Method and system for detecting and correcting malfunctions of a machine

Abstract

The invention relates to a method and system for detecting and correcting malfunctions of a machine, in particular a machine in a machine line for filling and packaging food and / or beverages. The method starts with receiving a sensor signal from the machine. The sensor signal can indicate a malfunction of the machine and include a context of the malfunction. An instruction for correcting the malfunction is then identified, which is selected from an instruction database based on the context of the malfunction. An operator of the machine is notified and receives information about the machine malfunction and, at the same time, the identified instruction for correcting the detected malfunction. The operator has the option of creating modified instructions for correcting the machine malfunction and sending them back to the database.The modified instruction is then stored in the instruction database and linked to the sensor signal, making it identifiable using the context of the malfunction.

Description

The invention relates to a method and system for detecting and eliminating malfunctions of a machine, in particular a machine in a machine line for filling and packaging food and / or beverages.The increasing complexity of machine lines, particularly in machine lines for filling and packaging food and / or beverages, presents a significant problem for maintenance and fault diagnosis. Modern filling systems are characterized by a large number of individual components and their interactions, which makes the localization and correction of defects a demanding and time-consuming task. The current methods for error correction are often not sufficient to meet the increased requirements.Known documentation systems, usually in the form of comprehensive PDF files, are not only awkward, but also static. They provide little support in interactive problem resolution and are often not up-to-date with the particular plant configuration. Operators often have to work through many pages of text and diagrams to find specific information, which takes valuable time and increases the risk of relevant information being overlooked. This procedure is particularly problematic in the case of unpredictable errors that are outside the standard procedures.Moreover, the process of debugging is often an isolated event. Even if an operator successfully solves a problem, the solution is rarely documented in a manner that is accessible or comprehensible to other operators. This results in inefficient knowledge propagation and repetition of the debugging processes with similar or identical problems in the future. The lack of an integrated system for error detection and documentation prevents the construction of a knowledge base that is essential for continuous improvements and fast response times.Current systems thus often have the disadvantages of a high time outlay for the problem diagnosis and solution, inefficient use of knowledge and experiences and a limited ability to adapt to new or modified installation configurations.These factors can result in longer downtime, increased operating costs, and less flexibility in production.There is therefore a need for a solution that overcomes these disadvantages in the prior art and thus provides an efficient dynamic technique for detecting and remedying malfunctions of machines for filling and packaging machines, lines and plants based on a digital twin.The object is achieved according to the invention by a method according to claim 1 and a system according to claim 10. Embodiments and developments are covered in the dependent claims.One embodiment of the invention relates to a method for detecting and eliminating malfunctions of a machine, in particular a machine in a machine line for filling and packaging food and / or beverages. The method starts with receiving a sensor signal from the machine. The sensor signal may indicate a malfunction of the machine and include a context of the malfunction. Then, a guidance for remedying the malfunction is identified, which is selected from a guidance database based on the context of the malfunction. An operator of the machine is notified and receives information about the malfunction of the machine and, at the same time, the identified instruction for remedying the detected malfunction. The operator has the possibility of creating a modified instruction for correction of the malfunction of the machine and sending it back to the database. The changed instruction is then stored in the instruction database and linked to the sensor signal, whereby it is identifiable by means of the context of the malfunction.A further embodiment of the invention relates to a system which implements the method.Exemplary aspects of the invention are illustrated in the drawings. The following are shown: FIG. 1 is a diagram showing an outline of the essential elements and the basic structure of the invention; FIG. 2 : shows an exemplary user interface for displaying and processing instructions; FIG. 3 : an exemplary plant configuration for PET containers and adhesive containers; FIG. 4 : shows an exemplary system configuration for PET containers and shrink packers; FIG. 5 shows an exemplary plant configuration for cans or glass bottles; and FIG. 6 : an exemplary plant configuration for cans.The invention aims to provide not only a more efficient fault detection, but at the same time also to provide a targeted solution strategy for remedying the detected machine fault. FIG. 1 shows an overview of the essential elements and the basic structure of the invention. A machine plant according to embodiments includes one or more machines 110 connected to an edge device 120 via a sensor connection.According to embodiments, machine problems or malfunctions of a machine 110 may be measured using sensors. The sensor values may then be transmitted via the edge device 120 to a computer 130, such as a server or cloud, for further processing.The sensors coupled to the machines 110 may sense various physical and operational parameters that may be used for condition monitoring and fault detection. Various sensors, such as vibration sensors, temperature sensors, pressure sensors, and / or flow meters, may be strategicly attached to critical points of the machine. For example, vibration sensors may detect irregularities in the operation of bearings or transmissions while temperature sensors may identify overheated components. It should be noted that these sensors are exemplary only and many other types of sensor fault detection may be used herein.When the captured data is transmitted to the edge device 120, it may provide the sensor signals with a context to effectively use the sensor data for fault diagnosis. According to embodiments, the sensor signals may already be provided with a context when being generated at the sensor of the corresponding machine 110.The context can be an indication of a type of fault, an indication of a component in question and / or a specific parameter which deviates from a normal state. The context may include identification of the specific machine 110 or component, environmental conditions, and / or operational state at the time of data acquisition. For example, each sensor may be provided with a unique ID that associates it with a particular machine or component. According to embodiments, sensors may also acquire additional data such as room temperature, humidity, or operating times to maintain the operating constraints or operating modes such as load conditions, speeds, or production counts may be recorded along with the sensor data. It can be determined by means of the sensor values in various ways that a malfunction of the machine is present.The sensor data may be sent from the edge device 120 to the server 130 along with the context in real-time or at fixed intervals. The data transmission can take place via wired networks (Ethernet), via the Internet or by means of wireless technologies (WLAN, Bluetooth, ZigBee).The logic for detecting machine problems and malfunctions of the machine 110 may be executed on the machine 110 itself or on the server 130. For example, software in server 130 may use machine learning and / or pattern recognition algorithms to draw conclusions from the sensor data and context and identify errors.To resolve and remedy the malfunction, the server 130 may access a guidance database 150 that includes a plurality of different guidances. The guidance database may be implemented in the server 130, or may be implemented separately in another computing device. By analyzing the context, one or more instructions suitable for remedying the malfunction of the machine 110 may be identified in the instruction database 150.According to embodiments, efficient and systematic error correction is thus made possible by virtue of the relevant repair and maintenance instructions for the problems determined being able to be provided automatically. The sensor data and / or the context of the sensor data can / can be matched to entries in the solution database 150, wherein the context of the sensor signals is taken into account in identifying instructions. The guidance database 150 comprises a plurality of approaches and guidance for various malfunctions, for example classified according to machine type, fault type and / or context.Once a matching solution or guidance is found in the guidance database 150, the corresponding guidance can be presented to an operator, for example by sending the guidance to the HMI or the mobile device 140 together with the error message. In this example, both the fault message about the detected fault at the machine 110 and the identified misfunction correction command may be sent to the operator's HMI or mobile device 140 simultaneously. However, error message and solution guidance need not necessarily be sent simultaneously to the operator. The operator may also manually select, via the HMI or the mobile device 140, to request a solution to the problem.The presentation can take place via a user interface of the mobile device, as shown by way of example in FIG. 2 and discussed further below. The guidance may be a step-by-step guidance in which clear, comprehensible instructions for error correction may be displayed one after another. The guidance can also include images, diagrams and videos, which illustrate the steps. The operator can thus see all steps of a manual one after the other in order to be able to focus on the step which is actually to be carried out. For example, an operator may navigate through a swiping gesture on a touch screen between the steps or jump to the next instruction when the steps of the previous instruction are completed.The instruction database 150 contains both individual recommendations and instructions created by customers and standard instructions based on the industry knowledge integrated by a manufacturer (technical documentation).These instructions can be accessed by operators and other workshop personnel during the fault finding process. In addition to accessing the entire database, the instructions are proposed in context-related fashion on the basis of the machine problems that have occurred / are pending, as described.According to embodiments, it is also possible that an operator himself can create or change instructions. For example, an operator may change and / or supplement a guidance of the guidance database via the HMI or the mobile device 140 and send it back to the guidance database 150. The changed guidance may be stored in the guidance database 150 such that the stored changed guidance is associated with the sensor signal and identifiable by the context of the malfunction.According to embodiments, the modified instruction may comprise one of the following features: an added and / or removed comment, and / or an additional action instruction to be performed by an operator, and / or a modified sequence of action instructions, and / or an alternative or corrective indication to an existing action instruction or an existing comment. Storing the changed guidance may include validating one or more correlated data points in the guidance. The associated rules can be adapted to the individual requirements and operating conditions in the production line.These extensions or changed instructions stored in the instruction database 150 can thus be linked by the operator to the original problem (e.g. to the sensor signal) and can be proposed by the system at the next occurrence of the problem.In general, when creating a manual, the operator can begin describing the first step, which can be automatically included in the title of the entire solution to the problem, in order to make creating contents as simple as possible. Guidance may be established generally or with respect to a particular context (relationship between touch messages and data point(s)). When created with respect to a machine problem, the created instruction may include a reference to the associated machine data (e.g., machine message, condition based maintenance (CbM) rule, line control event).According to embodiments, an approval process for later development steps may be implemented to sort out potentially defective or defective instructions. Altered instructions may be characterized visually or text-based as such to show an operator that the instruction is not an instruction generated by the manufacturer.According to embodiments, the instructions in the instruction database 150 may have a standardized shape and format. Guidance begins with a title that renders the content as clear and indiscernible as possible. This ensures that an operator immediately recognizes what the instructions are. An optional description may provide additional information and context that helps the operator to better understand the relevance and scope of the guidance.To facilitate the discoverability in the database, each instruction can be assigned to a specific category. These categories, such as debugging, maintenance, cleaning, lubrication, or changeover, may be defined by a user and help speed the search and discovery of the corresponding instructions.For further specification and for targeted use, references to the equipment can be included in the instructions. These references may extend over different levels of the enterprise, from the overall organization down to the specific machine within a production line. This allows granular assignment and application of the instruction.Attachments can supplement the instructions by providing further resources such as diagrams, photographs or videos. The instructions are structured in steps that guide the user through the process of error correction. This stepwise approach serves to reduce complexity and ensure that all necessary actions are performed in the proper order.The administrative information in the instructions which can be automatically filled in by the system includes the information about the creator of the instruction as well as the creation date and the last change date. This information is important not only for traceability and documentation, but also for quality assurance and management of maintenance histories. This structured approach guarantees that instructions for solving malfunctions are not only effective and efficient, but also better manage and search, which can support smooth operation and simplify maintenance work.Users can search the list of instructions available, for example, using search criteria. For example, free text search can be applied by title and description, and / or useful suggestions and auto-completion functions can be implemented by the system. This can help guide the users to the destination more easily. The search function within database 150 may also include various filtering criteria, such as "equipment" (each level of equipment may be selected (e.g., business, area, machine)), "category.".For a malfunction in a machine 110, several approaches or instructions may also be present in the database 150 or 160. Identifying the instruction(s) for remedying the malfunction may include the steps of identifying a plurality of instructions relevant to the context of the sensor signal and providing a selection of the plurality of instructions to the operator.In addition to individually generated instructions, there is also the possibility of integrating general solution descriptions from technical documentation by a manufacturer. Solution descriptions (for example from a global database 160) can be extracted from the technical documentation of a manufacturer (for example by means of artificial intelligence) and integrated as part of the instruction database 150.According to embodiments, a QR code may be automatically generated for instructions in the instruction database 150. These QR codes can be printed out, for example, and attached to corresponding machines. When scanning the QR code with the camera of the mobile device 140, the associated instruction can be opened.According to embodiments, the system may also automatically translate a selected command into a language the user has selected.According to further embodiments, the system may also include the possibility of evaluating instructions. In this case, the operating personnel can be given the option of evaluating the instructions. Assessment options can be, for example, "The instruction is useful (precipitated mirror)", "The instruction is generally useful", "The instruction is useful with regard to an occurring problem", etc. Furthermore, textual feedback can also be provided. These evaluations can provide ranking should there be multiple instructions for remedying a malfunction. Quality can also be improved.These user interactions may also be used as a basis for an incentive system. If there are instructions regarding problems that have occurred, the staff at the shop can evaluate the proposed instructions. Based on this assessment, the system lists the instructions sorted by the user's assessment (in most cases, the first displayed artifact solves the problem).The system will be all the better the more content available that is created by operators in production. An incentive for operators to make still further solutions could be sending push notifications in certain situations, such as "someone has fallen an instruction because it was helpful" or "someone has extended an instruction.". Also, user ranking may be implemented in some embodiments, which displays the users with the most-created instructions or best instructions.According to further embodiments, artificial intelligence may be implemented if the guidance database 150 does not suggest a matching solution. For example, with the aid of applications such as large language models, solutions could be sought which were previously trained with technical documentation of the machine 110.FIG. 2 shows two views 240 aand 240 bof a mobile device 140 according to example embodiments. As shown in view 240a, upon detecting a malfunction of a machine 110, a push message may be sent to the mobile device, thereby immediately notifying a responsible operator. The operator may select the message and thereby be provided with further information as seen in view 240b. At this time, the operator can immediately start the error correction by selecting and performing the simultaneously displayed guidance(s). In this case, the operator can change the instruction directly by means of the mobile device 140 and send it back to the database 150.Embodiments of the invention have several advantages over previous systems. The linking of instructions and malfunctions enables a faster and simpler access in the fault finding. Mobile access, for example via mobile device 140, enables location independence. Experience of workshop personnel has direct influence on the solution database. The operator can integrate his own instructions.In the following Figures 3 to 6, various exemplary system configurations for various bottle filling systems are described in which the invention or at least parts and aspects of the invention may be implemented. The description of Figures 3 to 6 is intended to provide a general overview of machines for which state data can be collected, on the basis of which the LLM can process user requests.FIG. 3 shows an example plant configuration 1000 for PET bottles or PET containers and adhesive containers. As can be seen in FIG. 3, the system configuration 1000 comprises a wide variety of modules which form a line at the end of which completely filled PET containers are discharged in the form of a pack on pallets. Some of the modules and machines may be optional and the invention is not limited to the exact shape and arrangement of the plant configurations.The system configuration 1000 comprises an oven 1002 for preforms, a preform sorting with a feed machine 1004 and a blow molding machine 1008. Modules 1002, 1004, and 1008 generally form a stretch blow molding machine in which PET containers are made and molded from a stock material. The produced PET containers are passed on into a filler 1010, in which the bottles are filled. The filler may optionally comprise a rinser. During storage or transport, various particles such as dust, cardboard or residues of wood pallets can be deposited in the preforms. These can be removed with the rinser. At the end of the filler, a closure device can be arranged, by means of which the PET containers are closed after filling.Optionally, the system configuration 1000 may comprise a turning device after the filler 1010, which is used when the PET containers are hot-filled. Via one or more conveyor belts 1016, which may also comprise a buffer 1018 for intermediate loading of filled containers, the filled PET containers are guided to a singler 1020 and further to a drying device 1024, in which the PET containers are dried.After drying has been completed, the PET containers are conveyed into a labeling machine 1026. The labeling machine 1026 can be designed for various labeling techniques, such as labeling using hot glue, cold glue, self-adhesive labels or Sleeves. After printing or labeling the PET containers, the PET containers are directed to a handle applicator through a second drying device 1028, a line spreader 1030, conveyor belts 1032, an adhesive container make 1034, and a curing line. In the adhesive package fabrication 1034, the PET containers are grouped together into certain group sizes and packaged into a package, such as a "sixpack.". In the handle applicator, a carrying handle is attached to the container, which enables a comfortable carrying of the container. The finished containers are then arranged accordingly by a robot 1042 for layer production and are packed on pallets by a palletizer 1044.In the system configuration 1000, so-called format carriages or format racks can be arranged on different modules and machines in order to provide rapidly exchangeable format sets for short changeover times and automatic tool changes. Examples of format carriages are the format carriage 1006 for the blow machine 1008, the format carriage 1012 for the filler 1010, the format carriage 1022 for the labeling machine 1026, the format carriage 1038 for the adhesive package production 1034 and the format carriage 1046 for the palletizer 1044.FIG. 4 shows a further exemplary system configuration 1100 for PET containers and shrink packers. The plant 1100 of FIG. 4 includes many of the modules and machines from the plant configuration 1000 of FIG. 3, but there are some differences. The description of the modules which have already been described in connection with FIG. 3 is therefore omitted for FIG. 4.A significant difference between the two example plant configurations 1000 and 1100 is that the labeling machine 1126 with the labeling modules 1127 may already be installed after the blowing machine 1008 and before the filler 1008. For this purpose, the plant configuration 1100 may comprise six transport tracks 1150 into which the PET containers may be inserted. After the PET containers have correspondingly entered one of the six lanes 1150, they are conveyed into the film tucking module 1152 and then into the shrink tunnel 1154.FIG. 5 shows an example plant configuration 1200 for cans or glass bottles. The example plant configuration 1200 of FIG. 5 again has some similarities to the plant configurations 1000 and 1100 of FIGS. 3 and 4, and the description of the plant configuration is therefore limited to the differences in the plant configurations.As shown in FIG. 5, the example plant configuration may include two separate feeds. A first feed, to the left in Figure 5, shows one branch at doses or optionally a subbranch at reusable bottles. The containers, i.e. cans or new bottles, are fed by a depalletizer 1302 into the machine, where they are fed via conveyor belts to the filler 1010. A second feed, to the right in Figure 5, shows a branch in reusable bottles which are introduced into the plant by a reusable sorting plant (not shown).In the case that the already used reusable bottles are introduced into the installation 1200 via the branch in the case of reusable bottles, the reusable bottles first pass through the cleaning machine or washing machine 1304. Another possible difference of the example plant configuration 1200 is the transfer packer 1306 downstream of the labeling machine 1026. The transfer packer may sort the bottles or cans into a carton clip application or boxes or both.FIG. 6 shows an example can plant configuration 1300 in which again the elements already described in the other plant configurations are no longer described. The cans in the facility configuration 1300 are introduced into the depalletizer 1302 from a can magazine 1402. The cans, after passing through the filler and being filled, are sealed by closure magazine 1404 and passed over the conveyor belts as described above along the line 1400.Optional pasteur 1408 may be bypassed via bypass 1412 if not needed. In the pasteur 1408, the freshly filled products can be pasteurized for preservation.In contrast to the plant configurations 1000, 1100, and 1200, in the example plant configuration 1300, various tanks for respective consumables are shown, such as the tanks 1410 with rinse liquid and / or the fill product and the tanks 1406 with belt lubricant. These tanks may also be included in the example plant configurations already described above. For example, the chemical products 106 that are directed from the mixer 110 to the machines may be stored in the tanks 1406 and 1410.

Claims

A method for detecting and remedying malfunctions of a machine (110), in particular a machine in a machine line for filling and packaging food and / or beverages, the method comprising: receiving a sensor signal from the machine, the sensor signal indicating a malfunction of the machine and comprising a context of the malfunction; identifying an instruction for remedying the malfunction, wherein identifying the instruction comprises a selection of the instruction from an instruction database (150) based on the context of the malfunction; creating a notification for an operator of the machine, wherein the notification comprises information about the malfunction of the machine and the identified instruction for remedying the detected malfunction; outputting the notification to the operator; receiving a modified instruction for remedying the malfunction of the machine from the operator; and storing the modified instruction in the instruction database, wherein the stored modified instruction is associated with the sensor signal and identifiable by the context of the malfunction.The method of claim 2, wherein the altered guidance comprises at least: an added and / or removed comment; and / or an additional action instruction to be performed by an operator; and / or an altered order of action instructions; and / or an alternative or corrective indication to an existing action instruction or an existing comment; and wherein storing the altered guidance comprises validating one or more correlated data points in the guidance.The method of claim 1 or 2, wherein identifying the guidance to remedy the malfunction comprises: identifying a plurality of guidances relevant to the context of the sensor signal; and providing a selection of the plurality of guidances to the operator.The method of any one of claims 1 to 3, further comprising: automatically translating the instruction issued to the operator into a language selected by the user.The method of any of claims 1 to 4, further comprising: validating the changed guidance before the changed guidance is identifiable using the context of the malfunction.The method of any of claims 1 to 5, wherein users have access to the guidance database and instructions in the guidance database are user-valable to create contextual ranking, and wherein instructions are selected based on the context and based on the ranking.The method of any of claims 1 to 6, further comprising: automatically generating a QR code that enables direct access to a command in the command database.The method of any of claims 1 to 7, wherein the guidance is a step-by-step guidance displayable on a mobile device.Method according to one of Claims 1 to 8, wherein the context comprises: an indication of a type of the fault, an indication of a component in question and / or a specific parameter which deviates from a normal state.A system for detecting and remedying malfunctions of a machine (110), in particular a machine in a machine line for filling and packaging food and / or beverages, the system comprising: one or more machines (110); one or more sensors configured to measure data of the one or more machines; an edge device (120) configured to receive the data from the one or more sensors as sensor data; a computer device (130) connected to the edge device, wherein the computer device comprises a guidance database (150), and wherein the computer device is configured to: receive the sensor signal from the machine, wherein the sensor signal indicates a malfunction of the machine and comprises a context of the malfunction; identifying a guidance for remedying the malfunction, wherein identifying the guidance comprises selecting the guidance from the guidance database based on the context of the malfunction; creating a notification to an operator of the machine, wherein the notification comprises information about the malfunction of the machine and the identified guidance for remedying the detected malfunction; outputting the notification to the operator; receiving a changed guidance for remedying the malfunction of the machine from the operator; and storing the changed guidance in the guidance database, wherein the stored changed guidance is linked to the sensor signal and is identifiable by means of the context of the malfunction.

Images