On-line monitoring and intelligent regulation and control system for temperature difference in precise injection molding mold

Abstract

The invention discloses an in-mold temperature difference online monitoring and intelligent regulation and control system for precise injection molding. A data acquisition module comprises pressure sensors arranged in a cavity and a runner and a temperature sensor arranged in the cavity; the data analysis module is used for constructing and training a prediction model to obtain a flow front maximum temperature difference prediction value, the prediction model is constructed based on a neural network, sensor data obtained through mold flow analysis under different process parameters are input, and the flow front maximum temperature difference serves as a label; the intelligent regulation and control module is used for adjusting process parameters according to a regulation and control critical point, a demarcation point and an alarm critical point which are set from small to large; the in-mold precise injection molding temperature uniformity control is broken through, the consistency and precision of products are improved, and the cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of precision in-mold injection molding, and more particularly to an online monitoring and intelligent control system for temperature difference within a precision injection molding mold. Background Technology

[0002] Precision in-mold injection molding, with its advantages of low cost, high efficiency, and high precision, is now widely used in many fields such as consumer electronics, automobiles, medical devices, and optical lenses. Applications in high-end fields bring even higher precision requirements. Controlling the precision of in-mold injection molded products is mainly achieved by adjusting process parameters, and this optimization process currently relies heavily on experience. At the same time, quality inspection of injection molded products is also primarily carried out through manual offline sampling methods, which suffer from low efficiency, high cost, and insufficient precision.

[0003] Numerous studies have shown that effective control of polymer melt pressure-volume-temperature (PVT) is crucial for improving the quality and consistency of injection molded parts. Among these three parameters, melt quality is primarily determined by temperature and pressure. However, existing monitoring and control methods based solely on temperature sensors are ineffective due to the delay in heat conduction. Furthermore, existing methods using a single pressure sensor to monitor process variables have correlation coefficients between 0.48 and 0.59, which are insufficient for production requirements. An excessive number of sensors can disrupt normal melt flow, thus affecting product quality. In addition, improper sensor placement will fail to characterize the complete flow history. Moreover, due to machine errors, mold cavity sealing, structural complexity, and uncertainties in operating conditions, monitoring and controlling set parameters cannot achieve precise control of the injection molding process. Further improving the precision and consistency of injection molded products requires precise control of key process parameters such as the temperature difference at the flow front, which is currently lacking in existing technologies. Summary of the Invention

[0004] To address the shortcomings of existing technologies in monitoring and controlling the precision in-mold injection molding process, this invention proposes an online monitoring and intelligent control system for temperature difference within the precision injection molding mold. Different sensors are installed at key locations, and a neural network is used to establish the relationship between process parameters, sensor data, and temperature uniformity, thereby achieving online monitoring and intelligent control of the precision in-mold injection molding process.

[0005] The specific technical solution is as follows:

[0006] A precision injection molding in-mold temperature difference online monitoring and intelligent control system includes: a data acquisition module, a data analysis module, an intelligent control module, and a visual human-computer interaction module;

[0007] The data acquisition module includes pressure sensors arranged in the cavity and flow channel, and temperature sensors arranged in the cavity;

[0008] The data analysis module is used to build and train the prediction model, and to receive sensor data collected by the data acquisition module, inputting it into the prediction model to obtain the predicted value of the maximum temperature difference at the flow front. The data is sent to the intelligent control module. The prediction model is based on a neural network. During training, the input to the prediction model is the data from three sensors obtained by the in-mold electronic injection molding device model under different process parameters. The maximum temperature difference at the flow front obtained by the mold flow analysis is used as its label. The process parameters include: melt temperature, mold temperature, and injection time.

[0009] The intelligent control module includes an alarm submodule and a device control submodule; it is used to control thresholds, boundary points, alarm thresholds, and other parameters set from small to large. Adjust the process parameters; if If it is within the control threshold, then intelligent control is not required; if If the device is located between the control threshold and the alarm threshold, the device control submodule will be invoked; if... Above the alarm threshold, the alarm submodule is invoked to trigger the alarm device and automatically shut down; in the equipment control submodule, if If the value is located between the control critical point and the boundary point, and the difference between the injection time and its optimal value exceeds a set step size, then the injection time is adjusted to approach its optimal value by a set step size; if... If the temperature is between the dividing point and the alarm threshold, or if the difference between the injection time and the optimal value is less than one set step, the melt temperature and the mold temperature will be adjusted synchronously.

[0010] The visualization human-computer interaction module provides users with visual sensor data and predicted maximum temperature difference at the flow front, and offers the function of manually modifying set parameters, including control critical points, boundary points, and alarm critical points.

[0011] Furthermore, in the intelligent control module, when simultaneously controlling the melt temperature and the mold temperature, the increase in melt temperature is: The temperature rise of the mold is k1 and k2 are different proportional coefficients set manually, T min To regulate the critical point.

[0012] Furthermore, the in-mold electronic injection molding device model includes a runner model and a part model, and the mold flow analysis performed on it includes the following operations:

[0013] (1) Create flow channel model and part model, and import them into the mold flow analysis software respectively;

[0014] (2) First perform 2D meshing on the imported flow channel model, then perform 3D meshing and repair; first perform 2D meshing on the imported part model, then perform 3D meshing and repair; combine the meshed part model and flow channel model; set boundary conditions according to the actual mold characteristics;

[0015] (3) Set the injection port position of the melt input channel, and set the melt material and mold material to be used;

[0016] (4) Select multiple sets of process parameters and obtain different mold flow analysis results through mold flow analysis software; the mold flow analysis results include: data detected by temperature sensor in cavity, data detected by pressure sensor in cavity, data detected by pressure sensor in flow channel, and maximum temperature difference at the flow front.

[0017] Furthermore, in step (2), a world coordinate system is used, and the boundary conditions include the following types:

[0018] Fixed constraints: nodal displacements and rotational constraints are fixed in all directions;

[0019] Unidirectional fixed constraint, nodal displacement and rotation constraints are fixed only in the Z direction, and free in the X and Y directions;

[0020] The positive direction is unilaterally constrained, with only the displacement constraint of the node in the +Z direction fixed, while the other five directions are free;

[0021] The negative direction is a unilateral constraint, which only fixes the displacement constraint of the node in the -Z direction, while the other five directions are free.

[0022] Furthermore, the mold flow analysis software used is Moldflow.

[0023] Furthermore, in the data analysis module, the Taguchi method is used to select three process parameters: melt temperature, mold temperature, and injection time, and a three-factor, four-level orthogonal experiment is designed; the melt temperature range is 340–370℃, the mold temperature range is 60–150℃, and the injection time range is 0.05–0.2s.

[0024] Furthermore, the prediction model is constructed based on a backpropagation (BP) neural network. The BP neural network has 3 input layer nodes, 10 hidden layer neurons, a hyperbolic tangent activation function for the hidden layer, and 1 output layer node. The training method adopts the Bayesian regularization algorithm, with the following training parameters: learning rate of 0.01, maximum number of iterations of 1000, performance error target of 1e-6, and minimum performance gradient of 1e-7.

[0025] Furthermore, the visual human-computer interaction module includes: a real-time display submodule, an operation panel submodule, and a one-click emergency stop submodule;

[0026] The real-time display submodule is used to provide users with real-time visual data of pressure in the cavity and flow channel, temperature data in the cavity, and predicted maximum temperature difference at the flow front.

[0027] The operation panel submodule is used to operate according to actual production conditions, including: ① adjusting the main process parameters of the injection molding process, including: melt temperature, mold temperature, injection pressure and time, holding pressure and time, and cooling rate; ② changing the values ​​of the control critical point, boundary point, and alarm critical point.

[0028] The one-click emergency stop submodule is used to manually stop the in-mold electronic injection molding device in emergency situations.

[0029] A method for online monitoring and intelligent control of temperature difference within a precision injection molding mold, based on the aforementioned online monitoring and intelligent control system for temperature difference within a precision injection molding mold, includes the following steps:

[0030] S1: Real-time acquisition of temperature and pressure data within the cavity of the in-mold electronic injection molding device, and pressure data within the runner; setting the control critical point, boundary point, and alarm critical point for the maximum temperature difference at the flow front;

[0031] S2: Input the sensor data obtained in S1 into the trained prediction model to obtain the predicted value of the maximum temperature difference at the flow front;

[0032] S3: If the predicted maximum temperature difference at the flow front is... If the value is within the control threshold, no intelligent control is needed; proceed to step S5. If the value is located between the control critical point and the boundary point, and the difference between the injection time and the optimal value exceeds a set step size, then the injection time is adjusted to approach its optimal value by a set step size, and S4 is executed; if If the temperature is between the boundary point and the alarm threshold, or if the difference between the injection time and the optimal value is less than one set step, then the melt temperature and mold temperature are simultaneously adjusted, and S4 is executed; if If the alarm threshold is exceeded, the alarm device will be triggered and the machine will automatically stop to investigate the abnormality. Once the cause of the abnormality has been repaired, the machine will return to S1.

[0033] S4: Produce using the adjusted process parameters. Once the production conditions are stable, return to execute S1-S3, acquire sensor data again, and obtain a new predicted value for the maximum temperature difference at the flow front to determine whether further adjustments are needed.

[0034] S5: Upload the pressure data and temperature data in the cavity and flow channel, as well as the maximum temperature difference at the flow front, to the user terminal and display them visually.

[0035] The beneficial effects of the present invention are:

[0036] This invention constructs and trains a BP neural network that uses data from three sensors as input parameters and the maximum temperature difference at the flow front as the target output function. Based on the actual characteristics and patterns of injection molding, different control schemes for process parameters are designed, providing logical support for online monitoring and intelligent control of the precision in-mold injection molding process. A system for online monitoring and intelligent control of in-mold temperature difference in precision injection molding is established, solving the problems of low efficiency, high cost, and insufficient accuracy of traditional monitoring and control methods. This improves the consistency and precision of precision injection molded parts and has a wide range of applicability to various injection molded parts. Attached Figure Description

[0037] Figure 1 This is a structural diagram of the precision injection molding mold temperature difference online monitoring and intelligent control system proposed in this embodiment of the invention.

[0038] Figure 2 This is a schematic diagram of the precision injection molded part and runner model in an embodiment of the present invention.

[0039] Figure 3 This is a simulation result of the flow front temperature field under actual production process parameters in an embodiment of the present invention.

[0040] Figure 4 This is a comparison curve of the predicted and simulated maximum flow front temperature difference of the test set in the monitoring part of this invention.

[0041] Figure 5 This is a flowchart of the online monitoring and intelligent control method for temperature difference within a precision injection molding mold proposed in this embodiment of the invention.

[0042] In the diagram, the flow channel is 1, the plastic part is 2-1, the metal insert is 2-2, the injection port is 3, and the nozzle is 4. Detailed Implementation

[0043] The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. The objectives and effects of the present invention will become clearer as a result. 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 merely illustrative of the invention and are not intended to limit the invention.

[0044] For precision in-mold injection molding scenarios, in order to overcome the difficulty of real-time temperature measurement in a closed cavity, control the temperature uniformity of the cavity during the injection molding process, and ensure high precision and consistency of the product, this invention proposes an online monitoring and intelligent control system for in-mold temperature difference in precision injection molding, such as... Figure 1 As shown, the system includes: a data acquisition module, a data analysis module, an intelligent control module, and a visual human-computer interaction module.

[0045] The data acquisition module includes a pressure sensor and a temperature sensor, and the two types of sensors are connected to the data analysis module.

[0046] Because in-mold electronic injection molding units, extreme temperatures tend to occur within a specific area rather than at a precise point, and due to the enclosed nature of the cavity and the small size of the parts, it's impossible to install a sufficient number of temperature sensors in the target area. Furthermore, temperature sensors suffer from latency due to heat conduction. Sensors in the runner provide more complete injection history information, but their timeliness is inferior to those in the cavity. Additionally, the sensor placement should avoid the dispensing gate and complex-shaped areas to minimize their impact on melt flow. Therefore, after structural analysis and considering the real-time nature, richness, and historical completeness of the collected data, a solution was adopted: one temperature sensor and one pressure sensor are installed inside the cavity of the in-mold electronic injection molding unit, and one pressure sensor is installed in the runner. The sensor signals from all three sensors are transmitted to the data analysis module in real time.

[0047] The data analysis module receives sensor signals from the data acquisition module, processes and analyzes them, predicts the maximum temperature difference at the flow front within the mold, and sends the predicted maximum temperature difference to the intelligent control module. The specific process for completing the data analysis and processing is as follows:

[0048] (1) Create models of the flow channel in the in-mold electronic injection molding device and the obtained precision injection molded parts (hereinafter referred to as parts) respectively, and then import them into the mold flow analysis software.

[0049] like Figure 2 As shown, in this embodiment, the melt inlet of runner 1 is injection port 3, and the outlet is connected to nozzle 4. Two types of melt are used, and the injection-molded part includes a plastic part 2-1 and a metal insert 2-2. The metal insert 2-2 is placed in the mold as a functional component beforehand, and the plastic part 2-1 is obtained by injection molding by injecting melt into the mold, forming a composite structure. Injection port 3 and nozzle 4 are vertically arranged, with the vertical upward direction being the positive direction of the Z-axis. According to the right-hand rule, the two directions perpendicular to the Z-axis are defined as the X-axis and Y-axis, respectively. Models of the metal insert 2-2, plastic part 2-1, and runner 1 are created in Unigraphics NX. The maximum size of plastic part 2-1 is 18.6mm (width) × 18.6mm (length) × 1.2mm (height), and the maximum size of metal insert 2-2 is 18.25mm (width) × 18.25mm (length) × 0.1mm (height). Moldflow is used as the mold flow analysis software.

[0050] (2) After performing 2D meshing on the imported part model and flow channel model respectively, perform 3D meshing and repair, combine the part model and flow channel model after meshing, and then set boundary conditions according to the actual mold characteristics.

[0051] In this embodiment, the boundary conditions include four types: the first type is fixed constraint, where the nodal displacement and rotation constraints are fixed in all directions; the second type is unidirectional fixed constraint, where the nodal displacement and rotation constraints are fixed only in the Z direction, and free in the X and Y directions; the third type is positive unilateral constraint, where only the displacement constraint of the node in the +Z direction is fixed, and the other five directions are free; the fourth type is negative unilateral constraint, where only the displacement constraint of the node in the -Z direction is fixed, and the other five directions are free.

[0052] (3) Set the position of the injection port 3 of the melt input channel 1, and set the corresponding melt, mold and insert materials.

[0053] In this embodiment, the thermoplastic material (i.e. melt) selected for the metal insert 2-2 of the part is SUS430, and the thermoplastic material selected for the plastic part 2-1 is liquid crystal polymer (LCP, XydarCM-529B).

[0054] (4) Select multiple sets of actual production process parameters (including melt temperature, mold temperature and injection time), and obtain different mold flow analysis results through mold flow analysis software. The mold flow analysis results include: data detected by temperature sensor in cavity, data detected by pressure sensor in cavity, data detected by pressure sensor in flow channel, and maximum temperature difference at the flow front.

[0055] The main process parameters affecting the injection molding process include: melt temperature, mold temperature, injection pressure and time, holding pressure and time, and cooling rate. In actual production, injection pressure is not a directly changeable control variable but is influenced by other process parameters. In this embodiment, since the melt has mostly solidified during the filling stage, the holding pressure and time have little effect on the flow front temperature difference. Furthermore, the ejection temperature of the part is fixed. Since the cooling channels are installed inside the mold, controlling the mold temperature controls the cooling rate; therefore, only the mold temperature needs to be studied. In summary, melt temperature, mold temperature, and injection time are selected as three process parameters as input parameters to obtain the mold flow analysis results for each group.

[0056] In this embodiment, the Taguchi method was adopted to select three process parameters: melt temperature, mold temperature, and injection time. A three-factor, four-level orthogonal experiment was designed, and different mold flow analysis results were obtained through mold flow analysis software. Based on actual production experience, a reasonable Taguchi method range was selected, in which the melt temperature range was 340–370℃, the mold temperature range was 60–150℃, and the injection time range was 0.05–0.2s, as shown in Table 1 below.

[0057] Table 1. Range and Level of Taguchi Process Parameters

[0058] ① ② ③ ④ Melt temperature (°C) 340 350 360 370 Mold temperature (°C) 60 90 120 150 Injection time (s) 0.05 0.1 0.15 0.2

[0059] The Moldflow simulation results corresponding to the above process parameters are shown in Table 2 below.

[0060] Table 2. Moldflow Injection Molding Simulation Analysis Results

[0061]

[0062]

[0063]

[0064] Taking group 38 in the table as an example, its model flow analysis results are as follows: Figure 3 As shown, the simulation results show that the lowest temperature is in region A and the highest temperature is in region B, while the temperature uniformity of other regions is good. Therefore, the objective can be simplified to control the temperature difference between these two regions within a certain range.

[0065] (5) Construct and train a prediction model based on a BP neural network: Use the sensor data from the multiple sets of model flow analysis results obtained in step (4) as input samples, and the maximum temperature difference at the corresponding flow front as the objective function. The training termination condition is that the number of training iterations reaches a set value or the average relative error between the predicted value and the model flow analysis value reaches a set threshold. Finally, the trained prediction model is obtained. The training of the prediction model is implemented through the following steps:

[0066] (5.1) The sensor data in the multiple sets of model flow analysis results obtained in step (4) are used as input samples, and the maximum temperature difference of the corresponding flow front is used as its expected output (i.e., label). The multiple sets of input sample-label pairs are divided into training set and test set. In this embodiment, the training set contains 51 sets of input sample-label pairs for training the BP neural network. The test set contains 13 sets of input sample-label pairs for detecting the accuracy of the BP neural network prediction data. The division of the training set and the test set is chosen to be random.

[0067] (5.2) The BP neural network was trained using the training set samples to obtain a pre-trained BP neural network. The input layer of the BP neural network has 3 nodes and the hidden layer has 10 neurons to prevent overfitting and underfitting. The activation function of the hidden layer is the hyperbolic tangent activation function, which is used to map the input data to a nonlinear space, control the range of activation values, and stabilize the training process. The output layer has 1 node. The training method used is the Bayesian regularization algorithm, and the training parameters are: learning rate of 0.01, maximum number of iterations of 1000, performance error target of 1e-6, and minimum performance gradient of 1e-7.

[0068] (5.3) Use the pre-trained BP neural network to predict the input samples of the test set and obtain the predicted output value of the test set.

[0069] (5.4) Calculate the average relative error between the predicted output value obtained in step (5.3) and the label value corresponding to the input sample. If the average relative error reaches a set threshold (15% in this embodiment), the training ends, and the trained BP neural network is obtained; if the set threshold is not reached, determine whether the number of training iterations has reached the set value. If yes, the training ends, and the trained prediction model is output; if no, the weights are updated, the number of training iterations is incremented by one, and steps (5.2) to (5.4) are repeated. Figure 4 As shown in this embodiment, the predicted results of the maximum flow front temperature difference of the test set and the simulation results of Moldflow have a high degree of agreement, with an average relative error of 13.12%.

[0070] (6) Based on real-time sensor data and the trained prediction model, predict the maximum temperature difference at the flow front of the current in-mold electronic injection molding device, and set the predicted value of the maximum temperature difference at the flow front. Send to the intelligent control module.

[0071] The intelligent control module includes an alarm submodule and an equipment control submodule. It sets the alarm threshold (i.e., the maximum value T) for the maximum temperature difference at the flow front. max ), and the control critical point (i.e., the minimum value T) min The intelligent control module receives the predicted maximum temperature difference at the flow front from the data analysis module. like If it is within the control threshold, then intelligent control is not required; if If the device is located between the control threshold and the alarm threshold, the device control submodule will be invoked; if... Above the alarm threshold, the alarm submodule is invoked. Because the adjustment is based on the predicted value, the adjustment error caused by delay is reduced.

[0072] When the maximum temperature difference at the flow front is predicted When the alarm threshold is reached, the alarm submodule triggers the alarm device and automatically shuts down the machine to investigate the cause of the abnormal temperature difference at the maximum flow front. Production resumes only after the cause of the abnormality has been repaired. The alarm threshold setting ensures system safety on the one hand, and on the other hand, because injection molding time is very short and factors such as heat conduction require a certain amount of time to adjust parameters in actual operation, emergency shutdown can reduce material waste, lower costs, and improve yield.

[0073] When the maximum temperature difference at the flow front is predicted When the temperature is between the control threshold and the alarm threshold, it enters the operating range of the equipment control submodule. This submodule controls the maximum temperature difference at the flow front by adjusting three process parameters: melt temperature, mold temperature, and injection time. The sensitivity of the maximum flow front temperature difference to different process parameters, from strongest to weakest, is: melt temperature, injection time, and mold temperature. Since adjusting the melt temperature and mold temperature requires significant time, and both need to be adjusted together to control warpage and other issues, and both require shutdown and preheating for temperature control, adjusting the injection time is relatively more convenient and is therefore the preferred method. Numerical simulation results show that within the recommended range of the three process parameters, both melt temperature and mold temperature are negatively correlated with the maximum flow front temperature difference. However, as the injection time increases, the maximum flow front temperature difference first decreases and then increases.

[0074] Specifically, a dividing point T is set between the control critical point and the alarm critical point. mid (Based on experience and manually set), if the predicted maximum temperature difference at the flow front is... Between the critical point and the dividing point, and if the difference between the injection time and the optimal value exceeds a set step size, then adjust the injection time to approach its optimal value (obtained through orthogonal experimental simulation) by a set step size (while minimizing changes to existing process parameters while meeting requirements); if the predicted maximum temperature difference at the flow front... If the difference between the dividing point and the alarm threshold, or the difference between the injection time and the optimal value, is less than one set step, then the melt temperature and mold temperature are simultaneously controlled. The melt temperature rise is... The temperature rise of the mold is k1 and k2 are different proportional coefficients set manually. Under the above conditions, production is carried out with the adjusted process parameters. After the production conditions stabilize, sensor data is collected again, and the data analysis module is called to obtain a new predicted value of the maximum temperature difference at the flow front. This value is then input into the intelligent control module for judgment until the predicted value of the maximum temperature difference at the flow front is within the control critical point.

[0075] In this embodiment, a control critical point T is set. min =5℃, dividing point T mid =8℃, alarm threshold T max=15℃, the optimal injection time is 0.15s, and the step size is set to 0.02s; correspondingly, if the predicted maximum temperature difference at the flow front is between 5-15℃, then it enters the working range of the equipment control submodule. The logic of the intelligent control module in determining the control scheme is as follows: if the predicted maximum temperature difference at the flow front output by the data analysis module is... If the temperature is below 5℃, no intelligent control is needed. If the temperature is between 5-8℃ and the injection time is less than 0.13s or greater than 0.17s, then the injection time should be adjusted in increments of 0.02s, approaching 0.15s. If If the temperature is between 8-15℃, or the injection time is between 0.13s-0.17s, the melt temperature should be increased synchronously. Mold temperature rise

[0076] The visual human-computer interaction module includes: a real-time display submodule, an operation panel submodule, and a one-click emergency stop submodule.

[0077] The real-time display submodule provides users with real-time pressure data within the cavity and flow channel, temperature data within the cavity, and maximum temperature difference at the flow front, based on information from the data acquisition and analysis modules. Based on the intelligent control module, it also provides real-time process parameter control values. Furthermore, the real-time display submodule can simultaneously display key information such as defective product number, remaining material quantity, and work progress.

[0078] The operation panel submodule provides the following functions: perform the following operations according to actual production conditions: ① manually adjust the main process parameters of the injection molding process, ② manually change the values ​​of the control critical point, boundary point and alarm critical point, ③ manually switch between different parts working condition subsystems to realize the memorization and switching of production, monitoring and control schemes for various commonly used parts.

[0079] The one-click emergency stop submodule is used to manually stop the in-mold electronic injection molding device in emergency situations, reducing safety hazards.

[0080] Based on the aforementioned precision injection molding in-mold temperature difference online monitoring and intelligent control system, this embodiment also proposes a method for precision injection molding in-mold temperature difference online monitoring and intelligent control, such as... Figure 5 As shown, the method includes the following steps:

[0081] S1: Real-time acquisition of temperature and pressure data within the mold cavity and pressure data within the runner of the in-mold electronic injection molding device via the data acquisition module; setting the control critical point T for the maximum temperature difference at the flow front from small to large. min Boundary point T mid Alarm threshold T max .

[0082] S2: The data analysis module receives sensor data, uses the trained prediction model to obtain the predicted value of the maximum temperature difference at the current in-mold flow front, and sends it to the intelligent control module.

[0083] S3: The intelligent control module determines the predicted maximum temperature difference at the flow front. The scope, if If the value is within the control threshold, no intelligent control is needed; proceed to step S5. If... If the value is between the control critical point and the boundary point, and the difference between the injection time and the optimal value exceeds a set step size, then the injection time is adjusted to approach the optimal value by a set step size, and S4 is executed. If the temperature is between the dividing point and the alarm threshold, or if the difference between the injection time and the optimal value is less than one set step, then the melt temperature and mold temperature will be adjusted synchronously, with the melt temperature increase being... The temperature rise of the mold is And execute S4. If Above the alarm threshold, the alarm device is triggered and the machine is automatically shut down to investigate the cause of the abnormal temperature difference at the maximum flow front. After the cause of the abnormality is repaired, production resumes and returns to S1 to collect sensor data again.

[0084] S4: Produce using the adjusted process parameters. Once the production conditions are stable, return to execute S1-S3, acquire sensor data again, and obtain a new predicted value for the maximum temperature difference at the flow front to determine whether further adjustments are needed.

[0085] S5: Upload the pressure data and temperature data inside the cavity and flow channel obtained by the data acquisition module, as well as the maximum temperature difference at the flow front output by the data analysis module, to the user terminal and visualize them.

[0086] To demonstrate the accuracy and effectiveness of the system of this invention, it was verified as follows: using melt temperature of 355℃, mold temperature of 105℃, and injection time of 0.125s as input parameters, mold flow analysis was performed. The maximum temperature difference at the flow front obtained by simulation was 5.78℃. The sensor data obtained by simulation under these process parameters was input into the trained prediction model, and the predicted value of the maximum temperature difference at the flow front was 5.42℃, with a relative error of 6.23%, which proved the accuracy of the data analysis module. Simultaneously, a set of sensor data collected during the actual injection molding experiment—370.6℃, 70.51MPa (runner pressure), and 29.5MPa (cavity pressure)—along with the set process parameters of melt temperature 360℃, mold temperature 60℃, and injection time 0.05s, were used as input. The predicted maximum flow front temperature difference was 10.42℃. The intelligent control module (with parameters such as control threshold, boundary point, and alarm threshold set as in the previous embodiment) proposed adjusting the melt temperature and mold temperature to 370.84℃ and 92.52℃ respectively. After the production process stabilized (generally taking the 11th product after the production switchover), the predicted maximum flow front temperature difference decreased to 3.92℃, meeting the set requirements, thus proving the effectiveness of the intelligent control module. This invention achieves a breakthrough in the control of in-mold precision injection molding temperature uniformity, improving yield and accuracy while reducing costs.

[0087] It will be understood by those skilled in the art that the above descriptions are merely preferred examples of the invention and are not intended to limit the invention. Although the invention has been described in detail with reference to the foregoing examples, those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. All modifications and equivalent substitutions made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. A precision injection molding mold temperature difference online monitoring and intelligent control system, characterized in that, include: Data acquisition module, data analysis module, intelligent control module, and visual human-computer interaction module; The data acquisition module includes pressure sensors arranged in the cavity and flow channel, and temperature sensors arranged in the cavity; The data analysis module is used to build and train the prediction model, and to receive sensor data collected by the data acquisition module, inputting it into the prediction model to obtain the predicted value of the maximum temperature difference at the flow front. The data is sent to the intelligent control module. The prediction model is based on a neural network. During training, the input to the prediction model is the data from three sensors obtained by analyzing the flow of the in-mold electronic injection molding device model under different process parameters. The maximum temperature difference at the flow front obtained by the flow analysis is used as its label. The process parameters include: melt temperature, mold temperature, and injection time; The intelligent control module includes an alarm submodule and a device control submodule; it is used to control thresholds, boundary points, alarm thresholds, and other parameters set from small to large. Adjust the process parameters; if If it is within the control threshold, then intelligent control is not required; if If the device is located between the control threshold and the alarm threshold, the device control submodule will be invoked; if... Above the alarm threshold, the alarm submodule is invoked to trigger the alarm device and automatically shut down; in the equipment control submodule, if If the value is located between the control critical point and the boundary point, and the difference between the injection time and its optimal value exceeds a set step size, then the injection time is adjusted to approach its optimal value by a set step size; if... If the temperature is between the dividing point and the alarm threshold, or if the difference between the injection time and the optimal value is less than one set step, the melt temperature and the mold temperature will be adjusted synchronously. The visualization human-computer interaction module provides users with visual sensor data and predicted maximum temperature difference at the flow front, and offers the function of manually modifying set parameters, including control critical points, boundary points, and alarm critical points.

2. The precision injection molding mold temperature difference online monitoring and intelligent control system according to claim 1, characterized in that, In the intelligent control module, when simultaneously controlling the melt temperature and the mold temperature, the increase in melt temperature is: The temperature rise of the mold is k1 and k2 are different proportional coefficients set manually, T min To regulate the critical point.

3. The precision injection molding mold temperature difference online monitoring and intelligent control system according to claim 1, characterized in that, The in-mold electronic injection molding device model includes a runner model and a part model. Mold flow analysis of the device includes the following operations: (1) Create flow channel model and part model, and import them into the mold flow analysis software respectively; (2) First perform 2D meshing on the imported flow channel model, then perform 3D meshing and repair; first perform 2D meshing on the imported part model, then perform 3D meshing and repair; combine the meshed part model and flow channel model; set boundary conditions according to the actual mold characteristics; (3) Set the injection port position of the melt input channel, and set the melt material and mold material to be used; (4) Select multiple sets of process parameters and obtain different mold flow analysis results through mold flow analysis software; The mold flow analysis results include: data detected by the temperature sensor in the cavity, data detected by the pressure sensor in the cavity, data detected by the pressure sensor in the flow channel, and the maximum temperature difference at the flow front.

4. The precision injection molding mold temperature difference online monitoring and intelligent control system according to claim 3, characterized in that, In step (2), the world coordinate system is used, and the boundary conditions include the following types: Fixed constraints: nodal displacements and rotational constraints are fixed in all directions; Unidirectional fixed constraint, nodal displacement and rotation constraints are fixed only in the Z direction, and free in the X and Y directions; The positive direction is unilaterally constrained, with only the displacement constraint of the node in the +Z direction fixed, while the other five directions are free; The negative direction is a unilateral constraint, which only fixes the displacement constraint of the node in the -Z direction, while the other five directions are free.

5. The precision injection molding mold temperature difference online monitoring and intelligent control system according to claim 3, characterized in that, The mold flow analysis software used is Moldflow.

6. The precision injection molding mold temperature difference online monitoring and intelligent control system according to claim 1, characterized in that, In the data analysis module, the Taguchi method is used to select three process parameters: melt temperature, mold temperature, and injection time, and a three-factor, four-level orthogonal experiment is designed. The melt temperature range is 340–370℃, the mold temperature range is 60–150℃, and the injection time range is 0.05–0.2s.

7. The precision injection molding mold temperature difference online monitoring and intelligent control system according to claim 1, characterized in that, The prediction model is based on a backpropagation (BP) neural network. The BP neural network has 3 input layer nodes, 10 hidden layer neurons, a hyperbolic tangent activation function for the hidden layer, and 1 output layer node. The training method uses the Bayesian regularization algorithm, with the following training parameters: learning rate of 0.01, maximum number of iterations of 1000, performance error target of 1e-6, and minimum performance gradient of 1e-7.

8. The precision injection molding mold temperature difference online monitoring and intelligent control system according to claim 1, characterized in that, The visual human-computer interaction module includes: a real-time display submodule, an operation panel submodule, and a one-click emergency stop submodule; The real-time display submodule is used to provide users with real-time visual data of pressure in the cavity and flow channel, temperature data in the cavity, and predicted maximum temperature difference at the flow front. The operation panel submodule is used to operate according to actual production conditions, including: ① adjusting the main process parameters of the injection molding process, including: melt temperature, mold temperature, injection pressure and time, holding pressure and time, and cooling rate; ② changing the values ​​of the control critical point, boundary point, and alarm critical point. The one-click emergency stop submodule is used to manually perform an emergency stop operation on the in-mold electronic injection molding device in emergency situations.

9. A method for online monitoring and intelligent control of temperature difference within a precision injection molding mold, implemented based on the online monitoring and intelligent control system for temperature difference within a precision injection molding mold as described in any one of claims 1-8, characterized in that, Includes the following steps: S1: Real-time acquisition of temperature and pressure data within the cavity of the in-mold electronic injection molding device, and pressure data within the runner; setting the control critical point, boundary point, and alarm critical point for the maximum temperature difference at the flow front; S2: Input the sensor data obtained in S1 into the trained prediction model to obtain the predicted value of the maximum temperature difference at the flow front; S3: If the predicted maximum temperature difference at the flow front is... If the value is within the control threshold, no intelligent control is needed; proceed to step S5. If the value is located between the control critical point and the boundary point, and the difference between the injection time and the optimal value exceeds a set step size, then the injection time is adjusted to approach its optimal value by a set step size, and S4 is executed; if If the temperature is between the boundary point and the alarm threshold, or if the difference between the injection time and the optimal value is less than one set step, then the melt temperature and mold temperature are simultaneously adjusted, and S4 is executed; if If the alarm threshold is exceeded, the alarm device will be triggered and the machine will automatically stop to investigate the abnormality. Once the cause of the abnormality has been repaired, the machine will return to S1. S4: Produce using the adjusted process parameters. Once the production conditions are stable, return to execute S1-S3, acquire sensor data again, and obtain a new predicted value for the maximum temperature difference at the flow front to determine whether further adjustments are needed. S5: Upload the pressure data and temperature data in the cavity and flow channel, as well as the maximum temperature difference at the flow front, to the user terminal and display them visually.

Images