A two-plate injection molding machine brake holding control method and system based on a prediction model
By analyzing historical data using predictive models and adjusting the position of the gateposts, the instability of the two-platen injection molding machine caused by external factors was resolved, improving the stability of the injection molding machine and the accuracy of its positional deviation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HAITIAN PLASTICS MACHINERY GRP
- Filing Date
- 2026-03-05
- Publication Date
- 2026-06-26
AI Technical Summary
The existing control algorithm for the clamping nut of a two-platen injection molding machine is affected by external factors, which can cause the gatepost to not clamp smoothly or even fail to clamp, thus affecting the operational stability of the injection molding machine.
Historical data is analyzed using a predictive model. By collecting production modulus, determining positional deviation, generating prediction formulas, calculating specific gravity coefficients and compensation distances, the position of the Goring column is adjusted to improve stability.
By analyzing historical deviations using predictive models, the position of the gate column can be accurately adjusted, thereby improving the operational stability and accuracy of positional deviations in two-platen injection molding machines.
Smart Images
Figure CN122275259A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of injection molding control, and in particular to a method and system for brake control of a two-platen injection molding machine based on a predictive model. Background Technology
[0002] Two-platen injection molding machines are injection molding equipment that uses a mold-closing system with two parallel templates (fixed template and moving template).
[0003] In the prior art, the fixed template is generally equipped with a clamping nut, and the movable template is generally equipped with a tie rod. The clamping nut and the tie rod are the pre-stage for establishing clamping force. The existing clamping nut control generally moves the tie rod to a target position and stops the movement after reaching the target position, without predictive adjustment of the tie rod position.
[0004] However, during the production process, the original clamping nut control algorithm may fail to clamp smoothly or even fail to clamp at all due to a series of external factors, such as changes in the template position caused by thermal expansion and contraction of the mold, hydraulic oil leakage caused by increased machine running time, and overshooting of the clamping column caused by changes in the pressure and flow settings of the clamping column movement. Summary of the Invention
[0005] To improve the operational stability of a two-platen injection molding machine, this invention provides a brake control method and system for a two-platen injection molding machine based on a predictive model, which analyzes historical data using a predictive model to optimize the control of the brake position.
[0006] In a first aspect, the present invention provides a brake control method for a two-platen injection molding machine based on a predictive model, employing the following technical solution: A brake control method for a two-platen injection molding machine based on a predictive model includes: Step 100: Collect production modules; Step 200: Determine the positional deviation based on the production module; Step 300: When the positional deviation does not fall within the preset deviation range, retrieve the historical deviation based on the production module; Step 400: Generate a prediction formula based on the historical deviation; Step 500: Determine the weight coefficient in response to the prediction formula; Step 600: Determine the compensation distance by combining the specific gravity coefficient and the prediction formula; Step 700: Determine the target location based on the compensation distance; Step 800: In response to the target location, generate and send a Goring column compensation command.
[0007] By adopting the above technical solution, the positional deviation of the gatepost in each mold is calculated. When the positional deviation is large, the historical deviation of the gatepost is retrieved. The historical deviation is then analyzed by the prediction model to obtain the deviation in the next mold. In this way, an appropriate compensation distance is selected to adjust the position of the gatepost and improve the stability of the two-platen injection molding machine.
[0008] Optionally, the method for determining the positional deviation includes: Step 201: Retrieve the actual position, target position, and template position of the guide column based on the production module; Step 202: Calculate the difference between the actual position and the target position, and define it as the Goring column deviation; Step 203: Calculate the difference between the deviation of the goring column and the position of the template, and define it as the position deviation.
[0009] By adopting the above technical solution, when the guide column is controlled to approach the clamping nut in the locking direction, the fixed template is prone to displacement under the action of the moving template, which aggravates the positional deviation of the guide column. At this time, the template position is retrieved, and the positional deviation is corrected according to the template position, thereby improving the accuracy of the positional deviation.
[0010] Optionally, the method for generating the prediction formula includes: Step 401: Determine the deviation modulus based on the historical deviation and the preset deviation range; Step 402: Determine the continuous modulus by combining the deviation modulus and the production modulus; Step 403: Retrieve the continuous deviation based on the continuous modulus, and generate autoregressive coefficients based on the continuous modulus; Step 404: Generate a prediction formula in response to the continuous deviation and autoregressive coefficients.
[0011] By adopting the above technical solution, the continuous moduli with consecutively large deviations in the historical deviations are checked. Then, the historical deviation and autoregression coefficient corresponding to each continuous moduli are multiplied as a term in the prediction formula to obtain a polynomial with the same number of terms as the continuous moduli, thereby improving the prediction accuracy of the deviation of the next moduli.
[0012] Optionally, the method for calculating the specific gravity coefficient includes: Step 501: Determine the prediction error by combining the production modulus and the prediction formula; Step 502: Calculate the sum of squared errors based on the prediction error and the deviation modulus; Step 503: Obtain the partial derivative formula by taking the partial derivative of the sum of squared errors with respect to the autoregressive coefficients; Step 504: Determine the minimum coefficient according to the partial derivative formula; Step 505: Determine the weight coefficient in response to the minimum coefficient.
[0013] By adopting the above technical solution, the sum of squared errors is first calculated according to the prediction formula. Then, the least squares method is used to find the partial derivative of the sum of squared errors with respect to the autoregressive coefficients to obtain multiple partial derivative formulas. Through algebraic operations of multiple partial derivative formulas, the autoregressive coefficient that minimizes the sum of squared errors is obtained as the minimum coefficient. In this way, an accurate prediction formula is obtained and the position deviation of the next mold is calculated, thereby improving the stability of the two-platen injection molding machine.
[0014] Secondly, this application provides a two-platen injection molding machine brake control system based on a predictive model, employing the following technical solution: A predictive model-based brake control system for a two-platen injection molding machine includes: The data acquisition module is used to collect production data. The memory is used to store the program of any of the above-mentioned two-platen injection molding machine brake control methods based on predictive models; The processor is the unit of memory that allows programs to be loaded and executed by the processor.
[0015] By adopting the above technical solution, the positional deviation of the gatepost in each mold is calculated. When the positional deviation is large, the historical deviation of the gatepost is retrieved. The historical deviation is then analyzed by the prediction model to obtain the deviation in the next mold. In this way, an appropriate compensation distance is selected to adjust the position of the gatepost and improve the stability of the two-platen injection molding machine.
[0016] In summary, this application includes at least one of the following beneficial technical effects: The positional deviation of the gatepost in each mold is calculated. When the positional deviation is large, the historical deviation of the gatepost is retrieved. The historical deviation is then analyzed by the prediction model to obtain the deviation in the next mold. Then, an appropriate compensation distance is selected to adjust the position of the gatepost and improve the stability of the two-platen injection molding machine. When the gatepost moves closer to the retaining nut in the locking direction, the fixed template is prone to displacement under the action of the moving template, which aggravates the positional deviation of the gatepost. At this time, the template position is adjusted to correct the positional deviation according to the template position, thereby improving the accuracy of the positional deviation. By examining the continuous moduli with consecutively large deviations in the historical deviations, the historical deviation and autoregressive coefficient corresponding to each continuous moduli are multiplied together as a term in the prediction formula to obtain a polynomial with the same number of terms as the continuous moduli. This improves the accuracy of predicting the deviation of the next moduli. Attached Figure Description
[0017] Figure 1This is a schematic diagram of a two-platen injection molding machine brake control based on a predictive model. Figure 2 This is a flowchart of a brake control method for a two-platen injection molding machine based on a predictive model.
[0018] The parts referred to by the numbers in the above attached diagrams are as follows: 1. Fixed template; 2. Moving template; 3. Clamping nut; 4. Tie rod. Detailed Implementation
[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the 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 and not intended to limit the invention.
[0020] This application discloses a brake control method for a two-platen injection molding machine based on a predictive model.
[0021] Reference Figure 1 and Figure 2 A method for brake control of a two-platen injection molding machine based on a predictive model, comprising: Step 100: Collect production modules.
[0022] Production modulus refers to the timing and production cycle value defined based on the injection molding production process. The t-th mold is the t-th injection molding production process. The production modulus can be collected from the control system of the injection molding machine. The method of collecting the production modulus is selected by the staff according to the actual situation, and will not be elaborated here.
[0023] Step 200: Determine the positional deviation based on the production module.
[0024] Positional deviation refers to the distance between the gatepost 4 and the retaining nut 3 after the gatepost 4 has moved in the mold-locking direction. For example, the positional deviation of the first mold is X1, and the positional deviation of the t-th mold is X. t The calculation method for position deviation is as described in steps 201 to 203 below, and will not be repeated here.
[0025] Step 300: When the positional deviation does not fall within the preset deviation range, retrieve the historical deviation according to the production module.
[0026] The deviation range refers to the allowable range of positional deviation, i.e., (-ε0, ε0). The deviation range is selected by the staff based on the actual situation and will not be elaborated here. If the positional deviation does not fall within the deviation range, it means that the positional deviation of the gatepost 4 is too large, and the position of the gatepost 4 needs to be adjusted to reduce the possibility of the gatepost 4 not engaging smoothly or even directly causing the engagement nut 3 to fail to engage.
[0027] Historical deviation refers to the positional deviation of the module before production. Historical deviation can be retrieved from the injection molding machine's control system. The method for retrieving historical deviation is selected by the staff based on the actual situation, and will not be elaborated here.
[0028] Step 400: Generate a prediction formula based on the historical deviation.
[0029] The prediction formula is a formula used to predict the position deviation of the next model. The method for generating the prediction formula is described in steps 401 to 404 below, and will not be repeated here.
[0030] Step 500: Determine the weight coefficient in response to the prediction formula.
[0031] The weight coefficient refers to the autoregressive coefficient value of each item in the prediction formula. The method for determining the weight coefficient is described in steps 501 to 505 below, and will not be repeated here.
[0032] Step 600: Determine the compensation distance by combining the specific gravity coefficient and the prediction formula.
[0033] The compensation distance refers to the distance at which the position of column 4 needs to be adjusted. The compensation distance can be calculated by substituting the specific gravity coefficient into the prediction formula. The calculation method for the compensation distance is selected by the staff according to the actual situation and will not be elaborated here.
[0034] Step 700: Determine the target location based on the compensation distance.
[0035] The target position refers to the position of the gate column 4 after adjusting its position according to the compensation distance. That is, the target position is calculated as the sum of the compensation distance and the initial position. The initial position refers to the current position of the gate column 4, which can be retrieved from the control system of the injection molding machine.
[0036] Step 800: In response to the target location, generate and send a Goring column compensation command.
[0037] The gatepost compensation command is a command that controls the movement of gatepost 4 according to the target position in the next module. The method of generating the gatepost compensation command is common knowledge to those in the field and will not be elaborated here.
[0038] The positional deviation of the gatepost 4 in each mold is calculated. When the positional deviation is large, the historical deviation of the gatepost 4 is retrieved. The historical deviation is then analyzed by the prediction model to obtain the deviation in the next mold. Then, an appropriate compensation distance is selected to adjust the position of the gatepost 4 and improve the stability of the two-platen injection molding machine.
[0039] Methods for determining positional deviation include: Step 201: Retrieve the actual position, target position, and template position of the column based on the production module.
[0040] The actual position refers to the real-time position of the gatepost 4, the target position refers to the predetermined position of the gatepost 4, i.e. the position of the retaining nut 3, and the template position refers to the position of the fixed template 1. The actual position, target position and template position can be retrieved from the control system of the injection molding machine, which will not be elaborated here.
[0041] Step 202: Calculate the difference between the actual position and the target position, and define it as the Goring column deviation.
[0042] The deviation of the guide column refers to the distance between the guide column 4 and the target position. The deviation of the guide column shows the positional deviation of the guide column 4 when the template 1 is fixed.
[0043] Step 203: Calculate the difference between the deviation of the goring column and the position of the template, and define it as the position deviation.
[0044] When the gatepost 4 and the movable template 2 move in the locking direction, the fixed template 1 is prone to move in the locking direction under the action of the gatepost 4 and the movable template 2, which leads to an increase in the positional deviation of the gatepost 4.
[0045] When the guide column 4 moves closer to the clamping nut 3 in the locking direction, the fixed template 1 is prone to displacement under the action of the moving template 2, which aggravates the positional deviation of the guide column 4. At this time, the template position is retrieved, and the positional deviation is corrected according to the template position, thereby improving the accuracy of the positional deviation.
[0046] Methods for generating prediction formulas include: Step 401: Determine the deviation modulus based on the historical deviation and the preset deviation range.
[0047] The deviation modulus refers to the timing and production cycle value that needs to be adjusted for the position of the guide column 4, i.e., the modulus corresponding to the historical deviation that does not fall into the deviation range. The method for determining the deviation modulus is selected by the staff according to the actual situation, and will not be elaborated here.
[0048] Step 402: Determine the continuous modulus by combining the deviation modulus and the production modulus.
[0049] The continuous modulus refers to the deviation modulus that is continuous with the production modulus. The method for determining the continuous modulus is selected by the staff based on the actual situation, and will not be elaborated here.
[0050] Step 403: Retrieve continuous deviations based on the continuous modulus, and generate autoregressive coefficients based on the continuous modulus.
[0051] Continuous deviation is the historical deviation corresponding to the continuous modulus. The method for determining continuous deviation is selected by the staff based on the actual situation, and will not be elaborated here.
[0052] Autoregressive coefficients are numerical values used to represent the strength of the influence of historical deviations on the positional deviation of the next model; they are generated in the form of Φ1, Φ2, ..., Φ m The coefficient of , where m refers to the number of consecutive moduli.
[0053] Step 404: Generate a prediction formula in response to the continuous deviation and autoregressive coefficients.
[0054] The sum of the products of continuous deviations and autoregressive coefficients is used as the prediction formula, i.e., X. t =Φ1X t-1 +Φ2X t-2 +......Φ m X t-m , where t refers to the module of the next module, i.e., t = production module + 1.
[0055] By examining the continuous moduli with consecutively large deviations in the historical deviations, the historical deviation and autoregressive coefficient corresponding to each continuous moduli are multiplied together as a term in the prediction formula to obtain a polynomial with the same number of terms as the continuous moduli. This improves the accuracy of predicting the deviation of the next moduli.
[0056] The methods for calculating the specific gravity coefficient include: Step 501: Determine the prediction error by combining the production modulus and the prediction formula.
[0057] Prediction error refers to the difference between the calculated position deviation for the next model and the actual value, i.e., prediction error e. t =X t -(Φ1X t-1 +Φ2X t-2 +......Φ m X t-m ).
[0058] Step 502: Calculate the sum of squared errors based on the prediction error and the deviation modulus.
[0059] The sum of squared errors refers to the sum of the squares of the prediction errors, i.e., the sum of squared errors S = Σ(t=1 to m)e t 2 The method for calculating the sum of squared errors is selected by the staff based on the actual situation, and will not be elaborated here.
[0060] Step 503: Obtain the partial derivative formula by taking the partial derivative of the sum of squared errors with respect to the autoregressive coefficients.
[0061] The partial derivative formula refers to the system of equations obtained by taking the partial derivatives of the autoregressive coefficients with respect to the sum of squared errors. The partial derivative formula contains m unknowns (Φ1, Φ2, ..., Φ). m A system of linear equations consisting of m equations, i.e., the partial derivative formula = ∂s / ∂Φ t (t=1 to m), the calculation method of the partial derivative formula is selected by the staff according to the actual situation, and will not be elaborated here.
[0062] Step 504: Determine the minimum coefficient according to the partial derivative formula.
[0063] The minimum coefficient is the autoregressive coefficient value when the sum of squared command errors is minimized. The method for determining the minimum coefficient is common knowledge among those in the field and will not be elaborated here.
[0064] Step 505: Determine the weight coefficient in response to the minimum coefficient.
[0065] The minimum coefficient is substituted into the corresponding position of the autoregressive coefficient in the prediction formula as the weighting coefficient.
[0066] First, the sum of squared errors is calculated according to the prediction formula. Then, the least squares method is used to find the partial derivative of the sum of squared errors with respect to the autoregressive coefficients to obtain multiple partial derivative formulas. Through algebraic operations of multiple partial derivative formulas, the autoregressive coefficient that minimizes the sum of squared errors is obtained as the minimum coefficient. In this way, an accurate prediction formula is obtained and the position deviation of the next mold is calculated, thereby improving the stability of the two-platen injection molding machine.
[0067] Based on the same inventive concept, embodiments of the present invention provide a two-platen injection molding machine brake control system based on a predictive model, comprising: The data acquisition module is used to collect production data. The memory is used to store the program of any of the above-mentioned two-platen injection molding machine brake control methods based on predictive models; The processor is the unit of memory that allows programs to be loaded and executed by the processor.
[0068] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0069] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.
Claims
1. A method for controlling a brake of a two-plate injection molding machine based on a prediction model, characterized by, The method comprises the following steps: Step 100: collecting production modulus; Step 200: determining position deviation based on the production modulus; Step 300: calling historical deviation according to the production modulus when the position deviation does not fall into a preset deviation interval; Step 400: generating a prediction formula based on the historical deviation; Step 500: determining a specific gravity coefficient in response to the prediction formula; Step 600: determining a compensation distance by combining the specific gravity coefficient and the prediction formula; Step 700: determining a target position according to the compensation distance; Step 800: generating and sending a Gelin column compensation instruction in response to the target position.
2. The method of claim 1, wherein the method further comprises: The method for determining the position deviation comprises the following steps: Step 201: calling the actual position, target position and template position of the Gelin column based on the production modulus; Step 202: calculating the difference between the actual position and the target position, and defining it as Gelin column deviation; Step 203: calculating the difference between the Gelin column deviation and the template position, and defining it as position deviation.
3. The method of claim 1, wherein the method further comprises: The method for generating the prediction formula comprises the following steps: Step 401: determining a deviation modulus according to the historical deviation and a preset deviation interval; Step 402: determining a continuous modulus by combining the deviation modulus and the production modulus; Step 403: calling a continuous deviation based on the continuous modulus, and generating an autoregressive coefficient according to the continuous modulus; Step 404: generating a prediction formula in response to the continuous deviation and the autoregressive coefficient.
4. The method of claim 3, wherein the method further comprises: The method for calculating the specific gravity coefficient comprises the following steps: Step 501: determining a prediction error by combining the production modulus and the prediction formula; Step 502: calculating error sum of squares according to the prediction error and the deviation modulus; Step 503: obtaining a partial derivative formula by taking partial derivative of the error sum of squares based on the autoregressive coefficient; Step 504: determining a minimum coefficient according to the partial derivative formula; Step 505: determining a specific gravity coefficient in response to the minimum coefficient.
5. A two-plate injection molding machine brake holding control system based on a prediction model, characterized by, The method comprises the following steps: A collection module is configured to collect production modulus; A memory is configured to store a program of a two-plate injection molding machine brake control method based on a prediction model according to any one of claims 1 to 4; A processor, and the program in the memory can be loaded and executed by the processor.