Method for operating a conveying device and conveying device

Abstract

The invention relates to a method for operating a conveyor system (1) in which push plates (3) for receiving in each case one component (4) can be conveyed along a conveyor path (5) by means of stationary conveyor elements (2), comprising the following steps: detecting a respective conveyor time for a respective conveyor element during the conveying process of each push plate (3); (S1) detecting at least one respective parameter value for the respective conveying element (2) at the respective conveying time by means of the sensor device (6); (S2) forming a respective datagram for the respective conveying element (2) at the respective conveying time by the respective control device (7) of the respective conveying element (2); (S3) summarizing the datagrams into a dataset by means of the central electronic computing device (8); (S4) evaluating the data set by means of an anomaly model; (S5) outputting a maintenance indication if the evaluation results that the anomaly of at least one of the parameter values exceeds the threshold value. (S6) The invention also relates to a conveying device (1).

Description

Technical Field

[0001] The invention relates to a method for operating a conveyor installation according to claim 1 and to a conveyor installation according to claim 9 . Background Art

[0002] Conveyor systems are often used for assembly or production. Therefore, different conveyor systems or conveyor technologies can be used to assemble vehicles. For example, a conveyor system includes a push plate. Here, the push plate does not have its own drive, but is moved along the track by a fixed conveyor element.

[0003] The corresponding conveying element thus comprises a drive and, for example, a drive wheel. The push plate is held or guided by guide rollers on the guide rails. Usually, vehicles or vehicle components, in particular vehicle bodies, are transported on the ground (overflur) by means of push plates, during which assembly work is performed on them. Near the end or reversal point of the conveying section, the components are removed from the push plate and the push plate can be transported back, in particular, under the ground. The conveying of the push plate is achieved by means of a plurality of fixed or stationary conveying elements. These conveying elements constitute a drive, which is fixedly mounted and moves the push plate via the drive wheels.

[0004] A failure of one of the conveyor elements or of a push plate usually leads to a stoppage of the conveyor system, which causes high costs in vehicle production.

[0005] EP 1 800 196 B1 discloses a device for predicting the mean time period between two failures of a technical system.

[0006] A disadvantage of the prior art is that, for example, the current state of the conveying system is unknown and therefore irregularities or impending failures cannot be predicted. Summary of the invention

[0007] The object of the present invention is to provide a method and a conveyor system by means of which malfunctions of conveyor elements of the conveyor system can be prevented in a particularly advantageous manner.

[0008] According to the invention, this object is achieved by the subject matter of the independent patent claim. Advantageous embodiments are the subject matter of the dependent claims and the description.

[0009] A first aspect of the invention relates to a method for operating a conveyor system in which push plates for receiving a component can be conveyed or moved along a conveyor section by means of fixed conveying elements. The conveyor section of the conveyor system can in particular form a closed loop, which extends above ground for transporting components and, for example, along assembly points at which components, in particular vehicle body components, are processed. In order to convey or move the respective push plate back to the starting point, a part of the conveyor section can be arranged below ground.

[0010] In order to be able to prevent malfunctions of the conveying system in a particularly advantageous manner, the method according to the invention comprises a plurality of steps:

[0011] In a first step, a corresponding conveying moment is detected for a corresponding conveying device during the conveying process of the corresponding push plate or each push plate. The corresponding conveying elements in particular each include a drive device, which drives, for example, a drive wheel, which conveys the push plate along the track facility while in contact with the push plate. At least during the conveying of one of the push plates along the conveying section by the conveying element currently conveying, a certain time interval passes at a certain moment. In other words, in the first step of the method, it is therefore determined which conveying element conveys which push plate at what time and / or for how long. Here, the moment, i.e. the conveying moment detected or determined here, in particular characterizes or describes the end of the conveying process.

[0012] In a second step of the method according to the invention, at least one corresponding parameter value is detected for the corresponding conveying element at the corresponding conveying moment by means of a sensor device, said parameter value characterizing the state of the corresponding conveying element and / or the state of the push plate conveyed by the corresponding conveying element during the conveying process. In other words, during each conveying process, in particular each push plate, on each conveyor element or a plurality of conveyor elements to be monitored in the conveying device, a parameter value or a measured value and thus a parameter is detected by means of a sensor device suitable for the detected measured variable, by means of which the associated conveying element and / or the push plate currently being conveyed therein can be characterized or described.

[0013] In a third step, a corresponding datagram or data block is formed for the corresponding conveying element at the corresponding conveying time by the corresponding control device of the corresponding conveying element, which is in particular designed as a programmable logic controller, and which includes the at least one corresponding parameter value, the conveying time, the push plate ID (i.e. push plate identification code or identification feature) of the respectively conveyed push plate and the conveying element ID (i.e. conveying element identification code or identification feature) of the corresponding conveying element. In other words, a data packet is compiled or formed for each conveying process of the monitored conveying element, which records the at least one corresponding parameter value and for this purpose contains not only the time but also the identification of the push plate or the conveying element involved in the conveying process, so that the datagram allows a unique assignment of the at least one corresponding parameter value.

[0014] In a fourth step, the datagrams are combined into a data set by the central electronic computing device. In other words, the respective control device of the respective conveying element transmits its datagram to the central electronic computing device or the central electronic computing device calls up the respective datagram. The individual datagrams are now combined into a complete data set by the central electronic computing device.

[0015] Subsequently, in a fifth step of the method according to the invention, the data set is evaluated by means of an anomaly model, which is designed to evaluate anomalies of the at least one corresponding parameter value of the corresponding data set. In other words, an anomaly model is provided, which has been trained, for example, by means of a machine learning method or is at least generally designed to recognize deviations of parameter values ​​from nominal values ​​or to interpret the data set in such a way that upcoming changes in the corresponding conveyor element and / or the corresponding push plate are recognized, which changes may lead to a malfunction of at least one part of the conveyor device.

[0016] Finally, in a sixth step, if the evaluation performed in the fifth step results in an anomaly of at least one of the parameter values ​​exceeding a threshold value, a maintenance instruction is output for the corresponding push plate and / or the corresponding conveying element, the anomaly of the parameter value provided by the conveying element being identified by the anomaly model or the parameter value of the conveying element having an anomaly or the parameter value having an anomaly being detected in the conveying element. In other words, in the sixth step, if the anomaly model has determined that there is an anomaly in at least one parameter value, which is detected in one of the conveying elements or in one of the push plates or in a combination of the conveying element and the push plate, an instruction, for example in the form of a signal, is output to the maintenance robot or maintenance unit, so that an inspection or inventory of the corresponding push plate or the corresponding conveying element can be initiated or performed.

[0017] The method should be used to find abnormal conveying elements among all conveying elements. Here, the conveying elements can be identified by their unique conveying element IDs, so each conveying element is unique in the conveying element set. In order to identify abnormal conveying elements, at least one of its parameter values ​​is compared with other detected parameter values. This method can determine which one or which type of parameter value should be used in the at least one parameter value. Subsequently, the parameter value is aggregated or detected. The parameter value can also depend on the position of the standard or corresponding conveying element in the conveying device, for example. The data set can be formed, for example, within a specific time period, such as one hour of operation of the conveying device, so that the datagrams contained therein are aggregated into a data set. Therefore, a data set containing n datagrams and m at least one detected parameter value can constitute an input for an abnormal model. In the abnormal model, the deviation of the average parameter value can now be detected by means of, for example, an n-by-m matrix, so as to identify anomalies and calculate, for example, abnormal frequencies. Here, for example, the deviation of the Gaussian distribution can be determined in the standard deviation in order to find anomalies.

[0018] The method according to the invention thus yields a number of advantages. On the one hand, malfunctions of the conveying system can be prevented in a targeted manner. Furthermore, more convincing data can be generated due to the large number of available parameter values ​​and / or comparison elements in the form of conveying elements. Furthermore, the method can be applied to different systems and is independent of, for example, the number of conveying elements and / or push plates.

[0019] In addition, the method can be particularly advantageously performed, for example, because existing sensor devices can be used to detect parameter values. In particular, for example, each conveying element has its own sensor device for state monitoring. In addition or alternatively, parameters or parameter values ​​from almost any installed component can be used, which parameters or parameter values ​​can be read, for example, by the sensor device or in particular another sensor device of the corresponding component, such as a corresponding control device. Therefore, these parameter values ​​can be advantageously used for the method. In addition, it can be used independently of the manufacturer, because the parameter values ​​of different components can be used in common for the method according to the invention without any problem. In addition, the method for operating a conveying device is also suitable as a method for debugging a conveying device or a new conveying device, because the method can quickly react to errors that occur during the debugging process.

[0020] In an advantageous design of the present invention, the data set and / or another data set is used as a training data set for an anomaly model using a machine learning method. In other words, the anomaly model is at least partially based on a self-learning algorithm and / or a neural network and is therefore at least partially constructed as artificial intelligence. Here, the neural network can be trained, for example, by means of a so-called deep learning method (English: Deep Learning). In order to train the self-learning algorithm and / or the neural network, a training data set that is particularly marked is used, that is, the parameter values ​​with abnormalities are manually marked in advance by a staff member, for example. In addition or alternatively, a second instance of the self-learning algorithm can be used, which monitors the training. In other words, the data obtained by the method and aggregated into a data set in the form of a datagram can be used to train the anomaly model. In addition, the anomaly model is advantageously constructed as artificial intelligence. The advantage obtained thereby is that the method can be applied particularly effectively. In addition, the method can be particularly advantageously applied to new parameter values.

[0021] In another advantageous embodiment of the invention, material and / or temperature are detected as the at least one parameter. In other words, the corresponding sensor device for detecting the parameter value is designed to detect, in particular, an active current flowing through the conveyor element drive, and / or, for example, the temperature of a bearing on which the push plate is guided. This has the advantage that parameter values ​​that can particularly accurately describe anomalies can be detected in a particularly advantageous manner.

[0022] In another advantageous embodiment of the invention, the respective conveying element is conveyed using a drive device, which is in particular an electric motor. In other words, the conveying element is designed in such a way that it has its own drive device for conveying the push plate, which is preferably in the form of an electric motor, which is operated in electric motor mode. In this case, for example, electrical energy or current can be supplied to the electric motor via a converter, so that the conveying element can be conveyed. This has the advantage that the conveying device can be operated particularly advantageously.

[0023] In another advantageous embodiment of the present invention, the average active current and / or the maximum active current and / or the variance of the active current of the corresponding drive device to which the corresponding conveying element belongs is detected for the corresponding datagram. In other words, in addition to the parameter value, an active current of the corresponding drive device or a parameter characterizing the active current is also detected or determined. For example, corresponding measurements for detecting the corresponding active current can be performed on the converter of the corresponding drive device. The average active current or the maximum active current or the variance of the active current is always limited to the time period of the corresponding conveying process, i.e., the conveying duration. The advantage obtained from this is that the anomaly model can particularly advantageously evaluate the relevant information.

[0024] In another advantageous embodiment of the invention, the corresponding datagrams are preprocessed before or in order to combine the data sets. In other words, the datagrams are preprocessed, for example, by a preprocessor or the like, which can, for example, remove unrealistic parameter values ​​or, for example, datagrams with missing or incomplete data. This has the advantage that the method can be carried out particularly advantageously and the conveyor system can therefore be operated particularly efficiently.

[0025] In an advantageous embodiment of the invention, a time interval is determined in the preprocessing and / or one of the parameter values ​​is selected in the case of a plurality of parameter values ​​and / or a target is set. In other words, the preprocessing serves to determine a time interval, which therefore determines which datagrams are retained for the data set. Additionally or alternatively, for example, if a plurality of corresponding parameter values ​​are detected for a conveying element or a corresponding conveying process, only certain parameter values ​​are allowed for further processing or analysis by an anomaly model. Additionally or alternatively, a target can be set, i.e., in the preprocessing, for example, it is decided which anomaly or the degree of the anomaly should be determined. The advantage resulting from this is that the method can be carried out particularly efficiently. In addition, the conveying device can be operated particularly advantageously or protected against a failure of one of the conveying elements.

[0026] In another advantageous embodiment of the invention, the threshold value is adjusted as a function of the outlier value. In other words, the threshold value for determining when to output a maintenance instruction is adjusted as a function of an outlier value, in particular calculated or determined by a model, which specifies the degree of tolerable anomalies or contains an anomaly score. Thus, the threshold value can be adjusted dynamically, for example, as a function of a model that is improved by training data. This results in the advantage that the method can be carried out particularly efficiently or the reliability of the conveying device is particularly high.

[0027] A second aspect of the invention relates to a conveying device having a fixed conveying element and a push plate, each of which is used to receive a component and can be conveyed or moved along a conveying path by the conveying element, the conveying device being operable by means of the method according to the first aspect of the invention.

[0028] Advantages and advantageous embodiments and refinements of the second aspect of the present invention are hereby regarded as advantages and advantageous embodiments and refinements of the first aspect of the present invention, and vice versa.

[0029] Further features of the invention are apparent from the claims, the drawings and the description of the drawings. The features and feature combinations mentioned above in the description and the features and feature combinations mentioned below in the description of the drawings and / or shown individually in the drawings can be used not only in the respectively given combination but also in other combinations or individually. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will now be described in detail according to preferred embodiments and with reference to the accompanying drawings. The accompanying drawings are as follows:

[0031] Figure 1 A schematic flow chart showing a method for operating a conveyor system; and

[0032] Figure 2 A schematic side view of a conveyor device with a stationary conveyor element for conveying a push plate along a conveying path is shown. DETAILED DESCRIPTION

[0033] Figure 1 Shown for running Figure 2 Schematic flow chart of a method of a conveyor device 1 schematically shown in . The conveyor device 1 has stationary conveying elements 2 , by means of which push plates 3 for each receiving a component 4 can be conveyed or moved along a conveying path 5 .

[0034] The method comprises a number of steps.

[0035] In a first step S1, a corresponding conveying time is detected for the corresponding conveying element 2 during the conveying process of the corresponding push plate 3. In a second step S2, at least one corresponding parameter value is detected for the corresponding conveying element 2 at the corresponding conveying time by means of at least one sensor device 6, the parameter value characterizing the state of the corresponding conveying element and / or the state of the push plate 3 conveyed by the corresponding conveying element 2 during the conveying process. In a third step S3, a corresponding datagram is formed for the corresponding conveying element 2 by the corresponding control device 7 of the corresponding conveying element 2 at the corresponding conveying time, the datagram comprising the at least one corresponding parameter value, the associated conveying time, the push plate ID of the respectively conveyed push plate 3 and the conveying element ID of the corresponding conveying element 2. In a fourth step S4, the datagrams are aggregated into a data set by a central electronic computing device 8. In a fifth step S5, the data set is evaluated by an anomaly model, which is constructed for evaluating anomalies of the at least one corresponding parameter value of the corresponding datagram. Finally, in a sixth step S6, if the evaluation shows that the anomaly of at least one of the parameter values ​​exceeds a threshold value, a maintenance instruction for the corresponding push plate 3 and / or the corresponding conveying element 2 in which the parameter value with an abnormality is detected is output.

[0036] The respective conveying elements 2 can each have, for example, a drive wheel 9, by means of which the push plate 3 can be moved along the conveying section 5. Guide rollers 10 can also be provided in order to guide the respective push plate 3 along, for example, a track forming the conveying section 5. For fixing the respective component 4, for example, a lifting platform 11 can be mounted on the respective push plate 3, which component can be designed as a body of a motor vehicle, in particular when the conveying device 1 is used in motor vehicle construction.

[0037] Advantageously, you can use Figure 2 A drive device 12 , which is shown by way of example in FIG. 1 , is designed, in particular, as an electric motor.

[0038] It is advantageous to detect the active current as the at least one corresponding parameter value, in particular when an electric motor is used as the drive 12 of the respective conveyor element 2. Additionally or alternatively, for example, the temperature can be detected as the at least one corresponding parameter value.

[0039] Figure 2 The dashed line shows the ground level 13, so that the components 4 are transported along the conveying section 5 on the ground or above the ground and the empty or unloaded push plate 3 is transported back below the ground or underground. The conveying section 5 forms a closed loop.

[0040] It can be advantageous for the method if, for the respective datagram, an average active current and / or a maximum active current and / or a variance of the active current of the respectively associated drive device 12 is detected.

[0041] It is also advantageous if, when or for the generation or compilation of the data set, the corresponding datagrams are preprocessed, for example in a preprocessor implemented on the electronic computing device 8. In this preprocessing, for example, time intervals can be determined which determine which datagrams are included in the data set. Additionally or alternatively, for example, one of the parameter values ​​can be selected in the case of a plurality of parameter values, which parameter value can be observed by an anomaly model. This can be done in particular in addition or in combination with a target setting, which is intended to advantageously detect anomalies during the operation of the conveyor system 1.

[0042] The anomaly model used in the method here uses in particular at least one machine learning method and is therefore in particular constructed as an artificial intelligence or a model using an artificial intelligence method. Thus, for example, a self-learning algorithm and / or at least one neural network can be used to evaluate the data set. It is advantageous here to use the data set and / or another data set as a training data set for the anomaly model, for example by manually marking the data set or the anomalies contained therein. Thus, the anomaly model can be particularly advantageously coordinated with the characteristics of the conveyor device 1 and the conveyor device can therefore be operated particularly advantageously. It can be advantageous here, for example, if the threshold value is adjusted according to the anomaly value.

[0043] In other words, the method provides a monitoring system which, by in particular continuous comparison of data and parameters from a corresponding controller or a corresponding control device 7, detects anomalies of the corresponding conveyor system or conveyor element 2 and can therefore initiate preventive measures. Figure 2 As shown in Figure 1, an abnormality early warning system can be provided. Thus, for example, production downtime can be avoided, because unforeseen failures caused by technical disturbances can be avoided or at least reduced.

[0044] Therefore, the corresponding control device 7 of the corresponding conveying element 2 (which is advantageously arranged under the ground) is especially configured as a programmable logic controller for each conveying process, that is, during the conveying of the corresponding identifiable push plate 3 by the corresponding conveying element 2. This datagram of the programmable logic controller of the fixed conveying element 2 contains the information already mentioned and summarized again below: in particular, the time when the corresponding conveying process ends, the unique ID of the fixed or fixed conveying element, the unique ID of the conveyed push plate 3. In particular, parameters such as active current are aggregated during the conveying process. In addition, the average active current, the maximum active current or the variance of the active current caused by the converter of the drive device 12 can be additionally detected, especially during each conveying process. Thus, by means of this information contained in the datagram, for example, the current consumption of a specific fixed conveying element 2 can be assigned to the corresponding conveyed push plate 3. From the datagram, a data set can be formed, which can be preprocessed by some process steps, for example, by means of an abnormal push plate 3 and / or by means of a comparison of parameters of the fixed conveying elements 2 with each other and / or by means of parameter values, in particular when a corresponding plurality of parameter values ​​are measured for each conveying element 2, a plurality of parameter values ​​for each conveying element 2 are averaged or, for example, a plurality of parameter values ​​within a time interval (in which the parameter values ​​are aggregated) are averaged. Thus, after these preprocessing process steps, a data set can be formed, which in particular represents the total data set. This data set constitutes an input for an anomaly model, in particular constructed as a machine learning model, for detecting anomalies. The processed data set can be shown in Table 1:

[0045] A machine learning model or anomaly model, in particular a trained one, is applied to the data set, whereby anomalies or anomaly scores can be calculated. In this case, the corresponding data set can be re-detected after training and processed or aggregated in a manner similar to the training data set, whereby labels can be eliminated. The calculation of the threshold value depends on the selection of the anomaly score or anomaly value. In particular, an anomaly score can be calculated according to the model type, which in practice is usually scaled in the interval of 0 to 1 to achieve comparability. Other models perform a binary classification of the input data and generate a label abnormal or no abnormality and, for example, have 0 and 1. A higher anomaly score indicates that one of the conveying elements deviates from the normal state to a greater extent and therefore corresponds to a higher probability of failure of the conveying device 1 and therefore a higher probability of downtime. This can therefore be advantageously avoided by the method.

[0046] Thus, further advantages are achieved in the targeted prevention of system failures, for example, more convincing data are generated, which characterize the operation of the conveying device 1 or a similar model, simply by generating corresponding datagrams through data aggregation in the form of detection. In addition, it can also be transferred to different types of equipment or conveying equipment. The cost can be particularly low, because the hardware costs are low or non-existent by cleverly utilizing the already installed programmable logic controller and electronic computing devices. In addition, it can be used independently of the manufacturer, because parameter data or parameter values ​​of different components and therefore different manufacturers can be shared. Therefore, it can also be used advantageously for commissioning new conveying equipment, because the errors that occur can be reduced more quickly.

[0047] The proposed method and the proposed conveying device 1 can provide predictive maintenance in a particularly advantageous manner.

[0048] Reference numerals list

[0049] 1Conveying equipment

[0050] 2Conveying elements

[0051] 3 Push Plate

[0052] 4 Components

[0053] 5Transportation sections

[0054] 6 Sensor device

[0055] 7. Control Device

[0056] 8 Electronic computing devices

[0057] 9 driving wheels

[0058] 10 Guide rollers

[0059] 11 Lifting platform

[0060] 12 Drive device

[0061] 13 Ground Level

[0062] S1 First step

[0063] S2 Second step

[0064] S3 Step 3

[0065] S4 Step 4

[0066] S5 Step 5

[0067] S6 Step 6

Claims

1. A method for operating a conveying device (1) in which push plates (3) for receiving a component (4) can be conveyed along a conveying path (5) by means of stationary conveying elements (2), the method comprising the following steps: - during the conveying process of each push plate (3), detecting a corresponding conveying time (S1) for the corresponding conveying device; - detecting at least one corresponding parameter value for the corresponding conveying element (2) at the corresponding conveying time by means of a sensor device (6), said parameter value characterizing the state of the corresponding conveying element (2) and / or the state of a push plate (3) conveyed by the corresponding conveying element (2) during the conveying process (S2); - generating a corresponding datagram for the corresponding conveyor element (2) at the corresponding conveying time by the corresponding control device (7) of the corresponding conveyor element (2), the datagram comprising the at least one corresponding parameter value, the conveying time, the push plate ID of the respectively conveyed push plate (3) and / or the conveyor element ID of the corresponding conveyor element (2) (S3); - aggregating the datagrams into a data set (S4) via a central electronic computing device (8); - evaluating the data set by an anomaly model, the anomaly model being configured to evaluate anomalies of the at least one corresponding parameter value of the corresponding datagram (S5); If the evaluation results in an abnormality of at least one of the parameter values ​​exceeding a threshold value, a maintenance instruction is output (S6) for the corresponding push plate (3) and / or the corresponding conveyor element (2) having the abnormality.

2. The method according to claim 1, characterized in that The data set and / or another data set is used as a training data set for an anomaly model using a machine learning method.

3. The method according to claim 1 or 2, characterized in that: Active current and / or temperature are detected as the at least one parameter value.

4. The method according to any one of the preceding claims, characterized in that The respective conveyor element (2) is conveyed using a drive device (12), which is in particular designed as an electric motor.

5. The method according to claim 4, characterized in that For the respective datagram, an average active current and / or a maximum active current and / or a variance of the active current of the respectively associated drive device (12) is detected.

6. The method according to any one of the preceding claims, characterized in that Prior to and / or in order to aggregate the data sets, the corresponding datagrams are pre-processed.

7. The method according to claim 6, characterized in that In the preprocessing, a time interval is determined and / or, in the case of a plurality of parameter values, one of the parameter values ​​is selected and / or a target is set.

8. The method according to any one of the preceding claims, characterized in that The threshold is adjusted according to the outlier.

9. A conveying device (1) comprising a stationary conveying element (2) and a push plate (3) which is used to receive a component (4) in each case and can be conveyed along a conveying path (5) via the conveying element (2), the conveying device (1) being operable by means of a method according to any of the preceding claims.

Images