An intelligent management method and system for assembly production of shading sheet products

Abstract

The present invention discloses an intelligent management method and system for the assembly production of a shading sheet product, comprising clustering historical assembly data and current assembly demand data of the shading sheet to obtain a first assembly parameter for the current assembly production; constructing a test task to obtain multiple groups of position data and angle data of the shading sheet to calculate the test assembly error; if the preset conditions are not met, establishing a state transformation matrix, calculating the rotation and translation combination error to construct an error correction model, solving the error correction model by the least squares method to obtain the second assembly parameter, and correcting the first assembly parameter by using the second assembly parameter; returning to execute the above steps until the test assembly error meets the preset conditions, generating a target assembly parameter, and executing the shading sheet assembly production process by using the target assembly parameter. The present invention takes into account the coupling relationship between translation and rotation errors, establishes a correction model by the rotation and translation combination error to iteratively correct the assembly parameters, and improves the efficiency and accuracy of error correction.

Description

Technical Field

[0001] The present invention relates to the technical field of shading sheet assembly, and in particular to an intelligent management method and system for assembly production of shading sheet products. Background Art

[0002] The assembly of the shading sheet involves multiple steps, including the positioning, fixing, rotation and translation of the shading sheet. The accuracy of these steps directly affects the quality and performance of the final product. There are still some shortcomings in the traditional assembly method: during the assembly process, it usually relies on manual experience or simple automated equipment to control the assembly, so it is difficult to achieve precise control of the position and angle of the shading sheet, and assembly errors are prone to occur. Once the assembly error is found, it is usually necessary to first determine the size of the translation error and the angle error respectively, and then correct the two errors separately according to their respective ideal ranges, and correct them as many times as they occur. Since the root cause of the error is ignored, not only can the overall assembly error not be eliminated, but the correction efficiency is low and the correction accuracy is difficult to guarantee, which affects the overall energy efficiency of the shading sheet assembly production. Summary of the invention

[0003] In order to solve at least one of the technical problems mentioned above, the present invention provides an intelligent management method and system for assembly production of shading sheet products.

[0004] In a first aspect, the present invention provides an intelligent management method for assembly production of a shading sheet product, the method comprising:

[0005] Step S10, obtaining historical assembly data and current assembly requirement data of the light shielding sheet; performing cluster analysis on the current assembly requirement data and the historical assembly data to obtain the first assembly parameter of the current assembly production;

[0006] Step S20, setting a number of test assembly tasks according to the first assembly parameter, collecting image data and sensor data of each test assembly task in real time, extracting features from the image data and fusing them with the sensor data, and obtaining position data and angle data of multiple groups of shading sheets under different test assembly tasks;

[0007] Step S30, according to the multiple sets of position data and angle data, respectively calculating the average translation error and the average angle error as the test assembly error; when the test assembly error does not meet the preset conditions, establishing a state transformation matrix according to the shading sheet position data and angle data, and calculating the rotation and translation combined error according to the state transformation matrix;

[0008] Step S40, constructing an error correction model according to the combined rotation and translation errors, solving the error correction model by the least square method to obtain second assembly parameters, and using the second assembly parameters to correct the first assembly parameters;

[0009] Step S50, returning to execute steps S20-S40, until the test assembly error meets the preset conditions, generating target assembly parameters, and executing the light shielding sheet assembly production process using the target assembly parameters.

[0010] Preferably, the step of establishing a state transformation matrix according to the position data and angle data of the light shielding sheet, and calculating the rotation and translation combined error according to the state transformation matrix includes:

[0011] Get the actual position coordinates (x, y) and ideal position coordinates (x0, y0) of the shading plate, the actual angle θ and the ideal angle θ0, and establish the actual transformation matrix H and the ideal transformation matrix H0:

[0012]

[0013] Calculate the state transformation matrix T from the actual transformation matrix H to the ideal transformation matrix H0:

[0014]

[0015] Determine whether the state transformation matrix is ​​the identity matrix. When it is determined that the state transformation matrix is ​​not the identity matrix, use the state transformation matrix to determine the rotation and translation combined error:

[0016]

[0017] In the formula, H -1 represents the inverse matrix of H, ε represents the combined rotation and translation error, and Δθ satisfies Δθ=θ-θ0.

[0018] Preferably, constructing an error correction model according to the combined rotation and translation error, and solving the error correction model by the least square method to obtain the second assembly parameter comprises:

[0019] Construct the error correction model:

[0020] ε=w1v+w2a+∈,

[0021] Where w1 and w2 represent weight coefficients, v and a represent the robot arm speed and the rotary motor acceleration, and ∈ is a random noise term;

[0022] Obtain N groups of test assembly parameters and the corresponding rotation and translation combined errors, and establish the objective function:

[0023]

[0024] In the formula, min{·} means minimization, ε i represents the i-th rotation and translation combined error, v i 、a i represents the robot arm speed and rotary motor acceleration in the i-th group of assembly parameters;

[0025] The objective function is solved by the least square method to obtain the optimal solution of w1 and w2. The first assembly parameter and the optimal solution of w1 and w2 are substituted into the error correction model to obtain the target error.

[0026] Based on target error Determine the second assembly parameters:

[0027]

[0028]

[0029] Where v′ and a′ are the robot arm speed and the rotation motor acceleration in the second assembly parameter, k v , k a Respectively represent proportional gains, satisfying 0.1<k v , k a <1.

[0030] Preferably, the cluster analysis of the current assembly demand data and the historical assembly data includes:

[0031] Based on historical assembly data, assembly results under different assembly parameters are obtained, and the current assembly requirement data and assembly results are clustered using the K-means algorithm; the assembly results include translation error and angle error;

[0032] The optimal cluster is identified from the clustering results, and the corresponding assembly data is determined as the first assembly parameter of the current assembly production.

[0033] Preferably, the method further comprises:

[0034] Acquire the operation cycle of each assembly task in real time, identify the assembly tasks whose operation cycle is greater than a first preset value, and determine the queuing time from the current assembly task to the subsequent assembly task;

[0035] Determine whether the queuing time is greater than a second preset value, if so, mark the subsequent assembly task as an abnormal task; if not, mark the current assembly task as an abnormal task;

[0036] Trace the abnormal task back to the previous process task until the associated abnormal interval is identified, and match the operation instructions to correct the assembly operation in the associated abnormal interval.

[0037] In a second aspect, the present invention further provides an intelligent management system for assembly and production of shading sheet products, the system comprising:

[0038] The cluster analysis unit is used to execute step S10 to obtain historical assembly data and current assembly requirement data of the light shielding sheet; perform cluster analysis on the current assembly requirement data and the historical assembly data to obtain the first assembly parameter of the current assembly production;

[0039] A data acquisition unit, configured to execute step S20, set a plurality of test assembly tasks according to the first assembly parameter, collect image data and sensor data of each test assembly task in real time, extract features from the image data and fuse them with the sensor data, and obtain position data and angle data of multiple groups of light shielding sheets under different test assembly tasks;

[0040] The error calculation unit is used to execute step S30, and respectively calculate the average translation error and the average angle error according to the multiple sets of position data and angle data as the test assembly error; when the test assembly error does not meet the preset conditions, establish a state transformation matrix according to the shading sheet position data and angle data, and calculate the rotation and translation combined error according to the state transformation matrix;

[0041] A parameter correction unit, configured to execute step S40, construct an error correction model according to the rotation and translation combined error, solve the error correction model by the least square method to obtain a second assembly parameter, and use the second assembly parameter to correct the first assembly parameter;

[0042] The iterative correction unit is used to execute step S50, return to execute steps S20-S40, until the test assembly error meets the preset conditions, generate target assembly parameters, and use the target assembly parameters to execute the light shielding sheet assembly production process.

[0043] Preferably, the system further comprises:

[0044] A queuing duration determining unit, configured to execute step S60, obtain the operation cycle of each assembly task in real time, identify the assembly tasks whose operation cycle is greater than a first preset value, and determine the queuing duration from the current assembly task to the subsequent assembly task;

[0045] The abnormal task marking unit is used to execute step S70 to determine whether the queuing time is greater than a second preset value. If so, the subsequent assembly task is marked as an abnormal task; if not, the current assembly task is marked as an abnormal task;

[0046] The abnormal task tracing unit is used to execute step S80, tracing from the abnormal task to the preceding process task until the associated abnormal interval is identified, and matching the operation instructions to correct the assembly operation of the associated abnormal interval.

[0047] In a third aspect, the present invention further provides an electronic device comprising a processor and a memory, wherein the memory is used to store computer program code, and the computer program code comprises computer instructions. When the processor executes the computer instructions, the electronic device executes the method as described in the first aspect above and any possible implementation thereof.

[0048] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program includes program instructions, and when the program instructions are executed by a processor of an electronic device, the processor executes the method as described in the first aspect above and any possible implementation thereof.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1) When determining assembly parameters, the present invention first obtains historical assembly data and current assembly demand data, and then performs cluster analysis on these data, so as to accurately predict the first assembly parameters required for the current assembly production, thereby improving the accuracy and rationality of the assembly plan.

[0051] 2) The present invention takes into account the connection between the error and the assembly parameters, that is, analyzes the root cause of the error. By first setting up a set of test tasks, the coupling effect between translation and rotation is comprehensively considered to analyze the combined error of rotation and translation, so that the assembly parameters are iteratively corrected during the test process, and the probability of assembly errors is fundamentally reduced from the perspective of assembly parameters. Compared with the existing method of respectively correcting the translation error and the rotation error with variable assembly edges, the present invention can greatly improve efficiency, ensure the accuracy of error correction, and improve the overall assembly production efficiency of the shading sheet.

[0052] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background technology, the drawings required for use in the embodiments of the present invention or the background technology will be described below.

[0054] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0055] Figure 1 A schematic diagram of a process flow of an intelligent management method for assembly production of a shading sheet product provided by an embodiment of the present invention;

[0056] Figure 2A schematic diagram of a process flow of an intelligent management method for assembly production of a shading sheet product provided by another embodiment of the present invention;

[0057] Figure 3 A schematic diagram of the structure of an intelligent management system for assembly and production of a shading sheet product provided by an embodiment of the present invention;

[0058] Figure 4 A schematic structural diagram of an intelligent management system for assembly and production of shading sheet products provided in another embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0061] In addition, in order to better illustrate the present invention, numerous specific details are provided in the following specific embodiments. It should be understood by those skilled in the art that the present invention can be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present invention.

[0062] See also Figure 1 , Figure 1 The present invention provides a flowchart of an intelligent management method for assembly production of a shading sheet product. Figure 1 As shown, the method includes:

[0063] Step S10, obtaining historical assembly data and current assembly requirement data of the shading sheet; performing cluster analysis on the current assembly requirement data and historical assembly data to obtain first assembly parameters of the current assembly production.

[0064] The historical assembly data of the shading sheet usually includes the assembly data and assembly results in the historical assembly process. The assembly production lines for shading sheets are mostly based on automated equipment, so the assembly data involves assembly parameters, such as the assembly speed of the robot arm, the acceleration of the rotating motor, etc., as well as assembly quantity, sequence and other data. The assembly result refers to the assembly quality of each assembly line, which is usually evaluated by indicators such as translation error and angle error. In the historical assembly process, a corresponding database is usually established for the data after assembly production to store various successful cases. Before the execution of this assembly task, in order to quickly and accurately determine the assembly parameters instead of relying on manual experience, the demand data related to this assembly, such as the error requirements in the assembly results, the assembly quantity and sequence, etc., will be clustered with the historical assembly data, and then a close case will be found based on the clustering results. The first assembly parameter of the current assembly production is determined with reference to the case, thereby improving the efficiency and rationality of parameter configuration.

[0065] In one embodiment, clustering the current assembly demand data and the historical assembly data includes:

[0066] Based on historical assembly data, assembly results under different assembly parameters are obtained, and the current assembly requirement data and assembly results are clustered using the K-means algorithm; the assembly results include translation error and angle error;

[0067] The optimal cluster is identified from the clustering results, and the corresponding assembly data is determined as the first assembly parameter of the current assembly production.

[0068] In terms of clustering algorithm, this embodiment mainly adopts the K-means algorithm, which is equivalent to other clustering algorithms. It only needs to set the number K of clusters to be formed, and the iterative process is relatively intuitive, which can ensure that the implementation process is relatively simple and the calculation efficiency is high.

[0069] This embodiment first selects the assembly parameters with the smallest error through cluster analysis, which can effectively improve the accuracy of the light shielding sheet assembly and reduce the production of defective products. Using the verified optimal assembly parameters can avoid the increase in debugging time and cost caused by incorrect assembly parameters, thereby accelerating the assembly process. Cluster analysis based on historical data can better adapt to different assembly requirements, and can quickly find suitable assembly parameters even when the requirements change, thereby improving the accuracy and rationality of the assembly plan.

[0070] Step S20, setting a number of test assembly tasks according to the first assembly parameters, collecting image data and sensor data of each test assembly task in real time, extracting features from the image data and fusing them with the sensor data, and obtaining position data and angle data of multiple sets of shading sheets under different test assembly tasks.

[0071] It is understandable that errors are almost inevitable during the assembly process because the assembly process is affected by many factors, including but not limited to material properties, mechanical wear, operating skills, environmental conditions, etc., so it is difficult to achieve precise control of the position and angle of the shading sheet.

[0072] For the error problem, two solutions are usually adopted in the existing assembly production process. One is to first determine the size of the translation error and the angle error respectively, and then correct the two errors separately according to their respective ideal ranges, and correct them as many times as they appear. The other is to set a reasonable error range. Once an abnormal task with an error exceeding the reasonable range is found, it will be directly eliminated, and the qualified assembly tasks will be retained. Although the correction process of the first method is intuitive, the correction process is relatively complicated. If a large number of error tasks need to be corrected on the production line, it will seriously affect the production assembly energy efficiency. Although the latter can alleviate the complexity of the correction process, the yield is relatively low, which will also affect the overall assembly energy efficiency. Since the above two methods focus on the intervention process after the error occurs, and ignore the root cause of the error, the correction operation process is cumbersome and inefficient. Secondly, these two methods usually consider translation and angle errors separately. For example, if the translation error is corrected first and then the angle error is corrected, the translation error may appear again after the angle error is corrected. Therefore, the separate corrections usually cannot eliminate the overall assembly error. This method ignores the coupling between the translation and rotation angles, and the correction accuracy is difficult to guarantee, which will eventually affect the overall energy efficiency of the assembly production of the light shielding sheet. Therefore, this embodiment considers first giving some test assembly task sets. In this test process, the assembly task execution is first controlled by the first assembly parameter, and then the errors are detected in sequence, and the rotation and translation combined errors are considered. The combined error is used to return to iteratively correct the first assembly parameter until the final assembly error can meet the requirements. The final target assembly parameter can be obtained, and the target assembly parameter can reduce the probability of error occurrence from the fundamental problem, thereby reducing the number and frequency of corrections and ensuring the production assembly accuracy. At the same time, the production method of first selecting part of the test set and then carrying out actual production and processing can avoid too many invalid assemblies, reduce the rework rate, improve the yield rate, and improve production energy efficiency, compared with directly investing in production and processing and then finding problems and remedying them.

[0073] Specifically, in step S20, a limited number of test assembly tasks are first designed. These tasks should cover various possible assembly situations so as to comprehensively evaluate the effectiveness of assembly parameters. When collecting data, the image data in each test assembly task is collected in real time using a machine vision system. These image data can be used to observe the position, posture and other information of the light shielding plate during the assembly process. At the same time, various sensors installed on the assembly line, such as displacement sensors and angle sensors, collect relevant sensor data in real time. The sensor data can provide accurate values ​​about the actual position and angle of the light shielding plate. The collected image data is preprocessed, such as denoising, contrast enhancement, etc., to improve the image quality. Then, algorithms such as edge detection and contour extraction can be used to extract key features of the light shielding plate from the image, such as edges, center points, etc. Finally, the features extracted from the image data are fused with the sensor data, and the visual information provided by the image is used to supplement the possible deficiencies of the sensor data, such as occlusion or signal interference, so as to obtain more accurate position and angle information. Preferably, a Kalman filter or a particle filter can be used to fuse these two types of data to improve the consistency and reliability of the data.

[0074] Therefore, the fusion of image data and sensor data provides more accurate position and angle information, which helps to accurately control the assembly position of the shading sheet and reduce assembly errors. This can improve the robustness of the assembly process. Even when some sensor data is unreliable, the assembly position can still be corrected through image data to ensure assembly quality, thereby providing a basis for subsequent error correction, helping to optimize assembly parameters and achieve data-driven decision-making.

[0075] Step S30, according to multiple sets of position data and angle data, respectively calculate the average translation error and the average angle error as the test assembly error; when the test assembly error does not meet the preset conditions, establish a state transformation matrix according to the shading plate position data and angle data, and calculate the rotation and translation combined error according to the state transformation matrix.

[0076] Assuming that N groups of test tasks are set, then N groups of position data and angle data will be collected accordingly, and the average translation error can be calculated separately and the average angle error Specifically:

[0077]

[0078] In the formula, Δd i , Δθ i are the translation error and angle error of the ith test assembly task, respectively. (x0, y0) and θ0 are the ideal position coordinates and angles of the shading sheet. (x i ,y i ),θ iThe actual position coordinates and rotation angle of the shading sheet in the i-th test assembly task.

[0079] Get the average translation error and the average angle error After that, they are compared with their respective preset thresholds. As long as one of the two errors exceeds the corresponding preset threshold, it is considered that the preset condition is not met, and the assembly parameters need to be corrected.

[0080] In one embodiment, the step of establishing a state transformation matrix according to the position data and angle data of the light shielding sheet and calculating the rotation and translation combined error according to the state transformation matrix includes:

[0081] Get the actual position coordinates (x, y) and ideal position coordinates (x0, y0) of the shading plate, the actual angle θ and the ideal angle θ0, and establish the actual transformation matrix H and the ideal transformation matrix H0:

[0082]

[0083] Calculate the state transformation matrix T from the actual transformation matrix H to the ideal transformation matrix H0:

[0084]

[0085] Determine whether the state transformation matrix is ​​the identity matrix. When it is determined that the state transformation matrix is ​​not the identity matrix, use the state transformation matrix to determine the rotation and translation combined error:

[0086]

[0087] In the formula, H -1 represents the inverse matrix of H, ε represents the combined rotation and translation error, and Δθ satisfies Δθ=θ-θ0.

[0088] In this embodiment, the inverse matrix H of the actual transformation matrix H is -1 Multiplying with the ideal transformation matrix H0, we get the state transformation matrix T, which is equivalent to the transformation matrix required to convert from the actual position and angle to the ideal position and angle. If the state transformation matrix is ​​a unit matrix, it means that the actual position and the ideal position are exactly the same and there is no error. If the state transformation matrix is ​​not a unit matrix, it means that there is an error, and the size of the combined rotation and translation error ε needs to be determined. In this way, the difference between the actual assembly position and the ideal position can be accurately quantified, which is convenient for subsequent error correction. By considering the coupling relationship between translation and rotation angle errors, through continuous monitoring and iterative optimization, the accuracy of the shading sheet assembly can be significantly improved and the generation of defective products can be reduced.

[0089] Step S40, constructing an error correction model according to the combined rotation and translation errors, solving the error correction model by the least square method to obtain second assembly parameters, and using the second assembly parameters to correct the first assembly parameters;

[0090] Specifically, the step of constructing an error correction model according to the combined rotation and translation error and solving the error correction model by the least square method to obtain the second assembly parameter includes:

[0091] Construct the error correction model:

[0092] ε=w1v+w2a+∈,

[0093] Where w1 and w2 represent weight coefficients, v and a represent the robot arm speed and the rotary motor acceleration, and ∈ is a random noise term;

[0094] Obtain N groups of test assembly parameters and the corresponding rotation and translation combined errors, and establish the objective function:

[0095]

[0096] In the formula, min{·} means minimization, ε i represents the i-th rotation and translation combined error, v i 、a i represents the robot arm speed and rotary motor acceleration in the i-th group of assembly parameters;

[0097] The objective function is solved by the least square method to obtain the optimal solution of w1 and w2. The first assembly parameter and the optimal solution of w1 and w2 are substituted into the error correction model to obtain the target error.

[0098] Based on target error Determine the second assembly parameters:

[0099]

[0100] Where v′ and a′ are the robot arm speed and the rotation motor acceleration in the second assembly parameter, k v , k a Respectively represent proportional gains, satisfying 0.1<k v , k a <1.

[0101] In this embodiment, the error correction model ε=w1v+w2a+∈ can associate the actual measured error with the speed v of the robot arm and the acceleration a of the rotating motor, and can accurately quantify the influence of the robot arm motion parameters on the assembly error, avoiding the uncertainty caused by blindly adjusting the parameters. iMinimization processing is performed and the optimal solution is obtained by the least square method to minimize the assembly error, thereby improving the assembly accuracy. v , k a The reasonable setting of makes the adjustment range moderate, which can quickly respond to error changes and avoid instability caused by over-adjustment. Therefore, by dynamically adjusting the assembly parameters and accurately controlling the speed of the robot arm and the acceleration of the rotary motor, the errors in the assembly process can be compensated in real time and the stability of the assembly can be improved. This adaptive adjustment mechanism enhances the robustness of the system and makes the assembly process more reliable.

[0102] Step S50, returning to execute steps S20-S40, until the test assembly error meets the preset conditions, generating target assembly parameters, and executing the light shielding sheet assembly production process using the target assembly parameters.

[0103] Therefore, this embodiment takes into account the connection between the error and the assembly parameters, that is, analyzes the root cause of the error. By first setting up a set of test tasks, the coupling effect between translation and rotation is comprehensively considered to analyze the combined error of rotation and translation, so as to iteratively correct the assembly parameters during the test process, and fundamentally reduce the probability of assembly errors from the perspective of assembly parameters. Compared with the existing method of correcting translation errors and rotation errors separately with variable assembly edges, the present invention can greatly improve efficiency, ensure the accuracy of error correction, and improve the overall assembly production efficiency of the shading sheet.

[0104] See also Figure 2 In one embodiment, the method further comprises:

[0105] S60: Acquire the operation cycle of each assembly task in real time, identify assembly tasks with an operation cycle greater than a first preset value, and determine the queuing time from the current assembly task to the subsequent assembly task.

[0106] Acquire the operation cycle data of each assembly task in real time. These data can be collected through various sensors and control systems deployed on the assembly line. The operation cycle refers to the time taken from the start of assembly to the completion of assembly. Set a first preset value as the upper limit of the normal operation cycle. When the operation cycle of an assembly task exceeds the first preset value, it is considered that the task may be abnormal. Record all assembly tasks whose operation cycles exceed the first preset value and their related information such as task number, start and end time, etc. Then, calculate the queue waiting time from the current assembly task to the subsequent assembly task. The queue length refers to the waiting time before the next assembly task starts after the current assembly task is completed.

[0107] S70, determining whether the queuing time is greater than a second preset value, if so, marking the subsequent assembly task as an abnormal task; if not, marking the current assembly task as an abnormal task.

[0108] The second preset value is a threshold of the queue time. Exceeding this threshold indicates that there may be a congestion or other abnormal conditions. When the queue time is greater than the second preset value, the subsequent assembly task is marked as an abnormal task; if the queue time is not greater than the second preset value, the current assembly task is marked as an abnormal task.

[0109] S80, tracing back from the abnormal task to the preceding process task until the associated abnormal interval is identified, and matching the operation instructions to correct the assembly operation of the associated abnormal interval.

[0110] Starting from the task marked as abnormal, trace back to its previous process tasks to find the root cause of the abnormality. The previous tasks that may be involved include raw material supply, parts processing, semi-finished product inspection and other links. Through traceability analysis, a series of related abnormal intervals are identified. The related abnormal intervals may be caused by a failure in a specific link, or the result of the combined action of multiple links.

[0111] Furthermore, according to the identified abnormal interval, corresponding operation instructions or solutions are found. These operation instructions can be pre-established emergency measures or specific suggestions generated by expert systems or artificial intelligence. According to the operation instructions, the assembly operation is adjusted to correct the problem of the abnormal interval.

[0112] Therefore, in addition to parameter testing and error correction before assembly production, the present invention will also monitor the operation cycle of the assembly task in real time after the actual production assembly process is put into use, so as to timely discover abnormal situations, reduce production stagnation or delays caused by abnormalities, and improve overall production efficiency. By judging whether the queuing time exceeds the preset value, the task scheduling on the assembly line can be effectively managed to reduce the waste of resources caused by long queuing time. Through tracing analysis and correcting assembly operations, abnormal tasks can be responded to in a timely manner, thereby reducing rework caused by assembly errors, saving costs, enhancing the stability and reliability of the assembly line operation, and improving production energy efficiency.

[0113] See also Figure 3 In one embodiment, the present invention further provides an intelligent management system for assembly and production of shading sheet products, the system comprising:

[0114] The cluster analysis unit 100 is used to execute step S10 to obtain historical assembly data and current assembly requirement data of the light shielding sheet; perform cluster analysis on the current assembly requirement data and the historical assembly data to obtain the first assembly parameter of the current assembly production;

[0115] The data acquisition unit 200 is used to execute step S20, set a plurality of test assembly tasks according to the first assembly parameter, collect image data and sensor data of each test assembly task in real time, extract features from the image data and fuse them with the sensor data, and obtain position data and angle data of multiple groups of light shielding sheets under different test assembly tasks;

[0116] The error calculation unit 300 is used to execute step S30, and respectively calculate the average translation error and the average angle error according to the multiple sets of position data and angle data as the test assembly error; when the test assembly error does not meet the preset conditions, a state transformation matrix is ​​established according to the shading sheet position data and angle data, and the rotation and translation combined error is calculated according to the state transformation matrix;

[0117] The parameter correction unit 400 is used to execute step S40, construct an error correction model according to the rotation and translation combined error, solve the error correction model by the least square method to obtain the second assembly parameter, and use the second assembly parameter to correct the first assembly parameter;

[0118] The iterative correction unit 500 is used to execute step S50, return to execute steps S20-S40, until the test assembly error meets the preset conditions, generate target assembly parameters, and use the target assembly parameters to execute the light shielding sheet assembly production process.

[0119] See also Figure 4 In one embodiment, the system further comprises:

[0120] The queuing duration determining unit 600 is used to execute step S60, obtain the operation cycle of each assembly task in real time, identify the assembly tasks whose operation cycle is greater than the first preset value, and determine the queuing duration from the current assembly task to the subsequent assembly task;

[0121] The abnormal task marking unit 700 is used to execute step S70 to determine whether the queuing time is greater than a second preset value. If so, the subsequent assembly task is marked as an abnormal task; if not, the current assembly task is marked as an abnormal task;

[0122] The abnormal task tracing unit 800 is used to execute step S80, tracing from the abnormal task to the preceding process task until the associated abnormal interval is identified, and matching the operation instructions to correct the assembly operation of the associated abnormal interval.

[0123] It can be understood that the functions or modules included in the system provided in this embodiment can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

[0124] The present invention also provides an electronic device, including a processor and a memory, wherein the memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes a method as described in any possible implementation manner.

[0125] The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a method as described in any possible implementation manner.

[0126] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0127] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. Those skilled in the art can also clearly understand that the descriptions of the various embodiments of the present invention have different focuses. For the convenience and brevity of description, the same or similar parts may not be repeated in different embodiments. Therefore, for parts not described or not described in detail in a certain embodiment, refer to the records of other embodiments.

Claims

1. An intelligent management method for assembly production of a shading sheet product, characterized in that: The method comprises: Step S10, obtaining historical assembly data and current assembly requirement data of the light shielding sheet; performing cluster analysis on the current assembly requirement data and the historical assembly data to obtain the first assembly parameter of the current assembly production; Step S20, setting a number of test assembly tasks according to the first assembly parameter, collecting image data and sensor data of each test assembly task in real time, extracting features from the image data and fusing them with the sensor data, and obtaining position data and angle data of multiple groups of shading sheets under different test assembly tasks; Step S30, according to the multiple sets of position data and angle data, respectively calculating the average translation error and the average angle error as the test assembly error; when the test assembly error does not meet the preset conditions, establishing a state transformation matrix according to the shading sheet position data and angle data, and calculating the rotation and translation combined error according to the state transformation matrix; The step of establishing a state transformation matrix according to the position data and angle data of the light shielding sheet and calculating the rotation and translation combined error according to the state transformation matrix includes: Get the actual position coordinates (x, y) and ideal position coordinates (x0, y0) of the shading plate, the actual angle θ and the ideal angle θ0, and establish the actual transformation matrix H and the ideal transformation matrix H0: Calculate the state transformation matrix T from the actual transformation matrix H to the ideal transformation matrix H0: Determine whether the state transformation matrix is ​​the identity matrix. When it is determined that the state transformation matrix is ​​not the identity matrix, use the state transformation matrix to determine the rotation and translation combined error: In the formula, H -1 represents the inverse matrix of H, ε represents the combined error of rotation and translation, and Δθ satisfies Δθ=θ-θ0; Step S40, constructing an error correction model according to the combined rotation and translation errors, solving the error correction model by the least square method to obtain second assembly parameters, and using the second assembly parameters to correct the first assembly parameters; The step of constructing an error correction model according to the combined rotation and translation errors and solving the error correction model by the least square method to obtain the second assembly parameter comprises: Construct the error correction model: ε=w1v+w2a+∈, Where w1 and w2 represent weight coefficients, v and a represent the robot arm speed and the rotary motor acceleration, and ∈ is a random noise term; Obtain N groups of test assembly parameters and the corresponding rotation and translation combined errors, and establish the objective function: In the formula, min{·} means minimization, ε i represents the i-th rotation and translation combined error, v i 、a i represents the robot arm speed and rotary motor acceleration in the i-th group of assembly parameters; The objective function is solved by the least square method to obtain the optimal solution of w1 and w2. The first assembly parameter and the optimal solution of w1 and w2 are substituted into the error correction model to obtain the target error. Based on target error Determine the second assembly parameters: Where v' and a' are the robot arm speed and the rotary motor acceleration in the second assembly parameter, k v , k a Respectively represent proportional gains, satisfying 0.1 <k v <1; 0.1 <k a <1; Step S50, returning to execute steps S20-S40, until the test assembly error meets the preset conditions, generating target assembly parameters, and executing the light shielding sheet assembly production process using the target assembly parameters.

2. The intelligent management method for assembly production of shading sheet products according to claim 1, characterized in that: The cluster analysis of the current assembly demand data and the historical assembly data includes: Based on historical assembly data, assembly results under different assembly parameters are obtained, and the current assembly requirement data and assembly results are clustered using the K-means algorithm; the assembly results include translation error and angle error; The optimal cluster is identified from the clustering results, and the corresponding assembly data is determined as the first assembly parameter of the current assembly production.

3. The intelligent management method for assembly production of shading sheet products according to claim 1, characterized in that: The method further comprises: Acquire the operation cycle of each assembly task in real time, identify the assembly tasks whose operation cycle is greater than a first preset value, and determine the queuing time from the current assembly task to the subsequent assembly task; Determine whether the queuing time is greater than a second preset value, if so, mark the subsequent assembly task as an abnormal task; if not, mark the current assembly task as an abnormal task; Trace the abnormal task back to the previous process task until the associated abnormal interval is identified, and match the operation instructions to correct the assembly operation in the associated abnormal interval.

4. An intelligent management system for assembly and production of shading sheet products, characterized in that: The system comprises: The cluster analysis unit is used to execute step S10 to obtain historical assembly data and current assembly requirement data of the light shielding sheet; perform cluster analysis on the current assembly requirement data and the historical assembly data to obtain the first assembly parameter of the current assembly production; A data acquisition unit, configured to execute step S20, set a plurality of test assembly tasks according to the first assembly parameter, collect image data and sensor data of each test assembly task in real time, extract features from the image data and fuse them with the sensor data, and obtain position data and angle data of multiple groups of light shielding sheets under different test assembly tasks; The error calculation unit is used to execute step S30, and respectively calculate the average translation error and the average angle error according to the multiple sets of position data and angle data as the test assembly error; when the test assembly error does not meet the preset conditions, establish a state transformation matrix according to the shading sheet position data and angle data, and calculate the rotation and translation combined error according to the state transformation matrix; The step of establishing a state transformation matrix according to the position data and angle data of the light shielding sheet and calculating the rotation and translation combined error according to the state transformation matrix includes: Get the actual position coordinates (x, y) and ideal position coordinates (x0, y0) of the shading plate, the actual angle θ and the ideal angle θ0, and establish the actual transformation matrix H and the ideal transformation matrix H0: Calculate the state transformation matrix T from the actual transformation matrix H to the ideal transformation matrix H0: Determine whether the state transformation matrix is ​​the identity matrix. When it is determined that the state transformation matrix is ​​not the identity matrix, use the state transformation matrix to determine the rotation and translation combined error: In the formula, H -1 represents the inverse matrix of H, ε represents the combined error of rotation and translation, and Δθ satisfies Δθ=θ-θ0; A parameter correction unit, configured to execute step S40, construct an error correction model according to the rotation and translation combined error, solve the error correction model by the least square method to obtain a second assembly parameter, and use the second assembly parameter to correct the first assembly parameter; The step of constructing an error correction model according to the combined rotation and translation errors and solving the error correction model by the least square method to obtain the second assembly parameter comprises: Construct the error correction model: ε=w1v+w2a+∈, Where w1 and w2 represent weight coefficients, v and a represent the robot arm speed and the rotary motor acceleration, and ∈ is a random noise term; Obtain N groups of test assembly parameters and the corresponding rotation and translation combined errors, and establish the objective function: In the formula, min{·} means minimization, ε i represents the i-th rotation and translation combined error, v i 、a i represents the robot arm speed and rotary motor acceleration in the i-th group of assembly parameters; The objective function is solved by the least square method to obtain the optimal solution of w1 and w2. The first assembly parameter and the optimal solution of w1 and w2 are substituted into the error correction model to obtain the target error. Based on target error Determine the second assembly parameters: Where v' and a' are the robot arm speed and the rotary motor acceleration in the second assembly parameter, k v , k a Respectively represent proportional gains, satisfying 0.1 <k v <1; 0.1 <k a <1; The iterative correction unit is used to execute step S50, return to execute steps S20-S40, until the test assembly error meets the preset conditions, generate target assembly parameters, and use the target assembly parameters to execute the light shielding sheet assembly production process.

5. The intelligent management system for assembly and production of shading sheet products according to claim 4, characterized in that: The system further comprises: A queuing duration determining unit, configured to execute step S60, obtain the operation cycle of each assembly task in real time, identify the assembly tasks whose operation cycle is greater than a first preset value, and determine the queuing duration from the current assembly task to the subsequent assembly task; The abnormal task marking unit is used to execute step S70 to determine whether the queuing time is greater than a second preset value. If so, the subsequent assembly task is marked as an abnormal task; if not, the current assembly task is marked as an abnormal task; The abnormal task tracing unit is used to execute step S80, tracing from the abnormal task to the preceding process task until the associated abnormal interval is identified, and matching the operation instructions to correct the assembly operation of the associated abnormal interval.

6. An electronic device, characterized in that: include: A processor and a memory, wherein the memory is used to store computer program codes, wherein the computer program codes include computer instructions. When the processor executes the computer instructions, the electronic device executes the intelligent management method for assembly production of shading film products as described in any one of claims 1 to 3.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes the intelligent management method for assembly production of shading film products as described in any one of claims 1 to 3.

Images