Injection mold man-machine interaction intelligent design optimization method and system
By collecting and processing data in real time through a sensor array, and combining regression analysis and machine learning models, a human-machine interaction system for injection molds was constructed. This system solved the problems of inaccurate parameter optimization and lagging quality monitoring in injection mold processing, and achieved efficient production process optimization and quality control.
Patent Information
- Application Number
- CN202411018377.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-29
AI Technical Summary
In the existing injection mold processing technology, inaccurate parameter optimization and lagging quality monitoring lead to uneven mold temperature, improper control of plastic material expansion, and unstable crystallization rate, affecting the quality of finished products and production efficiency.
A sensor array is used to collect mold processing data in real time. Data processing is performed by combining Kalman filter and wavelet transform. Regression analysis and machine learning models are established, and analysis formulas for the relationship between temperature, material change and crystallization are constructed. Optimization and adjustment are performed through a human-computer interaction interface, and quality thresholds are set for real-time evaluation and optimization.
It achieves high-precision data acquisition and processing, provides accurate mold processing datasets, improves finished product quality and production efficiency, and significantly enhances the control capability and efficiency of the production process through multi-dimensional analysis and automated optimization.
Smart Images

Figure CN119026195A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of injection molds, and in particular to a method and system for optimizing the human-machine interactive intelligent design of injection molds. Background Art
[0002] Injection mold human-computer interaction technology is an important part of realizing intelligent manufacturing and Industry 4.0. It improves user experience and operating efficiency by designing a friendly and intuitive interface. In the injection molding process, the HMI system can monitor and control key parameters in real time, provide intelligent alarm and fault diagnosis functions, optimize process parameters, and support remote monitoring and maintenance. In the future, with the development of technologies such as artificial intelligence, virtual reality and big data, injection mold human-computer interaction technology will develop in a more intelligent, automated and collaborative direction, further improving production efficiency and product quality.
[0003] At present, traditional injection mold processing technology has obvious defects in parameter optimization and quality control, mainly manifested in problems such as inaccurate parameter adjustment and lagging quality monitoring. These shortcomings are due to the lack of comprehensive and real-time data support for the processing process, which leads to abnormal conditions such as uneven mold temperature, improper control of plastic material expansion and unstable crystallization rate in actual production. These abnormalities will not only reduce the quality of the finished product, but also increase the scrap rate, affecting production efficiency and cost control. Summary of the invention
[0004] In view of the deficiencies in the prior art, the present invention provides an injection mold human-machine interactive intelligent design optimization method and system, which solves the problems mentioned in the background technology.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: including an injection molding parameter acquisition module, a finished product quality relationship module, a processing analysis algorithm module, a human-computer interaction module and an intelligent analysis and evaluation module;
[0006] The injection molding parameter acquisition module uses sensors and monitoring equipment to collect mold processing data in real time during the mold re-injection process, and pre-processes the collected mold processing data to obtain a mold processing data set;
[0007] The finished product quality relationship module is used to utilize the collected mold processing data set, historical data and experimental data, and divide them into a training set, a validation set and a test set to establish a regression analysis model and a machine learning model between the mold processing data and the finished product quality;
[0008] The processing analysis algorithm module is used to construct a mold temperature distribution analysis formula, a plastic material change analysis formula and a crystallization relationship analysis formula, and load the collected mold processing data set to calculate and obtain the temperature distribution T (x, y, z), the plastic material change rate Bh and the crystallization rate Jj, and perform correlation and summary to produce comprehensive processing data JG;
[0009] The human-computer interaction module is used to use development tools to build a visual interface, view optimization results and input parameters for optimization adjustment;
[0010] The intelligent analysis and evaluation module is used to preset the first processing quality threshold value Y1 and the second processing quality threshold value Y2 and the obtained comprehensive processing data JG, and compare and evaluate the quality of the current injection mold. When an abnormality is identified, a comprehensive optimization formula is further constructed to calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine.
[0011] Preferably, the injection molding parameter acquisition module includes a parameter acquisition module and a data preprocessing module;
[0012] The parameter acquisition module is used to install a sensor group around the re-injection molding machine, and the sensor group includes a roughness sensor, a thermocouple sensor, an infrared thermometer, a pressure sensor and a flow meter to collect mold processing data of the mold re-injection molding process in real time. The mold processing data includes mold surface roughness Ra, thermal conductivity Km, plastic material expansion coefficient Ap and shear rate Jq;
[0013] The data preprocessing module is used to filter the mold processing data collected in real time, use a Kalman filter to smooth the data, and use wavelet transform to remove the high-frequency noise of the mold processing data collected by the sensor, and normalize the mold processing data collected at home.
[0014] Preferably, the finished product quality relationship module includes a regression analysis model unit, a machine learning model unit and a model optimization and verification unit;
[0015] The regression analysis model unit is used to construct a multivariate linear regression model, identify the linear relationship between mold processing data and finished product quality, analyze the prediction performance and regression coefficient of the model, and transmit the preprocessed mold data to the regression analysis model to evaluate the impact of each parameter on the quality of the finished product;
[0016] The machine learning model unit is used to select a support vector machine SVM and a multi-layer perceptron, and respectively construct a support vector machine SVM model and a neural network NN model. The support vector machine SVM model is used to capture the nonlinear relationship between mold processing data and finished product quality, and the neural network NN model is used to capture complex nonlinear relationships. The constructed support vector machine SVM model and the neural network NN model are integrated and summarized to obtain a machine learning model. The collected mold processing data set, historical data and experimental data are pre-processed and then transmitted to the machine learning model. The prediction performance of the support vector machine SVM model and the neural network NN model are respectively evaluated, and the linear and nonlinear relationship between the data is analyzed;
[0017] The model optimization and verification unit uses recursive feature elimination (RFE) to identify variables that affect the model's prediction performance and eliminates irrelevant and redundant variables, thereby simplifying the model and improving the model's interpretability and prediction accuracy. K-fold cross validation is then used to divide the mold processing data into K subsets, one of the subsets is used for verification each time, and the remaining K-1 subsets are used for training, which is repeated K times. The result is the average of the K verifications. At the same time, grid search is used to traverse all hyperparameter combinations to select a combination that optimizes the model performance. After the machine learning model is optimized, it is necessary to verify the machine learning model to evaluate the performance of the machine learning model and ensure that it is suitable for practical applications.
[0018] Preferably, the processing analysis algorithm module includes a mold temperature analysis algorithm unit, a material change algorithm unit, a crystallization algorithm unit and a comprehensive analysis algorithm unit;
[0019] The temperature analysis algorithm unit is used to construct a mold temperature distribution analysis formula, and to analyze and calculate the temperature distribution T (x, y, z) according to the thermal conductivity Km and the mold surface roughness Ra in the mold processing data set after preprocessing;
[0020] The temperature distribution T(x, y, z) is calculated and obtained by the following mold temperature distribution analysis formula;
[0021]
[0022] Wherein, T0 represents the initial temperature, which indicates the temperature of the mold when there is no heat source, and is directly measured by an infrared thermometer. Q represents the intensity of the heat source, which indicates the heat intensity provided by the heater or cooling system, and is determined by calculating the thermal power provided by the injection molding machine per unit time. x, y and z are the horizontal axis, vertical axis and vertical axis respectively, representing the coordinates of the mold in three-dimensional space. These coordinates are usually related to the geometric shape and specific spatial position of the mold. The mold surface roughness Ra affects the heat conduction efficiency of the mold surface, thereby affecting the temperature distribution of the mold. The thermal conductivity Km determines the thermal conductivity of the mold material and affects the temperature gradient inside the mold. exp represents the exponential function, which is used to describe the attenuation characteristics of the temperature distribution.
[0023] Preferably, the material change algorithm unit is used to construct a plastic material change analysis formula, and to analyze and calculate the plastic material change rate Bh based on the plastic material expansion coefficient Ap in the preprocessed mold processing data set;
[0024] The plastic material change rate Bh is calculated and obtained by the following plastic material change analysis formula:
[0025]
[0026] In the formula, V0 represents the initial volume of the plastic, the expansion coefficient of the plastic material Ap represents the degree of volume expansion of the material when the temperature changes, △T represents the temperature change, which represents the change value of the material temperature relative to the initial temperature T0, βm represents the volume change rate coefficient, which represents the change of the material volume with the temperature change rate, d represents the integral variable, dt represents the variable per unit time t, and T represents the temperature, which is measured by the thermocouple and the infrared thermometer. It represents the rate of change, which indicates the change in temperature T per unit time t.
[0027] Preferably, the crystallization algorithm unit is used to construct a crystallization relationship analysis formula, and analyze and calculate the crystallization rate Jj based on the shear rate Jq in the mold processing data set after preprocessing;
[0028] The crystallization rate Jj is calculated by the following crystallization relationship analysis formula:
[0029]
[0030] In the formula, k1 represents the rate constant, which represents the crystallization rate without the action of shear rate, Ea represents the activation energy, which represents the energy barrier that needs to be overcome in the crystallization process, R represents the gas constant, Jq0 represents the reference shear rate, n represents the shear rate exponent, which represents the sensitivity of the material to the shear rate, and exp represents the exponential function.
[0031] Preferably, the comprehensive analysis algorithm unit is used to correlate and summarize the obtained temperature distribution T (x, y, z), plastic material change rate Bh and crystallization rate Jj to obtain comprehensive processing data JG, and analyze the current injection molding machine processing quality;
[0032] The comprehensive processing data JG is obtained by calculation through the following algorithm formula;
[0033]
[0034] Where V is the volume integral, covering the entire three-dimensional space, and S represents the overall performance of the system.
[0035] Preferably, the human-computer interaction module is used to design a preliminary interface using the QtDesigner interface design tool, and define the interface layout and the position and function of each control, use PyQt and PyGObject libraries to implement data input, optimization calculation and result display functions, and then embed graphics into the GUI through Matplotlib and D3.js to display data graphically and achieve dynamic interaction.
[0036] Preferably, the energy analysis and evaluation module includes a quality evaluation unit and an optimization algorithm unit;
[0037] The quality assessment unit is used to preset a first processing quality threshold value Y1 and a second processing quality threshold value Y2 based on the mean value of the injection mold processing quality, and then compare and evaluate with the acquired comprehensive processing data JG to analyze the current mold processing quality. The specific assessment scheme is as follows;
[0038] When the comprehensive processing data JG> the first processing quality threshold Y1, it means that the processing quality of the current injection molding machine is abnormal, and the optimization algorithm is further executed to optimize the parameters;
[0039] When the second processing quality threshold Y2≤comprehensive processing data JG≤first processing quality threshold Y1, it indicates that the processing quality of the injection molding machine is normal;
[0040] When the comprehensive processing data JG is less than the second processing quality threshold value Y2, it indicates that the processing quality of the injection molding machine is abnormal. Then, the optimization algorithm is further executed to optimize the parameters.
[0041] The optimization algorithm unit is used to construct a comprehensive optimization formula, substitute the obtained temperature distribution T(x, y, z), plastic material change rate Bh and crystallization rate Jj into the algorithm formula, calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine, find the best process parameter combination that minimizes the optimization target through iterative calculation, and adjust and optimize the parameters through the interactive interface according to the calculation results;
[0042] The comprehensive optimization coefficient Yh is obtained by the following algorithm formula;
[0043]
[0044] Where α, β and γ represent the weight coefficients of temperature optimization, volume change control and crystallization rate control, respectively. target Indicates the target temperature set during design, Bh target represents the target volume change, Jj target It represents the target crystallization rate Jj set during design;
[0045] α·∫ V T(x,y,z)-T target dV is used to optimize the temperature distribution inside the mold so that the processing temperature is close to the target temperature T target , by minimizing the error of temperature distribution to ensure the uniformity of mold heating, thereby improving the quality and performance of the injection molded product;
[0046] β·Bh-Bh target |Used to control the volume change of plastic during the injection molding process, so that it is close to the designed target volume V after control target , by minimizing volume variation errors to ensure that the size and quality of the finished product meet the design specifications, thereby improving production consistency and accuracy;
[0047] γ·|Jj-Jj target |Used to adjust the crystallization rate Jj of the plastic to make it close to the target crystallization rate Jj target , by minimizing the crystallization rate error to control the crystallization process of the plastic, thereby optimizing the performance of the material and the structural stability of the finished product.
[0048] The human-computer interaction intelligent design optimization method for injection molds includes the following steps:
[0049] S1. First, a sensor group is installed around the injection molding machine to collect mold processing data of the mold during the injection molding process in real time, and the collected mold processing data is preprocessed to obtain a mold processing data set;
[0050] S2. Using the collected mold processing data set, historical data and experimental data, and dividing them into training set, validation set and test set, a regression analysis model and a machine learning model are established between the mold processing data and the finished product quality;
[0051] S3, construct mold temperature distribution analysis formula, plastic material change analysis formula and crystallization relationship analysis formula, and load the collected mold processing data set to calculate and obtain temperature distribution T (x, y, z), plastic material change rate Bh and crystallization rate Jj, and correlate and summarize the production comprehensive processing data JG;
[0052] S4. Use development tools to build a visual interface, define the interface layout and the location and function of each control, implement data input, optimization calculation and result display functions, display data graphically to achieve dynamic interaction, view optimization results and input parameters for optimization adjustment;
[0053] S5. Finally, the first processing quality threshold value Y1 and the second processing quality threshold value Y2 are preset and compared with the obtained comprehensive processing data JG to evaluate and analyze the quality of the current injection mold. When an abnormality is identified, a comprehensive optimization formula is further constructed to calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine.
[0054] The present invention provides an injection mold human-machine interactive intelligent design optimization method and system. It has the following beneficial effects:
[0055] (1) The system can collect mold processing data including mold surface roughness Ra, thermal conductivity Km, plastic material expansion coefficient Ap and shear rate Jq in real time by installing a sensor group around the injection molding machine. These sensor data are processed by Kalman filter and smoothing, combined with wavelet transform to remove high-frequency noise, and normalized to ensure high accuracy and reliability of the data. Through these precise data collection and processing, the system provides a solid foundation for subsequent processing quality analysis and optimization. Through high-precision data collection and processing technology, the system can provide accurate mold processing data sets, ensuring the data quality in the subsequent analysis and modeling process, thus laying the foundation for improving the quality of finished products and improving production efficiency.
[0056] (2) The system uses regression analysis models and machine learning models, combined with training sets, validation sets, and test sets to conduct an in-depth analysis of the relationship between mold processing data and finished product quality. By establishing a multivariate linear regression model to identify the linear relationship between mold processing data and finished product quality, and using support vector machine SVM and neural network NN models to capture complex nonlinear relationships, the system can comprehensively evaluate the impact of mold processing data on finished product quality. By introducing multi-level analysis models, including regression analysis and machine learning models, the system can predict and optimize processing quality from multiple angles, providing more accurate and reliable analysis tools and improving the model's prediction ability and optimization effect.
[0057] (3) The system presets the first processing quality threshold Y1 and the second processing quality threshold Y2 through the intelligent analysis and evaluation module, and compares and evaluates them with the acquired comprehensive processing data JG. The system can analyze the processing quality of the current injection mold in real time. Once an abnormality is identified, it triggers the calculation of the comprehensive optimization formula and obtains the comprehensive optimization coefficient Yh to guide the injection molding machine to adjust and optimize parameters. The comprehensive optimization formula considers factors such as temperature distribution T(x, y, z), plastic material change rate Bh and crystallization rate Jj, and finds the optimal process parameter combination through iterative calculation, thereby maximizing the efficiency and quality of the mold processing process. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1 This is a schematic diagram of the process flow of the human-machine interactive intelligent design optimization system for injection molds of the present invention;
[0059] Figure 2 Schematic diagram of the steps of the human-computer interaction intelligent design optimization method for injection molds of the present invention. DETAILED DESCRIPTION
[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0061] Example 1
[0062] See also Figure 1 , the present invention provides an injection mold human-computer interaction intelligent design optimization system. To achieve the above purpose, the present invention is implemented through the following technical solutions: including an injection molding parameter acquisition module, a finished product quality relationship module, a processing analysis algorithm module, a human-computer interaction module and an intelligent analysis and evaluation module;
[0063] The injection molding parameter acquisition module uses sensors and monitoring equipment to collect mold processing data during the injection molding process in real time, and pre-processes the collected mold processing data to obtain the mold processing data set;
[0064] The finished product quality relationship module is used to utilize the collected mold processing data set, historical data, and experimental data, and divide them into training sets, validation sets, and test sets to establish a regression analysis model and a machine learning model between mold processing data and finished product quality;
[0065] The processing analysis algorithm module is used to construct the mold temperature distribution analysis formula, the plastic material change analysis formula and the crystallization relationship analysis formula, and load the collected mold processing data set to calculate and obtain the temperature distribution T (x, y, z), the plastic material change rate Bh and the crystallization rate Jj, and then summarize the production comprehensive processing data JG.
[0066] The human-computer interaction module is used to use development tools to build a visual interface, view optimization results, and input parameters for optimization adjustment;
[0067] The intelligent analysis and evaluation module is used to preset the first processing quality threshold Y1 and the second processing quality threshold Y2 and the obtained comprehensive processing data JG, and compare and evaluate the quality of the current injection mold. When an abnormality is identified, a comprehensive optimization formula is further constructed to calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine.
[0068] In this embodiment, the system realizes the comprehensive optimization of the injection molding process by integrating the injection molding parameter acquisition module, the finished product quality relationship module, the processing analysis algorithm module, the human-computer interaction module and the intelligent analysis and evaluation module. Data acquisition and data preprocessing technology provide high-quality processing data, and regression analysis and machine learning models improve the prediction accuracy of finished product quality. Compared with the traditional empirical adjustment method, the system provides a more scientific and accurate production process optimization scheme, thereby significantly improving production efficiency and product quality. The processing analysis algorithm module obtains the temperature distribution T (x, y, z), the plastic material change rate Bh and the crystallization rate Jj by calculating, and performs related calculations to obtain the comprehensive processing data JG, and deeply understands the key factors in the injection molding process. The human-computer interaction module simplifies the operation process through a visual interface, which is convenient for users to optimize and adjust. Compared with the traditional single factor analysis technology, the multi-dimensional analysis method and intuitive user interface of the system greatly improve the control ability of the production process and the convenience of user operation. The intelligent analysis and evaluation module automatically identifies production anomalies and performs optimization and adjustment by setting the first processing quality threshold Y1, the second processing quality threshold Y2 and the comprehensive optimization formula. This intelligent evaluation and optimization method is more efficient than manual inspection, and the ability of real-time monitoring and automatic adjustment significantly improves the efficiency and quality control level of the production process. Compared with existing technologies, this system has shown significant advantages in optimization strategy and automatic control, providing a more advanced solution for the injection molding industry.
[0069] Example 2
[0070] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the injection molding parameter acquisition module includes a ,parameter acquisition module and a data preprocessing module;
[0071] The parameter acquisition module is used to install a sensor group around the re-injection molding machine. The sensor group includes a roughness sensor, a thermocouple sensor, an infrared thermometer, a pressure sensor and a flow meter to collect the mold processing data of the mold re-injection molding process in real time. The mold processing data includes mold surface roughness Ra, thermal conductivity Km, plastic material expansion coefficient Ap and shear rate Jq;
[0072] The data preprocessing module is used to filter the mold processing data collected in real time, use the Kalman filter to smooth the data, and use the wavelet transform to remove the high-frequency noise of the mold processing data collected by the sensor, and normalize the mold processing data collected at home.
[0073] In this embodiment, the system significantly improves the accuracy of data acquisition and the efficiency of data processing by integrating multiple sensors and advanced data processing technologies. The parameter acquisition module obtains key mold processing data in real time and comprehensively monitors the status of the injection molding process. The data preprocessing module uses Kalman filter and wavelet transform to remove data noise and perform data normalization, providing a high-quality data foundation for subsequent analysis. This design not only improves the reliability of data acquisition and processing, but also lays a solid foundation for quality control and optimization analysis of the production process.
[0074] Example 3
[0075] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the finished product quality relationship module includes a regression analysis model unit, a ,machine learning model unit and a model optimization and verification unit;
[0076] The regression analysis model unit is used to construct a multivariate linear regression model, identify the linear relationship between mold processing data and finished product quality, analyze the model's prediction performance and regression coefficient, and transmit the preprocessed mold data to the regression analysis model to evaluate the impact of each parameter on the finished product quality;
[0077] The machine learning model unit is used to select a support vector machine (SVM) and a multi-layer perceptron, and respectively construct a support vector machine (SVM) model and a neural network (NN) model. The support vector machine (SVM) model is used to capture the nonlinear relationship between mold processing data and finished product quality, and the neural network (NN) model is used to capture complex nonlinear relationships. The constructed support vector machine (SVM) model and the neural network (NN) model are integrated and summarized to obtain a machine learning model. The collected mold processing data set, historical data, and experimental data are then preprocessed and transmitted to the machine learning model. The prediction performance of the support vector machine (SVM) model and the neural network (NN) model are evaluated respectively, and the linear and nonlinear relationships between the data are analyzed.
[0078] The model optimization and verification unit uses recursive feature elimination (RFE) to identify variables that affect the model's predictive performance and eliminates irrelevant and redundant variables, thereby simplifying the model and improving the model's interpretability and predictive accuracy. K-fold cross-validation is then used to divide the mold processing data into K subsets, one of which is used for verification each time, and the remaining K-1 subsets are used for training. This is repeated K times, and the result is the average of the K verifications. At the same time, grid search is used to traverse all hyperparameter combinations and select the combination that optimizes the model performance. After optimizing the machine learning model, it is necessary to verify the machine learning model to evaluate its performance and ensure that it is suitable for practical applications.
[0079] In this embodiment, the system realizes the in-depth analysis and optimization of the complex relationship between mold processing data and finished product quality by integrating regression analysis model unit, machine learning model unit and model optimization and verification unit. The regression analysis model unit uses the multivariate linear regression model to analyze the linear relationship between mold processing data and finished product quality in detail, clarifies the direct impact of each parameter on finished product quality, and provides basic data for optimization. The machine learning model unit uses two advanced machine learning technologies, support vector machine SVM and neural network NN model, to capture the nonlinear relationship between mold processing data and finished product quality, build a comprehensive prediction model, and improve the accuracy and robustness of prediction through model integration. Finally, the model optimization and verification unit combines recursive feature elimination RFE and K-fold cross validation technology to optimize the feature selection and hyperparameter adjustment of the model, and improve the prediction accuracy and applicability of the model. This multi-level modeling and optimization process not only improves the prediction ability of finished product quality, but also greatly enhances the reliability and practicality of the model, providing strong support for quality control and process improvement in injection molding production.
[0080] Example 4
[0081] This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the processing analysis algorithm module includes the mold temperature ,analysis algorithm unit, material change algorithm unit, crystallization algorithm unit and ,comprehensive analysis algorithm unit;
[0082] The temperature analysis algorithm unit is used to construct the mold temperature distribution analysis formula, and analyze and calculate the temperature distribution T (x, y, z) based on the thermal conductivity Km and mold surface roughness Ra in the mold processing data set after preprocessing;
[0083] The temperature distribution T(x, y, z) is calculated and obtained by the following mold temperature distribution analysis formula;
[0084]
[0085] Wherein, T0 represents the initial temperature, which indicates the temperature of the mold when there is no heat source, and is directly measured by an infrared thermometer. Q represents the intensity of the heat source, which indicates the heat intensity provided by the heater or cooling system, and is determined by calculating the thermal power provided by the injection molding machine per unit time. x, y and z are the horizontal axis, vertical axis and vertical axis respectively, representing the coordinates of the mold in three-dimensional space. These coordinates are usually related to the geometric shape and specific spatial position of the mold. The mold surface roughness Ra affects the heat conduction efficiency of the mold surface, thereby affecting the temperature distribution of the mold. The thermal conductivity Km determines the thermal conductivity of the mold material and affects the temperature gradient inside the mold. exp represents the exponential function, which is used to describe the attenuation characteristics of the temperature distribution.
[0086] The material change algorithm unit is used to construct a plastic material change analysis formula, and to analyze and calculate the plastic material change rate Bh based on the plastic material expansion coefficient Ap in the pre-processed mold processing data set;
[0087] The plastic material change rate Bh is calculated and obtained by the following plastic material change analysis formula;
[0088]
[0089] In the formula, V0 represents the initial volume of the plastic, the expansion coefficient of the plastic material Ap represents the degree of volume expansion of the material when the temperature changes, △T represents the temperature change, which represents the change value of the material temperature relative to the initial temperature T0, βm represents the volume change rate coefficient, which represents the change of the material volume with the temperature change rate, d represents the integral variable, dt represents the variable per unit time t, and T represents the temperature, which is measured by the thermocouple and the infrared thermometer. It represents the rate of change, which indicates the change in temperature T per unit time t.
[0090] The crystallization algorithm unit is used to construct a crystallization relationship analysis formula, and to analyze and calculate the crystallization rate Jj based on the shear rate Jq in the mold processing data set after preprocessing;
[0091] The crystallization rate Jj is calculated by the following crystallization relationship analysis formula;
[0092]
[0093] In the formula, k1 represents the rate constant, which represents the crystallization rate without the action of shear rate, Ea represents the activation energy, which represents the energy barrier that needs to be overcome in the crystallization process, R represents the gas constant, Jq0 represents the reference shear rate, n represents the shear rate exponent, which represents the sensitivity of the material to the shear rate, and exp represents the exponential function.
[0094] The comprehensive analysis algorithm unit is used to correlate and summarize the obtained temperature distribution T (x, y, z), plastic material change rate Bh and crystallization rate Jj to obtain comprehensive processing data JG and analyze the current injection molding machine processing quality;
[0095] The comprehensive processing data JG is obtained by calculation through the following algorithm formula;
[0096]
[0097] Where V is the volume integral, covering the entire three-dimensional space, and S represents the overall performance of the system.
[0098] In this embodiment, the system realizes comprehensive analysis and optimization of key factors of the injection molding process by integrating the mold temperature analysis algorithm unit, the material change algorithm unit, the crystallization algorithm unit and the comprehensive analysis algorithm unit. The temperature analysis algorithm unit uses the mold temperature distribution analysis formula to accurately calculate the temperature distribution T(x, y, z) inside the mold through the thermal conductivity Km and the mold surface roughness Ra, providing a quantitative basis for temperature control. The material change algorithm unit constructs a plastic material change analysis formula based on the plastic material expansion coefficient Ap, calculates the material expansion rate Bh of the plastic under temperature change, and reveals the influence of temperature change on the performance of the plastic. The crystallization algorithm unit calculates the crystallization rate Jj through the crystallization relationship analysis formula and the shear rate Jq, helping to optimize the crystallization process of the plastic. Finally, the comprehensive analysis algorithm unit associates the temperature distribution T(x, y, z), the plastic material change rate Bh and the crystallization rate Jj to obtain the comprehensive processing data JG for comprehensive processing quality evaluation. This module not only provides accurate physical process modeling and quantitative analysis for the injection molding process, but also improves the prediction ability of production quality through comprehensive calculation, and provides scientific data support for injection molding process optimization.
[0099] Example 5
[0100] This embodiment is explained in Example 1, please refer to Figure 1 Specifically: The human-computer interaction module is used to design the preliminary interface using the QtDesigner interface design tool, and define the interface layout and the position and function of each control. The PyQt and PyGObject libraries are used to implement data input, optimization calculation and result display functions. Then, Matplotlib and D3.js are used to embed graphics into the GUI to display data graphically and realize dynamic interaction.
[0101] In this embodiment, the system uses QtDesigner to design a preliminary interface, and combines PyQt and PyGObject libraries to implement data input, optimization calculation and result display. The human-computer interaction module provides a feature-rich and intuitive user interface. This design allows users to easily input data and adjust optimization parameters, and view calculation results in real time. Then, through the application of Matplotlib and D3.js, the data graphical display is embedded in the GUI, realizing dynamic data visualization and interactive operation functions, greatly improving the user's understanding and analysis of data. The human-computer interaction module provides intelligent production optimization and adjustment support through integrated data input, optimization calculation and result display functions, making the optimization of the production process more efficient and scientific.
[0102] Example 6
[0103] This embodiment is explained in Example 4. Please refer to Figure 1 ,Specifically: the intelligent analysis and evaluation module includes a quality ,assessment unit and an optimization algorithm unit;
[0104] The quality assessment unit is used to preset a first processing quality threshold value Y1 and a second processing quality threshold value Y2 based on the mean value of the injection mold processing quality, and then compare and evaluate with the acquired comprehensive processing data JG to analyze the current mold processing quality. The specific assessment scheme is as follows;
[0105] When the comprehensive processing data JG> the first processing quality threshold Y1, it means that the processing quality of the current injection molding machine is abnormal, and the optimization algorithm is further executed to optimize the parameters;
[0106] When the second processing quality threshold Y2≤comprehensive processing data JG≤first processing quality threshold Y1, it indicates that the processing quality of the injection molding machine is normal;
[0107] When the comprehensive processing data JG is less than the second processing quality threshold value Y2, it indicates that the processing quality of the injection molding machine is abnormal. Then, the optimization algorithm is further executed to optimize the parameters.
[0108] The optimization algorithm unit is used to construct a comprehensive optimization formula, substitute the obtained temperature distribution T(x, y, z), plastic material change rate Bh and crystallization rate Jj into the algorithm formula, calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine, find the best process parameter combination that minimizes the optimization target through iterative calculation, and adjust and optimize the parameters through the interactive interface according to the calculation results;
[0109] The comprehensive optimization coefficient Yh is obtained through the following algorithm formula;
[0110] Yh=min[α·∫ VT(x,y,z)-T target dV+β·Bh-Bh target |+γ·Jj-Jj target ];
[0111] Where α, β and γ represent the weight coefficients of temperature optimization, volume change control and crystallization rate control, respectively. target Indicates the target temperature set during design, Bh target represents the target volume change, Jj target It represents the target crystallization rate Jj set during design;
[0112] α·∫ V T(x,y,z)-T target dV is used to optimize the temperature distribution inside the mold so that the processing temperature is close to the target temperature T target , by minimizing the error of temperature distribution to ensure the uniformity of mold heating, thereby improving the quality and performance of the injection molded product;
[0113] β·|Bh-Bh target |Used to control the volume change of plastic during the injection molding process, so that it is close to the designed target volume V after control target , by minimizing volume variation errors to ensure that the size and quality of the finished product meet the design specifications, thereby improving production consistency and accuracy;
[0114] γ·|Jj-Jj target |Used to adjust the crystallization rate Jj of the plastic to make it close to the target crystallization rate Jj target , by minimizing the crystallization rate error to control the crystallization process of the plastic, thereby optimizing the performance of the material and the structural stability of the finished product.
[0115] In this embodiment, the system provides scientific quality monitoring and optimization means for the injection molding production process through the efficient combination of the quality assessment unit and the optimization algorithm unit. The quality assessment unit can accurately identify the abnormal situation of the mold processing quality by setting the first processing quality threshold Y1 and the second processing quality threshold Y2, and judge the state of the production process according to the comprehensive processing data JG, so as to ensure that problems are found and adjusted in time during the production process. This data-based automated quality assessment mechanism is more efficient and reliable than traditional manual inspection. The optimization algorithm unit systematically analyzes the temperature distribution T(x, y, z), the plastic material change rate Bh and the crystallization rate Jj by constructing a comprehensive optimization formula, calculates the comprehensive optimization coefficient Yh, and finds the best process parameter combination through iterative calculation. This algorithm-driven optimization method can not only optimize and adjust specific quality problems while finely managing production parameters, so as to effectively improve the processing quality of the mold, and enhance the final performance of the product and the consistency of production. Compared with the traditional single-factor adjustment method, the multi-dimensional analysis and automated optimization means of this module significantly improve the quality control level and production efficiency of the production process, and reflect the advanced technical means of modern manufacturing industry in production optimization.
[0116] Example 7
[0117] See also Figure 1 and Figure 2 , the human-computer interaction intelligent design optimization method for injection molds includes the following steps:
[0118] S1. First, a sensor group is installed around the injection molding machine to collect mold processing data of the mold during the injection molding process in real time, and the collected mold processing data is preprocessed to obtain a mold processing data set;
[0119] S2. Using the collected mold processing data set, historical data and experimental data, and dividing them into training set, validation set and test set, a regression analysis model and a machine learning model are established between the mold processing data and the finished product quality;
[0120] S3, construct mold temperature distribution analysis formula, plastic material change analysis formula and crystallization relationship analysis formula, and load the collected mold processing data set to calculate and obtain temperature distribution T (x, y, z), plastic material change rate Bh and crystallization rate Jj, and correlate and summarize the production comprehensive processing data JG;
[0121] S4. Use development tools to build a visual interface, define the interface layout and the location and function of each control, implement data input, optimization calculation and result display functions, display data graphically to achieve dynamic interaction, view optimization results and input parameters for optimization adjustment;
[0122] S5. Finally, the first processing quality threshold value Y1 and the second processing quality threshold value Y2 are preset and compared with the obtained comprehensive processing data JG to evaluate and analyze the quality of the current injection mold. When an abnormality is identified, a comprehensive optimization formula is further constructed to calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine.
[0123] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.
Claims
1. Injection mold human-computer interaction intelligent design optimization system, characterized by: It includes injection molding parameter acquisition module, finished product quality relationship module, processing analysis algorithm module, human-computer interaction module and intelligent analysis and evaluation module; The injection molding parameter acquisition module uses a sensor group to collect mold processing data in the mold re-injection process in real time, and pre-processes the collected mold processing data to obtain a mold processing data set; The finished product quality relationship module is used to utilize the collected mold processing data set, historical data and experimental data, and divide them into a training set, a validation set and a test set to establish a regression analysis model and a machine learning model between the mold processing data and the finished product quality; The processing analysis algorithm module is used to construct a mold temperature distribution analysis formula, a plastic material change analysis formula and a crystallization relationship analysis formula, and load the collected mold processing data set to calculate and obtain the temperature distribution T (x, y, z), the plastic material change rate Bh and the crystallization rate Jj, and then correlate and summarize them to generate comprehensive processing data JG; The human-computer interaction module is used to use development tools to build a visual interface, view optimization results and input parameters for optimization adjustment; The intelligent analysis and evaluation module is used to preset the first processing quality threshold value Y1 and the second processing quality threshold value Y2 and the obtained comprehensive processing data JG, and compare and evaluate the quality of the current injection mold. When an abnormality is identified, a comprehensive optimization formula is further constructed to calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine.
2. The injection mold human-computer interaction intelligent design optimization system according to claim 1, characterized in that: The injection molding parameter acquisition module includes a parameter acquisition module and a data preprocessing module; The parameter acquisition module is used to install a sensor group around the re-injection molding machine, and the sensor group includes a roughness sensor, a thermocouple sensor, an infrared thermometer, a pressure sensor and a flow meter to collect mold processing data of the mold re-injection molding process in real time. The mold processing data includes mold surface roughness Ra, thermal conductivity Km, plastic material expansion coefficient Ap and shear rate Jq; The data preprocessing module is used to filter the mold processing data collected in real time by using a Kalman filter to smooth the data, and at the same time use a wavelet transform to remove the high-frequency noise of the mold processing data collected by the sensor, and then normalize the collected mold processing data.
3. The injection mold human-computer interaction intelligent design optimization system according to claim 2, characterized in that: The finished product quality relationship module includes a regression analysis model unit, a machine learning model unit and a model optimization and verification unit; The regression analysis model unit is used to construct a multivariate linear regression model, identify the linear relationship between mold processing data and finished product quality, analyze the prediction performance and regression coefficient of the model, and transmit the preprocessed mold data to the regression analysis model to evaluate the impact of each parameter on the quality of the finished product; The machine learning model unit is used to select a support vector machine SVM and a multi-layer perceptron, and respectively construct a support vector machine SVM model and a neural network NN model. The support vector machine SVM model is used to capture the nonlinear relationship between mold processing data and finished product quality, and the neural network NN model is used to capture complex nonlinear relationships. The constructed support vector machine SVM model and the neural network NN model are integrated and summarized to obtain a machine learning model. The collected mold processing data set, historical data and experimental data are pre-processed and then transmitted to the machine learning model. The prediction performance of the support vector machine SVM model and the neural network NN model are respectively evaluated, and the linear and nonlinear relationship between the data is analyzed; The model optimization and verification unit uses recursive feature elimination (RFE) to identify variables that affect the model prediction performance and eliminates irrelevant and redundant variables. K-fold cross validation is then used to divide the mold processing data into K subsets, one of the subsets is used for verification each time, and the remaining K-1 subsets are used for training. This is repeated K times, and the result is the average of the K verifications. At the same time, grid search is used to traverse all hyperparameter combinations to select a combination that optimizes the model performance. After the machine learning model is optimized, it is necessary to verify the machine learning model, evaluate the performance of the machine learning model, and ensure that it is suitable for practical applications.
4. The injection mold human-computer interactive intelligent design optimization system according to claim 2, characterized in that: The processing analysis algorithm module includes a mold temperature analysis algorithm unit, a material change algorithm unit, a crystallization algorithm unit and a comprehensive analysis algorithm unit; The temperature analysis algorithm unit is used to construct a mold temperature distribution analysis formula, and to analyze and calculate the temperature distribution T (x, y, z) according to the thermal conductivity Km and the mold surface roughness Ra in the mold processing data set after preprocessing; The temperature distribution T(x, y, z) is calculated and obtained by the following mold temperature distribution analysis formula; Wherein, T0 represents the initial temperature, which is directly measured by an infrared thermometer, Q represents the intensity of the heat source, x, y and z represent the horizontal axis, vertical axis and vertical axis respectively, representing the coordinates of the mold in three-dimensional space, and exp represents the exponential function.
5. The injection mold human-computer interactive intelligent design optimization system according to claim 4, characterized in that: The material change algorithm unit is used to construct a plastic material change analysis formula, and to analyze and calculate the plastic material change rate Bh based on the plastic material expansion coefficient Ap in the preprocessed mold processing data set; The plastic material change rate Bh is calculated and obtained by the following plastic material change analysis formula: In the formula, V0 represents the initial volume of the plastic, △T represents the temperature change, which represents the change value of the material temperature relative to the initial temperature T0, βm represents the volume change rate coefficient, d represents the integral variable, dt represents the variable per unit time t, and T represents the temperature.
6. The injection mold human-computer interactive intelligent design optimization system according to claim 4, characterized in that: The crystallization algorithm unit is used to construct a crystallization relationship analysis formula, and analyze and calculate the crystallization rate Jj according to the shear rate Jq in the mold processing data set after preprocessing; The crystallization rate Jj is calculated by the following crystallization relationship analysis formula: Wherein, k1 represents the rate constant, Ea represents the activation energy, R represents the gas constant, Jq0 represents the base shear rate, n represents the shear rate exponent, and exp represents the exponential function.
7. The injection mold human-computer interactive intelligent design optimization system according to claim 4, characterized in that: The comprehensive analysis algorithm unit is used to correlate and summarize the obtained temperature distribution T (x, y, z), plastic material change rate Bh and crystallization rate Jj to obtain comprehensive processing data JG and analyze the current injection molding machine processing quality; The comprehensive processing data JG is obtained by calculation through the following algorithm formula; Where V is the volume integral, covering the entire three-dimensional space, and S represents the overall performance of the system.
8. The injection mold human-computer interactive intelligent design optimization system according to claim 1, characterized in that: The human-computer interaction module is used to design a preliminary interface using the QtDesigner interface design tool, define the interface layout and the position and function of each control, use the PyQt and PyGObject libraries to implement data input, optimization calculation and result display functions, and then embed graphics into the GUI through Matplotlib and D3.js to display data graphically and realize dynamic interaction.
9. The injection mold human-computer interactive intelligent design optimization system according to claim 1, characterized in that: The intelligent analysis and evaluation module includes a quality evaluation unit and an optimization algorithm unit; The quality assessment unit is used to preset a first processing quality threshold value Y1 and a second processing quality threshold value Y2 based on the mean value of the injection mold processing quality, and then compare and evaluate with the acquired comprehensive processing data JG to analyze the current mold processing quality. The specific assessment scheme is as follows; When the comprehensive processing data JG> the first processing quality threshold Y1, it means that the processing quality of the current injection molding machine is abnormal, and the optimization algorithm is further executed to optimize the parameters; When the second processing quality threshold Y2≤comprehensive processing data JG≤first processing quality threshold Y1, it indicates that the processing quality of the injection molding machine is normal; When the comprehensive processing data JG is less than the second processing quality threshold value Y2, it indicates that the processing quality of the injection molding machine is abnormal. Then, the optimization algorithm is further executed to optimize the parameters. The optimization algorithm unit is used to construct a comprehensive optimization formula, substitute the obtained temperature distribution T(x, y, z), plastic material change rate Bh and crystallization rate Jj into the algorithm formula, calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine, and adjust and optimize the parameters through the interactive interface according to the calculation results; The comprehensive optimization coefficient Yh is obtained by the following algorithm formula; Where α, β and γ represent the weight coefficients of temperature optimization, volume change control and crystallization rate control, respectively. target Indicates the target temperature set during design, Bh target represents the target volume change, Jj target It represents the target crystallization rate Jj set during design; α·∫ V T(x,y,z)-T target dV is used to optimize the temperature distribution inside the mold so that the processing temperature is close to the target temperature T target ; β·Bh-Bh target |Used to control the volume change of plastic during the injection molding process, so that it is close to the designed target volume V after control target ; γ·|Jj-Jj target |Used to adjust the crystallization rate Jj of the plastic to make it close to the target crystallization rate Jj target .
10. A human-computer interaction intelligent design optimization method for injection molds, applied to a human-computer interaction intelligent design optimization system for injection molds according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. First, a sensor group is installed around the injection molding machine to collect mold processing data of the mold during the injection molding process in real time, and the collected mold processing data is preprocessed to obtain a mold processing data set; S2. Using the collected mold processing data set, historical data and experimental data, and dividing them into training set, validation set and test set, a regression analysis model and a machine learning model are established between the mold processing data and the finished product quality; S3, construct mold temperature distribution analysis formula, plastic material change analysis formula and crystallization relationship analysis formula, and load the collected mold processing data set to calculate and obtain temperature distribution T (x, y, z), plastic material change rate Bh and crystallization rate Jj, and correlate and summarize the production comprehensive processing data JG; S4. Use development tools to build a visual interface, define the interface layout and the location and function of each control, implement data input, optimization calculation and result display functions, display data graphically to achieve dynamic interaction, view optimization results and input parameters for optimization adjustment; S5. Finally, the first processing quality threshold value Y1 and the second processing quality threshold value Y2 are preset and compared with the obtained comprehensive processing data JG to evaluate and analyze the quality of the current injection mold. When an abnormality is identified, a comprehensive optimization formula is further constructed to calculate and obtain the comprehensive optimization coefficient Yh to optimize the injection molding machine.
Citation Information
Patent Citations
Injection mold manufacturing method, system and equipment with high injection molding precision and stability
CN117359841A
Constraint tracking control method, system and equipment for batch injection molding process
CN117644623A
Injection molding parameter intelligent optimization system and method based on artificial intelligence
CN118107138A
Intelligent automatic control system of punching machine
CN118197497A
A system and method for constructing mathematical models of technological processes and training these models
WO2021015709A1
Cited By
Intelligent control method and system for mold maintenance precision
CN120122511A
