Device for weighing conveyed products in reverse operation

Abstract

Weighing device for automatic, especially dynamic, weighing of by means of a weighing device with products conveyed, in particular at high speed, in one conveying direction (x, x'), wherein the weighing device comprises one or more physical belt scales (S1, S2, S3, S1', S2', S3') arranged one behind the other in the conveying direction (x, x'), wherein the weighing device is designed such that the conveying direction of the weighing device can be changed between a forward direction (x) and a reverse direction (x') in order to weigh in both directions and wherein for the one physical belt scale (S1 or S2 or S3) or for at least one physical belt scale (S1, S2, S3) a respective first correction factor (K vi ) for weighing in the forward direction (x) is known or has been determined in order to reduce influences caused by a static-dynamic offset, characterized by the fact that the weighing device includes a control and / or evaluation device designed in such a way that for weighing in reverse direction (x') for the one physical belt scale (S1') or the at least one physical belt scale (S1', S2', S3') a respective second correction factor (K ri ) is used, the a) the respective first correction factor (K vi ) corresponds to or b) from the first correction factor (K vi ) was derived or c) from the first correction factor (K vi ) underlying measurement data in the forward direction (x) was derived or d) was determined using a model for weighing in the reverse direction (x'), to avoid a costly relearning process in reverse by using at least one of the alternatives a) to d). or e) was derived from measurement data obtained in the reverse direction (x').

Description

[0001] The invention relates to a method for the dynamic, automatic weighing of products conveyed at high speed in a conveying direction.

[0002] Such methods are used, for example, in logistics, particularly in heavy-duty transport, where products need to be conveyed at high speed and weighed while in motion. The weighing process is automated, i.e., without operator intervention, and highly accurate.

[0003] In contrast to static weighing, this saves a lot of time and money.

[0004] The present invention is therefore based on the objective of creating a method and a device for carrying out such a method, which also enables dynamic automatic weighing of products conveyed at high speed with high accuracy in the reverse direction.

[0005] This problem is solved according to the invention with a device having the features of claims 1 and 9.

[0006] According to the invention, products conveyed at high speed are dynamically and automatically weighed (i.e., without operator intervention) using at least one physical belt scale, in particular a single physical belt scale, preferably several physical belt scales arranged one behind the other in the conveying direction. The high speed can be, for example, greater than or equal to 1 m / s, in particular greater than or equal to 1.5 m / s or even greater than or equal to 3 m / s. The conveying direction can be changed between a forward direction (forward operation) and a reverse direction (reverse operation) to allow weighing in both directions.

[0007] In a product flow with products of different lengths and distances (mixed flow), the throughput can be increased or optimized by using several belt scales arranged in series, preferably with different lengths (by appropriately assigning one of the belt scales ideally suited for determining the weight of a product).

[0008] The products to be promoted are preferably heavy-duty products, such as logistics products like boxes, cartons, cartons, etc., which typically weigh several kilograms, for example more than 3 kg or 10 kg, preferably more than or equal to 15 kg or 30 kg and up to 120 kg.

[0009] In order to reduce the influence on measurement results caused by a static-dynamic offset, i.e., offset between dynamic measurement result and static reweighing, a respective first correction factor for weighing in the forward direction is known or has been determined for the physical belt scale, preferably at least one, in particular for each physical belt scale.

[0010] The cause of static-dynamic misalignment includes not only geometric properties of the conveyor belt (unevenness on the belt, especially unevenness / curvature on the upper surface of the belt, i.e., the platform for guiding / supporting / resting the underside of the upper run of a conveying device moving above it, such as a circulating belt, belt, chain, etc.) but also aerodynamic effects (lift effects) on the product being weighed, which is moving at high speed. Therefore, when correcting for static-dynamic misalignment, not only the geometry, especially the unevenness of the belt or platform, the position of the measuring window, etc., but also the speed, the mass of the product, as well as its geometry and the resulting aerodynamic effects must be taken into account.

[0011] In order to take into account all or at least some of the aforementioned influences for a static-dynamic offset, a model can be developed (derived from the influences) using, for example, neural networks or artificial intelligence (AI), and a correction factor for a (physical but also logical) balance can be determined using this model.

[0012] To reduce interference with measurement results, even when weighing in reverse, a second correction factor is used according to the invention. In the simplest case, the first correction factor is used for this purpose, so that the second correction factor has the same value as the first. A second correction factor can be used to correct the weight of a product, at the latest when determining its weight. The value of the first correction factor can then be used as the second correction factor. Alternatively, a separate second correction factor, set equal to the first, can be used.When using a storage device (local, on the local network or in the cloud), the correction factor can be stored at a single storage address, or two corresponding correction factors can be stored at two different storage addresses.

[0013] Since the aforementioned influences included in a correction factor, especially the unevenness of the belt body, are usually not symmetrical about the belt center in the conveying direction, the second correction factor can also differ from the first. This is due to a non-central upward curvature of the belt body surface that occurs during operation, for example, a parabolic shape, particularly when viewed in the longitudinal / conveying direction, which distorts the measurement result and makes the product appear lighter.

[0014] The second correction factor can be derived from the first correction factor, for example by taking into account the height and position of the curvature and their different influences on the measurement results depending on the conveying direction.

[0015] Alternatively, the second correction factor can be derived from the measurement data underlying the first correction factor in the forward direction, or it can be determined without derivation from the first correction factor using a separate model (for example, using mathematical modeling methods, neural networks, or AI) for weighing in the reverse direction, or it can be derived from measurement data obtained in the reverse direction. The separate model can preferably be derived from a model already created for determining the first correction factor.

[0016] By deriving a second correction factor, in whose determination the first correction factor is incorporated in the manner described above, the effort required to determine the second correction factor (for weighing in reverse) can advantageously be reduced or even completely avoided. In the latter case, no acceleration section is required in the reverse direction, which significantly shortens the required overall length of the weighing device and reduces the number of acceleration belts. The time-consuming calibration process (of a belt scale) for determining a second correction factor in reverse operation (using numerous trials with test specimens, as is conventionally required for determining a first correction factor for physical belt scales) is eliminated or can be avoided.

[0017] The above also expressly applies to a single belt scale, i.e., a single physical scale.

[0018] In an advantageous embodiment of the invention, at least two physical belt scales arranged successively in the conveying direction are connected to form a logic scale, wherein the weighing results or weighing signals of the individual belt scales are summed to obtain a total weight or total weight signal, preferably synchronously in time. This advantageously allows products that are longer than the length of a single physical belt scale to be weighed by having them rest on the logic scale for a sufficiently long time window for dynamic measurement.

[0019] In a preferred embodiment of the invention, in order to determine a first correction factor for weighing in the forward direction for the logic balance, the logic balance is trained using appropriate test specimens, or the correction factor is derived via a model (for example, using mathematical modeling methods, neural networks, or AI, taking into account existing influences (parameters for a static-dynamic offset) such as belt geometry, in particular curvature, conveying speed, etc.). For derivation via a model, it is also conceivable to derive the correction factor of the logic balance from the correction factors of the connected physical belt scales, as explained above for physical belt scales.

[0020] In a further embodiment of the invention, a second correction factor is used for weighing in the reverse direction for each logical balance, which – analogous to the second correction factor for weighing in the reverse direction for each physical belt balance explained above – corresponds to the respective first correction factor or was derived from the first correction factor or was derived from the measurement data underlying the first correction factor in the forward direction or was determined by means of a model for weighing in the reverse direction or was derived from measurement data determined in the reverse direction.

[0021] In a particularly preferred embodiment of the invention, the weighing device comprises at least three physical belt scales – preferably of different lengths – which can be connected to form several different logic scales. With three physical belt scales, this results in three additional logic scales in forward operation (first and second; second and third; first, second, and third scales) and also three logic scales in reverse operation (third and second; second and first; third, second, and first scales).

[0022] In this process, the physical belt scales and the logic scales for reverse weighing can be regrouped according to the opposite direction to facilitate their assignment. This does not involve any physical changes to the scales, but rather renaming them according to an internal numbering system.

[0023] Furthermore, the determined correction factors can be automatically assigned depending on the respective physical or logical scale and the direction of travel. This automatic assignment of correction factors occurs according to the specified or detected conveying direction, for example, directly from a storage device to a load cell of a scale, or during signal processing and weight determination in a control and / or evaluation device.

[0024] In the embodiment according to the invention, a weighing device has, in addition to several physical belt scales, a switching device for changing the direction of conveying, as well as a storage device (internal or external, for example cloud) with a first set of first correction factors for weighing in the forward direction and a second set of second correction factors for weighing in the reverse direction for each physical belt scale or for each physical belt scale and for each logical scale.

[0025] Furthermore, the weighing device according to the invention is designed such that, when determining weight depending on the conveying direction, the respective set is retrieved from the storage device and used.

[0026] In a further embodiment of the invention, the belt scales have deflection shafts on the inlet and outlet sides, both of which have a crown, in particular a double crown (two radial ridges spaced apart from each other in the axial direction), in order to prevent the belt from running laterally in both conveying directions.

[0027] In an advantageous embodiment of the invention, or as an independent invention, the weighing device is designed such that it can be operated and used for weighing in both the forward and reverse directions. A first reference point is provided to define the start of a first process section in the forward direction. Additionally, a second reference point is provided to define the start of a second process section in the reverse direction. The distance between these two reference points is determined by their definition. Preferably, the position of the second reference point, or the distance between the two reference points, is defined during the parameterization of the forward operation. This enables automatic calculation (without manual input) of the parameters for the reverse operation.

[0028] Alternatively, one of the two reference references and the distance to the other reference reference can be specified to define the other reference reference.

[0029] A process line within the meaning of the invention represents a digital model of the weighing device. This model replicates the individual physical components of the weighing device (such as weighing belts, infeed and outfeed belts, sorting devices, light barriers, barcode readers, cameras, and labeling systems), as well as the products conveyed on the weighing device, their movements, and product positions for data input and output, correctly taking into account the sequence of the process flow. The process line serves, among other things, for temporal and / or spatial product tracking.

[0030] This allows a control system to know at what time the individual components are in certain functional states (e.g. "belt occupied", "belt free"), or at what location (position) the products are currently located.

[0031] According to the invention, the weighing device for forward and reverse directions can have a common process section or different process sections, in particular of different lengths.

[0032] A reference point within the meaning of the invention is understood to be a local position. This serves as a reference point (zero point) for determining the local positions of the aforementioned physical components or parts thereof (e.g., the front and back edges of the weighing belts), as well as the conveyed products.

[0033] The leading edge or flank of the infeed conveyor in forward operation can be used as the first reference point. Alternatively, the light barrier in forward operation can also be used as the first reference point. Similarly, the leading edge or flank of the infeed conveyor in reverse operation, or the light barrier in reverse operation, can be used as the first reference point.

[0034] In a further embodiment, the weighing device is designed such that, for weighing in the reverse direction, parameters (position and / or time values) of the second process section are automatically calculated from parameters of the first process section. This is done in particular with the aid of the two reference points, preferably by the control system of the weighing device (for example, a control and evaluation device).

[0035] The calculation can be performed before the conveying direction is changed, or triggered during or by the change itself (e.g., by an external, particularly binary, switching or direction signal). Preferably, the change is not performed during the weighing process, but rather the conveying movement is stopped for this purpose.

[0036] In a preferred embodiment, the weighing device has a first light barrier at the entrance of the first belt scale and a second light barrier at the exit of the last belt scale, with only one of the two light barriers being used for a weighing process depending on the conveying direction.

[0037] The inventive method and the device for carrying out such a method advantageously reduce or even eliminate measurement errors or deviations due to static-dynamic offset. Furthermore, the effort required to determine the second correction factors (for weighing in the reverse direction) can be reduced or completely avoided. In the latter case, no acceleration section in the reverse direction is required, which significantly shortens the necessary overall length of the weighing device and reduces the number of acceleration belts.

[0038] Further advantageous embodiments of the invention will be found in the dependent claims.

[0039] The invention is explained in more detail below with reference to an embodiment shown in the drawing.

[0040] The drawing shows: Fig. 1 a schematic top view of a weighing device according to the invention with three physical belt scales in forward operation; Fig. 2 a schematic top view of the weighing device according to Fig. 1 in reverse operation and Fig. 3 Schematic representation in side view of a non-linear product movement on a single physical belt scale.

[0041] The Fig. Figure 1 shows a weighing device 1 in forward operation, i.e., with a conveying direction in the forward direction x. Viewed in the conveying direction, the weighing device 1 has a first belt scale S1, a second belt scale S2, and a third belt scale S3, each of which can have a different length (viewed in the conveying direction).

[0042] On the inlet and outlet sides, a feed area 9 and an outlet area 11 are provided, which are designed, for example, as belt conveyors.

[0043] If a product to be weighed (not shown in the drawing) is conveyed from the feed area 9 via the belt scales S1, S2 and S3 (S3 being the longest scale in this example) to the discharge area 11, the product is (temporarily) located on the belt scales S1, S2 and S3. Provided the product length is within the usable weighing length of any one of the scales S1, S2 and S3, the product is weighed on a single scale within a short time window (in the millisecond range) without resting on the adjacent scales S1, S2, S3.

[0044] To enable the weighing of products with a greater length (in the direction of travel) than the longest physical belt scale, the individual physical belt scales S1, S2, and S3 can be combined to form logical scales S4, S5, and S6. In this configuration, logical scale S4 comprises belt scales S1 and S2, logical scale S5 comprises belt scales S2 and S3, and logical scale S6 comprises belt scales S1, S2, and S3. Provided the length of the product to be weighed does not exceed the usable length of the entire combination of scales—in the illustrated embodiment with only three physical scales, this is the logical scale S6—the weight of this product can also be determined within a suitable timeframe.

[0045] In order to determine and track positions, in particular the leading edge of products (or their length), at a known speed in the conveying direction x, x' (process path with object tracking), a light barrier L (weighing in forward direction), L' (weighing in reverse direction) can be provided on the input side.

[0046] In order to have sufficient time for the allocation of one of the belt scales ideally suited for determining the weight of a product, the light barrier L, L' can be moved forward into the feed area 9, 9'.

[0047] As in Fig. As shown in Figure 1, the leading edge of the infeed conveyor 9 and the position of the light barrier L can coincide in forward operation and be used as the first reference point.

[0048] As in Fig. As shown in Figure 2, the leading edge of the infeed conveyor 9' and the position of the light barrier L' can coincide in reverse operation and be used as a second reference point.

[0049] Using a control and / or evaluation device (not shown in the drawing), the start and end of data processing of measurement signals (measuring window) can be set, given the known conveying speed, length of the weighing device, and lengths of the individual scales. The measuring window begins no earlier than when the product reaches the entry point of the first scale.

[0050] The final determined and output weight value should be available as early as possible (at the latest when the output position A, A' is reached by the front edge of the product) in order to be transmitted, for example, to subsequent data processing (e.g., label printer / labeler) (in order to enable a short design of the weighing device).

[0051] An output position A, A' in the run-out area can be used to determine the end of data processing of measurement signals from a measurement, whereby reaching the run-out area can also be determined via the time required for the path w, w'.

[0052] Fig. Figure 2 shows the aforementioned weighing device 1 in reverse operation, i.e., with a conveying direction in the reverse direction x'. In contrast to forward operation, when viewed in the conveying direction x', the weighing device 1 now has the following belt scales: belt scale S1' (the longest scale in the example) as the first belt scale, belt scale S2' as the second belt scale, and belt scale S3' as the third belt scale.

[0053] On the inlet and outlet sides, a feed area of ​​9' and an outlet area of ​​11' are now provided.

[0054] In the Fig. In the reverse operation shown in Figure 2, the individual physical belt scales S1', S2' and S3' are now connected to form logical scales S4', S5' and S6'. Here, the logical scale S4' comprises the belt scales S1' and S2', the logical scale S5' comprises the belt scales S2' and S3', and the logical scale S6' comprises the belt scales S1', S2' and S3'.

[0055] As can be seen, apart from the conveying direction x, x', the physical scales S1 and S3', the scales S2 and S2', and the scales S3 and S1' correspond. Likewise, the logical scales S4 and S5', the logical scales S5 and S4', and the logical scales S6 and S6' correspond.

[0056] Therefore, if the conveying direction of the weighing device 1 is changed from forward operation to reverse operation by means of a switching device not shown in the drawing, the physical scales and the logical scales must be regrouped or renamed accordingly for reassignment.

[0057] As from Fig. As can be seen in Figure 3, the belt guidance in the upper run does not necessarily correspond to the ideal of a linear guide, but runs along a curve (curvature), for example a parabola, with a maximum height δ relative to the horizontal.

[0058] The difference Δm in the weight or the determined mass of a product is represented by a function consisting of the mass m of the product, the conveying speed and the curve, in particular the height δ. Δm=f(m,v,δ)

[0059] A high velocity v is of considerable importance in dynamic weighing. Accordingly, the static-dynamic offset is influenced not only by buoyancy effects but also by the flatness of the belt body and the resulting motion curve of a product, especially a parabola (see Fig. 3) significantly influenced.

[0060] In order to obtain a correct measurement result despite the static-dynamic offset, a correction factor K is used according to the invention. vi for each physical scale S1, S2 and S3 - i.e. correction factor K v1 for scale S1, correction factor K v2 for scale S2 and correction factor K v3for scale S3 - used, which is or was determined, for example, by means of training with a larger number of runs with test weights or test packs.

[0061] For example, the same test pack is weighed 30 times, then a different pack with a different load is weighed 30 times at different speeds. The number of runs thus corresponds to, for example, 120 individual measurements (4 different loads, each weighed 30 times), which is then multiplied by the number of speeds.

[0062] Alternatively, such a correction factor K can be vi can also be derived from a model, for example using neural networks or AI, which takes into account the various aforementioned influences of the weighing device and, if applicable, the dimensions of the product (buoyancy).

[0063] A correction factor K can also be determined for the logic scales S4, S5 and S6 in the manner described above. vlogi to be determined, whereby it is also conceivable to use the correction factors of the physical scales S1 to S3 - i.e. correction factor K. vl1 for scale S1, correction factor K v2 for scale S2 and correction factor K v3 for scale S3 - to be used or at least considered.

[0064] Since the geometry of the belt body in a belt scale, especially the curvature on the surface, etc., is not necessarily symmetrical to the center of the belt body in the conveying direction in practice, the aforementioned curve is not identical in the forward and reverse directions.

[0065] For this reason, especially at high speeds, even first correction factors K vi and K vlogi , which were determined for forward operation, cannot be readily used for reverse operation.

[0066] A second correction factor K ri can be derived from the first correction factor K vi can be derived, for example by including the belt body, in particular its curvature and its position and height and their different influences on the measurement results depending on the conveying direction.

[0067] Alternatively, the second correction factor K can be ri from the first correction factor K vi The underlying measurement data can be derived in the forward direction, or without derivation from the first correction factor using a separate model (for example, using mathematical modeling methods, neural networks or AI) for weighing in the reverse direction, or derived from measurement data obtained in the reverse direction.

[0068] By using a second correction factor K riAdvantageously, the time-consuming learning process of a physical belt scale in reverse operation by means of a large number of passes with test specimens can be avoided, and an acceleration section in the reverse direction can be dispensed with.

[0069] In the course of this entire description, the term correction factor is understood to mean a quantity to which a fixed absolute value is assigned, taking into account the aforementioned influences (influencing a static-dynamic offset), in particular the belt geometry, the conveying speed, etc., in order to obtain a correspondingly corrected weight value of a product as a measurement result.

[0070] Of course, according to the invention, it is possible for several scales to have the same or a common correction factor, so that several respective correction factors correspond to each other in value. However, unless the belt scales are completely identical (with identical design), the respective correction factors of different scales preferably differ from one another.

[0071] For the purposes of the invention, dynamic weighing is understood to mean weighing during movement, preferably during uniform movement. In a broader sense, however, dynamic weighing is also understood to include weighing during accelerated or decelerated movement, or even during a momentarily stopped movement, insofar as static-dynamic effects (for example, product sloshing effects) occur and can affect the weighing accuracy.

[0072] Accordingly, the first and second correction factors K can also be used. vlogi and K rlogi to be determined, derived or taught for a logical scale.

[0073] Of course, the invention is not limited to the embodiment in question, namely a group of three scales, so that a different number (2, 4, 5 or more) of physical scales can also be used.

[0074] In the case of a single physical belt scale S1 (or S2 or S3), the aforementioned statements regarding the first and second correction factors apply accordingly with regard to their determination, derivation or derivation. Reference symbol list 1.1' Weighing device S1 first belt scale seen in the forward direction S2 second belt scale viewed in the forward direction S3 third belt scale viewed in the forward direction 9 Feed area in forward direction 11 Run-out area in forward direction S4 logic scale consisting of belt scales S1 and S2 S5 logic scale consisting of belt scales S2 and S3 S6 logic balance consisting of belt scales S1, S2 and S3 S1' first belt scale seen in reverse S2' second belt scale seen in reverse S3' third belt scale seen in reverse S4' logical balance made up of belt balances S1' and S2' S5' logical balance made up of belt balances S2' and S3' S6' logic balance consisting of belt balances S1', S2' and S3' 9' Feed area in reverse direction 11' Run-out area in reverse direction x Conveying direction forward (forward operation or forward weighing operation) x' Conveying direction backwards (reverse operation or reverse weighing operation) K vi first correction factor of a physical belt scale i in forward operation K risecond correction factor of a physical belt scale i in reverse operation K vlogi first correction factor of a logical tape scale i in forward operation K rlogi second correction factor of a logical belt scale i in reverse operation i 1, 2, 3 m mass of the product v Speed ​​of conveying δ maximum height of the curvature l Length of the band L Light barrier at the entrance, viewed in the forward direction L' Light barrier at the entrance seen in reverse A, A' Output position w Length of the path between light barrier L and output position A w' Length of the path between light barrier L' and output position A'

Claims

[1] Weighing device for automatic, in particular dynamic, weighing of by means of a weighing device with products conveyed, in particular at high speed, in one conveying direction (x, x'), wherein the weighing device comprises one or more physical belt scales (S1, S2, S3, S1', S2', S3') arranged one behind the other in the conveying direction (x, x'), wherein the weighing device is designed such that the conveying direction of the weighing device can be changed between a forward direction (x) and a reverse direction (x') in order to weigh in both directions and wherein for the one physical belt scale (S1 or S2 or S3) or for at least one physical belt scale (S1, S2, S3) a respective first correction factor (K vi ) for weighing in the forward direction (x) is known or has been determined in order to reduce influences caused by a static-dynamic offset, characterized by , that the weighing device includes a control and / or evaluation device designed in such a way that for weighing in reverse direction (x') for the one physical belt scale (S1') or the at least one physical belt scale (S1', S2', S3') a respective second correction factor (K ri ) is used, the a) the respective first correction factor (K vi ) corresponds to or b) from the first correction factor (K vi ) was derived or c) from the first correction factor (K vi ) underlying measurement data in the forward direction (x) was derived or d) was determined using a model for weighing in the reverse direction (x'), to avoid a costly relearning process in reverse by using at least one of the alternatives a) to d). or e) was derived from measurement data obtained in the reverse direction (x'). [2] Device according to claim 1, characterized by , that the weighing device comprises several physical belt scales (S1, S2, S3, S1', S2', S3') arranged one behind the other in the conveying direction (x, x'). [3] Device according to claim 2, characterized by , that the control and / or evaluation device is designed in such a way that at least two physical belt scales (S1-S2, S2-S3, S1-S2-S3, S1'-S2', S2'-S3', S1'-S2'-S3') following one another in the conveying direction can be connected to form a logical scale (S4, S5, S6, S4', S5', S6'). [4] Device according to claim 3, characterized by that the control and / or evaluation device is designed such that a first correction factor (K) is available for the logic balance (S4, S5, S6). vlogi ) to determine for weighing in the forward direction (x), the logical balance (S4, S5, S6) is trained for this purpose or the correction factor (K vlogi ) is derived via a model. [5] Device according to claim 3 or 4, characterized by , that the control and / or evaluation device is designed such that for weighing in the reverse direction (x') a respective second correction factor (K) is applied for each logical balance (S4', S5', S6'). rlogi ) is used, the a) the respective first correction factor (K vlogi ) corresponds to or b) from the first correction factor (K vlogi ) was derived or c) from the first correction factor (K vlogi ) underlying measurement data in the forward direction (x) was derived or d) was determined using a model for weighing in the reverse direction (x') or e) was derived from measurement data obtained in the reverse direction (x'). [6] Device according to any one of claims 3 to 5, characterized bythat the weighing device comprises at least three physical belt scales (S1, S2, S3, S1', S2', S3') which can be connected to form several different logical scales (S4, S5, S6, S4', S5', S6'). [7] Device according to any one of the preceding claims, characterized by , that the control and / or evaluation device is designed in such a way that the physical belt scales (S1', S2', S3') and / or the logical scales (S4', S5', S6') can be regrouped for weighing in the reverse direction (x') according to the opposite direction. [8] Device according to any one of the preceding claims, characterized by that the control and / or evaluation device is designed in such a way that the determined correction factors (K vi , K ri , K vlogi , K rlogi ) are automatically assigned depending on the respective physical or logical balance and the respective direction (x, x'). [9] Weighing device with a physical belt scale (S1 or S2 or S3, S1' or S2' or S3') with a switching device for changing the direction of conveying (x, x'), characterized by , that the weighing device has a storage device with a first set of first correction factors (K vi , K vlogi ) for weighing in the forward direction (x) and a second set of two correction factors (K ri , K rlogi ) for weighing in reverse (x') a) for the physical belt scale (S1 or S2 or S3, S1' or S2' or S3') and that the weighing device is designed such that, when determining weight depending on the conveying direction (x, x'), the respective set (K vi , K vlogi ; K ri , K rlogi ) is retrieved from the storage device and used. [10] Weighing device according to claim 9, characterized by, that the weighing device has several physical belt scales (S1, S2, S3, S1', S2', S3') arranged one behind the other in the conveying direction (x, x'), with a first set of first correction factors (K vi , K vlogi ) for weighing in the forward direction (x) and a second set of two correction factors (K ri , K rlogi ) for weighing in reverse (x') a) for at least one or each physical belt scale (S1, S2, S3, S1', S2', S3') or b) for at least one or each physical belt balance (S1, S2, S3, S1', S2', S3') and for at least one or each logic balance (S4, S5, S6, S4', S5', S6'). [11] Device according to claim 9 or 10, characterized by , that the belt scales (S1, S2, S3, S1', S2', S3') have deflection shafts on the inlet and outlet sides, both of which have a crown to prevent a belt or conveyor from running laterally in both conveying directions (x, x'). [12] Device according to any one of claims 8 to 10, characterized by , that the weighing device, viewed in the conveying direction (x, x'), has a first light barrier (L, L') at the entrance of the belt scale (S1) or the first belt scale (S1, S1') and a second light barrier (L', L') at the exit of the belt scale (S1) or the last belt scale (S3, S3'), wherein, depending on the conveying direction (x, x'), only one of the two light barriers is used for a weighing process.

Images