Paperboard pile control method and device based on warping estimation, equipment and medium
By collecting and processing multi-dimensional data to generate a quantitative value of warpage, the automated stacking equipment is guided to stack cardboard, which solves the problem of poor stability of cardboard stacking in the existing technology and improves the regularity and stability of cardboard stacking.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- DONGGUAN HOI FU PAPER PROD CO LTD
- Filing Date
- 2026-01-27
- Publication Date
- 2026-07-31
AI Technical Summary
The lack of precise warping judgment criteria in the current technology of cardboard stacking process leads to poor stacking stability and makes it difficult to achieve the expected results.
By collecting multi-dimensional raw datasets, progressive data processing is performed to generate quantified values of warpage. Based on these quantified values and constraints, a stacking strategy is generated, and stacking operations are executed using automated equipment. The model is then validated and optimized.
It achieves regularity and stability in cardboard stacking, improves the adaptability and reliability of stacking strategies, and ensures the accuracy and continuous optimization of stacking results.
Smart Images
Figure CN121929566B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of industrial data processing technology, and in particular to a method, apparatus, equipment and medium for controlling cardboard stacking based on warpage prediction. Background Technology
[0002] Cardboard often warps after processing. When stacking cardboard, some pieces should be stacked face up and some face down alternately to counteract the warping and ensure smooth stacking. Currently, this alternating stacking operation relies entirely on manual labor. Workers depend on their experience to determine the stacking method, which lacks precise judgment and cannot adapt to the actual warping condition. This often results in poor overall stability after stacking, failing to achieve the desired stacking effect. Summary of the Invention
[0003] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a method, apparatus, equipment, and medium for controlling cardboard stacking based on warpage prediction, which can ensure the regularity and stability of cardboard stacking.
[0004] In a first aspect, this application provides a method for controlling cardboard stacking based on warpage prediction, including:
[0005] Collect a multi-dimensional raw dataset of cardboard to be stacked; wherein, the multi-dimensional raw dataset includes basic material parameters, production process parameters, dimensional characteristic parameters and environmental related parameters; The multi-dimensional original dataset is subjected to progressive data processing to obtain an optimized feature set; The optimized feature set is input into a preset warp prediction model to obtain a quantitative value of the warp degree of the cardboard to be stacked. Based on the quantified warpage value and preset stacking constraints, a stacking strategy for the cardboard is generated; wherein, the stacking strategy is a matching rule for the number of cardboard stacks with the front side facing up and the number of cardboard stacks with the back side facing up. The stacking strategy is converted into control instructions for automated stacking equipment, and the control instructions are sent to enable the automated stacking equipment to complete the cardboard stacking. The overall effect after stacking is verified, and the warpage prediction model is optimized based on the verification and evaluation results.
[0006] The cardboard stacking control method based on warpage prediction according to the first aspect of this application has at least the following beneficial effects: First, a multi-dimensional original dataset of cardboard to be stacked is collected, which includes material basic parameters, production process parameters, dimensional characteristic parameters and environmental related parameters. Then, the multi-dimensional original dataset is subjected to progressive data processing to obtain an optimized feature set. After inputting the optimized feature set into a preset warpage prediction model, a quantitative value of the warpage degree of the cardboard to be stacked can be obtained. Then, a stacking strategy of the cardboard is generated based on the quantitative value of the warpage degree and the preset stacking constraints. The stacking strategy is a matching rule for the number of cardboard stacked face up and the number of cardboard stacked face down. Subsequently, the stacking strategy is converted into control instructions for automated stacking equipment and sent to complete the cardboard stacking operation. After the stacking is completed, the overall effect is verified, and the warpage prediction model is optimized based on the verification and evaluation results. By collecting and processing multi-dimensional data on cardboard, the degree of cardboard warping can be accurately quantified. Based on the quantification results, an appropriate stacking strategy can be formulated to guide automated stacking operations. This effectively solves the problem that in existing technologies, manual stacking of cardboard based on experience and inaccurate judgment methods leads to poor overall stability of the cardboard after stacking. It can control the matching method of alternating stacking based on the actual warping of the cardboard, ensuring the regularity and stability of the cardboard stack. At the same time, the warping prediction model can be continuously optimized based on the stacking effect verification, further improving the adaptability of the cardboard stacking strategy and the reliability of the stacking operation.
[0007] According to some embodiments of the first aspect of this application, the progressive data processing of the multi-dimensional original parameters to obtain an optimized feature set includes: Based on the multi-dimensional original dataset, feature derivation is performed to obtain an intermediate derived feature set; An initial fused feature set is obtained by combining the multi-dimensional original dataset and the intermediate derived feature set. According to the preset redundancy removal rules, the initial fused features are partially removed to obtain an optimized feature set.
[0008] According to some embodiments of the first aspect of this application, the step of performing feature derivation based on the multi-dimensional original dataset to obtain an intermediate derived feature set includes: Based on the material basic parameters and the production process parameters, the front equivalent shrinkage coefficient and the back equivalent shrinkage coefficient are obtained; The shrinkage difference is obtained based on the equivalent shrinkage coefficient of the front side and the equivalent shrinkage coefficient of the back side; The quantitative internal stress difference is obtained based on the material basic parameters, the dimensional characteristic parameters, and the shrinkage difference. Based on the environmental correlation parameters, the environmental adaptability coefficient is obtained; Based on the aforementioned size characteristic parameters, the geometric constraint factor is obtained; By integrating the positive equivalent shrinkage coefficient, the negative equivalent shrinkage coefficient, the shrinkage difference, the quantified internal stress difference, the environmental adaptation coefficient, and the geometric constraint factor, an intermediate derived feature set is obtained.
[0009] According to some embodiments of the first aspect of this application, the redundancy removal rule is established through the following steps: Obtain the initial historical fusion feature set; Calculate the similarity between each parameter in each of the historical initial fusion feature sets and other parameters, retain the parameter with the highest similarity, and remove and record the other parameters with the highest similarity to form the redundancy removal rule.
[0010] According to some embodiments of the first aspect of this application, generating a cardboard stacking strategy based on the quantified warpage value and preset stacking constraints includes: Set an overall warpage threshold, a height threshold, and an efficiency threshold; wherein, the overall warpage threshold is used to characterize the maximum degree of warpage after cardboard stacking, the height threshold is used to characterize the maximum height after cardboard stacking, and the efficiency threshold is used to characterize the total number of cardboard pieces that the automated stacking equipment must grab at least per unit time; The overall warpage threshold, the height threshold, and the efficiency threshold are used as stacking constraints. Based on the quantified value of the degree of warping, an objective function is constructed with the goal of optimizing the mutual cancellation of the warping stress generated by stacking the cardboard face up and the warping stress generated by stacking the cardboard face down. Based on the stacking constraints, the objective function is solved to obtain the first stack quantity with the front side of the cardboard facing up, the second stack quantity with the back side facing up, and the total stack quantity for each batch. A stacking strategy for cardboard is generated based on the first stack quantity, the second stack quantity, and the total stack quantity.
[0011] According to some embodiments of the first aspect of this application, the warpage prediction model is trained through the following steps: Collect optimized feature sets and corresponding actual warping measurements from multiple historical batches of cardboard, construct a labeled training dataset, and divide the training dataset into a training set, a validation set, and a test set according to a preset ratio; Construct an initial warp prediction model; wherein the warp prediction model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer matches the dimension of the optimized feature set, and the number of nodes in the output layer is 1 and corresponds to the quantized value of the warp degree; The initial warpage prediction model is trained using the training set, and the parameters of the warpage prediction model are adjusted using the validation set until the prediction error of the warpage prediction model on the validation set meets a preset error threshold. The trained warpage prediction model is tested using the test set. If the verification is successful, the preset warpage prediction model is obtained.
[0012] According to some embodiments of the first aspect of this application, verifying the overall effect after stacking is completed, and optimizing the warpage prediction model and the stacking strategy based on the verification evaluation results, includes: Obtain an overall image of the cardboard after it has been fully stacked, and preprocess the overall image of the cardboard; Feature extraction is performed on the overall image of the cardboard to obtain overall verticality features, top surface flatness features, and cardboard edge alignment features; The overall verticality feature, the top surface flatness feature, and the cardboard edge alignment feature are quantified to obtain the verticality deviation value, the maximum unevenness difference of the top surface, and the maximum misalignment distance of the edge, respectively. The verification evaluation result is obtained based on the verticality deviation value, the maximum unevenness difference of the top surface, the maximum misalignment distance of the edge, and the corresponding preset evaluation threshold. If the verification evaluation result is unsatisfactory, the optimized feature set of that batch and the corresponding warpage quantification value are added to the training set of the warpage prediction model to optimize the parameters of the warpage prediction model.
[0013] Secondly, this application also provides a cardboard stacking control device based on warpage prediction, comprising: The acquisition unit is used to acquire a multi-dimensional raw dataset of cardboard to be stacked; wherein the multi-dimensional raw dataset includes basic material parameters, production process parameters, dimensional characteristic parameters and environmental related parameters. The processing unit is used to perform progressive data processing on the multi-dimensional original dataset to obtain an optimized feature set; The quantization unit is used to input the optimized feature set into a preset warp prediction model to obtain a quantified value of the warp degree of the cardboard to be stacked. The generation unit is used to generate a stacking strategy for the cardboard based on the quantified warpage value and preset stacking constraints; wherein, the stacking strategy is a matching rule for the number of cardboard stacks with the front side facing up and the number of cardboard stacks with the back side facing up. A sending unit is used to convert the stacking strategy into control instructions for automated stacking equipment, and send the control instructions to enable the automated stacking equipment to complete the cardboard stacking. An optimization unit is used to verify the overall effect after stacking is completed, and to optimize the warpage prediction model based on the verification and evaluation results.
[0014] Thirdly, this application also provides an electronic device, comprising: At least one memory; At least one processor; At least one program; The program is stored in the memory, and the processor executes at least one of the programs to implement the cardboard stacking control method based on warpage prediction as described in any embodiment of the first aspect.
[0015] Fourthly, this application also provides a computer-readable storage medium storing a computer program for performing the cardboard stacking control method based on warpage prediction as described in any embodiment of the first aspect.
[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0017] Additional aspects and advantages of this application will become apparent and readily understood in conjunction with the following description of the embodiments, in which: Figure 1 Flowcharts of a cardboard stacking control method based on warpage prediction provided for some embodiments of this application; Figure 2 This is a schematic diagram of a cardboard stacking control method based on warpage prediction provided in some embodiments of this application. Detailed Implementation
[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.
[0019] In the description of this application, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.
[0020] In the description of this application, the use of "first" and "second" is for the purpose of distinguishing technical features only, and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.
[0021] In the description of this application, unless otherwise expressly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.
[0022] Cardboard often warps after processing. When stacking cardboard, some pieces should be stacked face up and some face down alternately to counteract the warping and ensure smooth stacking. Currently, this alternating stacking operation relies entirely on manual labor. Workers depend on their experience to determine the stacking method, which lacks precise judgment and cannot adapt to the actual warping condition. This often results in poor overall stability after stacking, failing to achieve the desired stacking effect.
[0023] Based on this, this application provides a method, apparatus, equipment and medium for controlling cardboard stacking based on warpage prediction to solve the above-mentioned technical problems. The technical solutions provided by this application will be described in detail below.
[0024] Firstly, referring to Figure 1 This application provides a method for controlling cardboard stacking based on warpage prediction, which may include, but is not limited to, the following steps: Step S110: Collect the multi-dimensional raw dataset of the cardboard to be stacked; wherein, the multi-dimensional raw dataset includes material basic parameters, production process parameters, dimensional characteristic parameters and environmental related parameters.
[0025] Step S120: Perform progressive data processing on the multi-dimensional original dataset to obtain an optimized feature set.
[0026] Step S130: Input the optimized feature set into the preset warp prediction model to obtain the quantified value of the warp degree of the cardboard to be stacked.
[0027] Step S140: Based on the warp degree quantification value and the preset stacking constraints, generate a stacking strategy for the cardboard; wherein, the stacking strategy is a matching rule for the number of cardboard stacked face up and the number of cardboard stacked face down.
[0028] Step S150: Convert the stacking strategy into control instructions for the automated stacking equipment and send the control instructions to enable the automated stacking equipment to complete the cardboard stacking.
[0029] Step S160: Verify the overall effect after stacking is completed, and optimize the warpage prediction model based on the verification and evaluation results.
[0030] In steps S110 to S160, a multi-dimensional original dataset of the cardboard to be stacked is first collected. This dataset includes basic material parameters, production process parameters, dimensional characteristic parameters, and environmental related parameters. Then, progressive data processing is performed on the multi-dimensional original dataset to obtain an optimized feature set. After inputting the optimized feature set into a preset warpage prediction model, the warpage degree of the cardboard to be stacked can be quantified. Then, based on the warpage degree quantification value and preset stacking constraints, a stacking strategy for the cardboard is generated. This stacking strategy is a matching rule for the number of cardboard stacked face up and the number of cardboard stacked face down. Subsequently, the stacking strategy is converted into control instructions for automated stacking equipment and sent to complete the cardboard stacking operation. After the stacking is completed, the overall effect is verified, and the warpage prediction model is optimized based on the verification and evaluation results. By collecting and processing multi-dimensional data on cardboard, the degree of cardboard warping can be accurately quantified. Based on the quantification results, an appropriate stacking strategy can be formulated to guide automated stacking operations. This effectively solves the problem that in existing technologies, manual stacking of cardboard based on experience and inaccurate judgment methods leads to poor overall stability of the cardboard after stacking. It can control the matching method of alternating stacking based on the actual warping of the cardboard, ensuring the regularity and stability of the cardboard stack. At the same time, the warping prediction model can be continuously optimized based on the stacking effect verification, further improving the adaptability of the cardboard stacking strategy and the reliability of the stacking operation.
[0031] It is understood that, in step S120, the following steps may be included, but are not limited to: Step S210: Perform feature derivation based on the multi-dimensional original dataset to obtain an intermediate derived feature set.
[0032] Step S220: Combine the multi-dimensional original dataset and the intermediate derived feature set to obtain the initial fused feature set.
[0033] Step S230: According to the preset redundancy removal rules, perform partial parameter removal on the initial fused features to obtain the optimized feature set.
[0034] In steps S210 to S230, a progressive data processing approach is adopted. First, an intermediate derived feature set is obtained by derivation from the multi-dimensional original dataset. Then, the multi-dimensional original dataset and the intermediate derived feature set are combined to form an initial fused feature set. Finally, partial parameter removal processing is performed on the initial fused features according to a preset redundancy removal rule. This approach fully mines the effective information in the multi-dimensional original dataset, achieving an organic fusion of the original basic data and derived feature data. This allows the feature data to more comprehensively and completely reflect the actual attributes of the cardboard to be stacked, significantly improving the richness and adaptability of the feature data. Simultaneously, the parameter removal processing on the initial fused features effectively filters out redundant information in the feature data, avoiding adverse interference from redundant parameters to subsequent data operations. The resulting optimized feature set simplifies the data dimensions and improves the efficiency of subsequent data operations.
[0035] It is understood that step S210 may include, but is not limited to, the following steps: Step S310: Based on the material basic parameters and production process parameters, obtain the front equivalent shrinkage coefficient and the back equivalent shrinkage coefficient.
[0036] Step S320: Obtain the shrinkage difference based on the front equivalent shrinkage coefficient and the back equivalent shrinkage coefficient.
[0037] Step S330: Based on the material's basic parameters, dimensional characteristic parameters, and shrinkage difference, obtain the quantitative internal stress difference.
[0038] Step S340: Obtain the environmental adaptation coefficient based on the environmental correlation parameters.
[0039] Step S350: Obtain the geometric constraint factor based on the dimensional characteristic parameters.
[0040] Step S360: Integrate the front equivalent shrinkage coefficient, the back equivalent shrinkage coefficient, the shrinkage difference, the quantified internal stress difference, the environmental adaptation coefficient, and the geometric constraint factor to obtain the intermediate derived feature set.
[0041] By relying on basic material parameters and production process parameters, the equivalent shrinkage coefficients of the front and back sides are obtained, and the shrinkage difference is further derived. Combining basic material parameters, dimensional characteristic parameters, and shrinkage difference, the internal stress difference is quantified. Simultaneously, environmental adaptation coefficients are obtained based on environmental correlation parameters, and geometric constraint factors are obtained based on dimensional characteristic parameters. Finally, various parameters are integrated to form an intermediate derived feature set. This allows for in-depth feature mining and transformation of multi-dimensional raw datasets from multiple key dimensions, including paperboard material properties, production processes, dimensional specifications, and environmental adaptation. This feature derivation method transforms previously scattered raw parameters into feature indicators that directly characterize the warp-related properties of paperboard. The resulting intermediate derived feature set comprehensively and accurately reflects the shrinkage difference, mechanical properties, environmental adaptation status, and geometric constraint conditions on both sides of the paperboard, fully demonstrating the core factors affecting paperboard warp. This makes the obtained intermediate derived feature set highly targeted and effective, further ensuring the accuracy of subsequent warp degree quantification.
[0042] Specifically, in step S310, based on the fiber basis weight and density of the front and back sides in the material's basic parameters, and the production process parameters including drying temperature, drying time, front-back drying temperature difference, and calendering pressure, the equivalent shrinkage coefficients of the front and back sides are calculated using the first formula, which is: S=a1×E1+a2×E2+a3×Y1+a4×Y2+a5×Y3+a6×Y4+b; Where S represents the equivalent shrinkage coefficient, E1 represents the fiber basis weight, E2 represents the density, Y1 represents the drying temperature, Y2 represents the drying time, Y3 represents the temperature difference between the front and back sides, Y4 represents the calendering pressure, a1, a2, a3, a4, a5, and a6 are the weighting coefficients of the corresponding parameters, and b is a constant term. When calculating the equivalent shrinkage coefficient S on the front side... front When calculating the equivalent shrinkage coefficient S on the reverse side, the fiber basis weight and tightness of the front side are substituted into the first formula. back Then, the fiber quantity and density on the reverse side are substituted into the first formula.
[0043] It should be noted that fiber basis weight refers to the weight of the paperboard base paper per unit area, which is the core indicator for measuring the fiber distribution density of the base paper. Density refers to the weight of the paperboard base paper per unit volume, reflecting the compactness of the base paper fibers. Drying time refers to the total time the paperboard remains in the preset drying environment during the production drying process. Front and back drying temperature difference refers to the difference in drying temperature experienced by the front and back sides of the paperboard during the drying process. Calendering pressure refers to the pressure applied to the surface of the paperboard by the calendering equipment during the calendering process.
[0044] Specifically, in step S320, the shrinkage difference is obtained using the second formula based on the equivalent shrinkage coefficients of the front and back sides. The second formula is: ΔS=|S front -S back |; Here, ΔS represents the degree of shrinkage difference. The absolute value of this degree of shrinkage difference reflects the degree of shrinkage imbalance. The larger the value, the higher the risk of warping.
[0045] Specifically, in step S330, based on the elastic modulus of the front and back materials (material basic parameters), and the cardboard thickness and shrinkage difference (dimensional characteristic parameters), the quantified internal stress difference is obtained through a third formula. The third formula is: σ unbal = (E front + E back )×h×ΔS / (2×(E front ×E back )); Where, σ unbal E represents the difference in quantified internal stress. front E represents the elastic modulus of the front material. back The modulus of elasticity represents the material on the reverse side, and h represents the thickness of the cardboard. The modulus of elasticity is a core mechanical parameter characterizing the cardboard material's resistance to elastic deformation, reflecting the magnitude of stress required to cause a unit deformation. Quantifying internal stress differences is a numerical indicator that visually represents the degree of unevenness in internal stress between the front and back sides of the cardboard due to differences in material properties and manufacturing processes.
[0046] Specifically, in step S330, based on the temperature difference and humidity difference of the environmental correlation parameters, the environmental adaptation coefficient is obtained through the third formula. The temperature difference represents the temperature difference between the production and storage environments, and the humidity difference represents the humidity difference between the production and storage environments. The fourth formula is: K env =1+c1×ΔT+c2×ΔRH; Among them, K env ΔT represents the environmental adaptability coefficient, ΔRH represents the temperature difference, c1 represents the temperature difference influence coefficient, and c2 represents the humidity difference influence coefficient.
[0047] Specifically, in step S340, the geometric constraint factor is obtained based on the cardboard length, cardboard width, and cardboard thickness, which are dimensional characteristic parameters. The fifth formula is as follows:
[0048] Among them, K geoLet L represent the geometric constraint factor, W represent the cardboard length, and h represent the cardboard thickness. The geometric constraint factor is a derived characteristic that quantifies the warping risk caused by the cardboard's inherent geometric properties. geo The higher the value, the stronger the warping tendency of the cardboard due to its length, width, and thickness characteristics.
[0049] It is understandable that the redundancy rules in step S230 are suggested through the following steps: Step S410: Obtain the historical initial fusion feature set; Step S420: Calculate the similarity between each parameter in each historical initial fusion feature set and other parameters, retain the parameter with the highest similarity, remove the other parameters with the highest similarity and record them to form a redundancy removal rule.
[0050] In steps S410 to S420, by acquiring historical initial fusion feature sets, the similarity between each parameter in each historical initial fusion feature set and other parameters is calculated. The parameter with the highest similarity is retained, and the remaining parameters with high similarity are removed. This is recorded to form a redundancy removal rule. This rule-based approach relies on real historical feature data, has sufficient actual data support, and ensures the objectivity and rationality of the resulting redundancy removal rule, effectively avoiding parameter selection bias caused by subjectively set removal rules. This method can accurately identify redundant parameters in the historical initial fusion feature set. When performing parameter removal processing on the initial fusion features according to this rule, redundant parts in the feature data can be efficiently and accurately removed. While retaining the core effective information of the feature data, the data dimensions are simplified, avoiding interference from redundant parameters to the input of the subsequent warpage prediction model, and improving the processing efficiency of the subsequent warpage prediction model.
[0051] It is understood that step S140 may include, but is not limited to, the following steps: Step S510: Set the overall warpage threshold, height threshold, and efficiency threshold; wherein, the overall warpage threshold is used to characterize the maximum degree of warpage after cardboard stacking, the height threshold is used to characterize the maximum height after cardboard stacking, and the efficiency threshold is used to characterize the total number of cardboard pieces that the automated stacking equipment must grab at least per unit time.
[0052] Step S520: Use the overall warpage threshold, height threshold, and efficiency threshold as stacking constraints.
[0053] Step S530: Based on the quantified warp value, construct an objective function with the goal of offsetting the warp stress generated by stacking cardboard face up with the warp stress generated by stacking back side up.
[0054] Step S540: Solve the objective function based on the stacking constraints to obtain the number of first stacks with the front side facing up, the number of second stacks with the back side facing up, and the total number of stacks for each batch.
[0055] Step S550: Generate a stacking strategy for the cardboard based on the first stack quantity, the second stack quantity, and the total stack quantity.
[0056] In steps S510 to S550, by setting an overall warp threshold, a height threshold, and an efficiency threshold as stacking constraints, and combining the quantified value of the warp degree of the cardboard to be stacked, an objective function is constructed with the goal of offsetting the warp stress generated by stacking the cardboard face up with the warp stress generated by stacking the cardboard face down. Then, based on the stacking constraints, the objective function is solved and a stacking strategy is generated. This allows the generation process of the stacking strategy to have a clear optimization direction and strict constraints, eliminating the need for subjective judgment based on human experience, and achieving precise matching between the stacking strategy and the actual warp condition of the cardboard. This method can accurately determine the number of sheets in the first stack with the front side facing up, the number of sheets in the second stack with the back side facing up, and the total number of sheets in each batch, while meeting the rigid requirements for the overall warp degree and stacking height of the cardboard after stacking, as well as the operating efficiency requirements of automated stacking equipment. For example, if the number of sheets in the first stack with the front side facing up is 10 sheets and the number of sheets in the second stack with the back side facing up is 8 sheets, the matching rule formed based on this can effectively achieve the cancellation of warp stress between the front and back stacking of the cardboard, ensuring the overall stability of the cardboard after stacking, while taking into account the actual production requirements of stacking operations, making the generated stacking strategy highly practical and reasonable.
[0057] It is understandable that the warpage prediction model in step S130 is trained based on the following steps: Step S610: Collect optimized feature sets and corresponding actual warping measurement values of multiple historical batches of cardboard, construct a labeled training dataset, and divide the training dataset into training set, validation set and test set according to a preset ratio.
[0058] Step S620: Construct an initial warp prediction model; wherein the warp prediction model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer matches the dimension of the optimized feature set, and the number of nodes in the output layer is 1 and corresponds to the quantized value of the warp degree.
[0059] Step S630: Train the initial warp prediction model using the training set, and adjust the parameters of the warp prediction model using the validation set until the prediction error of the warp prediction model on the validation set meets the preset error threshold.
[0060] Step S640: Test the trained warpage prediction model using a test set. If the verification is successful, the preset warpage prediction model is obtained.
[0061] In steps S610 to S640, the model training method relies on real historical cardboard data for modeling and training, providing ample real-world data support. This makes the model training process more standardized and scientific, effectively uncovering the intrinsic correlation between the optimized feature set and the quantified warpage value of the cardboard. The model is trained using a training set, and the parameters of the warpage prediction model are adjusted using a validation set until the prediction error meets a preset error threshold. Finally, the model is validated using a test set. This effectively controls the training effect and prediction accuracy of the warpage prediction model, ensuring that the obtained preset warpage prediction model can accurately output the quantified warpage value of the cardboard to be stacked. This provides accurate and reliable data support for generating suitable cardboard stacking strategies based on the quantified warpage value, ensuring the rationality and relevance of the generated stacking strategy.
[0062] It is understood that step S160 may include, but is not limited to, the following steps: Step S710: Obtain an overall image of the cardboard after it has been fully stacked, and preprocess the overall image of the cardboard.
[0063] Step S720: Extract features from the overall image of the cardboard to obtain overall verticality features, top surface flatness features, and cardboard edge alignment features.
[0064] Step S730: Quantify the overall verticality feature, top surface flatness feature, and cardboard edge alignment feature to obtain the verticality deviation value, the maximum unevenness difference of the top surface, and the maximum misalignment distance of the edge, respectively.
[0065] Step S740: Based on the verticality deviation value, the maximum unevenness difference of the top surface, the maximum misalignment distance of the edge, and the corresponding preset evaluation threshold, obtain the verification evaluation result.
[0066] Step S750: When the verification evaluation result is unsatisfactory, the optimized feature set of this batch and the corresponding warp degree quantification value are added to the training set of the warp prediction model to optimize the parameters of the warp prediction model.
[0067] In steps S710 to S750, the overall image of the cardboard after stacking is preprocessed and features are extracted to obtain overall verticality features, top surface flatness features, and cardboard edge alignment features. These features are then quantified to obtain verticality deviation values, maximum top surface unevenness differences, and maximum edge misalignment distances. Combined with corresponding preset evaluation thresholds, verification evaluation results are obtained. This allows for comprehensive and accurate quantitative verification of the overall effect of cardboard stacking from the core dimensions of stacking quality, replacing subjective human judgment with objective quantitative indicators, effectively ensuring the accuracy and reliability of the stacking effect verification results. When the verification evaluation result is unsatisfactory, the optimized feature set of that batch and the corresponding warpage quantification values are added to the training set of the warpage prediction model, and the model parameters are optimized. This enables iterative optimization of the warpage prediction model, continuously improving the accuracy of the warpage quantification values output by the model. Consequently, the subsequently generated cardboard stacking strategy better reflects the actual warpage condition of the cardboard, continuously optimizing the overall effect of cardboard stacking. This forms a closed-loop optimization technical system for the entire cardboard alternating stacking control method, significantly improving the method's adaptability and practicality.
[0068] It should be noted that the overall verticality feature characterizes the degree to which the stacked cardboard conforms to a preset vertical reference in the vertical direction. The top surface flatness feature characterizes the overall flatness of the top surface of the stacked cardboard, reflecting the unevenness of the top surface. The cardboard edge alignment feature characterizes the alignment of all cardboard edges in the horizontal direction after stacking.
[0069] Secondly, referring to Figure 2 This application also provides a cardboard stacking control device 800 based on warpage prediction, comprising: The acquisition unit 810 is used to acquire a multi-dimensional raw dataset of cardboard to be stacked; the multi-dimensional raw dataset includes basic material parameters, production process parameters, dimensional characteristic parameters and environmental related parameters.
[0070] The processing unit 820 is used to perform progressive data processing on the multi-dimensional original dataset to obtain an optimized feature set.
[0071] The quantization unit 830 is used to input the optimized feature set into the preset warp prediction model to obtain the quantized value of the warp degree of the cardboard to be stacked.
[0072] The generation unit 840 is used to generate a stacking strategy for cardboard based on the warp degree quantification value and preset stacking constraints; wherein, the stacking strategy is a matching rule for the number of cardboard stacks with the front side facing up and the number of cardboard stacks with the back side facing up.
[0073] The sending unit 850 is used to convert the stacking strategy into control instructions for the automated stacking equipment and send the control instructions to enable the automated stacking equipment to complete the cardboard stacking.
[0074] The optimization unit 860 is used to verify the overall effect after stacking is completed and to optimize the warpage prediction model based on the verification and evaluation results.
[0075] The specific implementation of the cardboard stacking control device 800 based on warp prediction is basically the same as the specific implementation of the cardboard stacking control method based on warp prediction described above, and will not be repeated here.
[0076] Thirdly, this application also provides an electronic device, comprising: at least one memory; at least one processor; at least one program; the program being stored in the memory, and the processor executing the at least one program to implement the cardboard stacking control method based on warp prediction as described in any embodiment of the first aspect.
[0077] In this electronic device, a multi-dimensional raw dataset of cardboard to be stacked is first collected. This dataset includes basic material parameters, production process parameters, dimensional characteristic parameters, and environmental related parameters. Then, the multi-dimensional raw dataset is progressively processed to obtain an optimized feature set. After inputting the optimized feature set into a preset warp prediction model, a quantitative value of the warp degree of the cardboard to be stacked can be obtained. Then, based on the quantitative value of the warp degree and preset stacking constraints, a stacking strategy for the cardboard is generated. This stacking strategy is a matching rule between the number of cardboard stacked face up and the number of cardboard stacked face down. Subsequently, the stacking strategy is converted into control commands for the automated stacking equipment and sent to complete the cardboard stacking operation. After the stacking is completed, the overall effect is verified, and the warp prediction model is optimized based on the verification and evaluation results. By collecting and processing multi-dimensional data on cardboard, the degree of cardboard warping can be accurately quantified. Based on the quantification results, an appropriate stacking strategy can be formulated to guide automated stacking operations. This effectively solves the problem that in existing technologies, manual stacking of cardboard based on experience and inaccurate judgment methods leads to poor overall stability of the cardboard after stacking. It can control the matching method of alternating stacking based on the actual warping of the cardboard, ensuring the regularity and stability of the cardboard stack. At the same time, the warping prediction model can be continuously optimized based on the stacking effect verification, further improving the adaptability of the cardboard stacking strategy and the reliability of the stacking operation.
[0078] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and signals, such as the program instructions / signals corresponding to the processing module in the embodiments of this application. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and signals stored in the memory, thereby implementing the cardboard stacking control method based on warpage prediction in the above-described method embodiments.
[0079] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store relevant data for the aforementioned cardboard stacking control method based on warp prediction. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processing module via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0080] One or more signals are stored in a memory, and when executed by one or more processors, the cardboard stacking control method based on warpage prediction in any of the above method embodiments is executed.
[0081] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that is executed by one or more processors, causing the one or more processors to perform the cardboard stacking control method based on warpage prediction in the above method embodiments.
[0082] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0083] Based on the above description of the embodiments, those skilled in the art will understand that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable signals, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible by a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable signals, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.
[0084] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0085] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0086] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0087] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0088] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0089] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application.
Claims
1. A method for controlling a paperboard pile based on warp estimation, characterized by, include: Collect a multi-dimensional raw dataset of cardboard to be stacked; wherein, the multi-dimensional raw dataset includes basic material parameters, production process parameters, dimensional characteristic parameters and environmental related parameters; The multi-dimensional original dataset is subjected to progressive data processing to obtain an optimized feature set; The optimized feature set is input into a preset warp prediction model to obtain a quantitative value of the warp degree of the cardboard to be stacked. Based on the quantified warpage value and preset stacking constraints, a stacking strategy for the cardboard is generated; wherein, the stacking strategy is a matching rule for the number of cardboard stacks with the front side facing up and the number of cardboard stacks with the back side facing up. The stacking strategy is converted into control instructions for automated stacking equipment, and the control instructions are sent to enable the automated stacking equipment to complete the cardboard stacking. The overall effect after stacking is verified, and the warpage prediction model is optimized based on the verification and evaluation results; The step of generating a cardboard stacking strategy based on the quantified warpage value and preset stacking constraints includes: Set an overall warpage threshold, a height threshold, and an efficiency threshold; wherein, the overall warpage threshold is used to characterize the maximum degree of warpage after cardboard stacking, the height threshold is used to characterize the maximum height after cardboard stacking, and the efficiency threshold is used to characterize the total number of cardboard pieces that the automated stacking equipment must grab at least per unit time; The overall warpage threshold, the height threshold, and the efficiency threshold are used as stacking constraints. Based on the quantified value of the degree of warping, an objective function is constructed with the goal of optimizing the mutual cancellation of the warping stress generated by stacking the cardboard face up and the warping stress generated by stacking the cardboard face down. Based on the stacking constraints, the objective function is solved to obtain the first stack quantity with the front side of the cardboard facing up, the second stack quantity with the back side facing up, and the total stack quantity for each batch. A stacking strategy for cardboard is generated based on the first stack quantity, the second stack quantity, and the total stack quantity.
2. The method of claim 1, wherein, The progressive data processing of the multi-dimensional original dataset to obtain an optimized feature set includes: Based on the multi-dimensional original dataset, feature derivation is performed to obtain an intermediate derived feature set; An initial fused feature set is obtained by combining the multi-dimensional original dataset and the intermediate derived feature set. According to the preset redundancy removal rules, the initial fused features are partially removed to obtain an optimized feature set.
3. The method of claim 2, wherein the method further comprises: The step of performing feature derivation based on the multi-dimensional original dataset to obtain an intermediate derived feature set includes: Based on the material basic parameters and the production process parameters, the front equivalent shrinkage coefficient and the back equivalent shrinkage coefficient are obtained; The shrinkage difference is obtained based on the equivalent shrinkage coefficient of the front side and the equivalent shrinkage coefficient of the back side; The quantitative internal stress difference is obtained based on the material basic parameters, the dimensional characteristic parameters, and the shrinkage difference. Based on the environmental correlation parameters, the environmental adaptability coefficient is obtained; Based on the aforementioned dimensional characteristic parameters, the geometric constraint factor is obtained; By integrating the positive equivalent shrinkage coefficient, the negative equivalent shrinkage coefficient, the shrinkage difference, the quantified internal stress difference, the environmental adaptation coefficient, and the geometric constraint factor, an intermediate derived feature set is obtained.
4. The method of claim 2, wherein the method further comprises: The redundancy removal rules are established through the following steps: Obtain the initial historical fusion feature set; Calculate the similarity between each parameter in each of the historical initial fusion feature sets and other parameters, retain the parameter with the highest similarity, and remove and record the other parameters with the highest similarity to form the redundancy removal rule.
5. The method of claim 1, wherein, The warpage prediction model is trained through the following steps: Collect optimized feature sets and corresponding actual warping measurements from multiple historical batches of cardboard, construct a labeled training dataset, and divide the training dataset into a training set, a validation set, and a test set according to a preset ratio; Construct an initial warp prediction model; wherein the warp prediction model includes an input layer, a hidden layer and an output layer, the number of nodes in the input layer matches the dimension of the optimized feature set, and the number of nodes in the output layer is 1 and corresponds to the quantized value of the warp degree; The initial warpage prediction model is trained using the training set, and the parameters of the warpage prediction model are adjusted using the validation set until the prediction error of the warpage prediction model on the validation set meets a preset error threshold. The trained warpage prediction model is tested using the test set. If the verification is successful, the preset warpage prediction model is obtained.
6. The method of claim 1, wherein, The process of verifying the overall effect after stacking is completed, and optimizing the warpage prediction model and the stacking strategy based on the verification and evaluation results, includes: Obtain an overall image of the cardboard after it has been fully stacked, and preprocess the overall image of the cardboard; Feature extraction is performed on the overall image of the cardboard to obtain overall verticality features, top surface flatness features, and cardboard edge alignment features; The overall verticality feature, the top surface flatness feature, and the cardboard edge alignment feature are quantified to obtain the verticality deviation value, the maximum unevenness difference of the top surface, and the maximum misalignment distance of the edge, respectively. The verification evaluation result is obtained based on the verticality deviation value, the maximum unevenness difference of the top surface, the maximum misalignment distance of the edge, and the corresponding preset evaluation threshold. If the verification evaluation result is unsatisfactory, the optimized feature set of that batch and the corresponding warpage quantification value are added to the training set of the warpage prediction model to optimize the parameters of the warpage prediction model.
7. A warp estimation based paperboard pile control apparatus, characterized by include: The acquisition unit is used to acquire a multi-dimensional raw dataset of cardboard to be stacked; wherein the multi-dimensional raw dataset includes basic material parameters, production process parameters, dimensional characteristic parameters and environmental related parameters. The processing unit is used to perform progressive data processing on the multi-dimensional original dataset to obtain an optimized feature set; The quantization unit is used to input the optimized feature set into a preset warp prediction model to obtain a quantified value of the warp degree of the cardboard to be stacked. The generation unit is used to generate a stacking strategy for the cardboard based on the quantified warpage value and preset stacking constraints; wherein, the stacking strategy is a matching rule for the number of cardboard stacks with the front side facing up and the number of cardboard stacks with the back side facing up. A sending unit is used to convert the stacking strategy into control instructions for automated stacking equipment, and send the control instructions to enable the automated stacking equipment to complete the cardboard stacking. An optimization unit is used to verify the overall effect after stacking is completed, and to optimize the warpage prediction model based on the verification and evaluation results. The step of generating a cardboard stacking strategy based on the quantified warpage value and preset stacking constraints includes: Set an overall warpage threshold, a height threshold, and an efficiency threshold; wherein, the overall warpage threshold is used to characterize the maximum degree of warpage after cardboard stacking, the height threshold is used to characterize the maximum height after cardboard stacking, and the efficiency threshold is used to characterize the total number of cardboard pieces that the automated stacking equipment must grab at least per unit time; The overall warpage threshold, the height threshold, and the efficiency threshold are used as stacking constraints. Based on the quantified value of the degree of warping, an objective function is constructed with the goal of optimizing the mutual cancellation of the warping stress generated by stacking the cardboard face up and the warping stress generated by stacking the cardboard face down. Based on the stacking constraints, the objective function is solved to obtain the first stack quantity with the front side of the cardboard facing up, the second stack quantity with the back side facing up, and the total stack quantity for each batch. A stacking strategy for cardboard is generated based on the first stack quantity, the second stack quantity, and the total stack quantity.
8. An electronic device, comprising: include: At least one memory; At least one processor; At least one program; The program is stored in the memory, and the processor executes at least one of the programs to implement the cardboard stacking control method based on warpage prediction as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program for performing the cardboard stacking control method based on warpage prediction as described in any one of claims 1 to 6.