveyor mat to generate a topographical map of a surface of the foam;
a computing unit configured to detect a defect of the foam based on the topography of the foam surface.
According to the embodiment of the invention, the topographic map is formed by acquiring the top profile data of the sections in a line scanning mode and splicing the top profile data of a series of sections. Because the scanning is only carried out along a single straight line, a complex motion correction algorithm does not need to be executed when the topography map is generated, and a sensor with high response speed does not need to be selected. In addition, according to the embodiment of the invention, the defects are detected by adopting a template comparison mode, so that the method has the advantages of simple judgment rule, strong customization, flexible configuration and the like. Moreover, in the embodiments of the present invention, the foam surface only needs to be scanned along a single straight line, and the appearance of the foam surface is relatively simple, which reduces the complexity of the point cloud image and is beneficial to selecting a simple recognition algorithm.
Drawings
FIG. 1 is a view of a typical continuous foam block process production apparatus.
FIG. 2 is a side view of a typical continuous foam block process production apparatus.
FIG. 3 is a schematic illustration of scanning the top of a foam with a laser sensor or an ultrasonic sensor to obtain a cross-sectional profile, in accordance with one embodiment of the present invention.
FIG. 4 is a schematic representation of a cross-sectional top profile of a foam article without surface defects scanned using a laser sensor or an ultrasonic sensor.
Fig. 5A-5C are schematic illustrations of a cross-sectional top profile of a foam article having surface defects scanned using a laser sensor or an ultrasonic sensor.
Fig. 6 exemplarily shows profile curves C1 and C2 taken on the topographical map along the first direction and the second direction, respectively.
FIG. 7 exemplarily shows a topography map of the foam surface, wherein the gray values of the points in the topography map represent the depths of the corresponding locations of the foam surface.
FIG. 8 is a schematic block diagram of an apparatus for on-line monitoring of foam quality during foam production according to one embodiment of the present invention.
Fig. 9 is a flow chart of a method for on-line monitoring of foam quality during foam production according to another embodiment of the present invention.
Fig. 10 is a flow chart of a method for on-line monitoring of foam quality during foam production according to another embodiment of the present invention.
Detailed Description
The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the invention are shown. This invention may, however, be embodied in different forms and should not be construed as limited to the embodiments set forth herein. The embodiments described above are intended to provide a full and complete disclosure of the present invention to more fully convey the scope of the invention to those skilled in the art.
In the present specification, words such as "comprise" and "comprises" mean that, in addition to elements and steps directly and unequivocally stated in the specification and claims, the technical solution of the present invention does not exclude other elements and steps not directly or unequivocally stated.
Terms such as "first" and "second" do not denote an order of the elements in time, space, size, etc., but rather are used to distinguish one element from another.
FIG. 2 is a side view of a typical continuous foam slabstock process production apparatus. As shown in fig. 2, the continuous foam block-foaming process production apparatus 20 includes a mixing head 210, an exhaust passage (not shown), a conveyor 220, and a control system (not shown), wherein the mixing head 210 is located at one end of the conveyor 220. The conveyor 220 includes a conveying base 221 and a power unit 222 for driving the conveying base 221 to travel in a direction indicated by an arrow in the figure. With further reference to fig. 2, the mixing head 210 injects a mixture of raw materials, such as polyol, isocyanate, water and auxiliary blowing agents, catalysts, stabilizers and other additives, into the conveyor floor 221 of the conveyor, which chemically reacts during the conveyance by the conveyor floor 221 via the exhaust channel, thereby forming a cake PF (e.g., flexible polyurethane foam) having a certain size. In the continuous foam slabstock process apparatus 20 shown in fig. 2, the control system controls and monitors the entire process in real time. For example, the control system controls the mixture injection timing and flow rate of the mixing nozzle 210 and the traveling speed of the delivery substrate 221.
According to one aspect of the invention, the foam's defects are determined based on a topographical map of the foam's surface. In one or more embodiments of the invention, the topographical map is a data set, each point in the data set corresponding to one of the points of the foam surface and representing spatial position information for the corresponding point of the foam surface in a three-dimensional array (x, y, z), where x, y, and z are spatial coordinate values of the corresponding point of the foam surface. Types of defects described herein include, for example, without limitation, surface cracking, slumping, surface blistering, and surface bulging.
According to another aspect of the invention, a location is selected on the travel path of the conveyor mat and the profile data of the foam at that location is continuously acquired. When the continuously acquired profile sections are spliced together, a topographic map of the foam surface can be obtained. The profile data described herein includes at least the spatial location coordinates of the points of the profile. -
Optionally, in one or more embodiments of the invention, the cross-section profile data is top profile data of the cross-section. It should be noted, however, that the top profile is merely an example of a cross-sectional profile. In other embodiments of the invention, the profile data may also include side profile data. Although the following description of the embodiments of the present invention takes defect detection of the top surface of the foam as an example, one skilled in the art will recognize on reading this specification that the principles of the present invention embodied by the following description are equally applicable to the detection of defects of other surfaces of the foam.
To acquire the top profile data, according to one embodiment of the present invention, a laser sensor or ultrasonic sensor 30 may be disposed above the foam at the cross-section as shown in fig. 2, the sensor being configured to line scan the foam in a direction perpendicular to the direction of travel of the conveyor floor, thereby acquiring the cross-sectional top profile data. Because the sensor only scans the foam surface along a single straight line, the complex motion correction algorithm is avoided when generating the topography map, and the sensor with high response speed is not required to be selected to meet the real-time monitoring requirement.
FIG. 3 is a schematic illustration of scanning the top of a foam with a laser sensor or an ultrasonic sensor to obtain profile data, in accordance with one embodiment of the present invention. In the embodiment shown in fig. 3, a dual sensor configuration is employed in which two laser or ultrasonic sensors 30a, 30b are arranged side by side in a direction perpendicular to the direction of travel of the sole plate, and the foam top surface is line-scanned or line-by-line scanned along the direction of merging arrangement to acquire top profile data of the cross-section. The scan areas of the dual sensors may overlap and the overlapping areas may be combined by image processing techniques to obtain complete top profile data of the cross-section.
It is noted that the number and arrangement of the above-mentioned sensors is merely illustrative. In practical applications, a larger number of sensors may be used depending on the situation. Fig. 4 is a schematic representation of a cross-sectional top profile of a foam article without surface defects scanned using a laser sensor or an ultrasonic sensor, where w represents the width of the conveyor mat or foam. During production, the top profile of the foam is not necessarily a flat surface, but is undulating as shown in fig. 4.
Fig. 5A-5C are schematic diagrams of a cross-sectional top profile of a foamed article having a surface defect scanned using a laser sensor or an ultrasonic sensor, wherein fig. 5A shows the top profile where a surface crack defect occurs, fig. 5B shows the top profile where surface blistering occurs, and fig. 5C shows the top profile of a surface bump. As can be seen from fig. 5A to 5C, the top profile of the surface crack defect locally exhibits a breakpoint, the top profile of the surface blister locally exhibits a downward concavity, and the top profile of the surface bump locally exhibits an upward convexity.
According to one or more embodiments of the invention, a topographical map containing only the spatial position information of the points of the foam top contour is generated by stitching together the top contours of the sections in a time sequence. For example, a laser sensor or an ultrasonic sensor may be used to scan the top of the foam at time intervals or sampling intervals to obtain a series of time-sequenced cross-sectional top profiles.
By way of example, assume that the following rectangular spatial coordinate system is used to describe the spatial position of points on the foam surface: the X axis of the space rectangular coordinate system is perpendicular to the advancing direction of the conveying bottom plate and the depth or thickness direction of the foam, the Y axis is parallel to the advancing direction of the conveying bottom plate, and the Z axis is parallel to the depth or thickness direction of the foam. For adjacent cross-sectional top profiles, there are pairs of points (X, y, z) and (X ', y ', z '), whose coordinate values along the X-axis remain constant (X ═ X '), y ═ y + v × Δ t, where v is the travel speed of the conveyor mat, Δ t is the set scanning time interval (i.e., sampling interval), and z ' are the actual measurements. When the scanning time interval is large, fitted cross-sectional top profile data may be interpolated between cross-sectional top profile data (between y and y 'and between z and z') adjacent in scanning time.
It should be noted that the conveying bottom plate can move at a constant speed or at variable speeds. Preferably, the above-mentioned sampling interval Δ t may be varied as a function of the travelling speed v of the conveyor mat, so as to ensure that the spacing of adjacent top profiles in the travelling direction of the conveyor mat is equal, i.e. to ensure that v × Δ t is always constant.
For the topographic map generated by stitching the top profile of the cross-section, a template comparison approach may be used to detect foam defects, as described further below.
Referring to fig. 6, profile curves C1 and C2 are illustratively shown on the topographical map taken along a first direction, such as an X-axis direction, and a second direction, such as a Y-axis direction, that are perpendicular to each other and to the depth direction of the foam, respectively.
In the determination of the defects, a series of contour curves taken in a first direction on the topographical map is compared on the one hand with a first standard contour template (fig. 6 shows, by way of example, only one contour curve C1 taken in a first direction), and on the other hand with a second standard contour template (fig. 6 shows, by way of example, only one contour curve C2 taken in a second direction). Since the topographic map is formed by stitching a plurality of sectional top profiles, the profile curve taken in the first direction is the top profile of the section.
Alternatively, the first standard outline template and the first standard outline template may be geometric feature thresholds determined based on production quality specifications. By comparing the geometric features of the profile curve obtained by actual measurement with the feature threshold, it is possible to determine whether a defect is present and the type of defect. For example, if the difference between the maximum height and the minimum height of the profile curve in the local range exceeds a corresponding threshold, it can be determined that the foam surface has a blister or bulge defect; or if the number of continuous break points of the profile curve exceeds a corresponding threshold value, judging that the foam surface has cracking defects.
Alternatively, the first standard contour template and the second standard contour template may also be standard contour curves (e.g., contour curves as shown in fig. 4) determined based on production quality specifications. By calculating the degree of matching (e.g., the degree of global deviation and the degree of local deviation) and the direction of deviation of the profile curve obtained by actual measurement with the standard profile curve, the presence and type of the defect can be judged.
The template comparison mode has the advantages of simple judgment rule, strong customization, flexible configuration and the like. For example, for different quality specifications, real-time monitoring of the foam quality can be carried out by using corresponding standard profile curves or characteristic thresholds.
In addition to the template comparison described above, image recognition may also be used to detect foam defects, as described further below.
Referring to fig. 7, a topographical map of the foam surface is exemplarily shown, wherein the gray value of each point in the topographical map represents the depth (also referred to as spatial depth value) of the corresponding location of the foam surface, i.e. the distance of the point with respect to the base plate. The grayscale representation of the topography map can provide a user with a visual illustration of the surface topography. Alternatively, the topographical map shown in FIG. 7 takes the form of a cloud of points, each point in the cloud of points representing a spatial location coordinate of each point of the foam top contour.
From the above, defects such as surface cracks, collapse, surface blistering, and surface bulging will result in abrupt changes in local geometric features. Therefore, in one or more embodiments of the present invention, a target region with a sudden depth change is first searched in a point cloud image, and then the target region is identified based on a neural network model to determine whether a defect exists and the type of the defect. The neural network model can be obtained by training by using a topographic map sample with marked defects.
Illustratively, in one or more embodiments of the invention, the neural network model may be trained in the following manner.
A plurality of point cloud picture samples are first acquired. In order to give consideration to the training difficulty and speed of the neural network model, certain limitation can be made on the physical size L multiplied by W corresponding to each point cloud image sample, wherein L is the physical size along the advancing direction of the conveying bottom plate corresponding to the point cloud image sample, and W is the width of the foam. Preferably, the physical dimension L may be set to not less than 100 mm.
And then marking the defects of the acquired point cloud picture sample according to a set quality standard. Similarly, to account for the training difficulty and speed of the neural network model, certain limits may be placed on the physical size of the individual defects being labeled. Preferably, the physical size may be set to not less than 100mm × 100mm, or may be set to not less than 104mm2。
Then, samples are selected from the multiple point cloud picture samples to obtain a training data set. Through intensive research, the inventor of the invention can well consider the training difficulty and speed of the neural network model when the proportion of the point cloud picture sample with the marked defect type in the training data set is larger. Particularly when the ratio is 10% or more.
And finally, training the neural network model by using the training data set. The number of exercises may be one or more. The inventor of the present invention has found through intensive research that, when a plurality of training is performed, if 20% of the point cloud image samples in the training data set are reserved for the next training after each training is completed, not only can the training speed be increased, but also the defect identification accuracy of the model is improved.
The image recognition mode has the advantages of convenience in implementation, high accuracy and the like. In particular, in various embodiments of the present invention, the foam surface only needs to be scanned along a single straight line, and the topography of the foam surface is relatively simple, which reduces the complexity of the point cloud image, so that a simple recognition algorithm can be selected to quickly generate a recognition result.
FIG. 8 is a schematic block diagram of an apparatus for on-line monitoring of foam quality during foam production according to one embodiment of the present invention.
As shown in fig. 8, the apparatus 80 includes a profile measurement unit 810 and a calculation unit 820 coupled to the profile measurement unit 810. Optionally, the apparatus 50 further comprises a storage unit 830 configured to store the contour data, a display device 840 configured to display the foam quality monitoring result, and a feedback unit 850.
The profile measurement unit 810 is configured to continuously acquire profile data (e.g., top profile data) of a cross section of the foam at predetermined locations to generate a topographical map of the surface of the foam, which may be, for example, a laser sensor or an ultrasonic sensor disposed above the foam at the cross section as described above. The calculation unit 820 is configured to detect defects of the foam based on the topography of the foam surface. The specific way of detecting defects is described above and will not be described further here.
The feedback unit 850 is communicatively coupled to the calculation unit 820, the storage unit 830 and the line control system (not shown), and is configured to send the defect detection results generated by the calculation unit 820 to the line control system, or to extract the defect detection results from the storage unit 830 and send the defect detection results to the line control system. Accordingly, the production line control system can optimize the process parameters and/or recipes in the product manufacturing scheme by comparing the defect detection results corresponding to the plurality of product manufacturing schemes. It is noted that the expression "and/or" herein means that both the process parameter and the recipe may be included, or one of the process parameter and the recipe may be included. Examples of process parameters include, but are not limited to, travel speed of the conveyor shoe and temperature and injection speed of the injected feedstock, among others.
It should be noted that the apparatus 80 of the embodiment shown in fig. 8 may be a component of the control system in the continuous foam slab-foaming apparatus 20, or may be a unit independent of the control system in the continuous foam slab-foaming apparatus 20.
Fig. 9 is a flow chart of a method for on-line monitoring of foam quality during foam production according to another embodiment of the present invention. The method of the present embodiment is illustratively implemented by means of the apparatus shown in fig. 8. It should be noted, however, that the practice of the method of this embodiment is not limited to devices having a particular configuration.
As shown in fig. 9, in step S910, the profile measuring unit 810 continuously acquires cross-sectional profile data (e.g., top profile data) of the foam at a predetermined position on the travel path of the conveying base, which can be used to generate a topographic map of the surface of the foam, and stores the acquired cross-sectional profile data in the storage unit 830.
Subsequently, at step S920, the calculation unit 820 acquires the traveling speed of the conveying base plate by the control system and retrieves the profile data from the storage unit 830, thereby generating a topography map of the foam surface.
Proceeding next to step S930, the calculation unit 820 compares the profile curve cut out on the topography map in the first direction with a first standard profile template, and compares the profile curve cut out on the topography map in the second direction with a second standard profile template.
Subsequently, in step S940, the calculation unit 820 determines the presence and type of the defect from the comparison result. The specific determination method has been described above, and is not described herein again.
Subsequently, proceeding to step S950, the calculation unit 820 outputs the determination result obtained in step S940 to the display unit 840.
Alternatively, in step S960, the feedback unit 850 transmits the defect detection result generated by the calculation unit 820 to the line control system, or extracts the defect detection result from the storage unit 830 and transmits the defect detection result to the line control system, thereby enabling the line control system to optimize the process parameters and/or recipes in the product manufacturing scheme by comparing the defect detection results corresponding to a plurality of product manufacturing schemes.
Fig. 10 is a flow chart of a method for on-line monitoring of foam quality during foam production according to another embodiment of the present invention. The method of the present embodiment is illustratively implemented by means of the apparatus shown in fig. 8. It should be noted, however, that the practice of the method of this embodiment is not limited to devices having a particular configuration.
As shown in fig. 10, the profile measuring unit 810 continuously acquires cross-sectional profile data (e.g., top profile data) of the foam at a predetermined position on the travel path of the conveying deck, which can be used to generate a topographic map of the surface of the foam, and stores the acquired cross-sectional profile data in the storage unit 830 at step S1010.
Subsequently, in step S1020, the calculation unit 820 acquires the travel speed of the conveying base plate by the control system and retrieves the top section data from the storage unit 830, thereby generating a topography map in the form of a point cloud map.
Then, in step S1030, the computing unit 820 searches the point cloud image for a target region with a sudden depth change.
Subsequently, in step S1040, the target area is identified based on the neural network model to determine whether there is a defect and the type of the defect. The neural network model can be obtained by training by using a topographic map sample with marked defects.
Subsequently, the flow proceeds to step S1050, and the calculation unit 820 outputs the determination result obtained in step S1040 to the display unit 840.
Alternatively, in step S1060, the feedback unit 850 sends the defect detection result generated by the calculation unit 820 to the line control system, or extracts the defect detection result from the storage unit 830 and sends the defect detection result to the line control system, thereby enabling the line control system to optimize the process parameters and/or recipes in the product manufacturing scheme by comparing the defect detection results corresponding to a plurality of product manufacturing schemes.
The foregoing has described the principles and preferred embodiments of the present invention. However, the invention should not be construed as being limited to the particular embodiments discussed. The preferred embodiments described above should be considered as illustrative and not restrictive, and it should be understood that variations may be made in those embodiments by workers skilled in the art without departing from the scope of the present invention as defined by the following claims.
Claims (20)
1. A method for on-line monitoring of foam quality during foam production, wherein raw materials for producing foam are injected onto a conveyor floor of a conveyor through a mixing nozzle provided at one end of the conveyor, the method comprising the steps of:
continuously acquiring section profile data of the foam at a preset position to generate a topography map of the surface of the foam, wherein the preset position is positioned on a traveling path of the conveying bottom plate; and
and detecting the defects of the foam based on the topography of the surface of the foam.
2. The method of claim 1, wherein the profile data is top profile data of a profile.
3. The method of claim 2, wherein the top profile data is acquired by line scanning a top surface of the foam with a laser sensor or an ultrasonic sensor disposed above the foam at the fracture.
4. A method as claimed in claim 2 or 3, wherein the top profile data comprises spatial position coordinates of points of the top profile of the foam, the topographical map is stitched from a plurality of top profiles acquired at sampling intervals which vary with the speed of travel of the conveyor mat to ensure equal spacing of adjacent top profiles in the direction of travel of the conveyor mat, and defects in the foam are detected in the following manner:
comparing a profile taken on the topographical map in a first direction with a first standard profile template, and comparing a profile taken on the topographical map in a second direction with a second standard profile template, wherein the first and second directions are perpendicular to a depth direction of the foam and to each other; and
and judging the existence and the type of the defect according to the comparison result.
5. The method of claim 4, wherein the first and second standard contour templates are standard contour curves or characteristic thresholds determined based on production quality specifications.
6. A method according to claim 2 or 3, wherein the top profile data comprises spatial position coordinates of points of the top profile of the foam, the topographical map is a point cloud map comprising the spatial position coordinates of the points of the top profile of the foam, and the defects of the foam are detected in the following manner:
searching a target area with a sudden depth change in the point cloud image; and
and identifying the target area based on a neural network model to judge whether the defect exists and the type of the defect.
7. The method of claim 6, wherein the neural network model is trained in the following manner:
acquiring a plurality of cloud image samples, wherein the physical dimension corresponding to each cloud image sample is L multiplied by W, L is the physical dimension corresponding to the cloud image sample along the advancing direction of the conveying bottom plate and is not greater than a set first threshold value, and W is the width of foam;
marking the point cloud picture sample according to a set quality standard, wherein for a marked single defect, the corresponding physical area is not larger than a second threshold value;
selecting a training data set from the plurality of point cloud picture samples, wherein the ratio of the number of point cloud picture samples marked with defect types to the total number of point cloud picture samples of the training data set is not less than a first proportion; and
and training the neural network model for multiple times by using the training data set, wherein after each training is finished, the point cloud picture sample with the second proportion in the training data set is reserved for the next training.
8. The method of claim 4 or 6, wherein the types of defects include surface cracking, slumping, surface blistering, and surface bulging.
9. A method according to claim 1 or 2, wherein the foam is a polyurethane soft foam.
10. The method of claim 1 or 2, further comprising the steps of:
and optimizing the process parameters and/or the formula in the product manufacturing scheme by comparing the defect detection results corresponding to a plurality of product manufacturing schemes.
11. An apparatus for on-line monitoring of foam quality during foam production, wherein raw materials for producing foam are injected onto a conveyor floor of a conveyor through a mixing nozzle provided at one end of the conveyor, the apparatus comprising:
a profile measuring unit configured to continuously acquire section profile data of the foam at a predetermined position on a traveling path of the conveyor base; a computing unit configured to generate a topography map of the foam surface from the profile data and to detect defects of the foam based on the topography map of the foam surface.
12. The apparatus of claim 11, further comprising:
a feedback unit communicatively coupled to the line control system and configured to send the defect detection results to the line control system to enable the line control system to optimize process parameters and/or recipes in a plurality of product manufacturing recipes by comparing the defect detection results corresponding to the product manufacturing recipes.
13. The apparatus of claim 11, wherein the profile data is top profile data of a profile.
14. The apparatus of claim 13, wherein the profile measuring unit comprises a laser sensor or an ultrasonic sensor disposed above the foam at the fracture surface configured to acquire the top profile data in a line scan over a top surface of the foam.
15. The apparatus of claim 13 or 14, wherein the top profile data comprises spatial position coordinates of points of a top profile of the foam, the topographical map is stitched from a plurality of top profiles acquired at sampling intervals that vary with the speed of travel of the conveyor mat to ensure equal spacing of adjacent top profiles in the direction of travel of the conveyor mat, and the computing unit is configured to detect defects in the foam in the following manner:
comparing a profile taken on the topographical map in a first direction with a first standard profile template, and comparing a profile taken on the topographical map in a second direction with a second standard profile template, wherein the first and second directions are perpendicular to a depth direction of the foam and to each other; and
and judging the existence and the type of the defect according to the comparison result.
16. The apparatus of claim 15, wherein the first and second standard contour templates are standard contour curves or characteristic thresholds determined based on production quality specifications.
17. The apparatus according to claim 13 or 14, wherein the top profile data comprises spatial position coordinates of points of a top profile of the foam, the topography map is a point cloud map comprising spatial position coordinates of points of the top profile of the foam, and the calculation unit is configured to detect defects of the foam in the following manner:
searching a target area with a sudden depth change in the point cloud image; and
and identifying the target area based on a neural network model to judge whether the defect exists and the type of the defect.
18. The apparatus of claim 17, wherein the neural network model is trained in the following manner:
acquiring a plurality of cloud image samples, wherein the physical dimension corresponding to each cloud image sample is L multiplied by W, L is the physical dimension corresponding to the cloud image sample along the advancing direction of the conveying bottom plate and is not greater than a set first threshold value, and W is the width of foam;
marking the point cloud picture sample according to a set quality standard, wherein for a marked single defect, the corresponding physical area is not larger than a second threshold value;
selecting a training data set from the plurality of point cloud picture samples, wherein the number of point cloud picture samples marked with defect types is not less than a first proportion of the total number of point cloud picture samples of the training data set; and
and training the neural network model for multiple times by using the training data set, wherein after each training is finished, the point cloud picture sample with the second proportion in the training data set is reserved for the next training.
19. The apparatus of claim 15 or 17, wherein the types of defects include surface cracks, slumping, surface blistering, and surface bulging.
20. The device of any one of claims 11-13, wherein the foam is a polyurethane soft foam.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010801935.5A CN114076765A (en) | 2020-08-11 | 2020-08-11 | Method and device for on-line monitoring of foam quality during foam production |
| EP21746058.3A EP4196745A1 (en) | 2020-08-11 | 2021-07-23 | Method and apparatus for online monitoring of foam quality during foam production process |
| PCT/EP2021/070647 WO2022033837A1 (en) | 2020-08-11 | 2021-07-23 | Method and apparatus for online monitoring of foam quality during foam production process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010801935.5A CN114076765A (en) | 2020-08-11 | 2020-08-11 | Method and device for on-line monitoring of foam quality during foam production |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114076765A true CN114076765A (en) | 2022-02-22 |
Family
ID=80279846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010801935.5A Pending CN114076765A (en) | 2020-08-11 | 2020-08-11 | Method and device for on-line monitoring of foam quality during foam production |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114076765A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120467183A (en) * | 2025-06-26 | 2025-08-12 | 南京南大工程检测有限公司 | Method, device and system for measuring linear dimensions of foam plastic products |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1820926A (en) * | 2005-01-25 | 2006-08-23 | 拜尔材料科学股份公司 | Method and device for the optimal production of foam in a continuous slabstock foaming process |
| CN107561092A (en) * | 2017-07-26 | 2018-01-09 | 天津大学 | A kind of surface quality of steel detection method |
| US20180106604A1 (en) * | 2016-10-19 | 2018-04-19 | Columbia Insurance Company | Three dimensional laser measurement device for quality control measurements |
| US20180211373A1 (en) * | 2017-01-20 | 2018-07-26 | Aquifi, Inc. | Systems and methods for defect detection |
| CN110136130A (en) * | 2019-05-23 | 2019-08-16 | 北京阿丘机器人科技有限公司 | A kind of method and device of testing product defect |
| CN110992337A (en) * | 2019-11-29 | 2020-04-10 | 添维信息科技(天津)有限公司 | Container damage detection method and system |
-
2020
- 2020-08-11 CN CN202010801935.5A patent/CN114076765A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1820926A (en) * | 2005-01-25 | 2006-08-23 | 拜尔材料科学股份公司 | Method and device for the optimal production of foam in a continuous slabstock foaming process |
| US20180106604A1 (en) * | 2016-10-19 | 2018-04-19 | Columbia Insurance Company | Three dimensional laser measurement device for quality control measurements |
| US20180211373A1 (en) * | 2017-01-20 | 2018-07-26 | Aquifi, Inc. | Systems and methods for defect detection |
| CN107561092A (en) * | 2017-07-26 | 2018-01-09 | 天津大学 | A kind of surface quality of steel detection method |
| CN110136130A (en) * | 2019-05-23 | 2019-08-16 | 北京阿丘机器人科技有限公司 | A kind of method and device of testing product defect |
| CN110992337A (en) * | 2019-11-29 | 2020-04-10 | 添维信息科技(天津)有限公司 | Container damage detection method and system |
Non-Patent Citations (1)
| Title |
|---|
| 连晓峰著: "《移动机器人及室内环境三维模型重建技术》", 31 August 2010, 国防工业出版社, pages: 20 - 22 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120467183A (en) * | 2025-06-26 | 2025-08-12 | 南京南大工程检测有限公司 | Method, device and system for measuring linear dimensions of foam plastic products |
| CN120467183B (en) * | 2025-06-26 | 2025-10-10 | 南京南大工程检测有限公司 | Method, device and system for measuring linear dimensions of foam plastic products |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112654249B (en) | Method and system for trimming a workpiece to optimize the division of the trimmed workpiece into finished pieces | |
| US8473141B2 (en) | Robot system | |
| CN106273446B (en) | A kind of slice path generating method and system for 3D printing | |
| US20030145699A1 (en) | Three axis portioning method | |
| US20180264742A1 (en) | Information processing device, solid object modeling system, and information processing method | |
| JP2000075032A (en) | Method for detecting obstacle on traveling road and method for estimating its existence | |
| CN112504184B (en) | Rapid online quality inspection system for three-dimensional size of steel plate | |
| CN113375566B (en) | Accurate measurement method and system for object size | |
| US9751329B2 (en) | Method for printing on elevation contours of the print object | |
| CN114076765A (en) | Method and device for on-line monitoring of foam quality during foam production | |
| JP2004130786A (en) | Apparatus and method for manufacturing foam in continuous slab stock foaming process | |
| CN117723547B (en) | Quality monitoring system and method for finished product preparation based on magnesia carbon bricks | |
| CN113295001B (en) | A detection system, method and device for material layer thickness of sintering machine trolley | |
| CN113320924A (en) | Belt longitudinal tearing detection device based on single line laser radar | |
| CN115808126B (en) | Lithium electrode sheet coating boundary positioning method based on machine vision scanning data | |
| JP2023541129A (en) | System and method for foamboard processing with the aid of computer vision | |
| US20200406344A1 (en) | Mold-shift detection device for upper and lower molds and mold-shift detection method for upper and lower molds | |
| US10151583B2 (en) | Method of measuring a 3D profile of an article | |
| EP4001834A1 (en) | Method and apparatus for online monitoring of foam quality during foam production process | |
| EP4196745A1 (en) | Method and apparatus for online monitoring of foam quality during foam production process | |
| CN110608684A (en) | Single-layer multi-channel weld accumulation deposition effect detection method and system | |
| CN109158597A (en) | Powdering quality detection method, equipment, readable storage medium storing program for executing and three-dimension object manufacturing method | |
| CN106370100A (en) | Vehicle body symmetry deviation detection method and system | |
| Guo et al. | Boolean operations of STL models based on loop detection | |
| JP2013151119A (en) | Method and device for measuring tread length |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20220222 |
|
| WD01 | Invention patent application deemed withdrawn after publication |