DE102009000938B4 - Method for controlling an extrusion line - Google Patents

Abstract

Method for controlling an extrusion line (1) as a function of material data of a discontinued material (2), wherein material data is continuously recorded via at least one detection unit (3) before the material is subjected to processing in the extruder (1), characterized in that By means of data comparison and / or mathematical processes, one or more characteristic values are ascertained from the acquired material data and then a data set with the required machine parameters is read from a database (4) and transmitted to the extrusion line (1) on the basis of the determined characteristic value (e).

Description

The invention relates to a method for controlling an extrusion line as a function of material data of a discontinued material, wherein material data is continuously recorded via at least one detection unit before the material is subjected to processing in the extruder.

The increasing quality requirements and the requirement for minimal reject production on the products require early intervention in the process parameterization, for example in extruder machines and their downstream units. To solve these requirements, methods that map by means of neural networks, neuro-fuzzy, a self-learning, correcting and regulating system, are known. There are also different control algorithms and combinations used to optimize the process behavior and approaches to the theory of evolution.

Typically, a wide variety of materials are processed on a production machine, to which the process parameters are adapted. However, it has also been shown that the input material data from batch to batch and manufacturer on a production plant are subject to scattering and that environmental influences such as raw material form, temperature and material moisture disturb the process and thus optimal process management to produce the required quality not without intervention the user is to be guaranteed. In particular, in the processing of recycled materials and recyclates, the optimal setting of a process, without the exact knowledge of the composition of materials, almost impossible.

The retrieval of machine data from a database for the production of injection molded parts is from various writings such as DE 10 2004 052 499 A1 , of the DE 10 2006 033 421 B3 or the DE 699 36 385 T2 known. All this state of the art has in common that the injection molding a finished product produced while extruding only semi-finished products. In addition, injection molding is a discontinuous process in which the feed screw acts as a piston, which is not the case with extrusion.

Due to the fact that a continuous detection is required in the extrusion, since not as in injection molding, first a detection, then a rating, then a setting and then the process can be set in motion, a 1: 1 transmission is not possible.

The object of the invention is therefore to provide a method by means of which the influence of the material which is given to the plastic processing machine is automatically taken into account.

The object is achieved in conjunction with the preamble of claim 1, characterized in that by means of the detected material data by data comparison and / or mathematical processes one or more characteristics are determined and then read from the database using the determined characteristic value a record with the required machine parameters and transferred to the extrusion line.

Thus, in the continuous process, the incoming specific material data are recorded and evaluated accordingly. Based on this acquired data, the corresponding control, which is realized for example via a PLC or a similar control, look up in a knowledge database to query the machine parameters required for this material data and transfer it to the plastic processing machine.

The idea is therefore based on combining known detection systems and existing databases in a suitable manner so that a largely automatic adjustment and correction of the setting of a machine can be carried out. The detection systems determine from the acquired data e.g. a kind of index that matches a specific record number in the database. If this data record is not found, an algorithm or other mathematical system is used to determine which nearby data record is read out and whose parameter is transmitted to the machine control. If the read-out data / parameters deviate too much from the desired characteristic values, a warning can be issued. Here is a warning in the display or as a light or in any acoustic form conceivable. The machine personnel can then observe the product result, correct the parameters manually and, if necessary, save them as a new data record with a new characteristic value. If analogue data is then acquired at a later date, the system then finds the desired data record with the machine parameters in the database.

In this way, the system is successively enriched with additional data and evolves into an independent expert system.

It is thus possible, for example, to increase the temperature of an extruder with extremely cold material and thus to achieve the required degree of melting without the influence of the operating personnel. As another example, the speed of the extruder could vary depending on the shape of the material being detected be adjusted accordingly, since in a round granule or a flake-shaped recyclate other extruder speeds are useful.

The proposed method is suitable both for recording the material data of typical plastics such as thermoplastics and for clay, ceramics, rubber, rubber, organic compounds, composites, biomaterials or even metals and similar materials.

The detection unit, which may be, for example, a material sensor, detects both the type of material and material color, the composition of the material, the shape of the input material, the temperature and the humidity. Certainly it is also conceivable to be able to record further material data. As a sensor, a Nahinfrarotspektroskop and / or humidity and temperature sensors can also be used. The data acquisition does not necessarily have to take place via a detection unit, it is also possible for several sensors to transmit the data to the process control.

In this control unit, which is either integrated directly in the plastic processing machine, but can also be housed within a company, in a data center or otherwise, now the material data are processed. As a function of this determined material data already stored machine parameters are retrieved in a knowledge database and, as already described above, transmitted to the plastic processing machine. The processing can be achieved, for example, by determining a specific code for this material data, in which case the machine parameters are queried on the basis of this code in the knowledge database. If the determined characteristic deviates, for example, from a stored characteristic value in the database, a stored mathematical algorithm decides which of the adjacent data sets is read out as a machine parameter.

The machine parameters transmitted to the plastic processing machine are output in parallel in a suitable manner, which is made possible for example via a display or touch screen on the machine.

Via an input location, which can be the touch screen field on the machine, for example, it is possible to manually correct the machine parameters. This may be necessary if the desired result is not yet optimal and a correction is required. This corrected data can then be stored as a new record of the machine parameters in the knowledge base and be available to the system for follow-up processes.

During the process within the plastics processing, further measured values can be recorded, which are also transmitted to the process control. This procedure makes it possible to check the machine parameters supplied by the database as part of a result check and thus to individually correct individual or all parameters. Of course, these correction values can be stored as new values in the database.

Thus, the comparison of the measured data with a knowledge database for optimal automatic process parameterization and control of the plastic processing machine is performed. Process data suitable for the parameterization are e.g. Torque and speed of the main engine and / or other auxiliary drives and the temperature profile of the process unit and the tool and the melt data. Furthermore, it is conceivable to optimally parameterize, for example, in the extrusion downstream of the extruder calibration units their cooling and pressure behavior. The knowledge database is therefore designed for all extrusion processes of mono- and multi-layer extrusion. As mentioned above, the knowledge base can be integrated into the extruder control as well as centrally available on an in-house server and serve single and / or multiple extruder machines or plastic processing machines simultaneously.

Based on the interaction of the knowledge base, acquisition of input material data and control, an active processing recommendation is achieved, which enables the operator to optimally adjust the processing process. Based on the learned processing recommendations, the successively trained system shown here is gradually able to perform the process independently optimally.

Based on an extrusion line, the method will be explained by way of example in a figure. The figure shows a plastic processing machine 1 , the example here as an extruder 6 is shown, which is a material feed hopper 9 having. The extruder 6 has a drive motor 7 and one on the extruder 6 arranged tool 10 , In the material feed hopper 9 becomes the material 2 given up. In this area of the material task is the registration unit 3 arranged. The registration unit 3 transmits the acquired material data to a computer unit, which is shown here by the dashed line. The data flow is indicated by the arrow from the capture unit 3 shown to the computer unit. Within the computer unit are now through the process control 5 in connection with the database 4 determined by suitable mathematical processes, the required machine parameters. The computer unit transmits the read-out machine parameters to a display 11 and the plastic processing machine 1 , The three double arrows between the plastic processing machine 1 and the computer unit in turn identify the possibility of data exchange, since, as described above, also return values from the processing process can influence the process control. About the display 11 , which can act as an input device at the same time, correction values or of course completely new records in the knowledge base 4 be entered.

Using the example of this extruder line, for example, a second material or product sensor may be placed prior to shaping and / or in succession in order to obtain a feedback between the product properties and the plasticizing or conditioning process. For example, the gloss level of the product could be determined by the second material sensor and returned to the controller. This could in turn be used to control the nozzle heater to automatically set or change the previously defined gloss of the product.

Here is a system that is trained by the experience and reactions of users and experts to different material properties. The process parameters thus found are stored as data sets in a database and can be recalled at any time. In combination with a material condition detection, an immediate intervention into the process is possible. It can be optimally adjusted processes for tube, profile, flat films, plates, granules, blown film, but also injection molding and compounding.

• • Mit der vorgeschlagenen Erfindung kann die kontinuierliche Erfassung der spezifischen Materialdaten, der in den Extruder bzw. die Plastifiziereinheit einlaufenden Materialien, durch einen Materialsensor, der die Messergebnisse auflöst nach Materialtyp, -zusammensetzung, -form, -temperatur und -feuchte, umgesetzt werden. Typisch für diese Anwendung ist z.B. bei Kunststoffen der Einsatz von Nahinfrarot-Spektroskopie, von Feuchte- und Temperatursensoren. Erfasst werden können Tone, Keramik, Gummi, Kautschuk, Thermoplastische Polymere, organische Verbindungen, Verbundwerkstoffe, Biowerkstoffe und Metalle.
• • Weiterhin ist der Abgleich der Messdaten mit einer Wissensdatenbank zur optimalen automatischen Prozessparametrierung und -steuerung der Extrudermaschine realisiert. Optimale Prozessdaten sind z.B. Drehmoment und Drehzahl des Hauptmotors oder/und weiterer Nebenantriebe sowie das Temperaturprofil der Verfahrenseinheit und des Werkzeugs. Weiterhin ist denkbar, z. B. die dem Extruder nachgeschalteten Kalibriereinheiten optimal in dem Kühl- und Druckverhalten zu parametrieren.
• • Die Wissensdatenbank ist ausgelegt für alle Extrusionsverfahren der Mono- und Mehrschichtextrusion. Die Wissensdatenbank kann sowohl in der Extrudersteuerung integriert sein als auch zentral auf einem firmeninternen Server zur Verfügung stehen und einzelne oder / und mehrere Extrudermaschinen gleichzeitig bedienen.
• • Eine aktive Verarbeitungsempfehlung auf Basis der Interaktion von Wissensdatenbank und Steuerung, die den Bediener in die Lage versetzt, den Verarbeitungsprozess zu optimieren, ist umgesetzt.
• • Darüber hinaus ist es denkbar, einen zweiten Material- bzw. Produktsensor vor der Formgebung oder/ und in der Nachfolge zu platzieren, um eine Rückkopplung zwischen den Produkteigenschaften und dem Plastifizier- bzw. Aufbereitungsprozess zu bekommen. Beispielsweise könnte der Glanzgrad des Produktes, durch den zweiten Materialsensor bestimmt und an die PLC weitergegeben, dazu genutzt werden, um die Düsenheizung so zu steuern, um den im Vorfeld definierten Glanz des Produktes automatisch einzustellen.

In summary, therefore, as already mentioned above, mentioned again:

• With the proposed invention, the continuous detection of the specific material data, the materials entering the extruder or the plasticizing unit, can be converted by a material sensor which dissolves the measurement results according to material type, composition, shape, temperature and moisture. For example, the use of near-infrared spectroscopy, moisture and temperature sensors is typical for this application. Clays, ceramics, rubber, rubber, thermoplastic polymers, organic compounds, composites, biomaterials and metals can be included.
• • Furthermore, the comparison of the measured data with a knowledge database for the optimal automatic process parameterization and control of the extruder machine is realized. Optimal process data include torque and speed of the main engine and / or other auxiliary drives, as well as the temperature profile of the processing unit and the tool. Furthermore, it is conceivable, for. B. the parameters downstream of the extruder calibration units optimally in the cooling and pressure behavior.
• • The knowledge database is designed for all extrusion processes of mono- and multi-layer extrusion. The knowledge database can be integrated into the extruder control as well as centrally available on a company-internal server and serve individual or / and several extruder machines simultaneously.
• • An active processing recommendation based on the knowledge database and control interaction that enables the operator to optimize the processing process has been implemented.
• In addition, it is conceivable to place a second material or product sensor before shaping or / and in succession in order to obtain a feedback between the product properties and the plasticizing or conditioning process. For example, the gloss level of the product, as determined by the second material sensor and passed to the PLC, could be used to control the nozzle heater so as to automatically adjust the pre-defined gloss of the product.

LIST OF REFERENCE NUMBERS

1

 Plastic processing machine

2

 material

3

 acquisition unit

4

 Database

5

 process control

6

 extruder

7

Drive from 6

8th

Temperature control of 6

9

 Material hopper

10

 Tool

11

 display

Claims (8)

Method for controlling an extrusion line ( 1 ) depending on material data of a discontinued material ( 2 ), whereby at least one registration unit ( 3 ) continuously records material data before the material is processed in the extruder ( 1 ), characterized in that one or more characteristic values are determined by means of the recorded material data by data comparison and / or mathematical processes, and then a data set with the required machine parameters from a database (FIG. 4 ) and sent to the extrusion line ( 1 ) is transmitted. A method according to claim 1, characterized in that in case of deviation of the determined characteristic value from the characteristic value stored in the database by means of an algorithm, the data set to be selected is determined. A method according to claim 2, characterized in that in case of deviation of the selected data set from the desired data set desired according to a visual and / or audible indication is triggered. A method according to claim 1, 2 or 3, characterized in that the transmitted machine parameters are output / displayed in a suitable manner. A method according to claim 4, characterized in that via an input point, the machine parameters can be changed manually. Method according to one of the preceding claims, characterized in that values recorded via measuring means within the plastics processing process are transmitted to the process control ( 5 ) and, depending on these values, a correction of the machine parameters is made possible. Method according to one of the preceding claims, characterized in that as material data, for example, data such as material type, composition, shape, color, temperature and / or moisture are detected. Method according to one of the preceding claims, characterized in that the detection of the material data for materials such as clay, ceramics, rubber, rubber, thermoplastics, composite materials is possible.

Images