Device and method for manufacturing containers

Abstract

The invention relates to a device and a method for producing containers, comprising: a control device configured to set at least one operating parameter for producing at least one first container; a measuring device configured to measure at least one container parameter of the manufactured at least one first container to obtain a measured value of the at least one container parameter; a measurement correction device configured to correct a measurement of the at least one container parameter based on the at least one model parameter of the numerical model; and a performance parameter determination device configured to determine at least one performance parameter value based on the corrected at least one measurement value. Furthermore, the control device is configured to control the manufacturing of at least one second container different from the at least one first container on the basis of at least one operating parameter adjusted on the basis of the determined at least one performance parameter value.

Description

Technical Field

[0001] The present invention relates to the field of container manufacturing, and more particularly to model-based container manufacturing. Background Art

[0002] Filling equipment for beverages and the like includes a plurality of production units connected in series, such as machines for manufacturing containers, (forming) filling machines, labeling machines, and packaging machines.

[0003] For example, when manufacturing a labeled plastic bottle, the plastic bottle is continuously produced from a blank in a rotary blow molding machine. The blow molding machine receives the heated blank (also called a preform) in a suitably configured blow mold and then blows it into a plastic bottle under high pressure and high temperature during the cycle of the blow molding machine. The plastic bottle produced in this way is then filled and labeled.

[0004] Generally, the finished container should meet predetermined quality requirements, that is, it should have predetermined values of performance parameters (quality parameter values), such as predetermined values of top load, burst pressure, stress crack, material distribution, and cross-sectional weight. The performance parameters are not only used for quality control but also to optimize the manufacturing process by adjusting working parameters, such as the pressure applied to the plastic preform, the time and duration of applying the pressure, and the movement parameters of the stretching rod for stretching the plastic preform.

[0005] The performance parameters cannot be measured online but are determined based on measured container parameters, such as wall thickness, wall thickness profile, and inspection measurement data of the bottom and side walls. However, the measured values of the container parameters are usually not reliable enough because they may be affected by interfering parameters, such as different surfaces, orientations, and geometries of the container, as well as different absorption / transmission coefficients of the container material. Therefore, for example, as Figure 1 shown, the accuracy of optical or ultrasonic measurement methods for measuring wall thickness may be impaired due to the non-uniform container geometry in the measurement region G1 where the measurement beam M is not perpendicularly incident, and in the measurement region G2 where the measurement beam M is perpendicularly incident, the measurement beam M thus passes through more material.

[0006] Therefore, an object of the present invention is to provide an apparatus and method for manufacturing a container (such as a plastic bottle or a glass bottle), in which the manufacturing process is optimized based on accurately determined values of performance parameters. Summary of the Invention

[0007] The above object is achieved by providing a device for manufacturing containers (such as plastic bottles or glass bottles), the device comprising control means configured to set at least one operating parameter for manufacturing at least one first container (thereby controlling the manufacture of at least one first container based on the set at least one operating parameter), and measuring means configured to measure at least one container parameter of the manufactured at least one first container to obtain a measured value of the at least one container parameter. The device further comprises measurement correction means configured to correct the measured value of the at least one container parameter based on at least one model parameter of a numerical model (which may in particular take into account measurement disturbance parameters and environmental data), and performance parameter determination means configured to determine at least one performance parameter value (at least one performance parameter) based on the corrected at least one measured value. In addition, the control means is configured to control the manufacture of at least one second container different from the at least one first container based on at least one operating parameter adjusted based on the determined at least one performance parameter value.

[0008] At least one performance parameter can also be determined using a model based on the corrected at least one measured value, the model may include the above numerical model.

[0009] The term "control means" used herein and hereinafter includes control and / or regulation means. Thus, the term "control" herein includes control and / or regulation.

[0010] The control means, the measurement correction means and the performance parameter determination means can be at least partially logically and / or physically separated from each other or integrated with each other. In particular, the measurement correction means can be integrated in the performance parameter determination means.

[0011] The device may include a blow molding machine, such as a rotary blow molding machine.

[0012] At least one operating parameter can be selected from the group including the pressure applied to the plastic preform, the time and duration of applying the pressure, and the movement parameters of the stretching rod for stretching the plastic preform. Additionally, the group can include the adjustment parameters or manipulated variables of the heating device for heating the plastic preform. The heating device can be an infrared furnace, a microwave oven, a laser furnace, or a combination thereof. The manipulated variables of the infrared furnace include, for example, the heating power of a single heating radiator or the adjustment of the entire heating radiator, the enabling / disabling of a single heating radiator, the cooling power (surface cooling), the individual adaptation of the cooling power along the heating channels in a single heating chamber, the reflector adjustment for making the heating channels smaller / larger, for example, by moving the bottom reflector upwards, the immersion depth of the heating mandrel into the plastic preform in the heating channel, and the adjustment of the focusing reflector to optimize the energy input below the support ring. The corresponding manipulated variables can be included in the group for the laser furnace. For the microwave oven, the group can include the microwave power, the adjustment of the elements in the applicator for influencing the microwave field, and the adjustment of the functional elements for the profile as manipulated variables.

[0013] At least one container parameter can be selected from the group including the parameter characterizing the container wall thickness (e.g., the wall thickness itself or the absorption or transmission ability - e.g., the thicker the wall, the more light of the sensor beam is absorbed), the absorption coefficient of the container material, the transmission coefficient of the container material, the translucency of the container material, the wall thickness profile of the container, the temperature of the container, the height of the container, the diameter of the container, and the inspection measurement data of the bottom and side walls of the container, eccentricity detection data, neck pull-out measurement data, information data from the production line, such as counting defective blow molding operations, burst bottles and lost bottles at the filling machine, and filling level control data, etc. The measurement of the container parameter can include direct measurement or determination through measured values. At least one performance parameter can be selected from the group including top load, ground clearance, burst pressure, stress crack, material distribution, and sectional weight.

[0014] The numerical model can include at least one of the information about the surface of the container (especially grooves, facets, and other surface elements), orientation, geometry, and the absorption / transmission coefficient of the container material, and the parameters of the preform of the container as at least one model parameter. Similarly, the model can include data from environmental parameters, such as temperature and humidity, as at least one model parameter. For example, the model can be established from a plurality of data points, and it can, for example, allow the precise determination of the material distribution or sectional weight based on the measured values of one or more container parameters characterizing the wall thickness. In particular, the model can allow the mathematical calculation of the performance parameters based on the at least one corrected measured value provided by the model. The model can include physical models, expert systems, the results of inverse modeling, and look-up tables.

[0015] Numerical models can generally take into account production parameters and other metadata, such as preform inlet temperature, Contiform setting parameters, classification data of bottles and preforms, batch information about preforms (such as production date), object drawings, or any desired product specification data.

[0016] At least one operating parameter can be adjusted, for example, by a control device based on the determined value of at least one performance parameter. The performance parameter determined from the corrected measurement values can be verified and adjusted through a calibration run.

[0017] By providing a model-based correction device, compared with the prior art, more accurate values of performance parameters can be determined based on the model, and these values can be used to adjust the operating parameters of the manufacturing process to optimize the manufacturing process. By determining performance parameters based on the model and operating parameters adjusted based on these model-determined performance parameters, predetermined quality standards can be reliably met.

[0018] According to an improvement scheme, the device can include an artificial neural network, which is configured to create a model (especially before manufacturing at least one first container). The learnable neural network is particularly suitable for creating the model required for correction. In addition, it can also be further dynamically trained during the actual manufacturing process. Therefore, it can be configured to train the model even after manufacturing at least one first container and / or at least one second container, in order to further optimize the model (for example, by dynamically adjusting correction parameters). The control device and / or the measurement correction device and / or the performance parameter determination device can also include a neural network. Generally, the neural network of the performance parameter determination device can learn and dynamically adapt to the transformation mapping from the corrected measurement values to the performance parameter values.

[0019] In addition, a filling device having a device according to one of the above examples is provided. The filling device includes many other machines, such as a filling machine, a labeling machine, and a packaging machine. The control device can be configured to control all the machines.

[0020] The above object is also achieved by providing a method for manufacturing a container, the method comprising the steps of: providing a container manufactured based on at least one operating parameter, measuring at least one container parameter (e.g., a parameter characterizing the wall thickness) of the provided container to obtain a measured value of the at least one container parameter, correcting the measured value of the at least one container parameter based on at least one model parameter of a numerical model (e.g., using information on the surface, orientation, geometry of the container and the absorption / transmission coefficient of the container material, as well as parameters of the preform of the provided container), determining at least one performance parameter value based on the corrected at least one measured value, adjusting at least one operating parameter based on the determined at least one performance parameter value, and manufacturing a container based on the adjusted at least one operating parameter. The above applies to models and various parameters. Similarly, at least one performance parameter value can be determined based on the corrected at least one measured value with the aid of a model that may include the numerical model.

[0021] The method may further include creating a numerical model using a neural network. After manufacturing the container, the neural network can also be (further) trained based on the adjusted at least one operating parameter. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Embodiments of the apparatus according to the invention and of the method according to the invention are described below with reference to the drawings. The described embodiments are to be considered illustrative in every respect and not restrictive, and various combinations of the features are included in the invention.

[0023] Figure 1 The influence of an uneven container geometry as an interfering parameter for measuring container parameters is shown.

[0024] Figure 2 A block diagram of an apparatus for manufacturing a container according to an embodiment of the invention is shown.

[0025] Figure 3 An exemplary filling device including an apparatus for manufacturing a container according to an embodiment of the invention is shown. DETAILED DESCRIPTION

[0026] The present invention relates to the manufacture of containers. According to the present invention, operating parameters for manufacturing a container can be adjusted based on performance parameter values determined based on measurement data corrected based on a model.

[0027] Figure 2 An apparatus 200 for manufacturing a container according to one embodiment is schematically shown. The apparatus 200 may include, for example, a blow molding machine for forming a container from a preform. According to Figure 2In the illustrated embodiment, the apparatus 200 for manufacturing a container includes a control device 210 for controlling the manufacturing process, and the control device 210 is configured to set the working parameters for manufacturing the container. The working parameters may include the pressure applied to the plastic preform, the time and duration of applying the pressure, and the movement parameters of the stretching rod for stretching the plastic preform.

[0028] The apparatus 200 further includes a measuring device 220, which is configured to determine at least one container parameter value of the container (or at least one container). The container parameters may include parameters characterizing the wall thickness of the container (e.g., the wall thickness itself or the absorption or transmission ability or absorption (quantity) or transmission (quantity)), the absorption coefficient of the container material, the transmission coefficient of the container material, the wall thickness profile of the container, and the bottom inspection measurement data of the container bottom.

[0029] In addition, the apparatus 200 includes a measurement correction device 230, which is configured to correct the measured value of at least one container parameter provided by the measuring device based on at least one model parameter of the numerical model. The at least one model parameter may include at least one of the information about the surface, orientation, geometry of the container and the absorption / transmission coefficient of the container material, and the parameters of the preform of the container as at least one model parameter.

[0030] The numerical model may be generated by artificial intelligence (AI) / neural network. The measurement correction device 230 may include such artificial intelligence or such neural network.

[0031] The neural network can be understood as a tool suitable for simulating any desired non-linear function, and thus can also simulate, for example, the rules of fuzzy logic if these functions are available based on the examples available for training the neural network. From a large number of examples, the regularities of the neural network can be learned / trained, and thus the weights of the neural network can be learned / trained, and then expressed by means of predetermined but also adaptable rules, such as fuzzy sets and rules. The combination of a fuzzy controller and a neural network allows the setting and parameterization of fuzzy rules based on intelligent learning.

[0032] The neural network can be one of a Transformer and / or a deep (learning) neural network, a multi-layer perceptron, a convolutional neural network, and a recurrent neural network. The deep (learning) neural network is characterized by multiple hidden layers. Through deep learning, the machine can improve its capabilities and make decisions on its own without human intervention by extracting and classifying patterns from existing data and information. The acquired cognition can then be associated with the data and related in a broader context. Finally, the machine can make decisions based on these associations. By constantly questioning the decisions, certain weights are obtained for the information associations. If the decision is confirmed, the weight increases, and if the decision is modified, the weight decreases. There are always multiple hidden intermediate layers and associations between the input layer and the output layer. The true output is determined by the number of intermediate layers and their current associations. The multi-layer perceptron is a relatively simple and stable neural network in which all nodes are fully connected. The convolutional neural network is based on convolution rather than matrix multiplication. The recurrent neural network allows feedback from one neural layer to the previous neural layer. The variant model neither has a convolutional neural network nor a recurrent neural network but is based on the concept of self-attention. However, the variant model can interact with the convolutional neural network or the recurrent neural network.

[0033] In addition, the device 200 includes performance parameter determination means 240 configured to determine at least one performance parameter value based on the at least one corrected measurement value / container parameter value. The at least one performance parameter may include top load, ground clearance, burst pressure, stress crack, material distribution, and sectional weight. The determination of the at least one performance parameter value is also based on a model, for example, a model including the above numerical model.

[0034] The control means 210 of the device 200 is further configured to control the manufacture of further containers based on at least one operating parameter, which is adjusted based on the determined at least one performance parameter value. Thus, the manufacture of further containers is controlled based on the very precise model-corrected measurements and the model-determined performance parameter values obtained therefrom, enabling the manufacturing process to be very finely optimized.

[0035] For example, the measured parameter value characterizing the container wall thickness can be corrected by means of a model that provides at least information about the geometry of the container and its orientation during the measurement. For example, infrared rays used for measurement that are perpendicularly incident on the side wall of the container pass through less container material than infrared rays incident on the conical head region of the same container at an angle other than 90°. This situation and the resulting measurement error can be taken into account and corrected by the model. Then, the corrected parameter value characterizing the container wall thickness (e.g., including the absorption (amount) of infrared rays) can be used to determine the material distribution or sectional weight, which is also based on a model.

[0036] Figure 3shows what can include Figure 2 The exemplary filling device 300 of the device 200 for manufacturing containers shown. The filling device 300 for filling containers 302, 303 (such as plastic bottles) with a liquid product (such as beverages, etc.) includes a filling machine 305 for filling and closing the containers 302, 303 and a distribution device 306 provided downstream of the filling machine 305 for distributing the containers 302, 303 onto two separately controllable transport paths 307, 308. In each case, at least one container buffer 309, 310 is provided with adjustable container guides 309, 310. Downstream of the container buffers 309, 310 are labelers 311, 312 and packaging machines 313, 314 for manufacturing container bundles 315. These are fed to a collection and distribution device 316 such that the container bundles 315 are distributed onto a sorting track 317 provided downstream of the collection and distribution device 316 and can be fed to a commissioning device 318.

[0037] The transport paths 307, 308 each include a first inlet-side portion 307a, 308a which is configured as a single track and is for the pressureless transport of the containers 302, 303. In addition, the transport paths 307, 308 include a second outlet-side portion 307b, 308b which is configured as a multi-track and is for the pressureless transport of the containers 302, 303. Diversion devices 307c, 308c or corresponding distribution means are provided for distributing the containers 302, 303 from the first single-track portion 307a, 308a onto the respective tracks of the second portion 307b, 308b, which second portion 307b, 308b is, for example, configured in the form of separate channels 307b1 to 307b3, 308b1 to 308b3.

[0038] In addition, the filling device 300 includes two devices for manufacturing containers 319, 320 in the form of blow molding machines 319, 320. In the example shown, separate blow molding machines 319, 320 are provided for manufacturing different containers 302, 303, such as containers of different geometries. At least one of the blow molding machines 319, 320 can be connected to the filling machine 3055 via an inlet-side transport path 321. Through an inlet-side diverter 305a, different incoming container streams can be fed for further processing. Additional production units 323, 324 can be provided, for example, in the form of shrink tunnels.

[0039] To control the filling device 300 according to the invention, a central control / regulation device 322 is provided which communicates, in particular, with the distribution device 306, the container buffers 309, 310, the labelers 311, 312 and the production units upstream of the distribution device 306, such as the filling machine 305 and the blow molding machines 319, 320. The central control / regulation device 322 can correspond to or include Figure 2The control device 210 of the illustrated device 200.

[0040] In the illustrated example, the labeling machines 311, 312 are connected to their own communication and control devices K1, K2, the filling machine 305 is connected to its own communication and control device K3, and the blow molding machines 319, 320 are connected to their own communication and control devices K4, K5. Each of the communication and control devices K1, K2, K3, K4 allows a user to operate the corresponding machine via a suitable interface. According to one example, the communication and control device K4 or K5 includes Figure 2 The control device 210 of the illustrated device 200. The communication and control devices K1, K2, K3, K4, K5 are logically assigned to the corresponding machines of the filling device 300. Of course, all the machines of the filling device 300 can be equipped with their own communication and control devices, and the communication and control devices can be networked with each other so that they can exchange the operating states of the machines and the requirements of the operator. Generally, for safety reasons, the networking of the communication and control devices with other machines, mobile collaborative robots (CRs), and the operator's smart phones, etc. can be restricted to a defined internal area (e.g., in the form of a company's proprietary network), while exchanges can be made over the Internet for the autonomous learning of the communication and control devices, e.g., regarding voice recognition or speaker recognition.

[0041] The central control / regulation device 322 is connected to the communication and control devices K1, K2, K3, K4, K5 and can at least partially take over the coordination of the machines and transport technology, e.g., during the organization of device production and the conversion of product types. Each machine can be logically and / or physically assigned to the communication and control devices K1, K2, K3, K4, K5. The operator can operate the corresponding machine via the communication and control devices K1, K2, K3, K4, K5, e.g., by means of voice input and voice dialogue. The communication and control devices K1, K2, K3, K4, K5 can use the display devices located on the machines to display information. The implementation of a central and distributed data processing and database accessible by the central control / regulation device 322 and the communication and control devices K1, K2, K3, K4, K5 is possible.

Claims

1. An apparatus (200, 319, 320) for manufacturing a container (302, 303), comprising: a control device (210, K4, K5) configured to set at least one operating parameter for producing at least one first container (302, 303); a measuring device (220) configured to measure at least one container parameter of at least one first container (302, 303) being manufactured to obtain a measured value of the at least one container parameter; a measurement correction device (230) configured to correct a measured value of the at least one container parameter based on at least one model parameter of the numerical model; a performance parameter determination device (240) configured to determine at least one performance parameter value based on the corrected at least one measured value; And among them A control device (210, K4, K5) is configured to control the production of at least one second container (302, 303) different from the at least one first container (302, 303) based on at least one operating parameter adjusted based on the determined at least one performance parameter value.

2. The apparatus (200, 319, 320) of claim 1, further comprising a neural network configured to establish the model.

3. The apparatus (200, 319, 320) according to claim 2, wherein the neural network is configured to train the model after manufacturing the at least one first container and / or the at least one second container (302, 303).

4. The device (200, 319, 320) according to any of the preceding claims, wherein the measuring device (220) is designed to measure a characteristic parameter of the light transmission, in particular the wall thickness of the first container (302, 303).

5. The device (200, 319, 320) according to any of the preceding claims, wherein the model comprises as the at least one model parameter at least one of information about the surface, orientation, geometry and absorption / transmission coefficient of the container material of the container (302, 303) and parameters of a preform of the container (302, 303).

6. A method for manufacturing a container (302, 303), comprising the following steps: Providing a container (302, 303) manufactured based on at least one operating parameter; measuring at least one container parameter of a provided container (302, 303) to obtain a measured value of the at least one container parameter; correcting the measured value of the at least one container parameter based on at least one model parameter of a numerical model; determining at least one performance parameter value based on the corrected at least one measured value; adjusting the at least one operating parameter based on the determined at least one performance parameter value; as well as The container (302, 303) is manufactured based on the adjusted at least one operating parameter.

7. The method of claim 6, further comprising establishing the numerical model using a neural network.

8. The method according to claim 7, further comprising training the neural network based on the adjusted at least one operating parameter after manufacturing the container (302, 303).

9. The method according to any one of claims 6 to 8, wherein the container parameter is a characteristic parameter of the light transmission, in particular the wall thickness of the container (302, 303).

10. The method according to any one of claims 6 to 9, wherein the model comprises at least one of information about the surface, orientation, geometry and absorption / transmission coefficient of the container material and parameters of a preform of the container (302, 303) as the at least one model parameter.

Images