Load-bearing integral hot runner system

Abstract

This invention relates to the field of injection molding control technology and discloses a load-bearing integral hot runner system. By extracting the cycle-by-cycle force impact time sequence of each bridging body, the valve needle pulse parameter set is calculated based on the nucleation kinetic equation. A dual-source displacement accumulation model is established to predict the alignment life by using dual-variable regression to separate the pulse impact displacement sensitivity and the basic displacement rate. Furthermore, the force impact constraint is introduced into a multi-objective optimization framework to solve for the corrected pulse parameter set, achieving coordinated control of valve needle pulse parameter optimization and bridging body displacement accumulation monitoring.

Description

Technical Field

[0001] This invention relates to the field of injection molding control technology, and more specifically, to a load-bearing integral hot runner system. Background Technology

[0002] In the application of a load-bearing integral hot runner system to microfoaming injection molding, each servo electric cylinder drives the valve needle to perform pulsed rapid retraction and forward movements to trigger bubble nucleation within the cavity. When the valve needle performs the pulsed retraction movement, it applies a force impact to the support pad on a millisecond timescale through melt pressure changes and servo electric cylinder reaction forces. The dynamic component of this force impact is amplified by the dynamic response of the bridging body, generating additional irreversible micro-displacements.

[0003] In existing technologies, the optimization of valve needle pulse parameters is solely aimed at improving the quality of bubble nucleation. Faster pulse retraction speed and greater retraction distance result in a steeper nucleation pressure drop and more uniform bubble formation, but also a greater impact on the bearing pad and bridging body. Furthermore, the prediction of the cumulative displacement rate of the bridging body is based on univariate linear regression of cold-state position data, which cannot distinguish the respective contributions of the gradual change force of conventional thermal cycling and the pulse impact force to the cumulative displacement.

[0004] The aforementioned existing technologies have the following drawbacks: when the pulse parameters become more aggressive due to the increased requirements for bubble density, the force impact intensifies, causing the actual cumulative displacement rate to deviate from the linear regression prediction value, and the alignment life prediction is too optimistic; pulse parameter optimization and displacement accumulation monitoring are not aware of each other's constraints in their respective control objectives, constituting a compound control misalignment problem. Summary of the Invention

[0005] This invention provides a load-bearing integral hot runner system, which solves the technical problem in related technologies where the force impact generated by the pulse action of the valve needle causes cumulative displacement of the bridging body, which in turn leads to a decrease in alignment accuracy and a deterioration in the quality of bubble nucleation due to the lack of a coordinated control mechanism.

[0006] This invention discloses a load-bearing integral hot runner system, comprising: performing time-domain analysis on the high-frequency pressure timing signal of each load-bearing pad during the valve needle pulse execution period of each injection cycle; calculating the force impact amount applied to each bridge body by each pulse action based on the force transmission relationship between each load-bearing pad and the corresponding bridging body; and generating the cycle-by-cycle force impact amount timing of each bridging body. Based on the micro-foaming nucleation kinetic equation and the wall thickness parameters and target cell density parameters of each gate, the required pressure drop amplitude and pressure drop rate at each gate position are calculated and converted into the valve needle pulse parameter set of each servo electric cylinder. The cumulative impact sequence is generated by summing the time series of the periodic force impact. The pulse impact displacement sensitivity and foundation displacement rate of each bridging body are extracted by using the thermal cycle number and the cumulative impact sequence as independent variables and the cold position offset as dependent variable. Based on the pulse impact displacement sensitivity and basic displacement rate, the remaining thermal cycle number of each bridging body is predicted by the dual-source displacement accumulation model, and the alignment life prediction value after pulse condition correction is generated. For bridging entities whose alignment life prediction value is lower than the preset safety threshold, a weighted and comprehensive objective function is constructed to perform multi-objective optimization with the first objective being to minimize the bubble nucleation quality deviation and the second objective being to minimize the force impact amount, and the corrected pulse parameter set under structural safety constraints is solved.

[0007] Furthermore, the method for calculating the force impact is as follows: For each bearing pad that has a force transmission relationship with the target bridging body, calculate the instantaneous force value obtained by subtracting the baseline pressure value before the pulse start from the instantaneous pressure value during the pulse period and multiplying it by the effective force-bearing area of ​​the bearing pad. Then multiply the result by the force transmission coefficient from the bearing pad to the target bridging body. Summate the integral results of all relevant bearing pads to obtain the force impact amount applied to the target bridging body by a single pulse action.

[0008] Furthermore, the force transmission coefficient is predetermined through finite element simulation or static loading calibration test. In the static loading calibration test, a static load of known amplitude is applied to the target bearing pad, and the response force on the target bridging body is measured simultaneously. The ratio of the response force to the applied load is used as the calibration value of the force transmission coefficient. For a bearing pad that has no direct structural connection with the target bridging body, the corresponding force transmission coefficient is zero.

[0009] Furthermore, the specific process of the bivariate linear regression includes: The position readings of the absolute position sensors on each bridging body are obtained in each thermal cycle cold state phase, and the position readings are compared with the initial installation reference position values ​​to obtain the cold state position offset. Before performing regression analysis, mean normalization based on range was applied to the thermal cycle number and cumulative impact sequence, and mean normalization based on range was applied to the cold position offset. The least squares method is used to solve the parameters of the bivariate linear regression model in the normalized space to obtain the regression coefficients of the basic displacement rate and the pulse impact displacement sensitivity in the normalized space. Then, the basic displacement rate and pulse impact displacement sensitivity are transformed back to the original dimensions according to the corresponding normalized proportional coefficients.

[0010] Furthermore, the dual-source displacement accumulation model expresses the cumulative displacement as the sum of the product of the basic displacement rate and the number of future thermal cycles plus the product of the pulse impact displacement sensitivity, the periodic average force impact, and the number of future thermal cycles, wherein the periodic average force impact is taken as the arithmetic mean of the force impact experienced by each bridging body in the most recent injection molding cycles. The remaining number of thermal cycles is calculated as follows: the remaining allowable displacement is obtained by subtracting the absolute value of the currently accumulated cold position offset from the upper limit of the alignment accuracy of each bridging body, and then dividing by the product of the safety factor and the sum of the basic displacement rate and the pulse impact displacement sensitivity multiplied by the periodic average force impact. The safety factor is greater than one and is determined based on the historical fluctuation amplitude of the pulse impact displacement sensitivity.

[0011] Furthermore, the weight coefficients in the weighted and comprehensive objective function are dynamically allocated based on the nucleation quality tolerance margin and the urgency of alignment life. The nucleation quality tolerance margin is the difference between the upper limit of tolerance and the current measured deviation of the nucleation quality of the bubble cells. The urgency of alignment life is measured by the relative value of the difference between the predicted value of alignment life and the preset safety threshold. The larger the nucleation quality tolerance margin or the closer the predicted value of alignment life is to the preset safety threshold, the greater the weight of the force impact quantity target. The sum of the two weighting coefficients is one and both are greater than zero; the multi-objective optimization also includes the constraint that the nucleation quality deviation does not exceed the upper limit of the nucleation quality deviation at the corresponding gate position.

[0012] Furthermore, after solving for the corrected pulse parameter set, the process also includes: The modified pulse parameter set is sent to the corresponding servo electric cylinder for execution. During the pulse execution, the high-frequency pressure signal of each bearing pad is subjected to short-time Fourier transform to extract the nucleation spectrum features of the bubble holes. The measured spectrum features are compared with the pre-calibrated qualified nucleation spectrum range. When the nucleation spectrum characteristics deviate from the qualified nucleation spectrum range, search for the nearest solution on the Pareto front of the multi-objective optimization in the neighborhood of the current corrected pulse parameter set along the direction of increasing target weight of nucleation quality deviation, and select the nearest solution that makes the nucleation spectrum characteristics return to the qualified nucleation spectrum range as the fine-tuned pulse parameters. The adjacent solutions on the Pareto front are pre-generated and stored during the multi-objective optimization process by solving the problem multiple times with different ratios of the weight coefficients.

[0013] Furthermore, for bridging entities whose alignment life prediction value is lower than a preset warning threshold, the following are also included: Based on the identification of the heat pipe segment that contributes most to the displacement direction vector of the bridging body, the asymmetric temperature offset that needs to be applied to the heat pipe segment in each thermal cycle heating stage is calculated so that the thermal expansion force of the heat pipe segment generates a correction component opposite to the cumulative displacement direction. The temperature offset is subject to an upper limit constraint, that is, the temperature set value after offset shall not cause the temperature of the heat pipe segment to exceed the allowable range of the melt process temperature window. In subsequent thermal cycles, the change value of the cold position offset of the affected bridging body is continuously acquired. If the measured displacement increment of multiple consecutive thermal cycles does not decrease to the expected level, the temperature offset is increased according to the preset step size. The preset warning threshold is greater than the preset safety threshold.

[0014] Furthermore, it also includes: after multiple consecutive evaluation intervals, constructing a time series of pulse impact displacement sensitivity values ​​calculated at different evaluation intervals, and performing linear regression on the time series to extract the slope of the change in pulse impact displacement sensitivity; When the slope of change is greater than zero, the baseline value of the force impact target weight in the multi-objective optimization is automatically increased, and the temperature offset correction of the corresponding heat pipe section is increased proportionally according to the magnitude of the slope of change. The updated slope of change and the alignment life prediction value are then written into the maintenance decision database.

[0015] This invention provides a control system for a load-bearing integral hot runner system, comprising: The force impact extraction module is used to perform time-domain analysis on the high-frequency pressure timing signal of each bearing pad during the valve needle pulse execution period of each injection cycle. Based on the force transmission relationship, it calculates the force impact amount applied to each bridging body by each pulse action and generates a cycle-by-cycle force impact timing sequence. The pulse parameter calculation module is used to calculate the required pressure drop amplitude and pressure drop rate at each gate position based on the micro-foaming nucleation kinetic equation, the wall thickness parameters of each gate, and the target cell density parameters, and convert them into the valve needle pulse parameter set of each servo electric cylinder. The displacement sensitivity regression module is used to accumulate and sum the time series of the cycle-by-cycle force impact to generate a cumulative impact sequence. With the thermal cycle number and the cumulative impact sequence as independent variables and the cold position offset as the dependent variable, a bivariate linear regression is used to extract the pulse impact displacement sensitivity and basic displacement rate of each bridging body. The alignment life prediction module is used to predict the remaining thermal cycle number of each bridging body based on the pulse impact displacement sensitivity and the basic displacement rate through a dual-source displacement accumulation model, and generate the alignment life prediction value after pulse condition correction. The multi-objective optimization module is used to optimize the bridging body whose alignment life prediction value is lower than the preset safety threshold. The first objective is to minimize the bubble nucleation quality deviation, and the second objective is to minimize the force impact. The module constructs a weighted and comprehensive objective function to solve for the set of corrected pulse parameters under structural safety constraints.

[0016] This invention establishes a dual-source displacement accumulation model and performs dual-variable regression analysis on the cumulative sequence of force impact and the cold-state position offset of the bridging body. This separates two independent parameters: pulse impact displacement sensitivity and basic displacement rate. This solves the technical problem that univariate linear regression cannot distinguish the contribution of conventional thermal cycle slow-changing force and pulse impact force to displacement accumulation. It achieves the technical effect that the alignment life prediction results can reflect the actual impact of pulse conditions when pulse conditions change. At the same time, by introducing the force impact constraint into the multi-objective optimization framework of valve needle pulse parameters, this invention solves the composite control mismatch problem where pulse parameter optimization and displacement accumulation monitoring are unaware of each other's constraints. This achieves the technical effect that valve needle pulse parameter optimization can perceive structural safety constraints and that displacement accumulation monitoring prediction results can be fed back to valve needle pulse parameter adjustment. Attached Figure Description

[0017] Figure 1 This is a flowchart of the load-bearing integral hot runner system provided in the embodiments of the present invention under micro-foaming injection molding conditions; Figure 2 This is a schematic diagram illustrating the cumulative trend of the cold-state position offset of each bridging entity provided in the embodiments of the present invention; Figure 3 This is a schematic diagram comparing the pulse characteristic parameters of each bearing pad in the 480th injection cycle provided in this embodiment of the invention; Figure 4 This is a schematic diagram comparing the parameter sets of various gate valve needles provided in the embodiments of the present invention; Figure 5 This is a schematic diagram of the alignment lifetime prediction results of each bridging body provided in the embodiments of the present invention; Figure 6 This is a schematic diagram of the optimization process of the comprehensive objective function of the candidate solution for the modified pulse parameter of gate 1 provided in an embodiment of the present invention; Figure 7 This is a schematic diagram showing the changes in force impact and remaining thermal cycle number before and after the parameter correction of gate 1 provided in this embodiment of the invention; Figure 8 This is a schematic diagram illustrating the evolution trend of the pulse impact displacement sensitivity of the bridging body 1 provided in this embodiment of the invention as a function of the evaluation interval; Figure 9 This is a screenshot of the three-dimensional model of the integral hot runner in an embodiment of the present invention; Figure 10This is a screenshot of a cross-sectional view of the three-dimensional model of the integral hot runner in an embodiment of the present invention; Figure 11 This is a schematic diagram of the integral hot runner connection method in an embodiment of the present invention; Figure 12 This is a schematic diagram of the hot and cold states of the integral hot runner in an embodiment of the present invention. Detailed Implementation

[0018] Integral hot runner systems are currently widely used in the market, typically in medium to large-sized plastic molds, offering the advantage of convenient installation compared to split systems. However, the development and widespread application of integral systems have remained largely unchanged for over twenty years. Products from various domestic and international hot runner companies are essentially the same, some problems persist, and there have been no major technological breakthroughs.

[0019] The existing integrated hot runner molds on the market have extremely large openings, resulting in a large amount of hollowing out inside the mold's A plate. The stress-bearing area between the templates is insufficient, and the thickness of the A plate needs to be very large.

[0020] Existing integral runner systems operate under heat, with only a few media contacts between the system and the mold. There are very few stress points, and once product defects occur, they cannot be resolved by increasing local stress.

[0021] The existing integral runner generates a huge amount of heat, and most of this heat is absorbed by the mold, which is ineffective heat. At the same time, the mold has to consume energy to absorb this ineffective heat to maintain the mold temperature, resulting in double loss, extremely low energy efficiency, and huge waste.

[0022] The existing integral flow channel has large lateral and longitudinal expansion, making installation difficult. Moreover, the system operates in a tortuous state for a long time, making disassembly and assembly extremely inconvenient in the event of after-sales service. Once leakage occurs, maintenance work is complex and time-consuming.

[0023] The existing integrated design uses a large plate connected by hot nozzles in a spiral manner. This large plate requires machining by a large machining center, which results in long processing time, low positional accuracy, and high cost.

[0024] Once the internal flow channels of existing integral flow channels are manufactured, surface treatment and flow channel dimensions cannot be modified. The tolerance for error is extremely low.

[0025] At least one embodiment of the present invention discloses a load-bearing integral hot runner, see Figures 9-12This system completely changes the traditional monolithic design approach, employing a cold-connection method. Most areas are connected using a cold structural connection, with internal points linked by expandable heat pipes. There are gaps between hot and cold parts, allowing the system to fully contact and bear force with the mold; hence the name "load-bearing system." This completely solves the existing problems of traditional systems. Although the system consists of small components, it can be assembled into a single, monolithic system for easy assembly and disassembly. Threaded connections are used at potential leakage points, minimizing the risk of glue leakage.

[0026] Hot runners can come into contact with the mold over a large area and directly bear the clamping force. Most of the external area of ​​the hot runner system does not generate heat (only internal heat is generated, and the outside is insulated). The hot runner system has no expansion in the lateral and longitudinal directions. Large systems do not require large equipment for processing, and the precision is more reliable. The local runner dimensions can be modified arbitrarily without increasing costs. The system has a very simple and pleasing appearance.

[0027] In the application of a load-bearing integral hot runner system to micro-foaming injection molding, each servo cylinder drives a valve needle to perform pulsed rapid retraction and forward movements to trigger bubble nucleation within the mold cavity. During each injection cycle, when the valve needle performs the pulsed retraction, it applies a sharp force impact to the support pad within a millisecond timescale through melt pressure changes and servo cylinder reaction forces. The dynamic component of this force impact is amplified by the dynamic response of the bridging body, generating additional irreversible micro-displacements. In existing technologies, the optimization of valve needle pulse parameters is solely aimed at improving bubble nucleation quality; faster pulse retraction speeds and greater retraction distances result in steeper nucleation pressure drops and more uniform bubble formation, but also greater force impacts on the support pad and bridging body. Furthermore, the prediction of the bridging body's displacement accumulation rate is based on univariate linear regression of cold-state position data, which cannot distinguish the respective contributions of the gradual change force of the conventional thermal cycle and the pulsed impact force to the displacement accumulation. When the pulse parameters become more aggressive due to increased pore density requirements, the intensified force impact causes the actual cumulative displacement rate to deviate from the linear regression prediction, resulting in insufficient temperature bias correction and overly optimistic alignment life prediction. Therefore, pulse parameter optimization and displacement accumulation monitoring are unaware of each other's constraints within their respective control objectives, constituting a complex control misalignment problem.

[0028] According to an embodiment of this invention, this embodiment provides a load-bearing integral hot runner system for micro-foaming injection molding. It should be understood that the hardware environment for implementing the load-bearing integral hot runner system includes: high-frequency pressure sensors mounted on each load-bearing pad in the load-bearing integral hot runner system, valve needle position encoders configured on each servo cylinder, absolute position sensors mounted on each bridging body, and a control processing unit with real-time data acquisition and calculation capabilities. The sampling frequency of the above sensors and encoders should meet the time-domain resolution requirements of millisecond-level pulse events.

[0029] At least one embodiment of the present invention discloses a load-bearing integral hot runner system, such as Figure 1 As shown, it includes the following steps: Step 1: Extract the time series of the periodic force impact of each bridging entity; High-frequency pressure timing signals of each support pad in a load-bearing integral hot runner system during micro-foaming injection molding were acquired. During the valve needle pulse execution period of each injection cycle, time-domain analysis was performed on the pressure timing signals of each support pad to extract the peak pressure change values. Peak rate of pressure change and pulse duration Among them, the peak pressure change The peak pressure rate is the maximum absolute value of the pressure timing signal during the pulse execution period minus the baseline pressure value before the pulse start. To obtain the maximum absolute value after performing a first-order difference on the pressure timing signal, the pulse duration is... The length of the continuous period during which the pressure deviates from the baseline pressure value by more than a preset noise threshold.

[0030] Furthermore, the preset noise threshold is the statistical upper bound of the fluctuation amplitude of the bearing pad pressure timing signal during the static period when the valve needle does not perform pulse action. This is used to distinguish the effective pressure change caused by the pulse from the sensor's background noise, thereby accurately defining the pulse duration. The start and end times.

[0031] Based on the force transmission relationship between each bearing pad and the corresponding bridging body, the force impact amount applied to each bridging body by each pulse action is calculated. Generate the time series of periodic force impact quantities for each bridging entity, where To bridge the main logo, This is the injection molding cycle number. Impact force. The integral of the dynamic pressure component measured on the bearing pad during a single pulse action over time, after being converted by the force transmission coefficient, is expressed as follows:

[0032] in, In order to be with the first A set of load-bearing pads that form a bridging structure and have a force transmission relationship. For the identification of the bearing pad block, For the first The first bearing pad block to the first Force transmission coefficient (dimensionless) of each bridging entity. For the first The instantaneous pressure value of each bearing pad during the pulse period. This is the baseline pressure value before the pulse begins. For the first The effective bearing area of ​​each bearing pad The pulse start time, This represents the pulse duration. In the above formula, The dimension of force is force, and after integrating over time... The dimension of the impulse is force multiplied by time. The contribution terms of each bearing pad have the same dimension, and the summation and conversion operations are valid in terms of dimension.

[0033] Furthermore, the force transmission coefficient Reflecting the bearing pad block With the bridging body The degree of structural coupling and force transmission coefficient between them The value is determined in advance through finite element simulation or static loading calibration test. Specifically, the bearing pad is tested in the calibration test. Apply a static load of known amplitude and simultaneously measure the bridging body. The response force is used as the force transfer coefficient, with the ratio of the response force to the applied load. The calibration value; for the bridging body The force transfer coefficient of a load-bearing pad without direct structural connection The value is zero and is not included in the summation.

[0034] Step 2: Calculate the valve needle pulse parameter set for each servo electric cylinder; Obtain the wall thickness parameters of the cavity region corresponding to each gate. and target pore density parameters ,in This serves as a gate location identifier. Based on the microfoaming nucleation kinetics equation, the pressure drop amplitude required to trigger ideal cell nucleation at each gate location is calculated. and the rate of pressure drop The nucleation kinetic equation describes the relationship between the bubble nucleation rate and the pressure drop. The expression for the nucleation kinetic equation is:

[0035] in, For the first Nucleation rate at each gate location, For frequency factors, Represents an exponential function. The interfacial tension between the bubble and the melt. Boltzmann's constant, The melting temperature is... This represents the pressure drop value. Pi is the mathematical constant. Based on the target bubble density... Inversely determine the required pressure drop. Specifically, the target pore density Divide by the cavity filling time to obtain the target nucleation rate Then the target nucleation rate Substituting into the above nucleation kinetic equations, for Perform the inverse solution, that is, let Taking the natural logarithm of both sides of the equation and simplifying, we get Combined with wall thickness parameters Determine the rate of pressure drop The lower limit value.

[0036] Furthermore, the rate of pressure decrease The lower limit is determined by the wall thickness parameter. The reason for the constraint is that when the wall thickness is smaller, the melt cools faster and the bubble nucleation window time is shorter, requiring the pressure drop to be completed in a shorter time. Therefore, the lower limit of the pressure drop rate increases as the wall thickness decreases.

[0037] Based on the current melt pressure and load-bearing capacity values ​​at each gate, the required pressure drop amplitude and pressure drop rate are converted into the valve needle retraction distance of each servo electric cylinder. Valve needle retraction speed and holding time Generate valve needle pulse parameter sets for each servo electric cylinder. ,in For the first Each gate position corresponds to the valve needle retraction distance of the servo electric cylinder. The valve needle retraction speed, To maintain the time frame, the above conversion is based on the geometric relationship between the valve needle retraction stroke and the change in gate cross-sectional area, as well as the rheological relationship between melt flow resistance and pressure. The valve needle retraction distance... With the magnitude of pressure drop Positive correlation, valve needle retraction speed With the rate of pressure drop Positive correlation.

[0038] Furthermore, the holding time The duration for which the valve needle remains stationary after retracting to the target position; the holding time. The lower limit of the value is determined by the nucleation kinetics constraint, that is, the holding time must not be shorter than the given pressure drop value. Lower bubble nucleation rate The integral reaches the target cell density The minimum required time is determined to ensure that there is a sufficient nucleation time window within the cavity to complete the formation of cells at the target density.

[0039] Step 3: Extract the pulse impact displacement sensitivity and basic displacement rate of each bridging entity; Acquire the position readings of the absolute position sensors on each bridging body during the cold phase of each thermal cycle. ,in To bridge the main logo, Number the thermal cycle. Record the position readings. Compared with the initial installation reference position value Compare and calculate the cold position offset. The time series of periodic force impacts of each bridging entity are accumulated and summed to generate a cumulative impact sequence. ,in For injection molding cycle numbering, The upper limit of the cumulative summation is the corresponding thermal cycle number.

[0040] Furthermore, the cumulative impact sequence Upper limit Offset from cold position thermal cycle number Maintain consistency, i.e., the cumulative impact sequence. For from the first The first injection cycle to the [number]th injection cycle At the end of each thermal cycle, the internal force impact of all injection cycles. The cumulative sum is used to align the cumulative impact with the cold position offset in the time dimension, ensuring that the two independent variables and the dependent variable have a corresponding relationship at the same thermal cycle node in the subsequent regression analysis.

[0041] Before performing regression analysis, the independent variables... and cumulative impact sequence Mean normalization based on range was applied to eliminate the influence of the dimensional difference between thermal cycle number (dimensionless quantity) and cumulative impact (impact dimension) on the regression coefficient estimation. The normalized independent variable is denoted as... and dependent variable (Displacement dimensions) Simultaneously undergo mean normalization based on the range, denoted as... .

[0042] Numbered according to normalized thermal cycle and normalized cumulative impact As the independent variable, the normalized cold-state position offset is used. Using the least squares method to solve for the parameters of the following bivariate linear regression model with the variable as the dependent variable:

[0043] in, For the first in the normalized space The regression coefficients of the basic displacement rate of the bridging main body The regression coefficients for pulse impact displacement sensitivity in the normalized space are denoted as . This is the regression residual. After regression, the regression coefficients of the basic displacement rate are... and pulse impact displacement sensitivity regression coefficient Inverse transformation using the corresponding normalized scaling factor restores the basic displacement rate to its original dimensions. (Unit: displacement / period) and pulse impact displacement sensitivity (Unit: displacement / impulse), for use in subsequent steps.

[0044] It should be noted that when the pulse parameters remain constant during operation, the cumulative impact sequence... and The two variables exhibit an approximately linear relationship, but collinearity exists, resulting in low separation of the regression results. This regression analysis is more effective during data accumulation periods where the impulse parameters have undergone over-adjustment, as parameter adjustments can disrupt the cumulative impulse sequence. and The linear relationship between the two contributing components is improved, thereby enhancing the separability of the two contributing components.

[0045] Step 4: Predict the alignment life of each bridging main body after pulse condition correction; Based on pulse impact displacement sensitivity and basic displacement rate The predicted total cumulative displacement of each bridging entity is calculated using the dual-source displacement accumulation model. The expression for the dual-source displacement accumulation model is:

[0046] in, For the first The first bridging entity will be in the future. Predicted cumulative displacement over one thermal cycle The number of future thermal cycles, counting from the current moment. The average force impact corresponding to the current valve needle pulse parameter set is calculated by taking the arithmetic mean of the force impact on each bridging body within the most recent injection cycles. When the valve needle pulse parameter set is updated, the calculation is recalculated based on the cycle data executed with the updated parameters. In the above formula, the basic displacement rate... The dimension is displacement / period, pulse impact displacement sensitivity. The dimensions are displacement / impulse, and the periodic average force impact quantity is... The dimension of is impulse / cycle, therefore and The dimensions of both are displacement, and the addition of the two terms is dimensionally valid.

[0047] Let the upper limit of the allowable alignment accuracy of each bridging body hot nozzle-gate be. The current accumulated cold position offset is Then the remaining allowable displacement is Introducing a safety factor To improve prediction robustness, a dual-source displacement accumulation model is used to predict the remaining number of thermal cycles required for the cold-state position offset to reach the upper limit of the allowable alignment accuracy.

[0048] in, For the first The number of remaining heat cycles for each bridging element. This is the remaining allowable displacement. For safety reasons, Based on the displacement rate, For pulse impact displacement sensitivity, This represents the periodic average impact force. Safety factor. The value is determined based on the pulse impact displacement sensitivity. The historical fluctuation range is determined; the greater the fluctuation, the lower the safety factor. The larger the value, the better to compensate for the prediction uncertainty caused by changes in sensitivity. This generates the alignment lifetime prediction values ​​after pulse condition correction for each bridging entity. .

[0049] Furthermore, pulse impact displacement sensitivity Historical fluctuation amplitude obtained from pulse impact displacement sensitivity at different evaluation intervals The range or standard deviation of the value series is used as a measure. When this measure exceeds a preset fluctuation tolerance, the safety factor is applied. A larger value is selected according to the preset mapping relationship to make the alignment life prediction result more conservative, thereby retaining a more sufficient structural safety margin when the uncertainty of sensitivity estimation is high.

[0050] Step 5: Solve for the modified impulse parameter set under structural safety constraints; Corrected alignment lifetime prediction Below the preset safety threshold The bridging body compares the force impact of the valve needle pulse action at each gate position in the direction of displacement of the bridging body, and extracts the servo electric cylinder identifier corresponding to the gate position with the largest component.

[0051] The optimization of the valve needle pulse parameters of the servo electric cylinder is extended to multi-objective optimization. The first objective is to minimize the bubble nucleation quality deviation based on the nucleation kinetic equation. The second objective is to minimize the force impact exerted by the pulse action on the critical bridging component. The weighted sum method is used to construct the comprehensive objective function:

[0052] in, and These are the weighting coefficients. For the valve needle pulse parameter set to be optimized, This represents the normalized deviation in bubble nucleation quality. This represents the normalized force impact. To ensure the two objective terms are comparable on a numerical scale, the bubble nucleation quality deviation is considered before constructing the comprehensive objective function. Impact force Perform mean normalization based on the range, and denote the normalized values ​​as follows: and The range and mean used for normalization are determined based on the statistical range of the corresponding quantities in historical operating data. (Bubble nucleation quality deviation) For a given set of valve needle pulse parameters The following is a description of the quality deviation of bubble nucleation. The calculation method is as follows: The valve needle pulse parameter set... Substituting into the nucleation kinetics equation, we obtain the corresponding nucleation rate. Then calculate the nucleation rate. With the target nucleation rate The absolute deviation between them, i.e. The target nucleation rate From the target pore density parameter Determined by dividing by the cavity filling time. Impact force. Valve needle pulse parameter set The corresponding impact force is determined by the impact force calculation formula in step 1. The greater the valve needle retraction speed and the greater the valve needle retraction distance, the greater the impact force. The weights are dynamically allocated based on the nucleation quality tolerance and the urgency of the alignment life: the greater the nucleation quality tolerance, the greater the impact force. Relative increase to favor structural protection, aligned with life prediction values The closer to the preset safety threshold but Relative increase.

[0053] Furthermore, the nucleation quality tolerance margin is defined as the deviation in bubble nucleation quality under current operating conditions. Distance to the maximum allowable limit The remaining space, i.e., the upper limit of allowance. Deviation from the currently measured bubble cell nucleation quality The difference; the urgency of alignment life versus the predicted alignment life. With preset safety threshold The relative value of the difference is a measure of urgency; the smaller the difference, the higher the degree of urgency. and satisfy Both are greater than zero, and the specific ratio between the two is determined according to the above two metrics based on a preset mapping relationship.

[0054] The optimization also includes the following constraints:

[0055] in, For the first The upper limit of the allowable deviation in nucleation quality at each gate location is a constraint that ensures that the cell nucleation quality still meets the basic process requirements even under optimized force impact. Solving the above weighted sum optimization problem outputs the set of corrected pulse parameters under structural safety constraints. Correct pulse parameter set Includes the corrected valve needle retraction distance Valve needle retraction speed and holding time It can be directly sent to the corresponding servo electric cylinder for execution.

[0056] Furthermore, the solution to the aforementioned weighted sum optimization problem is achieved using the valve needle pulse parameter set. Within the feasible region, a grid search or gradient descent method is performed to evaluate or iteratively optimize the comprehensive objective function value point by point, while satisfying the nucleation quality deviation constraint. The parameter set that minimizes the comprehensive objective function value among the feasible solutions is selected as the corrected impulse parameter set. Output.

[0057] Step 6: Monitor the nucleation spectrum characteristics in real time and fine-tune the pulse parameters at the Pareto front; Furthermore, in order to promptly compensate for possible nucleation quality shifts after the correction pulse parameters are executed, the following steps are also included in step 5: Correct the pulse parameter set The signal is sent to the corresponding servo cylinder for execution. During pulse execution, a short-time Fourier transform is performed on the high-frequency pressure signal of each bearing pad to extract the bubble nucleation spectral features at each gate location. The window length and frequency resolution of the short-time Fourier transform are set according to the characteristic time scale of the bubble nucleation event. The measured spectral features are compared with the pre-calibrated qualified nucleation spectral range. The qualified nucleation spectral range is the statistical distribution range of the spectral features obtained after short-time Fourier transform of the bearing pad pressure signal collected under known good bubble nucleation conditions.

[0058] When the nucleation spectrum characteristics deviate from the acceptable nucleation spectrum range, in the current corrected pulse parameter set within the neighborhood The search is performed in an increased direction to find neighboring solutions on the Pareto front. The nearest solution that allows the nucleation spectrum characteristics to regress to the acceptable nucleation spectrum range is selected as the fine-tuned pulse parameter. Simultaneously, the corresponding force impact change value is recorded after fine-tuning. .

[0059] Furthermore, the adjacent solutions on the Pareto front are optimized in step 5 by... and Different ratios are used to perform multiple solutions, which are pre-generated and stored. When step 6 requires... When increasing the directional search, directly search from the pre-stored solution set according to... The nucleation spectrum characteristics corresponding to each candidate solution are examined in ascending order. The first candidate solution that makes the measured spectrum characteristics return to the qualified nucleation spectrum range is selected as the fine-tuning result, so as to avoid repeating the complete optimization calculation during the real-time monitoring stage.

[0060] Step 7: Apply asymmetric temperature bias correction to the critical bridging components; Furthermore, in order to further slow down the rate of displacement accumulation, the following steps are included in addition to step 4: Corrected alignment lifetime prediction Below the preset warning threshold The bridging body is identified, and the heat pipe segment that contributes the most is identified based on the displacement direction vector of the bridging body. The asymmetric temperature offset that needs to be applied to this heat pipe segment during each thermal cycle heating phase is calculated. The thermal expansion force of the heat pipe section generates a correction component opposite to the direction of the cumulative displacement. The calculation is based on the thermal expansion coefficient of the heat pipe section material, the constraint boundary conditions of the heat pipe section, and the magnitude and direction of the target correction force. There is an upper limit constraint on the temperature offset, that is, the temperature set value after offset must not cause the temperature of the heat pipe section to exceed the allowable range of the melt process temperature window.

[0061] The temperature offset correction command is superimposed on the temperature control setpoint of the corresponding heat pipe section and executed. During subsequent thermal cycles, the change in cold-state position offset of the affected bridging entity is continuously acquired, and this change in cold-state position offset is compared with the applied temperature offset. The measured displacement increment is compared with the previous one to verify the correction effect. If the measured displacement increment does not decrease to the expected level after multiple consecutive thermal cycles, the temperature offset is increased by a preset step size. .

[0062] Furthermore, the expected level is based on the thermal expansion coefficient of the heat pipe section and the temperature offset amount after applying asymmetric temperature offset correction. The difference between the calculated theoretical corrected displacement and the single-cycle displacement increment predicted by the dual-source displacement accumulation model indicates the current temperature offset when the measured displacement increment consistently exceeds the expected level. The generated correction component is insufficient to offset the actual displacement accumulation, therefore the temperature offset is increased in increments of a preset step size. Continue until the correction effect meets the requirements or reaches the upper limit constraint of temperature offset.

[0063] Step 8: Adaptively update control parameters based on sensitivity trend analysis; Furthermore, to address the issue of pulse impact displacement sensitivity changing over time due to gradual factors such as fatigue of the bridging material or wear of the contact surface, the following steps are also included in addition to steps 3 and 5: After several consecutive evaluation intervals, the pulse impact displacement sensitivity of each bridging entity was assessed using accumulated dual-source displacement data. Trend analysis was performed. The pulse impact displacement sensitivity calculated at different evaluation intervals was compared. The values ​​form a time series, and linear regression is performed on this time series to extract the pulse impact displacement sensitivity. slope of change .

[0064] When the slope changes That is, pulse impact displacement sensitivity When the trend is upward, perform the following actions: automatically increase the multi-objective optimization... The baseline value makes subsequent optimization results more biased towards structural protection; simultaneously, according to the changing slope... The size ratio increases the corresponding temperature offset correction amount for the heat pipe section. The slope of the updated sensitivity trend data. and alignment lifetime prediction Write it into the maintenance decision database for use when formulating maintenance plans.

[0065] This implementation establishes a quantitative correlation between pulse events and structural loads by extracting the cycle-by-cycle force impact amount applied to each bridging body by the valve needle pulse action from the high-frequency pressure signal of the bearing pad in step 1. Based on this, step 3 performs a bivariate regression analysis on the cumulative sequence of force impact amounts and the cold-state position offset of the bridging body to separate the pulse impact displacement sensitivity. and basic displacement rate With two independent parameters, the dual-source displacement accumulation model in step 4 can distinguish the contribution of the gradual force of the conventional thermal cycle and the pulsed impact force to the displacement accumulation. When the pulse parameter becomes more aggressive due to the increased pore density requirements, the dual-source displacement accumulation model automatically adjusts the predicted value of the displacement accumulation rate by the change in the periodic average force impact, overcoming the limitation of univariate linear regression in predicting inaccuracies when pulsed operating conditions change. Therefore, the alignment life prediction results can reflect the actual impact of pulsed operating conditions.

[0066] Furthermore, step 5 incorporates force impact constraints into the nucleation kinetics optimization of the valve needle pulse parameters, forming a multi-objective optimization framework. This allows for a Pareto trade-off between the bubble nucleation quality objective and the safety objective of the bridging main structure in the same solution process. The weights are dynamically allocated based on the nucleation quality margin and the urgency of the alignment life. Therefore, when the alignment life prediction value... Approaching the preset safety threshold At the same time, the optimization results automatically bias towards reducing the impact force, rather than pursuing the ultimate optimization of nucleation quality without constraints. This synergy enables the valve needle pulse parameter optimization to perceive structural safety constraints, and the prediction results of displacement accumulation monitoring can be fed back to the valve needle pulse parameter adjustment, thus avoiding the problem that the two operate independently without being aware of each other's constraints.

[0067] The following is an example of an application of the present invention, such as... Figure 2-8 As shown, the implementation process is as follows: An automotive interior parts manufacturer uses a load-bearing integral hot runner system to produce micro-foamed polypropylene instrument panel frames. The mold has four gate positions (gate 1 to gate 4), corresponding to four sets of servo-driven electric cylinder valve needles. The load-bearing structure includes three bridging bodies (bridging body 1 to bridging body 3). The system has been running continuously for 480 thermal cycles, and each bridging body has accumulated a certain amount of cold-state positional shift. The current production task requires increasing the cell density to [a certain value]. For each unit per cm³, the process engineer needs to determine whether the alignment life of the bridging body meets the requirements of the next planned maintenance window (no less than 200 remaining thermal cycles) under more aggressive pulse parameters, and implement coordinated control intervention if necessary.

[0068] The initial installation reference positions for the system are: 0.000mm for bridging body 1, 0.000mm for bridging body 2, and 0.000mm for bridging body 3. The upper limit of the allowable alignment accuracy between the hot nozzle and the gate for each bridging body is 0.120mm.

[0069] During the injection cycle at the end of the 480th thermal cycle, the high-frequency pressure sensor acquired the pressure timing signals of each support pad. Taking the 480th injection cycle as an example, the pulse characteristic parameter extraction results of each support pad are shown in the table below. The force transmission relationship between the support pad and the bridging body is as follows: support pad 1 and support pad 2 correspond to bridging body 1, support pad 3 corresponds to bridging body 2, and support pad 4 corresponds to bridging body 3.

[0070] Table 1. Extraction results of pulse characteristic parameters of each bearing pad in the 480th injection cycle.

[0071] Based on the force impact calculation formula, taking bridging body 1 as an example, its associated bearing pad set is {bearing pad 1, bearing pad 2}, with force transmission coefficients of 0.73 and 0.68 respectively. The dynamic pressure component within the pulse period is integrated over time, approximated as the trapezoidal integral of the pressure change peak value multiplied by the equivalent pulse duration. The force impact of bridging body 1 is then calculated as follows:

[0072] The force impact of each bridging component during the 480th injection cycle is summarized as follows: Table 2 Impact Values ​​of Each Bridging Component in Injection Cycle 480

[0073] Obtain the wall thickness and target cell density parameters for each gate location. Gates 1 and 2 correspond to the central area of ​​the instrument panel and have thinner walls; gates 3 and 4 correspond to the mounting areas on both sides and have thicker walls. The target cell density is uniformly set to... Nucleation rate: [Number of nuclei / cm³], cavity filling time: 1.8s, target nucleation rate:

[0074] Nucleation kinetic parameters: N / m, J / K, K (melt temperature 220°C) The target nucleation rate is 1 / (cm³·s). Substituting the target nucleation rate into the inverse kinematics formula, the required pressure drop amplitude for each gate is obtained. Then, combined with the wall thickness of each gate, the lower limit of the pressure drop rate is determined, and finally converted into a valve needle pulse parameter set.

[0075] Table 3 Set of needle pulse parameters for each gate valve

[0076] Because gates 1 and 2 have smaller wall thicknesses, faster cooling rates, and shorter nucleation window times, they require higher valve needle retraction speeds to meet the lower limit of pressure drop rate. Consequently, the force impact on the bearing pad is also greater, which corroborates the result that the bearing pads 1 and 2 have the highest force impact in the bridging body 1 in step 1.

[0077] Using historical data accumulated by the system from the 1st to the 480th thermal cycle, including a pulse parameter adjustment (approximately 15% increase in pullback speed) near the 180th thermal cycle due to increased pore density requirements, the linear relationship between the cumulative impact sequence and the thermal cycle number was broken, effectively improving the separability of the bivariate regression. Taking bridging module 1 as an example, data from several representative thermal cycle nodes are selected as follows: Table 4. Input data for the bivariate regression of bridging subject 1 (representative nodes)

[0078] After the 180th thermal cycle, due to the adjustment of the pulse parameters, the cumulative impact per cycle increased from about 294 N·ms to about 318 N·ms. This resulted in a larger growth slope of the cumulative impact after the 180th thermal cycle, which was no longer strictly linear with the thermal cycle number, providing a separable condition for bivariate regression.

[0079] Least squares linear regression with two independent variables was performed on the normalized data, yielding a normalized basic displacement rate coefficient of 0.612 and a normalized pulse impact displacement sensitivity coefficient of 0.276. Inverse transformation using normalized proportionality coefficients (thermal cycle range 480°, cumulative impact range approximately 157,200 N·ms, offset range approximately 0.090 mm) yielded the parameters in their original dimensions.

[0080] The regression results for the three bridging subjects are summarized below: Table 5. Bivariate regression results for each bridging entity.

[0081] The current (end of the 480th thermal cycle) accumulated cold position offsets of each bridging element are: bridging element 1: 0.0793 mm, bridging element 2: 0.0541 mm, and bridging element 3: 0.0387 mm. Based on the force impact data of the last 50 injection cycles, the cycle average force impact corresponding to the current pulse parameter set is calculated as follows: bridging element 1: 318.4 N·ms / cycle, bridging element 2: 214.6 N·ms / cycle, and bridging element 3: 121.8 N·ms / cycle.

[0082] Taking the bridging body 1 as an example, the remaining allowable displacement is:

[0083] The range of the historical fluctuation of the pulse impact displacement sensitivity of bridging entity 1 (based on the first 4 evaluation intervals) is: mm / (N·ms), exceeding the preset fluctuation tolerance. mm / (N·ms), therefore a safety factor of 1.25 is taken. The predicted number of remaining thermal cycles is:

[0084] Table 6. Predicted lifespan of each bridging component.

[0085] The predicted number of remaining thermal cycles for bridging body 1 is 197, which is lower than the preset safety threshold of 200, triggering the collaborative control intervention in step 5.

[0086] The key gates corresponding to the bridging body 1 are gate 1 and gate 2. Among them, gate 1 corresponds to the valve needle retraction speed with the highest speed (186mm / s) and the force impact component in the displacement direction of the bridging body 1 is the largest. Therefore, the servo electric cylinder corresponding to gate 1 is determined to be the optimization object.

[0087] The nucleation quality deviation of the current gate 1 is calculated using the nucleation kinetic equation as follows: The per unit / (cm³·s) capacity is [number] units, with an upper limit of [percentage]. The nucleation rate is [number] per (cm³·s), indicating a relatively ample margin in nucleation quality. The alignment lifetime urgency is [high]. The situation is already below the safety threshold, indicating a high degree of urgency. Based on the preset mapping relationship, the weights are allocated as follows: nucleation quality weight 0.35, structural safety weight 0.65.

[0088] A grid search was performed on the feasible region defined by the valve needle retraction distance (0.48–0.62 mm) and retraction speed (140–190 mm / s), with step sizes of 0.02 mm and 5 mm / s, respectively, while ensuring that the nucleation quality deviation did not exceed [the specified value]. Among the feasible solutions to constraints of 1 / (cm³·s), the parameter combination that minimizes the comprehensive objective function value is selected.

[0089] Table 7. Optimization process of the modified pulse parameter set for gate 1 (partial candidate solutions)

[0090] The comprehensive objective function value of candidate solution D, 0.586, is the minimum value that satisfies the constraints. The output correction pulse parameter set is: pullback distance 0.50 mm, pullback speed 148 mm / s, and holding time 8.4 ms. After correction, the periodic average force impact of gate 1 is reduced to 231.8 N·ms, and the remaining thermal cycles of bridging body 1 are re-estimated.

[0091] The remaining number of thermal cycles has been increased to 215 after the correction, exceeding the safety threshold of 200 cycles, and meeting the requirements of the next planned maintenance window.

[0092] After the set of corrected pulse parameters (retraction distance 0.50mm, retraction speed 148mm / s, holding time 8.4ms) is sent to the servo electric cylinder corresponding to gate 1 for execution, a short-time Fourier transform is performed on the high-frequency pressure signal of the bearing pad 1. The window length is set to 4ms, the frequency resolution is 250Hz, and it covers the bubble nucleation characteristic frequency band (800~3200Hz).

[0093] The measured spectrum shows that the energy density in the 1600–2400 Hz frequency band is MPa² / Hz, while the qualified nucleation spectrum range requires an energy density of not less than [amount missing] in this frequency band. The measured value of MPa² / Hz is lower than expected, indicating that the pressure drop rate is slightly insufficient after the retraction speed is reduced, and the bubble nucleation spectrum characteristics deviate from the qualified range.

[0094] In the pre-stored Pareto front solution set, adjacent candidate solutions are examined along the direction of increasing nucleation mass weight: the candidate solution corresponding to a nucleation mass weight of 0.45 (retreat distance 0.52mm, retreat speed 158mm / s, holding time 8.2ms) is selected, corresponding to a force impact of 263.1 N·ms for bearing pad 1, and the measured spectral energy density recovers to MPa² / Hz, within the acceptable nucleation spectrum range. The recorded force impact change value is... N·ms. The remaining thermal cycles of the updated bridging body 1 were re-estimated to be approximately 207, which is still higher than the safety threshold of 200. The fine-tuning result was adopted and implemented.

[0095] The current alignment life prediction value of bridging body 1 is 207 cycles, which is lower than the preset warning threshold of 250 cycles, triggering temperature offset correction. The cumulative displacement direction of bridging body 1 (offset along the positive X-axis) is identified, and the heat pipe segment that contributes the most is heat pipe segment H1-2 (connecting bridging body 1 and the load-bearing structure on the side of gate 2).

[0096] The coefficient of thermal expansion of the material of heat pipe section H1-2 is / °C, effective constraint length is 120mm, current temperature setting is 238°C, melt process temperature window allowable range is 215~255°C, temperature offset upper limit is °C. Calculate the required temperature offset based on the target correction force:

[0097] The target single-cycle correction displacement of 0.0028 mm is 30% of the single-cycle displacement increment predicted based on the dual-source displacement accumulation model. The temperature offset correction command is superimposed on the temperature control setpoint of heat pipe section H1-2 and adjusted to... °C.

[0098] In the subsequent 5 thermal cycles, the change in the cold-state position offset of the bridging body 1 was continuously acquired. The measured single-cycle displacement increment changed from the original value. mm decreased mm, a decrease of approximately 5.4%, lower than the expected decrease. The temperature offset was increased to 1.86°C (corresponding to the temperature setpoint of 239.86°C) by a preset step size of 0.5°C, and the correction effect of subsequent thermal cycles was monitored.

[0099] After four evaluation intervals (120, 240, 360, and 480 thermal cycles), the estimated sequence of pulse impact displacement sensitivity values ​​for bridging body 1 is as follows: , , , mm / (N·ms), perform linear regression on this time series to obtain the slope of change:

[0100] Change slope The pulse impact displacement sensitivity shows an increasing trend when the mm / (N·ms·interval) value is greater than 0, indicating that the bridging body 1 may have slight wear on the contact surface, leading to increased structural coupling. The system performs the following adaptive update operations: the baseline value of the structural safety weight in the multi-objective optimization is increased from 0.65 to 0.72; the temperature offset of heat pipe section H1-2 is simultaneously increased by 0.3°C to 2.16°C proportionally to the slope; and the updated data (sensitivity change slope) is updated. mm / (N·ms·interval), remaining thermal cycles 207 times) are written into the maintenance decision database, and the bridging entity 1 is marked as needing to focus on checking the contact surface condition during the next planned maintenance.

[0101] The data flow throughout the implementation process demonstrates a clear logical chain: Step 1 extracts pulse characteristic parameters from the raw signal of the high-frequency pressure sensor, and converts them into the cycle-by-cycle force impact of each bridging body (340.5 N·ms for bridging body 1) through the force transmission coefficient, providing the load input basis for all subsequent steps; Step 2 determines the valve needle pulse parameter set based on the inverse solution of nucleation dynamics (retraction speed of gate 1 is 186 mm / s), the aggressiveness of which directly determines the magnitude of the force impact in Step 1; Step 3 utilizes the nonlinear characteristics brought about by the parameter adjustment of the 180th thermal cycle to successfully separate the basic displacement rate from the historical data of 480 thermal cycles. mm / cycle) and pulse impact displacement sensitivity ( Step 4 involves substituting the two parameters (mm / (N·ms)) and the current cycle average force impact into the dual-source displacement accumulation model, predicting that the remaining life of the bridging main body 1 is only 197 cycles, triggering subsequent intervention; Step 5 uses weighted multi-objective optimization (nucleation quality weight 0.35, structural safety weight 0.65) to reduce the pullback speed of gate 1 from 186mm / s to 148mm / s, increasing the life of the bridging main body 1 to 215 cycles while meeting the nucleation quality constraint; Step 6 uses real-time spectrum monitoring to detect a slight deficiency in nucleation quality, fine-tuning the pullback speed to 158mm / s along the Pareto front, executing at the equilibrium point where the remaining life of 207 cycles and nucleation quality are both satisfied; Steps 7 and 8 use temperature bias correction and sensitivity trend analysis as long-term means to continuously slow down the displacement accumulation rate and adaptively update the control parameter weights, forming a complete closed-loop data flow link from single-cycle load extraction to long-term structural health management.

[0102] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. A supported monoblock hot runner system characterized by, The system executes the following method: During the valve needle pulse execution period of each injection cycle, the high-frequency pressure timing signal of each bearing pad is analyzed in the time domain. Based on the force transmission relationship between each bearing pad and the corresponding bridging body, the force impact amount applied to each bridging body by each pulse action is calculated, and the cycle-by-cycle force impact amount timing of each bridging body is generated. Based on the micro-foaming nucleation kinetic equation and the wall thickness parameters and target cell density parameters of each gate, the required pressure drop amplitude and pressure drop rate at each gate position are calculated and converted into the valve needle pulse parameter set of each servo electric cylinder. The cumulative impact sequence is generated by summing the time series of the periodic force impact. The pulse impact displacement sensitivity and foundation displacement rate of each bridging body are extracted by using the thermal cycle number and the cumulative impact sequence as independent variables and the cold position offset as dependent variable. Based on the pulse impact displacement sensitivity and basic displacement rate, the remaining thermal cycle number of each bridging body is predicted by the dual-source displacement accumulation model, and the alignment life prediction value after pulse condition correction is generated. For bridging entities whose alignment life prediction value is lower than the preset safety threshold, a weighted and comprehensive objective function is constructed to perform multi-objective optimization with the first objective being to minimize the bubble nucleation quality deviation and the second objective being to minimize the force impact amount, and the corrected pulse parameter set under structural safety constraints is solved.

2. The supported monolithic hot runner system of claim 1, wherein, The method for calculating the force impact is as follows: For each bearing pad that has a force transmission relationship with the target bridging body, calculate the instantaneous force value obtained by subtracting the baseline pressure value before the pulse start from the instantaneous pressure value during the pulse period and multiplying it by the effective force-bearing area of ​​the bearing pad. Then multiply the result by the force transmission coefficient from the bearing pad to the target bridging body. Summate the integral results of all relevant bearing pads to obtain the force impact amount applied to the target bridging body by a single pulse action.

3. The supported monolithic hot runner system of claim 2, wherein, The force transmission coefficient is predetermined through finite element simulation or static loading calibration test. In the static loading calibration test, a static load of known amplitude is applied to the target bearing pad, and the response force on the target bridging body is measured simultaneously. The ratio of the response force to the applied load is used as the calibration value of the force transmission coefficient. For a bearing pad that has no direct structural connection with the target bridging body, the corresponding force transmission coefficient is zero.

4. The load-bearing integral hot runner system according to claim 1, characterized in that, The specific process of the bivariate linear regression includes: The position readings of the absolute position sensors on each bridging body are obtained in each thermal cycle cold state phase, and the position readings are compared with the initial installation reference position values ​​to obtain the cold state position offset. Before performing regression analysis, mean normalization based on range was applied to the thermal cycle number and cumulative impact sequence, and mean normalization based on range was applied to the cold position offset. The least squares method is used to solve the parameters of the bivariate linear regression model in the normalized space to obtain the regression coefficients of the basic displacement rate and the pulse impact displacement sensitivity in the normalized space. Then, the basic displacement rate and pulse impact displacement sensitivity are transformed back to the original dimensions according to the corresponding normalized proportional coefficients.

5. The load-bearing integral hot runner system according to claim 1, characterized in that, The dual-source displacement accumulation model expresses the cumulative displacement as the sum of the product of the basic displacement rate and the number of future thermal cycles, plus the product of the pulse impact displacement sensitivity, the periodic average force impact, and the number of future thermal cycles. The periodic average force impact is taken as the arithmetic mean of the force impact experienced by each bridging body in the most recent injection cycles. The remaining number of thermal cycles is calculated as follows: the remaining allowable displacement is obtained by subtracting the absolute value of the currently accumulated cold position offset from the upper limit of the alignment accuracy of each bridging body, and then dividing by the product of the safety factor and the sum of the basic displacement rate and the pulse impact displacement sensitivity multiplied by the periodic average force impact. The safety factor is greater than one and is determined based on the historical fluctuation amplitude of the pulse impact displacement sensitivity.

6. The load-bearing integral hot runner system according to claim 1, characterized in that, The weight coefficients in the weighted and comprehensive objective function are dynamically allocated based on the nucleation quality tolerance margin and the urgency of alignment life. The nucleation quality tolerance margin is the difference between the upper limit of tolerance and the current measured deviation of the nucleation quality of the bubble cells. The urgency of alignment life is measured by the relative value of the difference between the predicted value of alignment life and the preset safety threshold. The larger the nucleation quality tolerance margin or the closer the predicted value of alignment life is to the preset safety threshold, the greater the weight of the force impact quantity target. The sum of the two weighting coefficients is one and both are greater than zero; the multi-objective optimization also includes the constraint that the nucleation quality deviation does not exceed the upper limit of the nucleation quality deviation at the corresponding gate position.

7. The load-bearing integral hot runner system according to claim 1, characterized in that, After solving for the corrected pulse parameter set, the following steps are also included: The modified pulse parameter set is sent to the corresponding servo electric cylinder for execution. During the pulse execution, the high-frequency pressure signal of each bearing pad is subjected to short-time Fourier transform to extract the nucleation spectrum features of the bubble holes. The measured spectrum features are compared with the pre-calibrated qualified nucleation spectrum range. When the nucleation spectrum characteristics deviate from the qualified nucleation spectrum range, search for the nearest solution on the Pareto front of the multi-objective optimization in the neighborhood of the current corrected pulse parameter set along the direction of increasing target weight of nucleation quality deviation, and select the nearest solution that makes the nucleation spectrum characteristics return to the qualified nucleation spectrum range as the fine-tuned pulse parameters. The adjacent solutions on the Pareto front are pre-generated and stored during the multi-objective optimization process by solving the problem multiple times with different ratios of the weight coefficients.

8. The load-bearing integral hot runner system according to claim 1, characterized in that, For bridging entities whose alignment life prediction value is lower than a preset warning threshold, the following are also included: Based on the identification of the heat pipe segment that contributes most to the displacement direction vector of the bridging body, the asymmetric temperature offset that needs to be applied to the heat pipe segment in each thermal cycle heating stage is calculated so that the thermal expansion force of the heat pipe segment generates a correction component opposite to the cumulative displacement direction. The temperature offset is subject to an upper limit constraint, that is, the temperature set value after offset shall not cause the temperature of the heat pipe segment to exceed the allowable range of the melt process temperature window. In subsequent thermal cycles, the change value of the cold position offset of the affected bridging body is continuously acquired. If the measured displacement increment of multiple consecutive thermal cycles does not decrease to the expected level, the temperature offset is increased according to the preset step size. The preset warning threshold is greater than the preset safety threshold.

9. The load-bearing integral hot runner system according to claim 4, characterized in that, Also includes: After several consecutive evaluation intervals, the pulse impact displacement sensitivity values ​​calculated at different evaluation intervals are used to form a time series. Linear regression is then performed on this time series to extract the slope of the change in pulse impact displacement sensitivity. When the slope of change is greater than zero, the baseline value of the force impact target weight in the multi-objective optimization is automatically increased, and the temperature offset correction of the corresponding heat pipe section is increased proportionally according to the magnitude of the slope of change. The updated slope of change and the alignment life prediction value are then written into the maintenance decision database.

10. A load-bearing integral hot runner system, characterized in that, include: The force impact extraction module is used to perform time-domain analysis on the high-frequency pressure timing signal of each bearing pad during the valve needle pulse execution period of each injection cycle. Based on the force transmission relationship, it calculates the force impact amount applied to each bridging body by each pulse action and generates a cycle-by-cycle force impact timing sequence. The pulse parameter calculation module is used to calculate the required pressure drop amplitude and pressure drop rate at each gate position based on the micro-foaming nucleation kinetic equation, the wall thickness parameters of each gate, and the target cell density parameters, and convert them into the valve needle pulse parameter set of each servo electric cylinder. The displacement sensitivity regression module is used to accumulate and sum the time series of the cycle-by-cycle force impact to generate a cumulative impact sequence. With the thermal cycle number and the cumulative impact sequence as independent variables and the cold position offset as the dependent variable, a bivariate linear regression is used to extract the pulse impact displacement sensitivity and basic displacement rate of each bridging body. The alignment life prediction module is used to predict the remaining thermal cycle number of each bridging body based on the pulse impact displacement sensitivity and the basic displacement rate through a dual-source displacement accumulation model, and generate the alignment life prediction value after pulse condition correction. The multi-objective optimization module is used to optimize the bridging body whose alignment life prediction value is lower than the preset safety threshold. The first objective is to minimize the bubble nucleation quality deviation, and the second objective is to minimize the force impact. The module constructs a weighted and comprehensive objective function to solve for the set of corrected pulse parameters under structural safety constraints.

Images