Deformation prediction method and system for automobile injection molding parts

Abstract

The invention discloses a deformation prediction method and system for automobile injection molding parts, and relates to the technical field of automobile part manufacturing. Comprising the steps that historical process parameter data of a first injection molding machine monitoring device is obtained, and the first injection molding machine monitoring device is any one of a plurality of injection molding machine monitoring devices; and acquiring current abnormal process parameter data of a second injection molding machine monitoring device, wherein the second injection molding machine monitoring device is any device which monitors the abnormal process parameters. The system is composed of a plurality of injection molding machine monitoring devices and a part deformation monitoring device, the deformation of injection molding parts is predicted and controlled by collecting real-time injection molding process parameters and part deformation data, the injection molding machine monitoring devices are responsible for collecting process parameters such as temperature, pressure and injection speed, and the injection molding machine monitoring devices are responsible for monitoring the deformation of the injection molding parts. The deformation monitoring device is responsible for acquiring gear and deformation data of the parts, the data integration model synchronizes the data and marks space-time relevance, and subsequent analysis and processing are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive parts, and specifically to a method and system for predicting the deformation of automotive injection-molded parts. Background Art

[0002] In the automotive manufacturing industry, the injection molding process is one of the important links in the production of parts. However, due to fluctuations in process parameters or changes in material properties during the injection molding process, parts are prone to deformation problems such as shrinkage, warping, or dimensional tolerance issues. These problems not only affect the assembly accuracy and functional performance of the parts but may also lead to an increase in the scrap rate and production costs.

[0003] Traditional injection molding process monitoring methods mainly rely on post-inspection and manual experience adjustment, making it difficult to capture abnormal changes in process parameters in real time and unable to accurately predict and intervene in advance the deformation of parts. In addition, existing systems usually only focus on a single data source and lack the fusion analysis of multi-source data, resulting in insufficient prediction accuracy. The existing technology has the following deficiencies:

[0004] 1. Limitations of monitoring data, only collecting partial process parameters or deformation data, lacking the comprehensive integration of multi-source data;

[0005] 2. Lags in anomaly handling, relying on manual review for handling abnormal process parameters or deformation data, lacking an automated feedback control mechanism;

[0006] 3. Static nature of the prediction model, unable to dynamically adjust the prediction model based on real-time data and difficult to adapt to dynamic changes in process parameters;

[0007] 4. Ignoring data correlation, not fully considering the correlation between process parameters and deformation data, resulting in insufficient accuracy and reliability of prediction results.

[0008] Therefore, a method and system for predicting the deformation of automotive injection-molded parts are proposed. Summary of the Invention

[0009] The purpose of the present invention is to provide a method and system for predicting the deformation of automotive injection-molded parts to solve the problems raised in the above background art.

[0010] To achieve the above object, the present invention provides the following technical solution: A method for predicting the deformation of automotive injection-molded parts, which is applied to a deformation prediction system. The system includes multiple injection molding machine monitoring devices and part deformation monitoring devices. The injection molding machine monitoring devices are used to collect injection molding process parameter data, and the part deformation monitoring devices are used to monitor the real-time deformation data of the injection-molded parts. The method is characterized in that it includes:

[0011] Obtain the historical process parameter data of the first injection molding machine monitoring device, where the first injection molding machine monitoring device is any one of multiple injection molding machine monitoring devices;

[0012] Obtain the current abnormal process parameter data of the second injection molding machine monitoring device, where the second injection molding machine monitoring device is any device that monitors abnormal process parameters;

[0013] Obtain the historical deformation data of the first deformation monitoring device, where the first deformation monitoring device is any one of multiple deformation monitoring devices;

[0014] Obtain the current abnormal deformation data of the second deformation monitoring device, where the second deformation monitoring device is any device that monitors abnormal deformation.

[0015] As a specific solution of the technical solution of this application, the method for adjusting the prediction frequency based on the correlation between process parameters and deformation data includes:

[0016] Obtain the process parameter difference degree between the first injection molding machine monitoring device and the second injection molding machine monitoring device, and the parallel data correlation degree between the first deformation monitoring device and the second deformation monitoring device;

[0017] Based on the process parameter difference degree and deformation data correlation degree, calculate the first adjustment difference value and the second adjustment difference value;

[0018] Based on the first adjustment difference value and the historical process parameter data, adjust the real-time data acquisition frequency of the first injection molding machine monitoring device;

[0019] Based on the second adjustment difference value and the historical deformation data, adjust the deformation prediction model update frequency of the first deformation monitoring device.

[0020] As a specific solution of the technical solution of this application, the classification processing method based on abnormal process parameters and abnormal deformation data includes:

[0021] Based on the current process parameter data, perform cluster analysis to determine that the abnormal type is temperature fluctuation, pressure deviation or material filling deficiency;

[0022] Based on the current abnormal deformation data, perform pattern recognition to determine that the deformation type is shrinkage, warping or dimensional tolerance;

[0023] Based on the abnormal type, obtain the corresponding third adjustment difference value and fourth adjustment difference value from the preset weight table;

[0024] Combine the first adjustment difference value and the second adjustment difference value to dynamically optimize the correlation weight between the injection molding process parameters and the deformation prediction model.

[0025] As a specific embodiment of the technical solution of the present application, the confidence evaluation method based on historical anomalies includes:

[0026] Extract the historical time series data corresponding to the current abnormal process parameters and construct a process parameter change curve;

[0027] Perform exponential smoothing fitting based on the historical deformation trend corresponding to the current abnormal deformation data and calculate the fitting residuals;

[0028] If the standard deviation of the slope of the process parameter change curve exceeds the threshold or the fitting residuals are higher than the preset value, reduce the confidence of the abnormal data and trigger an artificial review process.

[0029] As a specific embodiment of the technical solution of the present application, the dynamic feedback control method based on real-time data includes:

[0030] Obtain the process parameters of the injection molding machine monitoring device and the deformation amount of the deformation monitoring device in real time;

[0031] Input the process parameters into a pre-trained deformation prediction model and output the real-time predicted deformation value;

[0032] Compare the deviation between the real-time predicted deformation value and the actual deformation value. If the deviation exceeds the operating range, generate a process parameter correction instruction and send it to the injection molding machine controller.

[0033] As a specific embodiment of the technical solution of the present application, the deformation warning method based on multi-source data fusion includes:

[0034] Fuse the mold temperature, holding pressure time, cooling rate parameters of the injection molding machine and the three-dimensional scan data of the components;

[0035] Calculate the influence weight of each process parameter on deformation through the grey relational analysis algorithm;

[0036] Generate a priority warning signal according to the weight distribution and push it to the production management system.

[0037] A deformation prediction system for automotive injection molded parts includes a plurality of injection molding machine monitoring devices and component deformation monitoring devices. The injection molding machine monitoring devices are used to collect temperature, pressure and injection speed data, and the deformation monitoring devices are used to obtain the dimensions and deformation data of the components. It is characterized in that the system further includes:

[0038] A data integration model for synchronizing the process parameters of the injection molding machine and the deformation monitoring data and marking the spatio-temporal correlation;

[0039] An intelligent analysis module for executing the calculation logic of any of the methods recited in claims 1 to 6 and outputting an adjustment difference and a prediction result;

[0040] The adaptive control unit dynamically adjusts the injection molding machine parameters or triggers the mold calibration instruction according to the prediction result.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0042] The method and system for predicting the deformation of automotive injection molded parts are composed of multiple injection molding machine monitoring devices and part deformation monitoring devices. By collecting real-time injection molding process parameters and part deformation data, the prediction and control of the deformation of injection molded parts are realized. The injection molding machine monitoring device is responsible for collecting process parameters such as temperature, pressure and injection speed, and the deformation monitoring device is responsible for obtaining the gear and deformation data of the parts. The data integration model synchronizes these data and marks the spatio-temporal correlation for subsequent analysis and processing.

[0043] At the same time, based on historical data and current real-time data, the intelligent analysis module uses technologies such as clustering analysis and grey correlation algorithm to identify abnormal process parameters and deformation types, and calculates the adjustment difference to optimize the correlation weight between the process parameters and the deformation prediction model, dynamically adjusting the injection molding machine parameters or issuing the mold calibration instruction, thereby improving the quality and consistency of the injection molded parts.

[0044] Secondly, by integrating multi-source data, the system can timely detect abnormal situations, generate priority warning signals, and push them to the production management system to ensure real-time monitoring and dynamic feedback control during the production process, thereby effectively predicting and preventing the deformation problems of injection molded parts and improving production efficiency and product quality. Brief Description of the Drawings

[0045] Figure 1 It is a schematic flow chart of the deformation warning method based on multi-source data fusion of the present invention;

[0046] Figure 2 It is the classification processing method based on abnormal process parameters and abnormal deformation data of the present invention

[0047] Schematic flow chart;

[0048] Figure 3 It is the method for adjusting the prediction frequency based on the correlation between process parameters and deformation data of the present invention

[0049] Schematic flow chart;

[0050] Figure 4 It is a schematic diagram of the deformation prediction system for automotive injection molded parts of the present invention. Detailed Description of the Invention

[0051] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention.

[0053] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0054] In the present invention, unless otherwise clearly specified and limited, the terms "mounted", "connected", "connected to", "fixed", etc. should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly limited. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0055] In the present invention, unless otherwise clearly specified and limited, the first feature being "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are indirectly in contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on top of" the second feature may be that the first feature is directly above or obliquely above the second feature, or merely indicates that the first feature has a higher horizontal height than the second feature. The first feature being "under", "beneath" and "underneath" the second feature may be that the first feature is directly below or obliquely below the second feature, or merely indicates that the first feature has a lower horizontal height than the second feature.

[0056] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0057] As Figures 1 to 4 shown, the present invention provides a technical solution: a method for predicting the deformation of automotive injection-molded parts, which is applied to a deformation prediction system. The system includes a plurality of injection molding machine monitoring devices and part deformation monitoring devices. The injection molding machine monitoring devices are used to collect injection molding process parameter data, and the part deformation monitoring devices are used to monitor the real-time deformation data of the injection-molded parts. The method is characterized in that it includes:

[0058] Obtaining the historical process parameter data of the first injection molding machine monitoring device, where the first injection molding machine monitoring device is any one of the plurality of injection molding machine monitoring devices;

[0059] Obtaining the current abnormal process parameter data of the second injection molding machine monitoring device, where the second injection molding machine monitoring device is any one that monitors abnormal process parameters;

[0060] Obtaining the historical deformation data of the first deformation monitoring device, where the first deformation monitoring device is any one of the plurality of deformation monitoring devices;

[0061] Obtain the current abnormal deformation data of the second deformation monitoring device, where the second deformation monitoring device is any device that detects abnormal deformation. It should be clear that in the embodiments of this application, the injection molding machine monitoring device is used to collect injection molding process parameter data, and the parameter data includes but is not limited to injection temperature, mold temperature, injection speed, holding pressure, and cooling time, etc. For example, in this application, the injection temperature is between 200°C and 300°C, the mold temperature is between 50°C and 100°C, the injection speed is between 500 mm / s and 200 mm / s, the holding pressure is between 50 Mpa and 150 Mpa, and the cooling time is between 10 s and 60 s. The component deformation monitoring device is used to monitor the real-time deformation data of the injection molded components, and the data includes the thickness, warpage, and surface defects of the components, etc. For example, the thickness change range is between 0.1 mm and 1.0 mm, and the warpage is between 0.5 mm and 5.0 mm. Then, clean the collected injection molding process parameters and deformation data, remove outliers and noise, and then perform standardization processing to make the data have a unified dimension and range, which is convenient for subsequent model training and analysis. Extract key features from the processed data. The key features include the average value, standard deviation, change rate, etc. of the injection molding process parameters, and the maximum value, minimum value, change trend, etc. of the component deformation data. Then, based on the processed injection molding process parameters and deformation data, construct a digital twin model. This model is used to simulate the physical behavior and response of automotive injection molded components during the production process. Through model training and verification, adjust parameters such as the deformation rate to obtain the optimal digital twin model that reflects the actual production process. Use the extracted key features and combine machine learning algorithms to analyze and predict the deformation trend of automotive injection molded components. Through algorithm training and feature importance evaluation, select the features that contribute the most to deformation prediction to obtain a deformation prediction model.

[0062] It should also be clear that during the injection molding production process, use the constructed digital twin model and deformation prediction model, combine the real-time collected injection molding process parameters and deformation data, and perform real-time prediction on the deformation trend of the components. Then, according to the deviation between the real-time monitoring data and the prediction result, use an adaptive algorithm to dynamically adjust the model parameters, and optimize through a feedback loop to continuously improve the accuracy and reliability of the prediction model, thereby realizing the optimization of the injection molding process and reducing the deformation defects of the components.

[0063] The method for adjusting the prediction frequency based on the correlation between process parameters and deformation data includes:

[0064] Obtain the process parameter difference degree between the first injection molding machine monitoring device and the second injection molding machine monitoring device, and the parallel data correlation degree between the first deformation monitoring device and the second deformation monitoring device;

[0065] Calculate a first adjustment difference and a second adjustment difference based on the process parameter difference degree and the deformation data correlation degree;

[0066] Adjust the real-time data acquisition frequency of the first injection molding machine monitoring device based on the first adjustment difference and the historical process parameter data;

[0067] Adjust the update frequency of the deformation prediction model of the first deformation monitoring device based on the second adjustment difference and the historical deformation data. It should be noted that in the embodiments of the present application, calculate the process parameter difference degree between the first injection molding machine monitoring device and the second injection molding machine monitoring device. For example, the differences in statistical quantities such as the average value and standard deviation of process parameters such as injection temperature, mold temperature, injection speed, holding pressure, and cooling time of the two devices can be calculated to obtain the process parameter difference degree. Calculate the deformation data correlation degree between the first deformation monitoring device and the second deformation monitoring device. For example, the correlation of deformation data such as part thickness, warpage, and surface curve of the two devices can be calculated to obtain the deformation data correlation degree. Based on the process parameter difference degree and the deformation data correlation degree, calculate the first adjustment difference through a certain algorithm or formula. For example, methods such as linear combination and weighted average can be used to synthesize the process parameter difference degree and the deformation data correlation degree to obtain the first adjustment difference. Similarly, based on the process difference degree and the deformation data correlation degree, calculate the second adjustment difference through a certain algorithm or formula. The first adjustment difference is calculated based on the weighted average of the process parameter difference degree and the deformation data correlation degree. The specific formula is as follows: , where represents the numerical value of the process parameter difference degree or the deformation data correlation degree, represents the weight corresponding to the numerical value. The second adjustment difference is also based on the weighted average of the process parameter difference degree and the deformation data correlation degree, but the weight can be adjusted according to specific requirements. The specific formula is as follows: . In the present application, adjust the data acquisition frequency, and dynamically adjust the real-time data acquisition frequency of the first injection molding machine monitoring device based on the first adjustment difference and the historical process parameter data. For example, if the first adjustment difference is large, it means that the change in process parameters has a greater impact on deformation, and the data acquisition frequency can be appropriately increased. Based on the second adjustment difference and the historical deformation data, dynamically adjust the update frequency of the deformation prediction model of the first deformation monitoring device. For example, if the second adjustment difference is large, it means that the change in deformation data has a greater impact on the prediction model, and the model update frequency can be appropriately increased.

[0068] The classification processing method based on abnormal process parameters and abnormal deformation data includes:

[0069] Perform cluster analysis based on the current process parameter data to determine that the abnormal types are temperature fluctuation, pressure deviation, or insufficient material filling;

[0070] Perform pattern recognition based on the current abnormal deformation data to determine that the deformation type is shrinkage, warping, or dimensional out-of-tolerance;

[0071] Based on the abnormal type, obtain the corresponding third adjustment difference and fourth adjustment difference from a preset weight table;

[0072] Combine the first adjustment difference and the second adjustment difference to dynamically optimize the correlation weight between the injection molding process parameters and the deformation prediction model. It should be clear that in this application, cluster analysis is performed based on the current process parameter data to determine the abnormal type. The cluster algorithm is used to perform cluster analysis on the current process parameters. For example, calculate the similarity of process parameters such as injection temperature, mold temperature, injection speed, holding pressure, and cooling time, and classify similar parameters into one category to determine the abnormal type as temperature fluctuation, pressure deviation, or insufficient material filling. The IQR method can also be used for anomaly detection, calculate the interquartile range of each process parameter, and determine whether the data point is abnormal data to determine the abnormal type. Based on the abnormal type, obtain the corresponding third adjustment difference and fourth adjustment difference from a preset weight table. According to the determined abnormal type, look up the corresponding third adjustment difference and fourth adjustment difference in the preset weight table. For example, if the abnormal type is temperature fluctuation, the corresponding third adjustment difference is 0.3 and the fourth adjustment difference is 0.2; if the abnormal type is pressure deviation, the corresponding third adjustment difference is 0.4 and the fourth adjustment difference is 0.3; if the abnormal type is insufficient material filling, the corresponding third adjustment difference is 0.5 and the fourth adjustment difference is 0.4. In this application, combine the first adjustment difference and the second adjustment difference to dynamically optimize the correlation weight between the injection molding process parameters and the deformation prediction model. Add the first adjustment difference and the third adjustment difference to obtain the adjustment difference of the new injection molding process parameters, which is used to adjust the correlation weight of the injection molding process parameters. For example, if the first adjustment difference is 0.72 and the third adjustment difference is 0.3, the adjustment difference of the new injection molding process parameters is 1.02, and the correlation weight of the injection molding process parameters can be appropriately increased. Add the second adjustment difference and the fourth adjustment difference to obtain the adjustment difference of the new deformation prediction model, which is used to adjust the correlation weight of the deformation prediction model. For example, if the second adjustment difference is 0.6 and the fourth adjustment difference is 0.2, the adjustment difference of the new deformation prediction model is 0.8, and the correlation weight of the deformation prediction model can be appropriately increased. Through this method, the influence of abnormal process parameters and abnormal deformation data on the injection molding process parameters and the deformation prediction model can be accurately reflected, thereby optimizing the performance and prediction frequency of the prediction model.

[0073] The confidence evaluation method based on historical anomalies includes:

[0074] Extract the historical time-series data corresponding to the current abnormal process parameters and construct a process parameter change curve;

[0075] Perform exponential smoothing fitting based on the historical deformation trend corresponding to the current abnormal deformation data, and calculate the fitting residual;

[0076] If the standard deviation of the slope of the process parameter change curve exceeds the threshold or the fitting residual is higher than the preset value, reduce the confidence level of the abnormal data and trigger the manual review process. It should be clear that in the embodiments of the present application, by extracting the historical time-series data corresponding to the current abnormal process parameters (such as injection temperature, mold temperature, injection speed, holding pressure, and cooling time, etc.), a time series is formed, and then a process parameter change curve is constructed, and linear regression or other fitting methods are used to represent the change trend of the process parameters over time. For example, for the time-series data of the injection temperature, a linear regression model can be used to fit its change trend to obtain the process parameter change curve. As can be seen from the foregoing, perform exponential smoothing fitting based on the historical deformation trend corresponding to the current abnormal deformation data, calculate the fitting residual, by extracting the historical deformation trend data corresponding to the current abnormal deformation data (such as part thickness, warpage, and surface defects, etc.), and use the exponential smoothing method to fit the historical deformation trend data to obtain the fitted deformation trend curve. The calculation formula of the exponential smoothing method is: , where is the predicted value at time t, is the actual value at time t-1, a is the smoothing factor, and its value range is between 0 and 1. Calculate the fitting residual, that is, the difference between the actual value and the predicted value. The calculation formula of the residual is: .

[0077] In the present application, calculate the standard deviation of the slope of the process parameter change curve. For example, for the injection temperature change curve, calculate the standard deviation of its slope. If the standard deviation exceeds the set threshold (such as 0.5), it is considered that the change of the process parameter is relatively drastic and there may be an abnormality. Determine whether the fitting residual is higher than the preset value. For example, if the absolute value of the residual exceeds 0.3, it is considered that the fitting effect is not good and there may be an abnormality. If any of the above conditions is met, reduce the confidence level of the abnormal data and trigger the manual review process to further determine the accuracy and reliability of the abnormal data. Through the above method, the confidence level of the abnormal data can be evaluated, potential abnormalities can be discovered and processed in a timely manner, and the stability of the production process and the product quality can be improved.

[0078] The dynamic feedback control method based on real-time data includes:

[0079] Obtain the process parameters of the injection molding machine monitoring device and the deformation amount of the deformation monitoring device in real time;

[0080] Input the process parameters into the pre-trained deformation prediction model and output the real-time predicted deformation value;

[0081] Compare the deviation between the real-time predicted deformation value and the actual deformation value. If the deviation exceeds the operating range, generate a process parameter correction instruction and send it to the injection molding machine controller. It should be noted that in the embodiments of the present application, the injection molding process parameters are collected in real time by the injection molding machine monitoring device, such as injection temperature, mold temperature, injection speed, holding pressure, and cooling time, etc. The deformation amount of the injection molded parts is collected in real time by the deformation monitoring device, such as part thickness, warpage, and surface defects, etc. The process parameters collected in real time are used as inputs and input into a pre-trained deformation prediction model. This model can be a model based on machine learning or deep learning, such as neural network, support vector machine, etc. The model outputs the corresponding real-time predicted deformation value according to the input process parameters, and is expressed by the following formula: , where is the predicted deformation value, x is the input process parameter, f is the deformation prediction model. Then, by comparing the deviation between the real-time predicted deformation value and the actual deformation value, if the deviation exceeds the operating range, generate a process parameter correction instruction and send it to the injection molding machine controller. Calculate the deviation between the real-time predicted deformation value and the actual deformation value, that is: , where is the actual deformation value, is the predicted deformation value. If the deviation e exceeds the preset operating range (such as ±0.5mm), generate a process parameter correction instruction. The correction instruction can be based on the PID control algorithm to adjust the injection molding process parameters, such as adjusting the injection temperature, injection speed, etc. Send the correction instruction to the injection molding machine controller, and the controller adjusts the process parameters of the injection molding machine according to the instruction to reduce the deviation and optimize the production process. For example: Assume that the currently collected injection temperature is 250°C, the mold temperature is 80°C, the injection speed is 150mm / s, the holding pressure is 100Mpa, and the cooling time is 30s. Input these parameters into the pre-trained deformation prediction model, and the predicted deformation value is 0.3mm. The actually measured deformation value is 0.5mm, then the deviation is 0.2mm. If the preset operating range is ±0.1mm, the deviation exceeds the range, and a correction instruction needs to be generated. According to the PID control algorithm, adjust the injection temperature to decrease by 5°C, the injection speed to decrease by 10mm / s, etc. to reduce the deformation value. The corrected process parameters are sent to the injection molding machine controller for real-time adjustment. Through the above method, dynamic feedback control based on real-time data can be realized, the injection molding process parameters can be adjusted in time, the production process can be optimized, and the product quality can be improved.

[0082] The deformation warning method based on multi-source data fusion includes:

[0083] Fuse the mold temperature, holding time, cooling rate parameters of the injection molding machine and the three-dimensional scan data of the parts;

[0084] Calculate the influence weight of each process parameter on deformation through the grey relational analysis algorithm;

[0085] Generate a priority warning signal according to the weight distribution and push it to the production management system. It should be clear that in the embodiments of this application, process parameters such as the mold temperature, holding pressure time, and cooling rate of the injection molding machine, as well as the three-dimensional scan data of the components, are collected, and these data are preprocessed, including data cleaning, standardization, etc., to ensure the consistency and availability of the data. The process parameters and three-dimensional scan data are fused to form a comprehensive data set for subsequent analysis, and the reference sequence and comparison sequence are determined. Calculate the grey relational degree between each comparison sequence and the reference sequence. The formula is as follows:

[0086] , where is the reference sequence, is the comparison sequence, p is the distinguishing coefficient (0.5). Calculate the mean value of the grey relational degrees of each comparison sequence as the influence weight of this parameter on deformation. According to the weight sizes of each parameter, determine the priority of the warning signal. For example, the warning signal corresponding to the parameter with the highest weight has the highest priority. Then, generate a warning signal and push it to the production management system through the communication interface of the production management system. According to the weight sizes, the weight of the mold temperature is the highest, so the corresponding warning signal has the highest priority. Generate a production warning signal and push it to the production management system.

[0087] A deformation prediction system for automotive injection molded parts, including a plurality of injection molding machine monitoring devices and component deformation monitoring devices. The injection molding machine monitoring devices are used to collect temperature, pressure, and injection speed data, and the deformation monitoring devices are used to obtain the size and deformation data of the components. It is characterized in that the system further includes:

[0088] A data integration model, used to synchronize the process parameters of the injection molding machine and the deformation monitoring data, and mark the spatio-temporal correlation. Through the data acquisition module, real-time obtain the process parameters such as the temperature, pressure, and injection speed of the injection molding machine, as well as the monitoring data such as the size and deformation amount of the components, and store these data in the database. At the same time, use the spatio-temporal data model to mark the time and space information of each data point and establish the spatio-temporal correlation between the data. For example, the time stamp of each injection cycle can be recorded, as well as the position information of each component in the mold.

[0089] An intelligent analysis module is used to execute the calculation logic of the method described in any one of claims 1 to 6, and output the adjustment difference and the prediction result. According to the preset algorithms and models, it analyzes and calculates the collected data. For example, it uses the grey relational analysis algorithm to calculate the influence weights of various process parameters on deformation, and uses the PID control algorithm to calculate the adjustment difference, etc. According to the calculation results, it outputs the adjustment difference and the prediction result. For example, it outputs that the adjustment difference of the injection molding temperature is -5°C, the adjustment difference of the injection speed is -10 mm / s, etc., and the predicted deformation is 0.3 mm, etc.

[0090] An adaptive control unit dynamically adjusts the injection molding machine parameters or triggers a mold calibration instruction according to the prediction result. According to the adjustment difference output by the intelligent analysis module, it dynamically adjusts the process parameters of the injection molding machine. For example, if it is predicted that the deformation amount is large, the injection molding temperature and the injection speed can be appropriately reduced to reduce the deformation. If the prediction result indicates that there may be a calibration problem with the mold, a mold calibration instruction can be generated to remind the operator to check and calibrate the mold.

[0091] In this application, the grey relational analysis algorithm calculates the weight as follows: Let the reference sequence be the deformation amount D = [0.1, 0.2, 0.3, 0.4, 0.5], and the comparison sequences be the injection molding temperature T = [80, 85, 90, 95, 100], the holding pressure time H = [5, 10, 15, 20, 25], and the cooling rate C = [2, 4, 6, 8, 10]. Calculate the grey relational degree between each comparison sequence and the reference sequence. For the injection molding temperature T, calculate the difference at each time point , and then calculate the grey relational degree according to the formula . Similarly, calculate the grey relational degrees of the holding pressure time H and the cooling rate C, and calculate the average value of the grey relational degrees of each comparison sequence as the influence weight of this parameter on deformation. For example, the weight of the injection molding temperature is 0.4, the holding pressure time is 0.3, and the cooling rate is 0.3.

[0092] In this application, the PID control algorithm calculates the adjustment difference: By setting the target deformation amount to 0.2 mm and the actual deformation amount to 0.3 mm, then the deviation e = 0.1 mm. According to the PID control algorithm, calculate the adjustment difference. For example, the proportional term Kp = 0.5, the integral term Ki = 0.1, and the derivative term Kd = 0.05, then the adjustment difference , According to the adjustment difference, adjust the process parameters of the injection molding machine. For example, adjust the injection molding temperature to decrease by 5°C and the injection speed to decrease by 10 mm / s. Through the above system and method, the deformation of automotive injection molded parts can be effectively predicted and controlled, improving production efficiency and product quality.

[0093] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended embodiments and their equivalents.

Claims

1. A method for predicting the deformation of automotive injection-molded parts, which is applied to a deformation prediction system. The system includes a plurality of injection molding machine monitoring devices and part deformation monitoring devices. The injection molding machine monitoring devices are used to collect injection molding process parameter data, and the part deformation monitoring devices are used to monitor the real-time deformation data of the injection-molded parts. It is characterized in that, The method includes: Obtaining historical process parameter data of a first injection molding machine monitoring device, where the first injection molding machine monitoring device is any one of multiple injection molding machine monitoring devices; Obtaining current abnormal process parameter data of a second injection molding machine monitoring device, where the second injection molding machine monitoring device is any device that monitors abnormal process parameters; Obtaining historical deformation data of a first deformation monitoring device, where the first deformation monitoring device is any one of multiple deformation monitoring devices; Obtaining current abnormal deformation data of a second deformation monitoring device, where the second deformation monitoring device is any device that monitors abnormal deformation.

2. The deformation prediction method of an automotive injection molded part according to claim 1, characterized in that: The method for adjusting the prediction frequency based on the relevance between process parameters and deformation data includes: Obtaining the process parameter difference degree between the first injection molding machine monitoring device and the second injection molding machine monitoring device, and the parallel data correlation degree between the first deformation monitoring device and the second deformation monitoring device; Calculating a first adjustment difference and a second adjustment difference based on the process parameter difference degree and the deformation data correlation degree; Adjusting the real-time data acquisition frequency of the first injection molding machine monitoring device based on the first adjustment difference and the historical process parameter data; Adjusting the deformation prediction model update frequency of the first deformation monitoring device based on the second adjustment difference and the historical deformation data.

3. A method for predicting deformation of an automotive injection molded part according to claim 1, characterized in that: The classification processing method based on abnormal process parameters and abnormal deformation data includes: Performing cluster analysis based on the current process parameter data to determine that the abnormal types are temperature fluctuation, pressure deviation, or insufficient material filling; Performing pattern recognition based on the current abnormal deformation data to determine that the deformation types are shrinkage, warping, or dimensional over-tolerance; Obtaining corresponding third adjustment difference and fourth adjustment difference from a preset weight table based on the abnormal type; Combining the first adjustment difference and the second adjustment difference to dynamically optimize the correlation weight between the injection molding process parameters and the deformation prediction model.

4. A method for predicting deformation of an automotive injection molded part according to claim 1, characterized in that: The confidence evaluation method based on historical anomalies includes: Extracting historical time series data corresponding to the current abnormal process parameters and constructing a process parameter change curve; Performing exponential smoothing fitting based on the historical deformation trend corresponding to the current abnormal deformation data and calculating the fitting residual; If the standard deviation of the slope of the process parameter change curve exceeds the threshold or the fitting residual is higher than the preset value, then reduce the confidence of the abnormal data and trigger an artificial review process.

5. A method for predicting deformation of an automotive injection molded part according to claim 1, characterized in that: The dynamic feedback control method based on real-time data includes: Obtaining in real time the process parameters of the injection molding machine monitoring device and the deformation amount of the deformation monitoring device; Inputting the process parameters into a pre-trained deformation prediction model and outputting a real-time predicted deformation value; Comparing the deviation between the real-time predicted deformation value and the actual deformation value. If the deviation exceeds the operating range, then generate a process parameter correction instruction and send it to the injection molding machine controller.

6. A method for predicting deformation of an automotive injection molded part according to claim 1, characterized in that: The deformation warning method based on multi-source data fusion includes: Fusing the mold temperature, holding pressure time, cooling rate parameters of the injection molding machine and the three-dimensional scan data of the components; Calculating the influence weight of each process parameter on deformation through a grey relational analysis algorithm; Generating a priority warning signal according to the weight distribution and pushing it to the production management system.

7. A deformation prediction system for automotive injection molded parts, comprising a plurality of injection molding machine monitoring devices and part deformation monitoring devices. The injection molding machine monitoring devices are used to collect temperature, pressure and injection speed data, and the deformation monitoring devices are used to obtain the dimensions and deformation data of the parts, characterized in that, The system further includes: A data integration model for synchronizing the process parameters of an injection molding machine and deformation monitoring data, and annotating the spatio-temporal correlation; An intelligent analysis module for executing the calculation logic of any of the methods described in claims 1 to 6, and outputting an adjustment difference and a prediction result; An adaptive control unit for dynamically adjusting the parameters of the injection molding machine or triggering a mold calibration instruction according to the prediction result.

Images