JP2017081071A - Molding support method of injection molder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve acceleration of optimization processing without reducing reliability related to optimization and surely and easily perform optimization to a predicted molding condition.SOLUTION: An invention is configured to: set a constraint condition Xc..., and an objective function Xp based on a mathematical programming to molding data related to an input parameter Df..., and an output parameter Ds..., in advance; set an optimization processing program Ps for determining an optimized molding condition related to the input parameter Df... satisfying the constraint condition Xc..., and the objective function Xp by the mathematical programming; during production utilization, acquire molding data related to the output parameter Ds... in a plasticization step; then, determine a prediction function F... and the optimized molding condition optimized by the optimization processing program Ps from the molding data related to the output parameter Ds; then, change a molding condition in other shot after an optional shot is completed by the optimized molding condition.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a molding support method for an injection molding machine suitable for use in optimizing molding conditions using a neural network.

  Conventionally, as a method (means) for supporting molding by an injection molding machine, a method (means) using a neural network has also been proposed. As this kind of method (means), a patent already proposed by the present applicant has been proposed. An injection molding machine control method disclosed in Document 1 and an injection molding machine support apparatus disclosed in Patent Document 2 are known.

  The control method of the injection molding machine disclosed in Patent Document 1 is intended to provide an injection molding method capable of high-speed molding and considering a plurality of variable factors. In a control method of an injection molding machine to which a control device is provided, trial molding is performed by the injection molding machine, and a measurement monitor value in the measurement process and an injection monitor value in the injection process are obtained, and the obtained measurement Using the monitor value as the input item and the injection monitor value as the output item, the prediction function determining process that determines the prediction function by the neural network of the control device, and mass production molding is started by the injection molding machine, and the weighing monitor at that process at the end of the weighing process The control device that has acquired the value substitutes the acquired measurement monitor value into a prediction function, thereby predicting a first value corresponding to the injection monitor value, and a prediction Based on the second value determining step of processing the first value by the control device and determining the second value which is one of the injection conditions, and the injection condition including the second value, the control device And an injection process for performing injection control and pressure holding control.

  The injection molding machine support device disclosed in Patent Document 2 is intended to provide a support device that is suitable for education of unskilled workers and is useful for the operation of an actual injection molding machine. Is a support device for an injection molding machine that assists an operator involved in the operation of the injection molding machine, and this support device uses a plurality of molding conditions used when a good product is obtained by trial molding as input items. The quality value obtained by measuring non-defective products is used as an output item, and a neural network that determines a prediction function based on these input items and output items, and at least one of a plurality of molding conditions are selected and selected. When a first input device that inputs a fixed value to a molding condition other than the molding condition and the molding condition selected by the first input device is called a specific molding condition, an output item is set as an unknown in the prediction function. By entering a fixed value in the input item and calculating with the specific molding condition in the form of a variable, the calculation unit that calculates the predicted quality value that is the predicted value of the quality value, and the relationship between the specific molding condition and the predicted quality value It includes a graph generation unit that graphs and a display unit that displays information on this graph generation unit, and allows the operator to visually observe the predicted quality value corresponding to the specific molding condition that he / she has selected. It is.

JP 2008-110485 A JP 2008-110486 A

  However, the conventional molding support method for the above-described injection molding machine has the following problems to be solved.

  First, when using a neural network, it is possible to learn data obtained by trial molding and obtain a prediction function by pattern recognition. Therefore, since it becomes a method of modeling a certain physical phenomenon in a pseudo manner, it is possible to simulate the output parameter from the input parameter without actually operating the molding machine. On the other hand, it is also possible to predict an input parameter for obtaining a target output parameter by switching and learning the input parameter and the output parameter. However, if you select multiple output parameters, for example, “plasticization time”, “resin temperature”, and “power consumption”, which are the three output parameters, you can calculate combinations that are not physically possible. Since the above holds, it is almost useless when there are many output parameters used for prediction. In addition, when changing the input parameter to be derived, it is necessary to reset the input / output relationship of the neural network and to re-learn it. For this reason, the range of use is limited and it is difficult to use.

  Second, when optimizing input parameters, the resolution is set for the input parameters that are real numbers, and by changing each parameter, simulation (prediction processing) is performed for all combinations, and the most desired The combination along is searched. Therefore, for example, when the resolution is 100 and the number of parameters is 4, an enormous amount of processing that is 100 to the fourth power is required. In the end, there is a limit from the viewpoint of realizing high-speed optimization processing, and although it can be used for analysis at the experimental level, the processing should be performed within one molding cycle at the actual production site and applied to subsequent production. Therefore, practical use is not easy, because it becomes difficult to use it.

  An object of the present invention is to provide a molding support method for an injection molding machine that solves the problems existing in the background art.

  In order to solve the above-described problems, the present invention obtains the data processing unit E having a computer function by molding data related to the input parameters Df including the molding conditions of the injection molding machine M and trial molding based on the input parameters Df. A molding support method for an injection molding machine that obtains a prediction function F... Based on learning of the neural network N from molding data related to the output parameter Ds... And predicts predetermined molding conditions using the prediction function F. In the processing unit E, the constraint condition Xc... And the objective function Xp based on the mathematical programming are set for the molding data related to the input parameter Df... And the output parameter Ds. Optimized molding conditions related to the input parameter Df ... satisfying the objective function Xp Optimization processing program Ps is set, and during the production operation, the data processing unit E acquires molding data related to the output parameter Ds... During the plasticizing step (S6-S7) at an arbitrary shot. During the period other than the conversion step (S6-S7), the molding condition optimized by the prediction function F ... and the optimization processing program Ps is obtained from the molding data related to the output parameter Ds ... The present invention is characterized in that a process for changing molding conditions in another shot after an arbitrary shot is completed is performed.

  In this case, according to a preferred aspect of the invention, a period from the end of the plasticizing process (S8) to the start of the mold clamping process (S10) can be applied as a period other than the plasticizing process (S6-S7). . Further, as the constraint condition Xc, it is possible to select molding data indicating the state of the resin during the plasticizing step (S6-S7), and it is possible to select power consumption as the objective function Xp. One or more of the screw rotation speed, screw rotation speed, screw position, resin pressure, resin temperature, plasticizing time, and / or one or more of these deviations can be selected. On the other hand, the next shot can be applied to other shots after the completion of an arbitrary shot. A nonlinear function can be applied to the function related to the objective function Xp.

  According to the molding support method of the injection molding machine according to the present invention by such a method, the following remarkable effects are achieved.

  (1) Molding data relating to the output parameter Ds... During the plasticizing step (S6-S7) at an arbitrary shot is acquired, and during the period other than the plasticizing step (S6-S7), the output parameter Ds. Optimized molding conditions are obtained from the molding data related to, by using the prediction function F... Based on learning of the neural network N, the constraint condition Xc based on mathematical programming, the objective function Xp, and the optimization processing program Ps. Since the process of changing the molding condition in another shot after the completion of an arbitrary shot depending on the molding condition is performed, optimization with respect to the predicted molding condition can be performed easily and reliably. As a result, the range of use can be expanded, versatility can be improved, and a molding support method excellent in usability can be provided.

  (2) Even when the input parameter Df... Is optimized, the processing amount related to the optimization can be effectively reduced without reducing the reliability related to the optimization. The processing speed can be increased. As a result, processing within one molding cycle at an actual production site can be performed, and a molding support method having high practicality at an actual production site can be constructed such that it can be used effectively in subsequent production.

  (3) According to a preferred embodiment, if a period from the end of the plasticizing process (S8) to the start of the mold clamping process (S10) is applied as a period other than the plasticizing process (S6-S7), the process is optimized. Even in the case of changing to the molding condition, it can be changed without affecting the plasticizing step (S6-S7). That is, even during production operation, the molding quality can be improved without affecting the production volume.

  (4) According to a preferred embodiment, if the molding data indicating the state of the resin in the plasticizing step (S6-S7) is selected as the constraint condition Xc and the power consumption is selected as the objective function Xp, the plasticizing step (S6 -Power consumption can be minimized while stabilizing molding data (resin temperature, etc.) indicating the state of the resin at S7), so energy saving and running cost can be reduced without deteriorating molding quality. Can do.

  (5) According to a preferred embodiment, as molding data indicating the state of the resin in the plasticizing step (S6-S7), one of screw rotation speed, screw rotation speed, screw position, resin pressure, resin temperature, plasticizing time or If two or more and / or one or two or more of these deviations are selected, it is possible to secure a sufficient type of molding data that influences the state of the resin during the plasticizing step (S6-S7). Optimization from the viewpoint of saving energy and reducing running costs can be performed easily and flexibly without lowering.

  (6) According to a preferred embodiment, if the next shot is applied to another shot after completion of an arbitrary shot, the input parameter Df... Since the change process of ... can be realized promptly, it is possible to improve the quality of each individual molded product in each shot, which can contribute to quality improvement and homogenization of the entire production.

  (7) If a non-linear function is applied to the objective function Xp according to a preferred embodiment, it is possible to cope with various problems expressed as an arbitrary non-linear function, which is most desirable in optimizing the molding conditions of the injection molding machine M. It can be used as an embodiment.

The flowchart for demonstrating the process sequence at the time of the actual production in implementing the shaping | molding assistance method which concerns on suitable embodiment of this invention, A flow chart for explaining a processing procedure before production when carrying out the molding support method, A block system diagram showing a processing system of an injection molding machine and a data processing unit capable of carrying out the molding support method; An explanatory diagram of the principle of a neural network used to implement the molding support method, Block system diagram for explaining a pre-production processing system when carrying out the molding support method, A display screen diagram of a display attached to a data processing unit for carrying out the molding support method; Another display screen diagram of the display attached to the data processing unit that implements the molding support method, Another display screen diagram of the display attached to the data processing unit that implements the molding support method, Another display screen diagram of the display attached to the data processing unit that implements the molding support method, A table showing related data when setting constraints and objective functions used in the implementation of the molding support method, A graph showing a change in injection peak pressure in each shot when the molding support method is carried out, A graph showing a change in plasticizing time in each shot when the molding support method is carried out,

  Next, preferred embodiments according to the present invention will be given and described in detail with reference to the drawings.

  First, configurations of the injection molding machine M and the data processing unit E that can implement the molding support method according to the present embodiment will be described with reference to FIGS.

  In FIG. 3, M is an injection molding machine partly drawn with phantom lines, and includes a machine base Mb, and an injection device Mi and a mold clamping device Mc mounted on the machine base Mb. The injection device Mi includes a heating cylinder 31, and a screw that rotates and moves back and forth is accommodated in the heating cylinder 31, and an injection nozzle that does not appear in the drawing is provided at the front end of the heating cylinder 31. A hopper 32 for supplying a molding material (resin pellet) is provided at the rear portion of the cylinder 31. The mold clamping device Mc includes a mold 33 composed of a combination of a movable mold and a fixed mold, and a side panel 34 is provided on the machine base Mb, and a touch panel using a liquid crystal display or the like is attached to the side panel 34. The display 35 is disposed.

  The injection molding machine M incorporates a molding machine controller 41 having a computer function for performing various control processes and arithmetic processes and performing communication processing with the outside, and the display 35 described above is provided in the molding machine controller 41. Connecting. Furthermore, a sensor group 42 including various sensors for measuring (detecting) various physical quantities such as temperature, rotational speed, electric power, pressure, speed, position, time and the like during operation of the injection molding machine M is provided. Sensor) is connected to an analog input port (or digital input port) of the molding machine controller 41. Therefore, the molding machine controller 41 is a functional unit used to implement the molding support method according to the present embodiment. At least a measurement result capturing process functional unit that captures a part or all of the measurement results (detection results) of the sensor group 42. In addition to Ui, a sampling processing function unit Us that obtains a measurement value (molded data) by sampling the measurement result at a constant period.

  On the other hand, Em is a general computer system partially drawn with virtual lines, Ed is a display, and Ek is a keyboard. In addition, the display Ed includes a computer main body Em, and the computer main body Em has a general-purpose computer function for performing various control processes and arithmetic processes and for performing communication processes with the outside. The data processing unit E used in the molding support method according to the present embodiment is configured.

  Therefore, the computer main body Em includes hardware such as a CPU and an internal memory, and in this internal memory, the processing program used for the implementation of the molding support method according to the present embodiment, that is, the optimum based on the mathematical programming method. And a prediction function creation processing program Pn for creating a prediction function F... Based on learning of the neural network N is stored. Further, in relation to the optimization processing program Ps, it has a function of setting a constraint condition Xc and an objective function Xp based on mathematical programming. The computer main body Em and the molding machine controller 41 can communicate with each other by being connected by a LAN system or the like.

  Hereinafter, basic processing functions of the prediction function creation processing program Pn and the optimization processing program Ps will be described with reference to FIGS.

  First, the prediction function creation processing program Pn has, as a basic processing function, molding data relating to input parameters Df including the molding conditions of the injection molding machine M, and output parameters Ds obtained by trial molding based on the input parameters Df. A prediction function F... Based on learning of the neural network N is obtained from the molding data relating to the above, and a processing function for predicting a predetermined molding condition by the prediction function F.

FIG. 4 shows a multilayer structure model of the neural network N. A pattern to be processed is input to the input layer Ni, and the input is weighted by the intermediate layer Nm and transmitted to the output layer Mo. The total transmission amount o _i is expressed by [Equation 1]. In [Formula 1], x _i represents an input signal from the intermediate layer Nm, that is, an output signal from the intermediate layer Nm to the output layer No, and w _ij represents a weight.

  In this case, the intermediate layer Nm receives a signal input from the input layer Ni and outputs a signal to the output layer No. Further, the output layer No that receives signals from the plurality of intermediate layers Nm performs final output in response to each signal. As the output function at this time, the sigmoid function shown in [Expression 2] is used. As the output function, a Gaussian function, a trigonometric function, or the like is used in addition to the illustrated sigmoid function.

  FIG. 5 shows a pseudo plasticization model for obtaining the prediction function F... In plasticization according to the present embodiment by the neural network N. That is, measurement data 52 is obtained by a plasticization test 51 by trial molding. The measurement data 52 includes molding data relating to the input parameter Df... (Input layer Ni) including the molding conditions of the injection molding machine M, specifically, molding data relating to the heating cylinder temperature which is the heating temperature of the heating cylinder 31. 53, molding data 54 related to the rear 2 temperature at the rear 2 position of the heating cylinder 31, molding data 55 related to the rotational speed of the screw, and the like, and output parameters Ds obtained by the plasticizing test 51 based on this input parameter Df. Molding data relating to (output layer No.), specifically, molding data 56 relating to resin temperature, molding data 57 relating to standard deviation, molding data 58 relating to power consumption, and the like are included. Then, based on these molding data, the prediction function F... Is obtained by learning of the neural network N.

  6 to 9 show specific operation screens Vc and result screens Vm used when the prediction function F (plasticization model) in plasticization according to the present embodiment is obtained by the neural network N. FIG.

  FIG. 6 shows an operation screen Vc, which displays (selects) an input display unit Vci that displays (selects) an input parameter Df... That becomes a plasticizing condition in a plasticizing process during injection molding, and displays (selects) an output parameter Ds. Output display section Vco. In the case of the example, the input display unit Vci includes, as input parameters Df, from the top, the basic temperature of the heating cylinder, the temperature at the rear 1 of the heating cylinder, the temperature at the rear 2 of the heating cylinder, the measured value (SM) of the resin, the screw The back pressure, screw rotation speed, intermediate time, plasticizing time average, average resin temperature, and power consumption are displayed in order, and the basic temperature of the heating cylinder to the intermediate time are selected as the input parameter Df... . On the other hand, on the output display unit Vco, the plasticizing time average, the average resin temperature, and the power consumption are sequentially displayed from the top as the output parameter Ds..., And all of these indicate the selected state as the output parameter Ds. Yes.

  Further, a predetermined number of molding data (measurement data 52) corresponding to each input parameter Df... And each output parameter Ds... Is obtained in advance from a plasticizing step in trial molding performed a plurality of times. Therefore, when the learning key Vcw on the operation screen Vc shown in FIG. 6 is turned ON, the prediction function creation processing program Pn using the neural network N learns a predetermined number of molding data obtained by trial molding and recognizes the pattern. Thus, a certain physical phenomenon is modeled in a pseudo manner. Thus, the plasticization result can be simulated according to the plasticizing conditions by the input parameters Df... Without actually operating the injection molding machine M.

  FIG. 8 is a result screen Vm corresponding to the operation screen Vc of FIG. The result screen Vm includes a prediction function display unit Vmx for displaying the prediction function F learned by the prediction function creation processing program Pn using the neural network N, and an X-axis selection unit Vmx for selecting a parameter to be displayed on the X-axis. And a Y-axis selection unit Vmy for selecting a parameter to be displayed on the Y-axis. In the case of illustration, the case where the basic temperature of the heating cylinder is selected by the X axis selection unit Vmx and the average resin temperature is selected by the Y axis selection unit Vmy is shown. As a result, the prediction data display unit Vmf plots the shaping data, which is the learned result, and displays a pseudo correlation curve La. This correlation curve La becomes the prediction function F, and the average resin temperature can be predicted from the basic temperature of the heating cylinder.

  On the other hand, the molding data in FIG. 6 and FIG. 8 is left as it is, and the molding conditions (for obtaining the desired result) are learned by switching the input layer Ni (input parameter Df...) And the output layer (output parameter Ds. A prediction function F (plasticization model) for predicting the input parameter Df.

  7 changes the plasticizing time average, average resin temperature, and power consumption used as the output parameters Ds... In FIG. 6 to the input parameters Df... And the basic temperature and heating of the heating cylinder used as the input parameters Df. This shows a case where the temperature at the rear part 1 of the cylinder and the temperature at the rear part 2 of the heating cylinder are changed to output parameters Ds.

  FIG. 9 shows a result screen Vm corresponding to the operation screen Vc in FIG. 7, a case where the average resin temperature is selected by the X axis selection unit Vmx, and the average resin temperature of the heating cylinder is selected by the Y axis selection unit Vmy. Show. As a result, the shaping data, which is the learned result, is plotted and the pseudo correlation curve Lb is displayed on the prediction function display unit Vmf. This correlation curve Lb becomes the prediction function F, and the basic temperature of the heating cylinder capable of obtaining the average resin temperature can be predicted.

  On the other hand, the optimization processing program Ps has, as a basic processing function, a processing function for obtaining an optimized forming condition related to the input parameter Df... Satisfying the constraint condition Xc. In this process, the constraint condition Xc... And the objective function Xp set based on the mathematical programming method are used for the molding data related to the input parameter Df... And the output parameter Ds.

  In order to facilitate understanding of the processing function of the optimization processing program Ps, an algorithm for optimizing the molding conditions based on mathematical programming will be described. Various algorithms can be considered as an algorithm to be optimized using the constraint condition Xc... Set based on mathematical programming and the objective function Xp. As an example suitable for this embodiment, one of the interior point methods is used. An algorithm using the trust region method is described.

  Now consider the optimization problem of the nonlinear function shown in [Equation 3]. F (x) in [Equation 3] indicates one output layer No in the neural network N.

  In the trust region method, optimization can be performed by minimizing a quadratic approximation model of the objective function Xp.

First, the search direction is determined by provisionally giving the size (reliability radius) of a region (reliability region) where the quadratic approximation model is considered appropriate. When the confidence radius in the k-th iterative calculation is Δ ^k , the search direction can be obtained as a solution to the subproblem of [Equation 4], considering a quadratic approximation model of f (x).

Next, an initial point x ⁰ and an initial confidence radius Δ ⁰ are given. The parameters 0 <η ₁ ≦ η ₂ <1, 0 <γ ₁ <1 <γ ₂ are determined, and k = 0.

As a result, if it can be determined that x ^k is an approximation of the local optimal solution, the process ends.

On the other hand, p ^k is ^obtained by solving the subproblem [Formula 4]. Thus, if [Equation 5] holds, x ^{k + 1} = x ^k + ^pk is set.

At this time, if [Equation 6] holds, Δ ^{k + 1} ∈ [Δ ^k , γ ₂ Δ ^k ] is used, and the trust region is expanded. Otherwise, Δ ^{k + 1} = Δ ^k Maintain the current status.

If [Equation 5] does not hold, Δ ^{k + 1} ∈ [γ ₁ Δ ^k , Δ ^k ] is used to reduce the trust region.

  Then, k = k + 1 is set and the end condition is checked.

  The above is an algorithm using the trust region method which is one of the interior point methods in mathematical programming, and the data processing unit E uses the above-described prediction function F (prediction function) from the molding data related to the acquired output parameter Ds. The molding conditions optimized by the creation processing program Pn) and the optimization processing program Ps can be obtained.

  Next, the processing procedure of the molding support method according to the present embodiment will be described according to the flowcharts shown in FIGS. 1 and 2 with reference to FIGS.

  First, a processing procedure before production operation when the molding support method according to the present embodiment is performed will be described with reference to a flowchart shown in FIG.

  The process before the production operation basically performs trial molding, collects a predetermined number of molding data related to the input parameters Df... And a predetermined number of molding data related to the output parameters Ds. Thus, the prediction function F... Is created and the constraint condition Xc and the objective function Xp are set.

  The molding support according to this embodiment is intended to optimize the plasticizing condition among the molding conditions. Therefore, first, the current state of the target injection molding machine M is grasped, and control factors related to optimization of plasticizing conditions, that is, input parameters Df... And output parameters Ds... Are set (steps S21 and S22). ). For example, grasping the current situation such as “there is a relationship between the stability of the rear 2 temperature and the plasticizing time” and “the plasticization state tends to be unstable when the cycle time is long”, and the related rear 2 temperature and The cycle time or the like is used as a control factor, and the parameters are assigned to input parameters Df... And output parameters Ds.

  When the setting of the control factor is completed, trial molding is performed using the input parameters Df... (Steps S23 and S24). Further, from the trial molding, a predetermined number of molding data related to the set control factor, that is, the input parameter Df... Related to the plasticizing condition among the molding conditions and the predetermined number of molding data related to the output parameter Ds. Step S25).

  When the trial molding is performed a predetermined number of times and the collection of the molding data is completed, the prediction function creation processing program Pn is executed, and the prediction function F... Based on learning of the neural network N is obtained from the collected molding data (steps S26, S27). ). When the prediction function F... Is obtained, it is incorporated into the optimization processing program Ps as a plasticization model (steps S28 and S29).

  On the other hand, a constraint condition Xc and an objective function Xp based on mathematical programming necessary for functioning the optimization processing program Ps are set (steps S30 and S31). In the case of the example, as the constraint condition Xc, the measured value is 20 [mm], the back pressure is 8 [MPa], the cycle time (filling time + plasticization time + intermediate time) is 12.5 [seconds], and the maximum plasticization delay The time was set to 0.1 [second] or less, the average resin temperature was set to 255 to 258 [° C.], the resin temperature shot standard deviation was set to 0.1 or less, and the purge pressure shot standard deviation was set to 0.08 or less. Moreover, the minimization of the power consumption per shot was set as the objective function Xp.

  Thus, if at least the resin temperature is included in the constraint condition Xc and the power consumption is applied to the objective function Xp, for example, the power consumption can be minimized while stabilizing the resin temperature during production operation. There is an advantage that energy saving and running cost can be reduced without deteriorating the quality. In this case, the objective function Xp is a nonlinear function. Thus, a nonlinear function can be applied to the objective function Xp. Therefore, various problems expressed as an arbitrary non-linear function can be dealt with, and there is an advantage that it can be used as the most desirable mode in optimizing the molding conditions of the injection molding machine M.

  FIG. 10 shows actual items of items related to the input parameters Df... And output parameters Ds... And objective functions and constraint conditions for each item, and related reference conditions, optimization conditions, and input conditions when trial molding is performed. The obtained data table is shown.

  FIG. 11 shows the magnitude of the injection peak pressure [MPa] for each shot. The black dots show the results of trial molding under the reference conditions shown in FIG. 10, and the white dots are optimum. The result of trial molding according to the conditions after the crystallization treatment is shown. FIG. 12 shows the length of plasticization time [seconds] for each shot. Black dots indicate the results of trial molding under the reference conditions shown in FIG. 10, and white dots indicate optimization processing. The result of having performed trial molding under the following conditions is shown.

  Next, a processing procedure during actual production operation when the molding support method according to the present embodiment is performed will be described with reference to a flowchart shown in FIG.

  First, the operation of the injection molding machine M is started (step S1). In this case, the injection molding machine M is in the automatic operation mode, and the mold clamping device Mc is in the mold open state. Further, the injection device Mi is in a plasticization end state (measurement end state).

  On the other hand, a mold clamping process is performed at the start of the molding process (steps S2 and S3). In the mold clamping step, the movable mold in the mold open state moves forward with respect to the fixed mold, and the mold 33 is clamped by the set mold clamping force. An injection process is performed upon completion of the mold clamping process (steps S4 and S5). In the injection process, the screw in the heating cylinder 31 moves forward. Thereby, the plasticized (measured) molten resin is injected from the injection nozzle and filled in the cavity of the mold 33.

  By the end of the injection process, a plasticizing process is performed in the injection device Mi (steps S6 and S7). In the plasticizing step, the screw in the heating cylinder 31 rotates at the set rotation speed, and therefore the molding material (resin pellet) in the hopper 32 is taken into the heating cylinder 31. As a result, the molding material is plasticized and melted by the temperature of the heating cylinder 31 and the rotation of the screw, and is transferred to the front to be measured and accumulated in the front part of the heating cylinder 31.

  In the mold clamping device Mc, a mold opening process is performed (step S8). In the mold opening process, after the mold 33 is cooled, the movable mold in the mold clamping state moves backward from the fixed mold to the mold opening position, and the mold 33 is opened. When the mold opening process is completed, a molded product removing process is performed (step S9). In the molded product removal step, the molded product attached to the fixed mold is taken out by an ejector mechanism attached to the mold 33.

  Thus, the molding process (molding cycle) for one shot of the injection molding machine M is completed. Thereafter, the same molding process is repeated until the production (production plan) is completed (steps S10, S2,...). If the planned production is completed, the operation of the injection molding machine M is stopped (steps S10 and S11).

  On the other hand, in each molding cycle, molding support according to the molding support method according to the present embodiment is performed.

  First, in the plasticizing process (steps S6-S7), molding data relating to the output parameter Ds... In this process is acquired. That is, at the start of the plasticizing process (step S6), the sensor group 42 of the injection molding machine M measures the change waveform of the physical quantity accompanying plasticization, and the molding machine controller 41 samples the measurement result. By performing, the molding data relating to the plasticizing time average, average resin temperature, and power consumption is collected (steps S12 and S13). Then, the collected molding data is transmitted to the data processing unit E (computer system Em) (step S14).

  Further, the data processing unit E performs the optimization process described above by inputting the received molding data to the plasticization model stored in the optimization process program Ps. That is, first, a molding condition related to the input parameter Df... Is obtained from the received molding data by a plasticization model (prediction function F), and a molding condition that satisfies the constraint condition Xc and the objective function Xp based on mathematical programming is derived. (Steps S15 and S16). By repeatedly performing this process on the corresponding molding data, an optimized molding condition can be obtained (step S17).

  On the other hand, the optimized molding condition is transmitted from the data processing unit E to the molding machine controller 41 (step S18). Thereby, the molding machine controller 41 determines the received molding conditions, and performs a change process on the preset molding conditions (step S19). In this case, the changing process may be executed for all optimized forming conditions, or may be executed only for forming conditions that perform level determination and exceed a certain level (change level). .

  The execution timing of the change process is the next shot for any shot that has been measured, and during the period other than the plasticizing step (steps S6-S7), specifically, the end of the plasticizing step It is performed within a period from (Step S8) to the start of the mold clamping process (Step S10).

  As described above, if the change process is performed in the next shot after an arbitrary shot is completed, the input parameter Df... Is optimized and the input parameter Df. Since the processing can be realized quickly, it is possible to improve the quality for each individual molded product in each shot, and there is an advantage that it can contribute to quality improvement and homogenization of the entire production.

  Further, if a period from the end of the plasticizing process (step S8) to the start of the mold clamping process (step S10) is applied as a period other than the plasticizing process (steps S6-S7), the optimized molding conditions are obtained. Even if it is changed, it can be changed without affecting the plasticizing process (steps S6 to S7). That is, even during production operation, there is an advantage that the molding quality can be improved without affecting the production amount.

  In the above embodiment, as an example, the case where the constraint condition Xc includes at least the resin temperature and the power consumption is applied to the objective function Xp has been described. However, there are various restrictions on the constraint condition Xc and the objective function Xp. Items (factors) can be applied.

  For example, even when power consumption is selected as the objective function Xp, the screw rotation speed, screw rotation speed, screw position, and resin pressure indicating the resin state at the plasticizing step (S6-S7) as the constraint condition Xc. , Resin temperature, one or more plasticization times, and / or any molding data that provides one or more of these deviations can be selected. Even in this case, since it becomes possible to minimize power consumption while stabilizing the molding data indicating the state of the resin during the plasticizing process, it is possible to save energy and reduce the molding quality. The objective of reducing running costs can be achieved. Further, as the types of molding data indicating the state of the resin during the plasticizing process, one or more of screw rotation speed, screw rotation speed, screw position, resin pressure, resin temperature, plasticizing time, and / or each of these By selecting any molding data with one or more deviations, it is possible to secure a sufficient type of molding data that influences the state of the resin during the plasticizing process, thus saving energy without reducing molding quality. In addition, there is an advantage that optimization from the viewpoint of reducing the running cost can be easily and flexibly performed.

  Therefore, according to the molding support method of the injection molding machine according to this embodiment, the data processing unit E having a computer function and the molding data related to the input parameters Df including the molding conditions of the injection molding machine M and this A molding function of an injection molding machine that obtains a prediction function F based on learning of the neural network N from molding data related to an output parameter Ds obtained by trial molding based on the input parameter Df, and predicts a predetermined molding condition using the prediction function F. Assuming the support method, as a basic forming support method, the data processing unit E previously stores the constraint data Xc based on mathematical programming and the objective function on the forming data related to the input parameter Df... And the output parameter Ds. Xp is set, and the constraint Xc... And the objective function X are set by the mathematical programming. An optimization processing program Ps for obtaining an optimized molding condition related to the input parameter Df... Satisfying the condition is set, and the data processing unit E relates to the output parameter Ds. Molding data is acquired, and during the period other than the plasticizing step, molding conditions optimized by the prediction function F ... and the optimization processing program Ps are obtained from the molding data related to the output parameter Ds ... Since the process of changing the molding condition in another shot after the completion of an arbitrary shot depending on the molding condition is performed, the predicted input parameter Df (molding condition) can be easily and reliably optimized. As a result, the range of use can be expanded, versatility can be improved, and a molding support method excellent in usability can be provided. In addition, even when the input parameters Df... Are optimized, the processing amount related to the optimization can be effectively reduced without reducing the reliability related to the optimization. Can be speeded up. As a result, processing within one molding cycle at an actual production site becomes possible, and it can be effectively used in subsequent production, and can be constructed as a molding support method with high practicality at an actual production site.

  As mentioned above, although preferred embodiment was described in detail, this invention is not limited to such embodiment, It does not deviate from the summary of this invention in a detailed structure, a shape, a raw material, quantity, a numerical value, etc. It can be changed, added, or deleted arbitrarily.

  For example, although the data processing unit E has shown the case where the computer system Em separately prepared for the injection molding machine M is used, the data processing unit E may be configured integrally with the molding machine controller 41. Moreover, although the trust region method which is one of the interior point methods is exemplified as the mathematical programming method, various other well-known mathematical programming methods such as linear programming and branch and bound methods can be used. On the other hand, the optimization of the plasticizing conditions has been exemplified, but it can be applied to various molding conditions such as the injection conditions, and various operating physical quantities in the injection molding machine M can be applied to the constraint condition Xc and the objective function Xp.

  The molding support method according to the present invention can be used for various injection molding machines having a function of optimizing molding conditions using a neural network.

  E: Data processing unit, M: Injection molding machine, Df ...: Input parameter, Ds ...: Output parameter, N: Neural network, Xp: Objective function, Xc: Restriction condition, Ps: Optimization processing program, (S6-S7) ): Plasticization process, (S8): End of plasticization process, (S10): Start of mold clamping process

Claims (6)

  A data processing unit with a computer function obtains a prediction function based on learning of the neural network from molding data related to input parameters including molding conditions of the injection molding machine and molding data related to output parameters obtained by trial molding based on these input parameters. A molding support method for an injection molding machine that predicts a predetermined molding condition by using the prediction function, in which the data processing unit is previously constrained based on mathematical programming for molding data related to the input parameter and the output parameter. In addition to setting conditions and an objective function, an optimization processing program for obtaining the optimized molding conditions related to the input parameters satisfying the constraint conditions and the objective function according to the mathematical programming is set, and the data processing is performed during production operation. Depending on the part, any shot Molding data related to the output parameter during the plasticizing process is obtained, and molding is optimized by the prediction function and the optimization processing program from the molding data related to the output parameter during a period other than the plasticizing process. A molding support method for an injection molding machine, characterized in that a condition is obtained and a process for changing molding conditions in another shot after the completion of the arbitrary shot is performed according to the optimized molding conditions.   2. The molding support method for an injection molding machine according to claim 1, wherein the period other than the plasticizing process is a period from the end of the plasticizing process to the start of the mold clamping process.   2. The molding support method for an injection molding machine according to claim 1, wherein molding data indicating the state of the resin during the plasticizing process is selected as the constraint condition, and power consumption is selected as the objective function.   Molding data indicating the state of the resin during the plasticizing process includes one or more of screw rotation speed, screw rotation speed, screw position, resin pressure, resin temperature, plasticizing time, and / or one of these deviations. 4. The molding support method for an injection molding machine according to claim 3, wherein two or more are selected.   2. The molding support method for an injection molding machine according to claim 1, wherein another shot after the completion of the arbitrary shot is a next shot.   2. The molding support method for an injection molding machine according to claim 1, wherein the objective function is a nonlinear function.

Images