Automatic error correction method and system for robot box folding production

Abstract

The invention relates to the technical field of box folding automatic production, in particular to an automatic error correction method for robot box folding production, which comprises the following steps: acquiring a current production model and matching corresponding production standard parameters, then monitoring production links in real time, generating error data, comparing the error data with a preset error threshold value, and once an error is found to exceed the threshold value, outputting error correction results. And the system marks as abnormal and records related parameter data. Using a machine self-learning algorithm, the system analyzes the anomaly data to determine an anomaly type and generate a variation trend. And finally, according to the abnormal type and the error data, the system formulates and executes a robot error correction control strategy, including an error correction link and parameter adjustment, so as to automatically correct the deviation in the production process. According to the method, through an intelligent monitoring and response mechanism, manual intervention is reduced, and the accuracy and stability of box folding production are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automated box folding production, and in particular to an automatic error correction method and system for robot box folding production. Background Art

[0002] In the field of industrial automation, robotic box folding production has been widely used, but it still faces many technical challenges. Although traditional industrial robots used for box folding production can improve production efficiency and reduce labor costs, they have shortcomings in real-time monitoring and automatic error correction during the production process.

[0003] Currently, in the production of folding boxes, staff are required to conduct random inspections on the finished folding box products to determine whether there are any abnormalities in the folding box production, which requires manual intervention. This reduces production efficiency and increases labor costs. In addition, it is easy for the folding boxes to be of unqualified quality due to untimely detection, which reduces the quality of folding box production. Therefore, improvement is needed. Summary of the invention

[0004] In order to improve the quality of box folding production, the present application provides an automatic error correction method and system for robot box folding production.

[0005] In the first aspect, the above-mentioned invention objective of the present application is achieved through the following technical solutions: An automatic error correction method for robot folding box production, the method comprising the steps of: Obtain the current production model of the folding box and input the model into the production database to match the corresponding production standard parameters of the folding box in each production link; Acquire real-time detection data of each link of box folding, and compare the real-time detection data with corresponding production standard parameters to generate production error data; Comparing the production error data with a preset error threshold, and determining whether the production error data exceeds the preset error threshold; If the production error data exceeds a preset error threshold, the production error data is marked as production abnormality data, and the production parameter data during the production abnormality data period is recorded based on a preset time interval to construct an abnormal production parameter data set; The preset production anomaly analysis model analyzes the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determines the anomaly type based on the production parameter change trend; The preset correction model produces a robot error correction control strategy according to the abnormality type and production error data, and the robot error correction control strategy includes an error correction link and an error correction control parameter.

[0006] By adopting the above technical solution, it is possible to monitor the production process in real time, quickly identify and respond to production anomalies, and automatically adjust production parameters to correct deviations. Overall, this method improves production efficiency, reduces costs, improves product quality, and enhances production flexibility and adaptability.

[0007] In a preferred example, the present application can be further configured as follows: in the step of analyzing the abnormal production parameter data set based on the machine self-learning algorithm in the preset production abnormality analysis model to generate a production parameter change trend, and determining the abnormality type based on the production parameter change trend, the steps are as follows: If the difference value of adjacent production parameter data in the production parameter change trend is less than the preset mutation threshold, the abnormality type is determined to be a gradual production factor type, wherein the gradual production factor type includes mechanical wear and production environment changes; If in the production parameter change trend, the difference value of adjacent production parameter data is greater than the preset mutation threshold, the abnormality type is judged as a mutation production factor type, wherein the mutation production factor type includes folding box material change and machine failure.

[0008] By adopting the above technical solution, it is possible to quantify the judgment criteria, make reasonable judgments on abnormal situations, improve the adjustment accuracy of automatic error correction, and enhance the flexibility and adaptability of production.

[0009] In a preferred example, the present application may be further configured as follows: after the preset production abnormality analysis model analyzes the abnormal production parameter data set based on the machine self-learning algorithm to generate the production parameter change trend, the following steps are included: The preset trend prediction model analyzes the production parameter change trend based on a machine self-learning algorithm to generate a predicted production change trend, wherein the predicted production change trend includes predicted production parameters for folding box production in a future time period; Based on the predicted production change trend, the predicted production error data corresponding to the predicted production parameters in the future time period is calculated, and a predicted error trend is generated based on the predicted production error data. The predicted error trend is used to regulate the robot in advance for automatic error correction.

[0010] By adopting the above technical solution, not only can the current production status be monitored and analyzed in real time, but also future production change trends can be predicted, and automatic regulation can be performed in advance accordingly. Overall, this method improves the predictability and adaptability of production, reduces errors caused by changes in production parameters, reduces scrap rates and production costs, and improves production efficiency and product quality. Through the application of machine self-learning algorithms, this method can continuously learn and optimize, automatically adapt to changes in production environment and conditions, and realize intelligent closed-loop control.

[0011] In a preferred example, the present application may be further configured as follows: after marking the production error data as production abnormality data and recording the production parameter data during the production abnormality data based on a preset time interval to construct an abnormal production parameter data set, the following steps are included: Acquire historical environment data, and associate the historical environment data with production parameter data in the abnormal production parameter data set based on a preset common time axis to construct an environment-production data set; The preset environmental analysis model analyzes the environment-production data set based on a machine self-learning algorithm to generate environmental impact factors, and the environmental impact factors are used to predict production parameter data corresponding to different environmental data; The trend prediction model predicts the real-time detection data based on the environmental influencing factors to generate environmental prediction production data.

[0012] By adopting the above technical solution, the production process can be monitored in real time, and the impact of environmental changes on production can be analyzed, and future production trends can be predicted accordingly. Overall, this method improves the adaptability and predictability of production, reduces costs and quality problems caused by environmental changes and production errors, and improves production efficiency and product quality.

[0013] In a preferred example, the present application may be further configured as follows: after the preset production abnormality analysis model analyzes the abnormal production parameter data set based on the machine self-learning algorithm to generate a production parameter change trend, and determines the abnormality type based on the production parameter change trend, the following steps are included: Classify the abnormality types and construct a production abnormality profile; Associating the corresponding machine parameters in the historical abnormal events with the abnormal types in the abnormal portraits to construct a production-abnormal type data set; The production anomaly analysis model analyzes the production-anomaly type data set to generate a production anomaly attention coefficient, where the production anomaly attention coefficient is used to represent the frequency of production anomalies occurring in different machine parameters.

[0014] By adopting the above technical solutions, the production abnormality attention coefficient helps production managers understand the impact of different parameters on production abnormalities, so as to give priority to those parameters that are more likely to cause abnormalities, thereby improving the safety and stability of production. In a preferred example, the present application can be further configured as follows: in the step where the preset correction model produces a robot error correction control strategy according to the abnormal type and the production error data, the robot error correction control strategy includes an error correction link and an error correction control parameter, the following steps are also included: When the abnormal type is determined to be a mutated production factor type, the folding box production model is re-obtained and it is determined whether to change the model; If the judgment result is yes, the production standard parameters are re-matched to control the robot to work; If the judgment result is no, it is judged as a machine failure, and the robot data corresponding to the abnormal type is obtained to generate a machine failure report for sending to the operation and maintenance personnel.

[0015] By adopting the above technical solution, a detailed machine failure report is generated, and operation and maintenance personnel can quickly locate the problem and make repairs or adjustments, which reduces the impact of machine failure on production and improves production efficiency and equipment reliability. It can not only monitor and analyze anomalies in the production process in real time, but also respond quickly according to the type of anomaly, whether it is changing the production model or handling machine failures. Overall, this method improves the flexibility, adaptability and robustness of production, reduces production interruptions and scrap rates, and improves production efficiency and product quality.

[0016] In a preferred example, the present application can be further configured as follows: the machine self-learning algorithm includes isolation forest, support vector data description and autoencoder.

[0017] In the second aspect, the above invention objective of the present application is achieved through the following technical solutions: An automatic error correction device for robot folding box production, the device comprising: a production standard parameter matching unit, used to obtain the current production model of the folding box and input the model into a production database to match the production standard parameters corresponding to the folding box in each production link; A production error data generating unit, used to obtain real-time detection data of each link of box folding, and compare the real-time detection data with corresponding production standard parameters to generate production error data; an error threshold determination unit, used to compare the production error data with a preset error threshold, and determine whether the production error data exceeds the preset error threshold; A production abnormality data marking unit, used for marking the production error data as production abnormality data when the production error data exceeds a preset error threshold, and recording the production parameter data during the production abnormality data period based on a preset time interval to construct an abnormal production parameter data set; A production parameter change trend generating unit, which is pre-set with a production anomaly analysis model to analyze the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determine the anomaly type based on the production parameter change trend; A robot error correction control strategy generating unit is used to pre-set a correction model to generate a robot error correction control strategy according to the abnormality type and production error data, wherein the robot error correction control strategy includes an error correction link and an error correction control parameter.

[0018] On the third aspect, the above-mentioned purpose of the present application is achieved through the following technical solutions: An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned automatic error correction method for robot folding box production when executing the computer program.

[0019] Fourthly, the above-mentioned purpose of the present application is achieved through the following technical solutions: A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned automatic error correction method for robot folding box production.

[0020] In summary, the present application includes at least one of the following beneficial technical effects: 1. It can monitor the production process in real time, quickly identify and respond to production anomalies, and automatically adjust production parameters to correct deviations. Overall, this method improves production efficiency, reduces costs, improves product quality, and enhances production flexibility and adaptability; 2. It can quantify the judgment criteria, make reasonable judgments on abnormal situations, improve the adjustment accuracy of automatic error correction, and enhance the flexibility and adaptability of production; 3. It can not only monitor and analyze the current production status in real time, but also predict future production change trends and make automatic adjustments in advance accordingly. Overall, this method improves the predictability and adaptability of production, reduces errors caused by changes in production parameters, reduces scrap rates and production costs, and improves production efficiency and product quality. Through the application of machine self-learning algorithms, this method can continuously learn and optimize, automatically adapt to changes in production environment and conditions, and realize intelligent closed-loop control; 4. Generate detailed machine failure reports, so that operation and maintenance personnel can quickly locate problems and make repairs or adjustments, reducing the impact of machine failures on production and improving production efficiency and equipment reliability. It can not only monitor and analyze anomalies in the production process in real time, but also respond quickly according to the type of anomaly, whether it is changing the production model or handling machine failures. Overall, this method improves the flexibility, adaptability and robustness of production, reduces production interruptions and scrap rates, and improves production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1This is a flowchart of the steps of an automatic error correction method for robot folding box production in one embodiment of the present application; Figure 2 This is a principle block diagram of an automatic error correction device for robot box folding production in one embodiment of the present application; Figure 3 It is a schematic diagram of an electronic device in an embodiment of the present application.

[0022] Figure Number: 1. Production standard parameter matching unit; 2. Production error data generation unit; 3. Error threshold judgment unit; 4. Production abnormal data marking unit; 5. Production parameter change trend generation unit; 6. Robot error correction control strategy generation unit. DETAILED DESCRIPTION

[0023] The present application is further described in detail below in conjunction with the accompanying drawings.

[0024] In one embodiment, if Figure 1 As shown, the present application discloses an automatic error correction method for robot folding box production, which specifically includes the following steps: S10: obtaining the current production model of the folding box and inputting the model into the production database to match the corresponding production standard parameters of the folding box in each production link; Specifically, the current production model of the folding box is obtained by inputting or checking the scan. In an embodiment of the present application, the model is input into the production database by scanning the QR code affixed to the folding box, and the corresponding production standard parameters in each production link are automatically output according to the production database. Thus, through automated model recognition and database query, manual input errors are reduced and the efficiency and accuracy of production preparation are improved.

[0025] It should be noted that different production standard parameters correspond to different robot control parameters, and this application generates corresponding robot control parameters based on the production standard parameters.

[0026] S20: acquiring real-time detection data of each link of the box folding process, and comparing the real-time detection data with corresponding production standard parameters to generate production error data; Specifically, in step S20, there are the following detection methods: Photoelectric detection: During the folding process of the folding box, a photoelectric sensor is used to detect whether the edges of the folding box are aligned. The photoelectric sensor can detect slight deviations of the edges of the folding box and provide real-time feedback.

[0027] Beneficial effects: Photoelectric detection can quickly and accurately detect the alignment of the folding box edges, improve folding accuracy, and reduce the scrap rate caused by poor alignment.

[0028] Laser detection: In the folding box size detection process, a laser sensor is used to measure the length, width and height of the folding box to ensure that the size meets the standard.

[0029] Beneficial effects: Laser detection provides high-precision measurement results, helping to ensure the consistency of folding box dimensions and improve product quality.

[0030] Machine vision inspection: During the appearance inspection of the folding box, a high-resolution camera and image processing algorithm are used to detect whether there are scratches, dents or other surface defects on the surface of the folding box.

[0031] Beneficial effects: Machine vision inspection can achieve a comprehensive inspection of the appearance of the folding box, timely discover and eliminate appearance defects, and improve the appearance quality of the product.

[0032] Ultrasonic testing: In the material thickness detection process of the folding box, an ultrasonic sensor is used to detect the thickness of the material to ensure that the material thickness is uniform.

[0033] Beneficial effects: Ultrasonic testing can measure material thickness without contact, which helps to ensure the uniformity and strength of the folding box material.

[0034] Sensor detection: During the forming process of the folding box, a displacement sensor is installed to monitor the position of the forming mold to ensure that the mold is accurately in place.

[0035] Beneficial effects: Sensor detection can monitor the mold position in real time, reduce molding defects caused by mold position deviation, and improve production efficiency and product quality.

[0036] 3D point cloud registration technology: During the assembly of the folding box, a 3D scanner is used to obtain the point cloud data of the folding box components, and the point cloud registration technology is used to ensure that the components are accurately docked.

[0037] Beneficial effects: 3D point cloud registration technology provides accurate spatial positioning capabilities, which helps to improve the accuracy and efficiency of folding box assembly.

[0038] The real-time monitoring in step S20 ensures that any deviation in the production process can be discovered in time, providing a data basis for subsequent error analysis and correction.

[0039] S30: comparing the production error data with a preset error threshold, and determining whether the production error data exceeds the preset error threshold; Specifically, take the error threshold of ±0.5 degrees as an example. Since the actual bending angle of 90.2 degrees deviates from the standard angle of 90 degrees by 0.2 degrees, which does not exceed the threshold, the system determines it as normal. By comparing with the preset threshold, the system can quickly determine whether the production process is within the control range, reducing the reliance on manual monitoring and improving the response speed to abnormal situations through digital monitoring.

[0040] S40: if the production error data exceeds a preset error threshold, the production error data is marked as production abnormality data, and production parameter data during the production abnormality data period is recorded based on a preset time interval to construct an abnormal production parameter data set; Specifically, if in another production, the bending angle is monitored to be 89.5 degrees, which exceeds the preset error threshold, the system will mark it as abnormal and start recording the production parameter data during the abnormal period. It should be noted that in the embodiment of the present application, the abnormal period includes a period of time before the abnormality occurs and a period of time after the abnormality occurs. Recording abnormal data will help to subsequently analyze the cause of the abnormality, optimize the production process, and improve product quality.

[0041] S50: The preset production anomaly analysis model analyzes the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determines the anomaly type based on the production parameter change trend; Specifically, in step S50, the following machine learning methods are used to judge and detect anomalies: Isolation Forest, Support Vector Data Description (SVDD) and Autoencoder.

[0042] The detection parameters may include but are not limited to the following: Bending angle: detect whether the bending angle during the folding process meets the standard parameters; Pressure reading: detect whether the pressure applied to the folding box is within the preset range; Speed: detect whether the running speed of the folding machine is consistent with the standard speed; Material thickness: detect whether the thickness of the material used is uniform and meets production requirements.

[0043] The criteria for determination can be based on the following aspects: Statistical standards: According to the characteristics of normal distribution, the probability that the eigenvalue falls within the preset range is normal, and those outside this range can be regarded as abnormal.

[0044] Historical data comparison: Real-time detection data is compared with historical data. If obvious deviations are found, they can be determined as abnormal.

[0045] Machine learning model prediction: Use machine learning models to predict data. If the prediction results deviate significantly from the actual data, they can be judged as abnormal.

[0046] Trend analysis: Analyze the changing trends of production parameters. If the parameter trends are found to be obviously abnormal or have abnormal fluctuations, they can be judged as abnormal.

[0047] In an embodiment of the present application, the system analyzes the abnormal production parameter data set through a machine learning algorithm and finds that the deviation of the bending angle is related to the changing trend of the material thickness, and determines the abnormality type as "material thickness change". By automatically identifying the abnormality type, the system can provide more targeted solutions to reduce downtime and material waste.

[0048] S60: The preset correction model produces a robot error correction control strategy according to the abnormality type and the production error data, wherein the robot error correction control strategy includes an error correction link and an error correction control parameter; Specifically, for anomalies of the "material thickness change" type, the correction model automatically adjusts the robot's bending force and speed to adapt to materials of different thicknesses and ensure the quality of the folding boxes. It should be noted that the adjusted control parameters need to be further optimized according to the adjusted effects until the production standards are met.

[0049] Automatic error correction control strategy reduces manual intervention, improves production efficiency and product quality, and reduces production costs.

[0050] In summary, this application provides an automated and intelligent solution for automatic error correction in robot folding box production. This method can monitor the production process in real time, quickly identify and respond to production anomalies, and automatically adjust production parameters to correct deviations. Overall, this method improves production efficiency, reduces costs, improves product quality, and enhances production flexibility and adaptability, bringing innovation and technological progress to the field of robot folding box production.

[0051] After the step S40: marking the production error data as production abnormality data, and recording the production parameter data during the production abnormality data based on a preset time interval to construct an abnormal production parameter data set, the following steps are included: S401: Acquire historical environment data, and associate the historical environment data with production parameter data in the abnormal production parameter data set based on a preset common time axis to construct an environment-production data set; Specifically, during the box folding production process, the system records abnormal production parameter data sets, including parameters such as temperature, humidity, and bending angle. At the same time, the system obtains environmental data at the same time point from the historical environment database, such as temperature and humidity records. By matching these historical environmental data with the data in the abnormal production parameter data set, an environment-production data set is constructed.

[0052] S402: The preset environment analysis model analyzes the environment-production data set based on a machine self-learning algorithm to generate an environment impact factor, where the environment impact factor is used to predict production parameter data corresponding to different environment data; Specifically, using machine learning algorithms such as support vector data description (SVDD), the system analyzes the environmental-production data set to identify the degree of influence of temperature and humidity on the bending angle. The system generates an environmental influence factor model that can predict the expected change in bending angle under different temperature and humidity conditions.

[0053] By identifying environmental influencing factors, the system can predict the impact of environmental changes on production parameters and provide a scientific basis for automatic adjustment of the production process.

[0054] S403: The trend prediction model predicts the real-time detection data based on the environmental influencing factors to generate environmental prediction production data; Based on the environmental impact factor model, the trend prediction model predicts that in the next week, due to the upcoming high temperature weather, the bending angle of the box folding machine may increase by 0.2 degrees. Based on this prediction, the system automatically adjusts the setting parameters of the bending machine to offset the impact of the temperature increase.

[0055] By predicting the impact of environmental changes on production in advance, the system can automatically adjust production parameters, reduce production errors caused by environmental changes, and improve production efficiency and product quality.

[0056] In summary, this solution monitors the production process in real time, analyzes the impact of environmental changes on production, and predicts future production trends accordingly. Overall, this method improves the adaptability and predictability of production, reduces costs and quality issues caused by environmental changes and production errors, and improves production efficiency and product quality.

[0057] In S50: the preset production anomaly analysis model analyzes the abnormal production parameter data set based on the machine self-learning algorithm to generate a production parameter change trend, and determines the anomaly type based on the production parameter change trend. The steps include the following: S51: If the difference value of adjacent production parameter data in the production parameter change trend is less than a preset mutation threshold, the abnormality type is determined to be a gradual production factor type, wherein the gradual production factor type includes mechanical wear and production environment changes; Specifically, during the folding box production process, the system discovered through continuous monitoring that the average rate of change of the bending angle gradually decreased from 90 degrees to 89.5 degrees, and it was judged that an abnormality existed. Moreover, this change occurred gradually over a long period of time. The mutation threshold set by the system was 0.5 degrees / hour, while the actual change rate was 0.1 degrees / hour, which was less than the preset mutation threshold. The abnormality type was judged to be a gradual production factor type.

[0058] S52: If the difference value of adjacent production parameter data in the production parameter change trend is greater than a preset mutation threshold, the abnormality type is determined to be a mutation production factor type, wherein the mutation production factor type includes folding box material change and machine failure; Specifically, in another production batch, the system suddenly detected that the bending angle dropped sharply from 90 degrees to 85 degrees. This change occurred in a very short period of time. The system set the mutation threshold to 0.5 degrees / hour, while the actual change rate was 5 degrees / hour, which was much larger than the preset mutation threshold.

[0059] The system quickly identifies this mutation-type anomaly, which may be caused by a sudden change in the folding box material or a machine failure. It immediately suspends production and triggers a fault diagnosis program to quickly locate the problem and repair it, reducing production interruptions and material waste.

[0060] In summary, it is possible to quantify the judgment criteria, make reasonable judgments on abnormal situations, improve the adjustment accuracy of automatic error correction, and at the same time enhance the flexibility and adaptability of production.

[0061] After the step S50: the preset production abnormality analysis model analyzes the abnormal production parameter data set based on the machine self-learning algorithm to generate the production parameter change trend, the following steps are included: S501: The preset trend prediction model analyzes the production parameter change trend based on the machine self-learning algorithm to generate a predicted production change trend, wherein the predicted production change trend includes predicted production parameters for folding box production in a future time period; Specifically, suppose that in the production of folding boxes, historical data analysis shows that in summer, due to the increase in temperature, the ductility of the material increases, resulting in a larger bending angle. The trend prediction model uses a machine self-learning algorithm to analyze historical data and predicts that in the coming summer, the bending angle may increase by 0.3 degrees.

[0062] By predicting the changing trends of production parameters in advance, the production management system can adjust the production parameters in advance, such as appropriately reducing the bending force to offset the impact of temperature changes, thereby ensuring the stability of product quality.

[0063] S502: Based on the predicted production change trend, calculate the predicted production error data corresponding to the predicted production parameters in the future time period, and generate a predicted error trend based on the predicted production error data, wherein the predicted error trend is used to regulate the robot in advance to automatically correct errors; Specifically, based on the output of the trend prediction model, the system predicts that in the next week, due to changes in material supply, the material thickness may increase slightly, resulting in the need to adjust the forming pressure of the folding box. The system calculates the predicted production error data and generates a predicted error trend, indicating that the forming pressure needs to be increased by 5% to accommodate the increase in material thickness.

[0064] By predicting error trends, the system can perform automatic controls in advance, such as automatically adjusting pressure settings, to reduce production errors caused by material changes, improve production efficiency and product quality, and reduce scrap rates.

[0065] In summary, a forward-looking automatic error correction solution for robot folding box production is provided. This method can not only monitor and analyze the current production status in real time, but also predict future production change trends and perform automatic regulation in advance accordingly. Overall, this method improves the predictability and adaptability of production, reduces errors caused by changes in production parameters, reduces scrap rates and production costs, and improves production efficiency and product quality. Through the application of machine self-learning algorithms, this method can continuously learn and optimize, automatically adapt to changes in production environment and conditions, and realize intelligent closed-loop control.

[0066] After the step S50: the preset production abnormality analysis model analyzes the abnormal production parameter data set based on the machine self-learning algorithm to generate a production parameter change trend, and determines the abnormality type based on the production parameter change trend, the following steps are also included: S503: Classify the abnormality types and construct a production abnormality profile; Specifically, in the production of folding boxes, the system analyzes historical data and identifies two common types of anomalies: mechanical wear and changes in the production environment. Mechanical wear is usually manifested as a gradual decrease in the bending angle, while environmental changes such as increased humidity may cause the material to soften and affect the bending quality. The system classifies these features and constructs a production anomaly portrait that includes these features. By constructing the production anomaly portrait, the system can quickly identify and respond to specific types of production anomalies, improving the efficiency and accuracy of anomaly handling.

[0067] S504: Correlate the corresponding machine parameters in the historical abnormal events with the abnormal types in the abnormal portrait to construct a production-abnormal type data set; Specifically, the system associates all records of production anomalies caused by mechanical wear and material problems in the past year with the newly constructed production anomaly portrait. For example, the system found that the failure rate of bending machines increased by 20% in high humidity environments, and associated this data with the abnormal type of environmental changes.

[0068] By associating historical abnormal events with abnormal portraits, the system can more accurately predict production anomalies in different environments and machine conditions, providing data support for prevention and control.

[0069] S505: the production anomaly analysis model analyzes the production-anomaly type data set to generate a production anomaly attention coefficient, where the production anomaly attention coefficient is used to indicate the frequency of production anomalies occurring in different machine parameters; Based on the production-abnormal type data set, the system calculated that when the humidity exceeds 60%, the frequency of bending machine failure is 1.5 times that under normal humidity. The system uses this statistical result as the production abnormality attention coefficient to guide production adjustments.

[0070] The production abnormality attention coefficient helps production managers understand the impact of different parameters on production abnormalities, so that they can give priority to those parameters that are more likely to cause abnormalities, thereby improving the safety and stability of production.

[0071] Through steps S503-S505, the anomalies in the production process are monitored and analyzed in real time, and anomalies that may occur in the future can be predicted by constructing anomaly portraits and production-anomaly type data sets. Overall, this method improves the predictability and adaptability of production, reduces costs and quality problems caused by production anomalies, and improves production efficiency and product quality.

[0072] In S60: the preset correction model generates a robot error correction control strategy according to the abnormal type and the production error data, wherein the robot error correction control strategy includes an error correction link and an error correction control parameter. This step also includes the following steps: S61: when the abnormal type is determined to be a sudden change in the production factor type, the folding box production model is re-acquired and it is determined whether to change the model; Specifically, during the production process, the system detects a sudden change in production parameters, such as a sudden change in the bending angle. The system re-acquires the current production model information and finds that the model is "HD-456", which is a new model and different from the previous "HD-123".

[0073] By timely acquiring and confirming production models, the system can quickly respond to changes in production demand and ensure the continuity of the production process and the consistency of the products.

[0074] S62: If the judgment result is yes, re-match the production standard parameters to control the robot to work; Specifically, after confirming the model change, the system retrieves the new production standard parameters corresponding to the "HD-456" model from the production database, such as the new bending angle, pressure and speed settings.

[0075] Re-matching production standard parameters ensures that the robot can work according to the requirements of the new model, improves production flexibility and adaptability, and reduces production errors caused by parameter mismatch.

[0076] S63: If the judgment result is no, it is judged as a machine failure, and the robot data corresponding to the abnormal type is obtained to generate a machine failure report for sending to the operation and maintenance personnel; Specifically, if the system determines that the model has not been changed, the system will determine this abnormality as a machine failure. The system collects machine parameter data related to the "HD-123" model, such as motor current, temperature, and vibration data, and generates a machine failure report.

[0077] By generating detailed machine failure reports, operation and maintenance personnel can quickly locate problems and make repairs or adjustments, reducing the impact of machine failures on production and improving production efficiency and equipment reliability. It can not only monitor and analyze anomalies in the production process in real time, but also respond quickly according to the type of anomaly, whether it is changing the production model or handling machine failures. Overall, this method improves the flexibility, adaptability and robustness of production, reduces production interruptions and scrap rates, and improves production efficiency and product quality.

[0078] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0079] In one embodiment, an automatic error correction device for robot box folding production is provided, and the automatic error correction device for robot box folding production corresponds one-to-one with the automatic error correction method for robot box folding production in the above embodiment. Figure 2 As shown, the automatic error correction device for robot folding box production includes a production standard parameter matching unit 1, which is used to obtain the current production model of the folding box and input the model into the production database to match the production standard parameters corresponding to the folding box in each production link; The production error data generating unit 2 is used to obtain the real-time detection data of each link of the box folding process, and compare the real-time detection data with the corresponding production standard parameters to generate the production error data; The error threshold judgment unit 3 is used to compare the production error data with a preset error threshold and judge whether the production error data exceeds the preset error threshold; The production abnormal data marking unit 4 is used to mark the production error data as production abnormal data when the production error data exceeds a preset error threshold, and record the production parameter data during the production abnormal data period based on a preset time interval to construct an abnormal production parameter data set; A production parameter change trend generating unit 5 is used to pre-set a production anomaly analysis model to analyze the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determine the anomaly type based on the production parameter change trend; The robot error correction control strategy generating unit 6 is used to pre-set a correction model to generate a robot error correction control strategy according to the abnormality type and production error data, and the robot error correction control strategy includes an error correction link and an error correction control parameter.

[0080] For the specific definition of the automatic error correction device for robot folding box production, please refer to the definition of the automatic error correction method for robot folding box production above, which will not be repeated here. Each module in the above-mentioned automatic error correction device for robot folding box production can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0081] In one embodiment, an electronic device is provided. The electronic device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The electronic device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the electronic device is used to store a database. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, an automatic error correction method for robot folding box production is implemented.

[0082] In one embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Obtain the current production model of the folding box and input the model into the production database to match the corresponding production standard parameters of the folding box in each production link; Acquire real-time detection data of each link of box folding, and compare the real-time detection data with corresponding production standard parameters to generate production error data; Comparing the production error data with a preset error threshold, and determining whether the production error data exceeds the preset error threshold; If the production error data exceeds a preset error threshold, the production error data is marked as production abnormality data, and the production parameter data during the production abnormality data period is recorded based on a preset time interval to construct an abnormal production parameter data set; The preset production anomaly analysis model analyzes the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determines the anomaly type based on the production parameter change trend; The preset correction model produces a robot error correction control strategy according to the abnormality type and production error data, and the robot error correction control strategy includes an error correction link and an error correction control parameter.

[0083] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Obtain the current production model of the folding box and input the model into the production database to match the corresponding production standard parameters of the folding box in each production link; Acquire real-time detection data of each link of box folding, and compare the real-time detection data with corresponding production standard parameters to generate production error data; Comparing the production error data with a preset error threshold, and determining whether the production error data exceeds the preset error threshold; If the production error data exceeds a preset error threshold, the production error data is marked as production abnormality data, and the production parameter data during the production abnormality data period is recorded based on a preset time interval to construct an abnormal production parameter data set; The preset production anomaly analysis model analyzes the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determines the anomaly type based on the production parameter change trend; The preset correction model produces a robot error correction control strategy according to the abnormality type and production error data, and the robot error correction control strategy includes an error correction link and an error correction control parameter.

[0084] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0085] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0086] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An automatic error correction method for robot folding box production, characterized in that: The method comprises the steps of: obtaining the current production model of the folding box and inputting the model into a production database to match the corresponding production standard parameters of the folding box in each production link; Acquire real-time detection data of each link of box folding, and compare the real-time detection data with corresponding production standard parameters to generate production error data; Comparing the production error data with a preset error threshold, and determining whether the production error data exceeds the preset error threshold; If the production error data exceeds a preset error threshold, the production error data is marked as production abnormality data, and the production parameter data during the production abnormality data period is recorded based on a preset time interval to construct an abnormal production parameter data set; The preset production anomaly analysis model analyzes the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determines the anomaly type based on the production parameter change trend; The preset correction model produces a robot error correction control strategy according to the abnormality type and production error data, and the robot error correction control strategy includes an error correction link and an error correction control parameter.

2. The automatic error correction method for robot folding box production according to claim 1 is characterized in that: In the step of analyzing the abnormal production parameter data set based on the machine self-learning algorithm to generate a production parameter change trend and determining the abnormal type based on the production parameter change trend, the following steps are included: If the difference value of adjacent production parameter data in the production parameter change trend is less than the preset mutation threshold, the abnormality type is determined to be a gradual production factor type, wherein the gradual production factor type includes mechanical wear and production environment changes; If in the production parameter change trend, the difference value of adjacent production parameter data is greater than the preset mutation threshold, the abnormality type is judged as a mutation production factor type, wherein the mutation production factor type includes folding box material change and machine failure.

3. The automatic error correction method for robot folding box production according to claim 2 is characterized in that: After the preset production abnormality analysis model analyzes the abnormal production parameter data set based on the machine self-learning algorithm to generate the production parameter change trend, the following steps are included: The preset trend prediction model analyzes the production parameter change trend based on a machine self-learning algorithm to generate a predicted production change trend, wherein the predicted production change trend includes predicted production parameters for folding box production in a future time period; Based on the predicted production change trend, the predicted production error data corresponding to the predicted production parameters in the future time period is calculated, and a predicted error trend is generated based on the predicted production error data. The predicted error trend is used to regulate the robot in advance for automatic error correction.

4. The automatic error correction method for robot folding box production according to claim 3 is characterized in that: After marking the production error data as production abnormality data and recording the production parameter data during the production abnormality data based on a preset time interval to construct an abnormal production parameter data set, the following steps are included: Acquire historical environment data, and associate the historical environment data with production parameter data in the abnormal production parameter data set based on a preset common time axis to construct an environment-production data set; The preset environmental analysis model analyzes the environment-production data set based on a machine self-learning algorithm to generate environmental impact factors, and the environmental impact factors are used to predict production parameter data corresponding to different environmental data; The trend prediction model predicts the real-time detection data based on the environmental influencing factors to generate environmental prediction production data.

5. The automatic error correction method for robot folding box production according to claim 1, characterized in that: After the preset production anomaly analysis model analyzes the abnormal production parameter data set based on the machine self-learning algorithm to generate a production parameter change trend, and determines the anomaly type based on the production parameter change trend, the following steps are included: Classify the abnormality types and construct a production abnormality profile; Associating the corresponding machine parameters in the historical abnormal events with the abnormal types in the abnormal portraits to construct a production-abnormal type data set; The production anomaly analysis model analyzes the production-anomaly type data set to generate a production anomaly attention coefficient, where the production anomaly attention coefficient is used to represent the frequency of production anomalies occurring in different machine parameters.

6. The automatic error correction method for robot box folding production according to claim 5, characterized in that: In the step where the preset correction model produces the robot error correction control strategy according to the abnormal type and the production error data, the robot error correction control strategy includes an error correction link and an error correction control parameter, the following steps are also included: When the abnormal type is determined to be a mutated production factor type, the folding box production model is re-obtained and it is determined whether to change the model; If the judgment result is yes, the production standard parameters are re-matched to control the robot to work; If the judgment result is no, it is judged as a machine failure, and the robot data corresponding to the abnormal type is obtained to generate a machine failure report for sending to the operation and maintenance personnel.

7. The automatic error correction method for robot box folding production according to claim 1, characterized in that: The machine self-learning algorithm includes isolation forest, support vector data description and autoencoder.

8. An automatic error correction device for robot box folding production, applied to an automatic error correction method for robot box folding production according to any one of claims 1 to 7, characterized in that: The device comprises: A production standard parameter matching unit (1) is used to obtain the current production model of the folding box and input the model into the production database to match the corresponding production standard parameters of the folding box in each production link; A production error data generating unit (2) is used to obtain real-time detection data of each link of the box folding process, and compare the real-time detection data with corresponding production standard parameters to generate production error data; An error threshold determination unit (3), used for comparing the production error data with a preset error threshold, and determining whether the production error data exceeds the preset error threshold; A production abnormality data marking unit (4) is used to mark the production error data as production abnormality data when the production error data exceeds a preset error threshold, and to record the production parameter data during the production abnormality data period based on a preset time interval to construct an abnormal production parameter data set; A production parameter change trend generating unit (5), which is used to pre-set a production anomaly analysis model to analyze the abnormal production parameter data set based on a machine self-learning algorithm to generate a production parameter change trend, and determine the anomaly type based on the production parameter change trend; A robot error correction control strategy generating unit (6) is used to pre-set a correction model to generate a robot error correction control strategy according to the abnormality type and production error data, wherein the robot error correction control strategy includes an error correction link and an error correction control parameter.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the automatic error correction method for robot folding box production as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the steps of the automatic error correction method for robot folding box production as described in any one of claims 1 to 7 are implemented.

Images