Method of operating a carding machine, carding machine and spinning preparation apparatus

Abstract

This invention relates to a method of operating a carding machine in spinning preparation equipment, wherein fiber bundles are loosened, oriented, and cleaned between a cylinder with a carding cloth and a carding element, resulting in a fiber layer transferred from the cylinder to the doffer and subsequently becoming a fiber sliver. The carding machine has a carding machine control device with an operating unit that determines a reference cylinder speed after the operator inputs the raw materials and output. The feature is that the operator initiates an optimization program that allows selection between clumping optimization and / or energy optimization operating modes, whereby the carding machine control device executes an automatic measurement sequence in which multiple sensors detect the number of clumps in the fiber layer at various predetermined durations at different cylinder speeds, simultaneously determining the drive power of the carding machine. The sensor data and the determined drive power are transmitted to a higher-level control device in the spinning preparation workshop, which, with the aid of at least said data, suggests operating modes for different quality categories to the operator using a mathematical algorithm.

Description

Technical Field

[0001] This invention relates to a method for operating a carding machine, the carding machine, and spinning preparation equipment. Background Technology

[0002] In yarn manufacturing, the number of knots in the fiber sliver in the spinning preparation workshop is a decisive quality standard. A knot is understood as an accumulation of knotted fibers that occurs during cotton harvesting, ginning, and mechanical processing. Here, there is a distinction between husk knots and fiber knots. Husk knots are formed when a seed coat is attached to a fiber. Fiber knots are formed by the accumulation of short fibers that shrink into knots or clusters on the drawing frame or roving frame after the carding process. During subsequent spinning, knots cause uneven color distribution in the yarn, which is undesirable. Furthermore, fiber knots also contribute to yarn quality degradation and must be removed from these disruptive locations. In the following text, the concept of knots encompasses husk knots, short fiber accumulations, and fiber knots.

[0003] According to existing technology, modifying the carding machine settings in this way minimizes the number of clumps. However, these settings also vary depending on the raw materials, requiring operators to be quite experienced to achieve optimal results.

[0004] EP 0409772A1 discloses a method for determining the number of clumps at the exit of a carding machine, wherein the fiber pile is measured simultaneously. If the measured value deviates from a preset rated value, an attempt is first made to improve the corresponding value by resetting the cleaning machine. If this is unsuccessful, then the blending ratio needs to be changed, which requires operation of the unpacking machine and ultimately affects the storage of the cotton bales. In this method, individual parameters of the number of clumps and the fiber pile are measured, and new carding machine settings are assigned separately for each parameter of the number of clumps or the fiber pile. By using a specific carding machine setting, the number of clumps can be reduced on the one hand, or the fiber pile can be changed by using another carding machine setting on the other. It is possible that by changing the adjustment parameters, such as carding machine settings (carding speed, carding gap), the number of clumps may be significantly reduced, but at the same time, the fiber pile may be significantly negatively changed, and vice versa. Furthermore, it has been found in known methods that the improvement of the two separately measured parameters (i.e., the reduction of the number of clumps and the improvement of the fiber pile) cannot be achieved on the carding machine itself, but requires changing the blending ratio.

[0005] DE 19651893B4 specifies that the measurement of the number of clusters and the measurement of fiber length or fiber shortening are correlated, thereby achieving, unlike known methods, the minimum number of clusters while simultaneously minimizing fiber damage (fiber shortening). The measurements of cluster number and fiber length are combined and used for adjustment interventions. This achieves optimization in a particularly advantageous manner. The disadvantage is that a portion of the fibers must be manually removed from the carded fiber sliver, making the process cumbersome. Before this portion can be analyzed and the carding machine settings adjusted, a large amount of potentially poor-quality fiber material is generated. Summary of the Invention

[0006] The objective of this invention is to improve the method of operating a carding machine or the carding machine itself in such a way that the operator obtains the optimal settings for the carding machine based on the raw materials, and using these settings, it is possible to operate the machine with minimal clumping and / or energy savings. Here, clumping is a general term encompassing shell clumping, short fiber accumulation, and fiber knotting.

[0007] This invention relates to a method of operating a carding machine, wherein a fiber bundle between a rotating cylinder with a carding cloth and a fixed or surrounding carding element is loosened into individual fibers, oriented, and cleaned. The resulting fiber layer is transferred from the cylinder to the doffer and subsequently becomes a fiber strip.

[0008] This invention includes the following technical teachings: an operator inputs raw materials and output into a carding machine control unit equipped with an operating unit, and the carding machine control unit determines a reference cylinder speed. The operator can initiate an optimization program, which allows the carding machine to operate between agglomeration optimization and / or energy optimization modes. In this case, the carding machine control unit executes a measurement sequence in which multiple sensors detect the number of agglomerates in the fiber layer at various predetermined durations at different cylinder speeds and simultaneously determine the drive power of the carding machine. The sensor data and the determined drive power are transmitted to a higher-level control unit in the spinning preparation workshop. Using this data, the higher-level control unit uses mathematical algorithms to suggest operating modes for the carding machine under different quality categories to the operator.

[0009] Operators can choose to operate the carding machine with optimized clumping, optimized energy, or a combination of both, depending on the quality category. The advantage of this method is that the higher-level control unit can process large amounts of data from various sensors using mathematical algorithms. The sensor's computer processes image data in parallel for its respective track width and forwards this data to the higher-level control unit of the carding machine and then to the control unit of the spinning preparation equipment. Processing speed is increased when processing large amounts of data in parallel, and carding machine settings are made more quickly without handling excessively clumped fibers or consuming excessive energy. Provided it is technically feasible and the reference cylinder speed is already at its limit, automatic measurement sequences can be executed only at speeds higher or lower than the reference cylinder speed. For example, when operating temperatures are high due to the fiber material being processed or due to hot and humid conditions in the spinning process, it is reasonable to execute measurement sequences only at speeds lower than the reference cylinder speed, as increasing the cylinder speed would cause a further increase in temperature, necessitating adjustments to the carding gap.

[0010] Alternatively, different cylinder speeds can be higher or lower than the reference cylinder speed, thereby determining the number of agglomerates and energy consumption for different speed ranges near the reference cylinder speed.

[0011] In this configuration, different cylinder speeds are positioned at the same cylinder speed interval above and below the reference cylinder speed. This equidistant cylinder speed interval is particularly beneficial when only a limited amount of data is available, as it improves processing speed and accuracy when evaluating data.

[0012] Since the duration of measurement is the same regardless of the cylinder rotation speed, a simplified mathematical algorithm can be used.

[0013] If new raw materials are fed into the carding machine, or the carding cloth is changed, or the basic settings of the carding machine are different, the agglomeration and drive power are measured multiple times at each cylinder speed. The duration (t) of each measurement is added together to obtain the total duration (T).

[0014] The reduction of agglomerates or energy-optimized operation in carding machine strips can be further improved by configuring the carding machine's control unit to automatically set the carding gap at different carding machine temperatures. This means that even when the carding machine temperature increases but the rotation speed remains the same, the carding gap remains constant despite cylinder expansion. Because the carding gap remains the same under different carding machine temperatures, the minimum amount of agglomeration is always achieved regardless of the operating mode (energy optimization or agglomeration optimization), independent of the carding machine temperature.

[0015] Depending on the configuration of the carding machine control device and the spinning preparation equipment control device, the optimization program can be started on at least one carding machine control device or the spinning preparation equipment control device.

[0016] By transmitting the results of the optimization process to the control unit of at least one carding machine, the operator can manually confirm the selection of the quality category. This allows for the prevention of erroneous starts, especially when data is insufficient or data transmission errors occur between control units, thanks to the operator's experience.

[0017] Alternatively, after the operator selects and confirms the quality category, the results of the optimization program can be obtained through the control device of the spinning preparation equipment, thereby automatically starting all the carding machines of the spinning preparation equipment in the selected operating mode.

[0018] By enabling operators to directly select whether at least one comber is to operate in a clumping-optimized or energy-optimized manner when initiating an optimization program, the selected operating mode is given higher weight when classifying or clustering data.

[0019] The carding machine according to the invention has a feeding side for fiber bundles, wherein the carding mechanism causes the fiber bundles to be conveyed to a rotating cylinder by means of at least one licker roller, wherein the fiber bundles are loosened into individual fibers, oriented and cleaned between a fixed carding element and an encircling cover strip and the cylinder, and the resulting fiber layer can be transferred from the cylinder to the doffer, and a device for converting the fiber layer into fiber strips is arranged downstream of the doffer.

[0020] Here, the carding machine has at least three sensors configured to detect clumps or short fiber accumulations in the fiber layer. Furthermore, the carding machine has a control unit with an operating mechanism. This mechanism allows the operator to select whether to activate the carding machine's optimization program after inputting raw material and output values. The carding machine operates in a clumping-optimized and / or energy-optimized mode. If, from the operator's perspective, high raw material quality is anticipated and the produced carded sliver contains few clumps, the operator can directly select the energy-optimized mode, allowing the control unit in the spinning preparation workshop to evaluate data with a higher energy optimization weight. Conversely, if the raw material is poor or dusty, the operator can directly select the clumping-optimized mode, allowing the control unit in the spinning preparation workshop to evaluate data with a higher quality weight. If the operator has limited experience or cannot estimate the number of clumps, they can activate the optimization mode to obtain a neutral recommendation regarding energy-optimized or clumping-optimized operation. Alternatively, this selection can also be made within the control unit in the spinning preparation workshop.

[0021] With the activation of the optimization mode, at least one carding machine starts with a measurement sequence in which only the setpoints are checked if the carding machine settings and raw materials remain unchanged. If the raw materials change, the carding machine initiates multiple measurement sequences, in which at least three sensors detect the fiber layer, such as the fiber layer on the doffer. The sensors are configured to detect at least the clumps or short fiber accumulations in the fiber layer by means of prior calibration and transmit this data to the carding machine control unit. The carding machine control unit simultaneously monitors the power consumption of the drive in parallel and transmits both sets of data to the higher-level control unit in the spinning preparation room.

[0022] Here, the sensor can be positioned within the fiber web guiding element, which guides the fiber layers between the rollers. However, the sensor can also be positioned at any other location as long as it detects a cluster of flat fiber layers between the cylinder and the flared end of the fiber layer assembly.

[0023] The carding machine is preferably configured to automatically set the carding gap. Therefore, the number of agglomerates is further reduced based on the cylinder speed. When the cylinder speed is low and energy consumption is also low, the carding gap can be reduced by automatically setting the carding gap, thereby reducing the number of agglomerates.

[0024] By adjusting the energy consumption of the carding mechanism at different speeds, the carding machine can operate in an energy-optimized mode, a clumping-reducing mode, or a combination of both.

[0025] The spinning preparation apparatus according to the present invention includes a control device and multiple carding machines. The control device is configured to process data from one carding machine, preferably multiple carding machines, using a mathematical algorithm. In a first step, the data from at least one carding machine is clustered or classified to generate quality categories. In a second step, the data is further processed using a regression model, wherein the results of the mathematical algorithm are passed to each carding machine to set settings for clumping optimization and / or energy optimization. The advantage of the mathematical algorithm is that the data is grouped into quality categories, and these quality categories are then assigned to the optimal cylinder speed using a regression model. This method provides reliable results with a high probability, even with small amounts of data.

[0026] Since the first step of clustering is determined through the K-Means algorithm, fuzzy c-Means algorithm, or hierarchical cluster analysis, data can be unambiguously assigned to the selected quality categories. If the operating method has been selected beforehand based on prior knowledge of the raw material quality, this can influence the weighting of the data assignments.

[0027] Alternatively, the first step in grouping or classifying data can be determined through label propagation using a semi-supervised learning model, allowing the system to train itself.

[0028] In the second step, the categorized data can be further processed using a regression model, designed as a random forest regression model, a multinomial classifier, or an artificial neural network. Here, the allocation from cluster number to rotation speed is achieved with optimized energy consumption. Allocation based on these mathematical models provides good results with high probability, even with small datasets.

[0029] If data from multiple carding machines are processed in the control unit of the spinning preparation equipment and compared with each other in the first step, the control unit can output optical or acoustic alarm signals when the clustered or classified data deviates from a preset reference value. This allows for the identification and subsequent correction of erroneous settings for each carding machine. Attached Figure Description

[0030] Other measures to improve the invention are described in more detail below with the aid of the accompanying drawings, together with the description of preferred embodiments of the invention.

[0031] In the picture:

[0032] Figure 1 A schematic side view of a spinning preparation machine in the form of a carding machine is shown, wherein the device according to the invention is used;

[0033] Figure 2 An enlarged view of the installed fiber web guide element is shown;

[0034] Figure 2a A detailed diagram of the fiber web guiding element is shown;

[0035] Figure 3 A first embodiment of a sensor device in a fiber web guiding element is shown;

[0036] Figure 3a A second embodiment of the sensor device in the fiber web guiding element is shown;

[0037] Figure 4 A spinning preparation apparatus with an arrangement of multiple carding machines is shown, the multiple carding machines having control devices for summarizing and evaluating data. Detailed Implementation

[0038] The solutions according to the present invention include different combinations of features, particularly defined by the following sequentially numbered embodiments:

[0039] 1. A method for operating at least one carding machine in a spinning preparation apparatus, wherein fiber bundles are loosened to individual fibers, oriented, and cleaned between a rotating cylinder (4) with a carding cloth and fixed and surrounding carding elements, and the resulting fiber layer is transferred from the cylinder (4) to a doffer (5) and subsequently becomes a fiber sliver, said carding machine having a carding machine control device with an operating unit (18), wherein, after the operator inputs the raw material and output, the carding machine control device determines a reference cylinder rotation speed (n). R The feature is that the operator initiates an optimization program that allows the operator to select between agglomeration optimization and / or energy optimization operating modes, thereby enabling the carding machine control device to execute an automatic measurement sequence, wherein multiple sensors (30) detect the number of agglomerates in the fiber layer and simultaneously determine the drive power of the carding machine at different cylinder speeds within each predetermined duration (T), wherein the data from the sensors (30) and the drive power determined therein are transmitted to a higher-level control device (43) in the spinning preparation workshop, which, with the aid of at least the data, suggests to the operator, based on a mathematical algorithm, the operating mode of the carding machine under different quality categories.

[0040] 2. The method according to embodiment 1, characterized in that different cylinder rotation speeds are higher or lower than a reference cylinder rotation speed (n R ).

[0041] 3. The method according to embodiment 2, characterized in that different cylinder rotation speeds are based on a reference cylinder rotation speed (n). R The cylinders at the top and bottom have the same rotational speed spacing.

[0042] 4. The method according to embodiments 1 to 3, characterized in that the duration (T) for different cylinder rotation speeds is the same.

[0043] 5. The method according to embodiments 1 to 4 is characterized in that the determination of agglomeration and driving power is performed multiple times at each cylinder rotation speed, wherein the duration (t) of each measurement is added together to obtain the total duration (T).

[0044] 6. The method according to one of the above embodiments, characterized in that the control device of the carding machine is configured to automatically set the carding gap at different rotation speeds of the cylinder (5).

[0045] 7. The method according to one of the above embodiments, characterized in that the optimization program is started on the control device of at least one carding machine (100) or the control device (43) of the spinning preparation equipment.

[0046] 8. The method according to one of the above embodiments, characterized in that the result of the optimization procedure is transmitted by the control device (43) of the spinning preparation equipment to the operation unit (18) of the at least one carding machine (100) so that the operator can manually select the quality category and confirm it.

[0047] 9. The method according to any one of embodiments 1 to 7 above is characterized in that, after the operator selects and confirms the quality category, the result of the optimization program is obtained through the control device (43) of the spinning preparation equipment, thereby starting all the carding machines (100) of the spinning preparation equipment in an automatic operation mode.

[0048] 10. The method according to embodiment 1, characterized in that the operator can directly select whether at least one combing machine is operated in a clumping-optimized manner or an energy-optimized manner by means of starting the optimization program.

[0049] 11. A carding machine having a feeding side for fiber bundles, wherein the carding mechanism causes the fiber bundles to be conveyed to a rotating cylinder (4) by means of at least one licker roller (3a, 3b, 3c), wherein the fiber bundles are loosened into single fibers, oriented and cleaned between a fixed carding element and a surrounding cover strip (17) and the cylinder (4), and the resulting fiber layer can be transferred from the cylinder (4) to a doffer (5), wherein a device for converting the fiber layer into fiber strips is arranged downstream of the doffer, the carding machine having at least three sensors (3) configured to detect clumps or short fiber accumulations in the fiber layer, characterized in that the carding machine has a control device with an operating device (18), wherein an operator initiates an optimization program by means of the operating device after inputting raw materials and output, the carding machine operating in a clumping-optimized and / or energy-optimized mode using the optimization program.

[0050] 12. The carding machine according to embodiment 11, characterized in that the at least three sensors (30) detect the fiber layer on the doffer (5).

[0051] 13. The carding machine according to one of embodiments 11 to 12, characterized in that the sensor (30) is disposed in the fiber web guiding element (20).

[0052] 14. The carding machine according to embodiment 11, characterized in that the carding mechanism automatically sets the carding gap.

[0053] 15. The carding machine according to embodiment 11, characterized in that the carding mechanism is configured to determine energy consumption at different rotational speeds.

[0054] 16. A spinning preparation apparatus having a control device (43) having a plurality of carding machines (100), wherein the carding machines operate according to the method described in any one of embodiments 1 to 9, wherein the control device (43) is configured to process data of at least one carding machine (100), preferably a plurality of carding machines, with a mathematical algorithm, wherein in a first step, the data of at least one carding machine (100) is clustered or classified to generate quality categories, and in a second step, the data is further processed with a regression model, wherein the results of the mathematical algorithm are passed to each carding machine (100) to set settings for clumping optimization and / or energy optimization.

[0055] 17. The spinning preparation equipment according to embodiment 16, characterized in that the first step of determining clusters is by means of K-Means algorithm or fuzzy c-Means algorithm or hierarchical cluster analysis.

[0056] 18. The spinning preparation apparatus according to embodiment 16, characterized in that the first step of grouping is determined by label propagation using a semi-supervised learning model.

[0057] 19. The spinning preparation apparatus according to embodiment 16, characterized in that a random forest regression model, a multinomial classifier, or an artificial neural network is used as the regression model.

[0058] 20. The spinning preparation equipment according to embodiment 16 is characterized in that the data of multiple carding machines are processed in the control device (43) of the spinning preparation equipment and compared with each other in a first step, so that the control device outputs an optical or acoustic alarm signal when the clustered or classified data deviates from a preset reference value.

[0059] Figure 1 A carding machine 100 according to the prior art is shown, in which fiber bundles are guided through channels to feed rollers 1 and feed plate 2, and then guided to cylinder 4 or drums via at least one licker-in roller 3a, 3b, 3c. On cylinder 4, the fibers of the fiber bundle are made parallel and cleaned by means of fixed carding elements 13, suction hoods and cutters, and by means of surrounding carding elements arranged on a rotating cover plate system 17, these surrounding carding elements being constructed as cover strips 14. Subsequently, the resulting fiber layer 16 is conveyed to a fiber web guiding element 9 by a doffer 5, a stripping roller 6 and a plurality of squeeze rollers 7, 8, which transforms the fiber layer into fiber slivers by means of a gathering bell 10, the fiber slivers being passed to subsequent processing equipment or sliver cans 15 by separating rollers 11, 12. Adjustment of the cover strips 14 and carding elements 13 relative to cylinder 4 (carding gap) is made by sliding strips (not shown) having elements that can be wedge-shaped oriented relative to each other. The combing machine 100 is set and operated by means of the operation unit 18, which can be a monitor with an integrated input device.

[0060] exist Figure 2 and Figure 2a The diagram shows an exemplary arrangement of the fiber web guiding element 20 between the doffer 5, the stripping roller 6, and the extrusion roller 7, wherein the fiber layer 16 is removed from the doffer 5 by the stripping roller 6 and guided along the concave upper side 20c of the fiber web guiding element 20 to the collection flare 10.

[0061] Independent of this embodiment, the sensor 30 for detecting clumps, described subsequently, does not need to be integrated precisely into the fiber web guiding element. Importantly, the sensor 30 is aligned with the flat fiber layer 16, located between the cylinder 4 and the assembling horn 10 that transforms the fiber layer 16 into fiber strips.

[0062] In this embodiment, the fiber web guiding element 20 is mainly formed by four sides 20a, 20c, 20d, and 20e, which surround the cavity 20f. The front side 20a has at least partially a transparent element 20b configured to allow the sensor 30 located in the cavity 20f to detect or observe the fiber layer 16 located in the carding cloth 5a of the doffer 5. The transparent element 20b may be positioned only within the viewing angle of the sensor 30, or it may extend at least partially or completely across the working width of the carding machine 100 as a continuous transparent element 20b. Thus, the front side 20a of the fiber web guiding element 20 is oriented with a small gap from the surface of the doffer 5. The concave upper side 20c of the fiber web guiding element 20 guides the fiber layer 16 from the stripping roller 6 to the extrusion rollers 7 and 8. The difference from the prior art is that the sensor 30 detects the fiber layer 16 still within the teeth of the carding cloth 5a of the doffer 5. In this way, the fiber layer 16 rubs against the approximately vertically positioned front side 20a of the fiber web guide element 20 and also continuously passes over the transparent element 20b, which is thus significantly less contaminated than the horizontally positioned concave upper side 20c. According to the prior art, the transparent element is positioned within the horizontally positioned concave upper side 20c and is not fully contacted by the fiber layer 16, which may allow dirt to accumulate. This results in a longer service life before the fiber web guide element 20 or the transparent element 20b needs to be cleaned. A single sensor 30 can be movably positioned inside the fiber web guide element 20, or at least one sensor 30 can be fixedly positioned. In the case of multiple fixedly positioned sensors 30 within the fiber web guide element, these sensors can be arranged at regular intervals. The transparent element 20b can be constructed as a glass or plastic sheet, with a polarizing filter 31 positioned behind it. Therefore, the polarizing filter 31 is positioned between the transparent element 20b and the sensor 30. The polarizing filter 31 is configured for circular polarization, thereby blocking reflected light from the bright surface of the sensor 30 (e.g., the metal needle cloth 5a). Reflected light from the dull fibers and interfering particles remains visible to the sensor 30. Thus, the sensor 30 detects only the fibers of the fiber layer 16 and the interfering particles contained therein, and can identify thick sections, fiber knots, clumps, shell particle clusters, or foreign objects through image evaluation. White light is generated for this purpose, which, along with the polarizing filter, blocks the needle cloth 5a. Inside the fiber web guiding element 20, the polarizing filter 31 is framed by a reference film 32, which enables white balance. Figure 2a Sensor 30 is shown only schematically.

[0063] Each sensor 30 is equipped with its own computer 35, which can immediately evaluate the detected data. The configuration of each sensor 30 with its own computer 35 allows for parallel processing of the determined data, thus providing determined values ​​more quickly. The sensor 30 can be configured, for example, as a CCD sensor or a CMOS sensor, thereby enabling the detection of individual images. Simple binary allocation of the data is performed in the computer 35. For example, specific sizes of shell agglomerates, or specific sizes of fiber agglomerates or knots, can be defined as boundary conditions. This is achieved through image evaluation of the sensor data, either in the individual computers 35 of the sensors 30, or together in the control unit of the carding machine 100, or together in the control unit 43 of the spinning preparation equipment. This calibrates the sensitivity of the measurement data in the sense of conventional measurement techniques. Evaluating the data in the individual computers 35 of the sensors 30 enables faster data processing based on parallel processing. The evaluated data is transferred from the computer 35 to the control unit 43 of the spinning preparation equipment via the control unit of the carding machine for classification or clustering and further processing.

[0064] exist Figure 3 In the fiber web guiding element 20, for example, five sensors 30 are preferably evenly distributed at a spacing a across the working width A of the carding machine. With the cylinder 4 width, for example, being 1280 mm, the working width A is approximately 1180 mm, which is detected by five sensors, each with a detection width of 20 to 30 mm. Therefore, during evaluation by the carding machine's control unit, the detected magnetic track is obtained for each sensor. With the sensors 30 fixed, at least three sensors 30 within the fiber web guiding profile 20 have proven advantageous when the cylinder 4 width is 1000 mm, allowing for the determination of sufficiently accurate agglomerate counts. Considering the cost of the sensors 30 and the width of the cylinder 4, setting five sensors 30 has proven optimal, allowing for the determination of agglomerate counts with considerably high accuracy. The five sensors 30 are preferably fixedly positioned at the same spacing a across the working width A of the carding machine within the fiber web guiding element 20. To detect agglomerates, approximately 10,000 images must be determined and evaluated for each measurement by the sensors. With a carding machine output of, for example, 80 kg / h, the amount of 100 meters of fiber layer 16 passing through doffer 5 forms a measurement value, in which approximately 10,000 images are determined. In cases of irregular width distribution, the number of magnetic tracks (in this case, five sensors 30) determines the intermediate value of fiber sliver clumping with an error of 3%. With only three sensors 30, the error of the intermediate value of fiber sliver clumping increases to 12%. Using nine sensors 30 (… Figure 3aFurthermore, with these sensors arranged at a spacing b, the error in the intermediate value of fiber sliver clumping is reduced to 1%. Using a large number of sensors 30 not only improves the accuracy in determining clumping but also shortens the time required to determine the measurement value, because multiple images are detected simultaneously and processed in parallel by the computer 35. With three sensors 30, the measurement time is, for example, approximately 40 seconds; with five sensors 30, approximately 30 seconds; and with nine sensors 30, approximately 15 seconds. Since the internal structural space of the fiber web guiding element 20 is limited when the cylinder width is 1000mm to 1500mm, arranging three to nine fixed sensors for determining clumping has proven optimal. This provides sufficient structural space for the distribution of the sensors 30 across the working width A, ensures sufficient accuracy and measurement speed, and ultimately keeps costs within an affordable range.

[0065] Even though only regular spacing a, b of the sensors 30 is shown in this embodiment, this spacing can be irregular. The data evaluation algorithm needs to be adaptively adjusted if necessary. Therefore, it is advantageous for the sensors 30 in the middle of the fiber layer 16 to be arranged with larger or smaller spacing, because due to the specific carding machine structure, the detection of certain clumps, thick sections, or foreign objects occurs more frequently in the edge areas (due to lateral fluff) or more frequently in the middle of the fiber layer 16 (due to differences in carding gaps across the cylinder width).

[0066] If sensors 30 are also used to detect foreign objects, all sensors 30 together need to generate approximately 25,000,000 images. Therefore, with five sensors 30 used in the fiber web guiding element 20, each sensor must generate 5,000,000 images until a reliable indication of foreign object presence is possible. Consequently, a corresponding amount of fiber layer must be detected, extending the measurement process with evaluation to approximately 18 hours. Even with nine sensors 30, the measurement and evaluation time remains at 10 hours, meaning that carding machine production must at least be assessed while handling roving and must be destroyed in the event of a serious malfunction.

[0067] To shorten this process, the present invention proposes that data from the sensors 30 of at least one carding machine 100, preferably at least two carding machines, be aggregated and jointly evaluated in a control device 43 in the spinning preparation workshop. Unlike DE102019115138A1, the objective here is to detect the precise number of clumps (= hull clumps, short fiber accumulations, and fiber knots) and thereby optimally adjust all carding machines in the carding workshop. Alternatively or additionally, the carding machines can operate with energy optimization when the number of clumps is at its maximum. This does not preclude the use of sensors to detect foreign matter and other undesirable components and the adjustment of the upstream cotton cleaner via control device 43, as described in DE 102019115138A1.

[0068] Figure 4 A carding workshop with multiple carding machines 100 is shown, and the machine control unit of the carding workshop is connected to a higher-level control unit 43 in the spinning preparation workshop. The higher-level control unit 43 accumulates various data from the sensors 30 of each carding machine 100 and evaluates them according to different criteria. The evaluation may involve the proportion of short fibers, foreign matter, the number of clumps or knots, or characteristic values ​​such as turbidity, thin / thick sections, or fiber orientation. The display is configured to show different fiber types, such as foreign matter, clumps, or dirt particles. Through the input module 44, which can be configured as a mobile input terminal with a keyboard (such as a laptop, tablet, or smartphone), preset waste amounts or rejection rates can be entered, thereby enabling data optimization within the control unit 43. This can initiate an optimization program, and after the optimization program ends, the carding machines can be operated in a clumping optimization and / or energy optimization manner.

[0069] Control device 43 transmits new preset parameters with optimized data to the control devices of each carding machine, thereby allowing the operators or the machine to adjust accordingly based on quality requirements and energy consumption. The aim is to optimize the settings to achieve and maintain the required quality of the output material (fiber slivers formed from fiber layers) from the carding machine.

[0070] Data from sensors 30 at multiple carding machines 100 is aggregated in controller 43, which reduces the time required to detect the amount of images needed and decreases data processing time. This allows for the early prevention of potentially faulty carding machine production before further processing begins in subsequent spinning stages.

[0071] Summarizing data from sensors 30 at at least two carding machines in control unit 43 yields another advantage. If control unit 43 determines that the sensor measurements 30 change simultaneously or similarly across all carding machines, it can be assumed that there is a common cause, such as changes in raw materials, changes in one or more machines or the yarn being cleaned, or common changes in processing conditions (such as temperature or humidity in the spinning preparation room). Conversely, if a change is observed only on one carding machine, the cause is likely within the carding machine itself. Therefore, comparing measurements from at least two carding machines in control unit 43 can be used to more accurately determine which machine requires intervention to achieve the desired quality.

[0072] When the carding machine is started, the operator inputs the raw material and desired output into the control device of each carding machine 100 via the operation unit 18. The output, expressed in kg / h, is simultaneously associated with either a reference cylinder speed nR or a cylinder velocity, thus they can be considered as two equivalent input options. However, these vary depending on the type of carding machine because the cylinder diameter and working width differ. According to existing technology, this data is stored in the control device of the carding machine, thus only the input needs to be changed. Accordingly, with the same output, the carding gap can be input or adjusted by the operator or through the control device of the carding machine 100. If a higher cylinder speed is desired, a larger carding gap must be adjusted when the machine is cold, as the subsequent heat release increases and the carding gap changes significantly. However, a larger carding gap is detrimental to reducing agglomeration. Therefore, as the carding gap decreases, the number of agglomerates can be reduced, but fiber damage increases. In the case of automatic carding gap setting, the carding gap can be automatically determined by inputting the raw material. Although the operator also inputs the output into the operating unit 18, the combing gap remains constant because the combing gap is adjusted to be narrower or wider by the temperature level of the cylinder 4.

[0073] The optimization program according to the invention can be initiated by the operator using the operating unit 18 on the carding machine 100 or via the control device 43 or input module 44 of the spinning preparation equipment, i.e., starting from the reference cylinder speed n. R Initially, the combing machine 100 gradually deviates from the reference cylinder speed n upwards or downwards within a preset time window. R The purpose is to determine the detected clumping and energy consumption at each cylinder rotation speed, allowing the operator to choose between clumping-optimized and / or energy-optimized operating modes. For example, initially, referencing cylinder rotation speed n... R =500U / min (duration T1), then n R -10U / min (duration T2), then n R -20U / min (duration T3), then nR -30U / min (duration T4). Next, the method uses the cylinder rotation speed n as a reference. R Run at +10U / min (duration T5), then n R +20U / min (duration T6), then n R +30U / min (duration T7). Duration T1...T for each step of the carding machine's operation at speeds above and below the reference cylinder speed. x It depends on the magnitude of the data determined by sensor 30, but it is the same for all measurement processes. Relative to the reference cylinder rotation speed n R The speed difference can be achieved at the same interval, such as 10 or 20 levels, or at different intervals and in different sequences.

[0074] From an operator's perspective, if high-quality raw materials are anticipated and the number of clumps in the produced carding slivers is low, the operator can directly select the energy-optimized operating mode, allowing the spinning preparation workshop's control system to evaluate data with higher energy optimization weighting. Conversely, if the raw materials are of poor quality or contain dust, the operator can directly select the clumping-optimized operating mode, allowing the spinning preparation workshop's control system to evaluate data with higher quality weighting. If the operator has limited experience or cannot estimate the number of clumps, they can activate the optimization mode to obtain a neutral recommendation regarding either energy-optimized or clumping-optimized operating mode. Alternatively, this selection can also be made within the spinning preparation workshop's control system.

[0075] Two sets of data were used to determine the detected clumping and energy consumption. One set involved production-related data, which was also determined by sensor 30. The other set involved machine status data, where energy consumption was determined at a specific cylinder speed. The current power of the drive was also determined during each measurement to determine the number of clumping. The two sets of data were then combined. Here, for each duration T1 to T2... x Execution with a single time or duration t1...t x Multiple measurement processes are performed to determine highly reliable results. If the raw material has been used unchanged for several hours, then one measurement process is sufficient for inspection. If new raw materials are used or other machine settings are changed, then multiple measurement processes at the same cylinder speed are reasonable. For example, at a cylinder speed of n... RAt a speed of 500 U / min and a duration of T1 = 30 min, five measurement processes are initiated, each lasting t = 6 min. This is predicated on calibrating the sensitivity of the sensor 30 (i.e., setting parameters for the level of short fiber accumulation or clumping from which the measured values ​​are calculated). Sensitivity can be checked by manually evaluating the extracted fiber sliver in a textile technology laboratory, thereby evaluating the data from all sensors 30 in the same manner based on this feedback. Because at least three fixed sensors 30 are used according to the invention, at least three sets of data with the number of clumping are obtained for each sensor 30 and measurement process, exemplarily presented in the table below over the working width. Therefore, for each measurement process 4.1 to 4.5, a measurement value for a sub-width or track of the working width A is provided based on the number of sensors 30 (here, three sensors 30), this is extrapolated to the entire working width A and thus to the yield.

[0076] Table 1

[0077] Serial Number Duration t [min] Cylindrical rotation speed [n] Supply speed [m / min] Power [kW] Number of clumps [1 / g] 4.1 6 500 350 11.5 93 4.2 6 500 350 11.2 92 4.3 6 500 350 11.4 98 4.4 6 500 350 11.5 95 4.5 6 500 350 11.8 97

[0078] The multiple measurement procedures listed in Table 1 are performed the same number of times for each cylinder speed. For example, all measurement procedures for each cylinder speed are evaluated based on the measured values ​​or clustering or classification of the data, where measured values ​​4.1 to 4.5 are summarized under measurement number 4 in the subsequent Table 2. Here, the duration T and the duration t of all individual measurements are the same. The duration or time T used here for each speed range is equivalent to the sum of the durations t of each measurement.

[0079] Table 2

[0080] Serial Number Duration T [min] Cylindrical rotation speed [n] Supply speed [m / min] Power [kW] Number of clumps [1 / g] 1 30 470 350 10.5 156 2 30 480 350 10.8 130 3 30 490 350 11.2 112 4 30 500 350 11.5 95 5 30 510 350 11.8 75 6 30 520 350 12.1 79 n 30 530 350 12.5 87

[0081] As can be seen from the table, energy consumption increases with increasing cylinder speed, but at a certain point, even with the combing gap remaining unchanged, the number of detected agglomerates decreases. From the point where the cylinder speed increases (for example, given as 520 U / min), the number of detected agglomerates can increase again.

[0082] The evaluation of the measured values ​​is achieved through a mathematical algorithm in the control device 43 of the spinning preparation workshop, which operates as follows:

[0083] The data is evaluated using clustering. Each cluster is assigned a quality category: high, medium, or low. The data for each cluster can be processed using the K-Means algorithm, where a regression model predicts cylinder speeds for each quality category. The K-Means algorithm is used for cluster analysis of data, where a certain number of similar objects form a known number of k groups. The value of k is equal to the number of quality categories and can be between 1 and X, meaning it can include 5 or 10 categories. In this embodiment, k=3 is chosen for the three quality categories: high, medium, and low. In this case, the value of k can be determined using the elbow method or by the silhouette coefficient. In the elbow method, each k value is calculated by summing the squared distances between a data point and its nearest cluster center, while the silhouette coefficient is calculated by comparing the similarity of a data point to its cluster with other clusters. Alternatives to the K-Means algorithm can be other methods for data classification or clustering, such as fuzzy c-means or hierarchical cluster analysis.

[0084] Alternatively, a semi-supervised learning model, label propagation, can be used to aggregate measurements into quality categories. For this, the user must set boundary conditions, i.e., labels. This can be understood as predefined target variables corresponding to the class labels. Operators, based on experience, label a small amount of data as boundary conditions by predefining or inputting a small range of cluster numbers for each quality category. These n labels are then propagated, ultimately generating a total of n classes.

[0085] The data from both concepts can be further processed using a regression model in the control unit 43 of the spinning preparation workshop to determine or predict the optimal cylinder speed for each quality category. This could be a random forest regression model, which provides reliable results with high probability even with limited data. In this case, a tree-based algorithm is used, which can be tuned quite well in terms of the number and depth of decision nodes using its parameters.

[0086] Alternatively, multinomial classifiers or artificial neural networks can also be used.

[0087] Therefore, the control device 43 in the spinning preparation workshop is configured to transmit measurement data evaluated by an algorithm to at least one carding machine 100, so that the carding machine can operate with the fewest possible clumps or in an energy-optimized manner according to the desired quality grade. The control device of the carding machine can automatically set the associated cylinder speed and operate in the desired mode, or display the suggested value on the operation unit 18 and allow the operator to select it using a confirmation button or through the carding machine's adjustment device. In this case, the data for subsequent quality grades determined by the evaluation model need not be consistent with the actual measured values, but are associated with the determined data:

[0088] Table 3

[0089] Quality grade Cylindrical rotation speed [n] Supply speed [m / min] Power [kW] Number of clumps [1 / g] high 510 350 11.8 75 middle 490 350 11.2 108 Low 470 350 10.5 135

[0090] Table 3 shows recommendations when the operator does not select either cluster optimization or energy optimization operation mode at the start of optimization mode.

[0091] If the operator selects, for example, the clumping optimization setting, then the weighting coefficients in the K-Means method should be changed, and the recommended settings are shown in Table 4:

[0092] Table 4

[0093] Quality grade Cylindrical rotation speed [n] Supply speed [m / min] Power [kW] Number of clumps [1 / g] high 520 350 12.0 68 middle 500 350 11.5 92 Low 480 350 10.8 115

[0094] When setting or selecting energy optimization by the operator, the weighting coefficients in the K-Means method can be changed, and the recommended settings are shown in Table 5:

[0095] Table 5

[0096] Quality grade Cylindrical rotation speed [n] Supply speed [m / min] Power [kW] Number of clumps [1 / g] high 480 350 10.8 115 middle 470 350 10.5 135 Low 450 350 10.3 263

[0097] Without needing to pre-select a desired operating mode, after initiating the optimization program and measuring at different speeds at intervals, the operator can choose whether to operate the carding machine with the fewest possible clusters or with an acceptable cluster count within preset limits, thus minimizing energy consumption. For this purpose, the operator receives a suggestion from the control device 43 in the spinning preparation workshop, which is displayed on the operating unit 18 and requires only confirmation. Alternatively, the control device 43 of the spinning preparation equipment can be configured to intervene in the control of at least one carding machine 100 and automatically initiate the desired operating mode based on the operator's preset parameters. The operator then makes the selection at the level of the control device 43 of the spinning preparation equipment.

[0098] Table 3 above can be shown as a suggestion or presented graphically. Alternatively, the values ​​in Table 3 can also be displayed on the control device 43 in the spinning preparation workshop, thereby automatically or manually starting at least one carding machine 100. Similarly, the operator can directly select the clumping-optimized or energy-optimized operating mode and display or graphically present the data in Table 4 or Table 5.

[0099] If these carding machines are ensured to have identical settings, such as licker-in rollers, adjustable cutters, fixed carding elements, doffer spacing, etc., then a test run on one carding machine is sufficient to set all other carding machines with determined values ​​via the control device 43 in the spinning preparation workshop. However, the method completes more quickly in a time-optimized manner with at least two carding machines because more data can be processed in parallel.

[0100] With the carding gap set automatically, the carding gap remains constant, unaffected by the wear condition of the carding cloth, and only related to the raw material and the temperature of the carding machine. In this respect, although worn carding cloth increases the number of clumps, the optimized settings for the carding machine recommended in Tables 3 to 5 also apply here.

[0101] Laboratory tests of the combing strips showed that the deviations between different quality grades were 1% and 3% compared to the data determined by the K-Means algorithm using a random forest regression model.

[0102] List of reference numerals

[0103] 100 combing machine

[0104] 1. Feed Lola

[0105] 2 feed plate

[0106] 3a, 3b, 3c licker rollers

[0107] 4 Xilin

[0108] 5 dolf

[0109] 5a needle cloth

[0110] 6. Peeling Laura

[0111] 7 and 8 extrusion rollers

[0112] 9 Fiber mesh guiding elements

[0113] 10-unit funnel mouth

[0114] 11, 12 Separation Roller

[0115] 13 Combing components

[0116] 14 cover strips

[0117] 15 tubes

[0118] 16 fiber layers

[0119] 17-turn cover system

[0120] 18 operating units

[0121] 20 Fiber mesh guiding element

[0122] 20a front side

[0123] 20b component

[0124] 20c upper side

[0125] 20d rear side

[0126] 20e bottom side

[0127] 20f cavity

[0128] 30 sensors

[0129] 31 polarizing filter

[0130] 32 Reference Thin Film

[0131] 35 Computers

[0132] 43 Control Device

[0133] 44 Input Module

[0134] A working width

[0135] a spacing

[0136] b spacing

[0137] n R Reference cylinder speed

[0138] t1- t x time

[0139] T1-T x time

Claims

1. A method for operating at least one carding machine in spinning preparation equipment, wherein, The fiber bundles are loosened, oriented, and cleaned between the rotating cylinder (4) with needle cloth and the fixed and surrounding carding elements, and the resulting fiber layer is transferred from the cylinder (4) to the doffer (5) and subsequently becomes a fiber strip. The carding machine has a carding machine control device with an operating unit (18), wherein, after the operator inputs the raw materials and output, the carding machine control device determines a reference cylinder rotation speed (n). R The feature is that the operator initiates an optimization program that allows the operator to select between agglomeration optimization and / or energy optimization operating modes, thereby enabling the carding machine control device to execute an automatic measurement sequence, wherein multiple sensors (30) detect the number of agglomerates in the fiber layer and simultaneously determine the drive power of the carding machine at different cylinder speeds within each predetermined duration (T), wherein the data from the sensors (30) and the drive power determined therein are transmitted to a higher-level control device (43) in the spinning preparation workshop, which, with the aid of at least the data, suggests to the operator, based on a mathematical algorithm, the operating mode of the carding machine under different quality categories.

2. The method according to claim 1, characterized in that, Different cylinder speeds are higher or lower than the reference cylinder speed (n) R ).

3. The method according to claim 2, characterized in that, Different cylinder speeds are reference cylinder speeds (n) R The cylinders at the top and bottom have the same rotational speed spacing.

4. The method according to claim 1, characterized in that, The duration (T) for different cylinder speeds is the same.

5. The method according to claim 3, characterized in that, The duration (T) for different cylinder speeds is the same.

6. The method according to claim 1, characterized in that, Multiple measurements of agglomeration and drive power were performed at each cylinder speed, and the duration (t) of each measurement was summed to obtain the total duration (T).

7. The method according to claim 5, characterized in that, Multiple measurements of agglomeration and drive power were performed at each cylinder speed, and the duration (t) of each measurement was summed to obtain the total duration (T).

8. The method according to claim 1, characterized in that, The control device of the carding machine is designed to automatically set the carding gap at different speeds of the cylinder (5).

9. The method according to claim 7, characterized in that, The control device of the carding machine is designed to automatically set the carding gap at different speeds of the cylinder (5).

10. The method according to claim 1, characterized in that, The optimization program is initiated on the control device of at least one carding machine (100) or the control device (43) of the spinning preparation equipment.

11. The method according to claim 9, characterized in that, The optimization program is initiated on the control device of at least one carding machine (100) or the control device (43) of the spinning preparation equipment.

12. The method according to claim 1, characterized in that, The results of the optimization process are transmitted by the control device (43) of the spinning preparation equipment to the operation unit (18) of the at least one carding machine (100) so that the operator can manually select and confirm the quality category.

13. The method according to claim 11, characterized in that, The results of the optimization process are transmitted by the control device (43) of the spinning preparation equipment to the operation unit (18) of the at least one carding machine (100) so that the operator can manually select and confirm the quality category.

14. The method according to claim 1, characterized in that, After the operator selects and confirms the quality category, the results of the optimization program are obtained through the control device (43) of the spinning preparation equipment, thereby starting all the carding machines (100) of the spinning preparation equipment in an automatic operation mode.

15. The method according to claim 11, characterized in that, After the operator selects and confirms the quality category, the results of the optimization program are obtained through the control device (43) of the spinning preparation equipment, thereby starting all the carding machines (100) of the spinning preparation equipment in an automatic operation mode.

16. The method according to claim 1, characterized in that, With the help of the optimization program, the operator can directly select whether at least one combing machine is to be operated in a clumping-optimized or energy-optimized manner.

17. A carding machine having a feeding side for the fiber bundle, wherein, The carding mechanism causes the fiber bundle to be conveyed to the rotating cylinder (4) by means of at least one licker roller (3a, 3b, 3c), wherein the fiber bundle is loosened into individual fibers, oriented and cleaned between the fixed carding element and the surrounding cover strip (17) and the cylinder (4), and the resulting fiber layer can be transferred from the cylinder (4) to the doffer (5), downstream of the doffer is a device for converting the fiber layer into fiber strips, the carding machine having at least three sensors (3) configured to detect clumps or short fiber accumulations in the fiber layer, characterized in that the carding machine has a carding machine control device with an operating unit (18), which the operator initiates an optimization program by means of the operating unit after inputting raw materials and output, the carding machine operating in a clumping-optimized and / or energy-optimized mode using the optimization program.

18. The carding machine according to claim 17, characterized in that, The at least three sensors (30) detect the fiber layer on the doffer (5).

19. The carding machine according to any one of claims 17 to 18, characterized in that, The sensor (30) is disposed in the fiber web guiding element (20).

20. The carding machine according to claim 17, characterized in that, The combing mechanism automatically sets the combing gap.

21. The carding machine according to claim 17, characterized in that, The comb mechanism is designed to determine energy consumption at different speeds.

22. A spinning preparation device having a control unit (43) and having multiple carding machines (100), wherein, The carding machine operates according to any one of claims 1 to 15, wherein the control device (43) of the spinning preparation equipment is configured to process the data of at least one carding machine (100) with a mathematical algorithm, wherein in a first step, the data of at least one carding machine (100) is clustered or classified to generate quality categories, and in a second step, it is further processed with a regression model, wherein the results of the mathematical algorithm are passed to each carding machine (100) to set settings for clumping optimization and / or energy optimization.

23. A spinning preparation device having a control unit (43) and having multiple carding machines (100), wherein, The carding machine operates according to any one of claims 1 to 15, wherein the control device (43) of the spinning preparation equipment is configured to process data from at least a plurality of carding machines using a mathematical algorithm, wherein in a first step, data from at least one carding machine (100) is clustered or classified to generate quality categories, and in a second step, further processed using a regression model, wherein the results of the mathematical algorithm are passed to each carding machine (100) to set settings for clumping optimization and / or energy optimization.

24. The spinning preparation equipment according to claim 22 or 23, characterized in that, The first step in determining clusters is through the K-Means algorithm, the fuzzy c-Means algorithm, or hierarchical cluster analysis.

25. The spinning preparation equipment according to claim 22 or 23, characterized in that, The first step in grouping is determined through label propagation using a semi-supervised learning model.

26. The spinning preparation equipment according to claim 22 or 23, characterized in that, Random forest regression models, multinomial classifiers, or artificial neural networks can be used as regression models.

27. The spinning preparation equipment according to claim 22 or 23, characterized in that, Data from multiple carding machines are processed in the control unit (43) of the spinning preparation equipment and compared with each other in the first step, so that the control unit of the spinning preparation equipment outputs an optical or acoustic alarm signal when the clustered or classified data deviates from the preset reference value.

Images