JP2017124498A - Method of acquiring manufacturing parameters used for injection molding - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method of acquiring manufacturing parameters used for injection molding.SOLUTION: At least two manufacturing parameters affecting product quality to be measured and a plurality of characteristic parameters used for defining the product quality to be measured are selected and standard values of each characteristic parameters are set. Error function is formed. The error function satisfies the following mathematical formula 1. Err is the error function. Qis the j characteristic error and error of a feature value is in direct proportion with a value of the j characteristic parameter minus the standard value. acorresponds to one weight function of the j characteristic error and j is positive integer.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a method for setting manufacturing parameters in an injection molding manufacturing process, and in particular, a user can quickly and accurately obtain appropriate manufacturing parameter values by quantizing the sensitivity between manufacturing parameters and product quality. High quality products can be manufactured based on the obtained manufacturing parameter values.

  Injection molding is a manufacturing process widely used for manufacturing various plastic members. When a defect occurs in an injection molded product, in many cases, it is necessary for an experienced engineer to adjust the manufacturing parameters of the machine over time again by experience. In addition, after performing injection molding again, it is confirmed whether or not the product quality matches the designated quality by experience or a detection device. However, such solutions may not have uniform quantization data for machine tuning parameters and quality, and users often rely on experience and based on the location and amount of error where product defects occur, There is no choice but to adjust the manufacturing parameter values roughly. That is, the numerical values of the manufacturing parameters of each item cannot be adjusted effectively and accurately by a scientific quantization method.

JP2013-67150A

  The present invention has been made in view of the above-described problems, and an object thereof is to solve the above-described problems. Therefore, this invention provides the acquisition method of the manufacturing parameter used for injection molding.

The manufacturing parameter acquisition method according to the present invention includes the following steps. At least two manufacturing parameters that affect the product quality to be measured and a plurality of feature parameters used to define the product quality to be measured are selected, and a standard value for each feature parameter is set. Create an error function. The error function satisfies Equation 1 below. Of these, Err is an error function. Q _j is the jth feature value error, and the feature value error is directly proportional to the value obtained by subtracting the jth standard value from the jth feature parameter. a _j corresponds to one weight function of the j-th feature value error, and j is a positive integer.

  Numeric values that can be set for individual manufacturing parameters are paired to obtain a plurality of sets of setting values. In addition, the above set values are set in the injection molding machine, and production is actually performed to acquire actual feature values of each feature parameter.

The error of each feature value is calculated from the actual feature value described above. In addition, the sensitivity relationship between the set values and the individual feature parameters described above is calculated.
One of the above set values is selected to obtain a corresponding actual feature value. Each weight function of the error function is set based on the sensitivity relationship described above. The total error is calculated from the error function.

  It is determined whether the total error is lower than a threshold value. If the total error is below the threshold, complete all steps. If the total error is higher than the threshold value, the setting value selection and weight function setting steps described above are repeated.

  In this way, the user can manufacture a high-quality product with low error by obtaining numerical values of appropriate manufacturing parameters quickly and accurately using an error function.

  The production parameters selected are preferably the holding pressure and the screw position of the injection molding machine.

  In the above-described method, it is preferable that a sensitivity relationship is obtained from a numerical value in ascending order of each manufacturing parameter and a corresponding change rate of the actual feature value after creating a drawing with the manufacturing parameter and the actual feature value.

It is a flowchart which shows the step of one Embodiment of this invention. It is a figure which shows the position of the characteristic parameter in one Embodiment of this invention. It is a figure which shows the position of the characteristic parameter in one Embodiment of this invention. It is a figure which shows the relationship between the holding pressure in one Embodiment of this invention, and the difference | error of a feature value.

  In order to further understand the features of the present invention, an embodiment will be given and described with reference to the drawings. The production parameter acquisition method used for injection molding in the present invention is exemplified by producing gears by an injection molding machine. Each step will be described below. Please refer to Fig.1.

  Step S1: Select two manufacturing parameters that affect gear quality and a plurality of feature parameters used to define gear quality. In this embodiment, the holding pressure P and the screw position Q of the injection molding machine are selected as manufacturing parameters. In addition, the pixel area value A, the protruding height B, the root circle diameter C, the tip circle diameter D, and the tooth feature parameters E, F, and G at seven different positions were selected. The position and range of each characteristic parameter are shown in FIGS. Moreover, the standard value of each characteristic parameter described above is set. The standard value may be a numerical value of each characteristic parameter of a perfect gear.

  Next, step S2 is performed. The definition of the error function is as shown in Equation 2 below.

Of these, Err (P _i , Q _i ) is an error function. Q (ΔA _i ) is an error of the feature value corresponding to the feature parameter A, that is, an error of the feature value corresponding to the pixel area value A of the gear photograph, and its definition is as shown in Equation 3. . Of these, Q (A _st ) corresponds to the standard value in which the characteristic parameter A is set in step S1. Q (A _i ) is an actual feature value of the feature parameter A measured by the video. i is 1-n surveys. In addition, Q (ΔG _i ) can be inferred from Q (ΔB _i ) and Q (ΔC _i ).

  Next, step S3 is performed. A plurality of sets of setting values can be obtained by making the manufacturing parameter a pair with a numerical value that can be set in the injection molding machine. For example, the holding pressure P can be divided into 11 stages of pressure values (bar). The screw position Q of the injection molding machine can be divided into 11-stage numerical values (cm). In this way, 121 sets of setting values can be combined. Each set of setting values corresponds to one stage of holding pressure P and one stage of screw position. It is not necessary to incorporate all set values for the holding pressure P and screw position Q of the injection molding machine into the experimental design of step S3. In other words, a sufficient quantity of set values may be defined.

  Step S4: The above set values are set in the injection molding machine, and gears are actually produced.

  Step S5: A photograph of the gear produced in step S4 is taken, and the actual feature value corresponding to each feature parameter of the gear is automatically surveyed using a computer.

  Thereafter, step S6 is performed. A set of other set values for which step S4 is not performed is set in the injection molding machine, and gear production (step S4) and actual characteristic value surveying (step S5) are performed. After the gears corresponding to the set values defined in step S3 are produced, step S7 is entered.

  Step S7: Each feature value error is calculated from each actual feature value measured in step S5. After that, the holding pressure P is set as the X axis, the screw position Q is set as the Y axis, the error of the characteristic value of one characteristic parameter (for example, the characteristic parameter A) is selected, and the drawing is generated as the Z axis. Therefore, the user can know the sensitivity relationship of the feature parameter A to the holding pressure P and the screw position Q from the curved surface having the three-dimensional feature in the figure. Thereafter, the user can fix the value of one of the manufacturing parameters. For example, by fixing the numerical value of the screw position Q, a relational diagram of the characteristic parameter A with respect to the holding pressure P can be obtained. As shown in FIG.

Next, step S8 is performed. One set of setting values is selected. That is, one of the holding pressure P and the numerical value of the screw position Q is selected. In addition, each weight function α _ι of the error function is set according to the sensitivity relationship obtained from step S7. Combined with the feature error of each feature parameter corresponding to the selected holding pressure P and screw position Q, the total error is calculated by the error function. Thereafter, the process enters step S9. In the present embodiment, the numerical values of the weighting functions α _ι are set equal and the sum is 1.

  In step S9, it is determined whether or not the total error is lower than a preset threshold value. If the total error is higher than the threshold, the process returns to step S8, the numerical values of the holding pressure P and the screw position Q are adjusted again, and the error function is converged by performing the above calculation. If the total error is lower than the threshold value, the process proceeds to step S10. All the steps are completed, and the user applies the production on the production line based on the values of the holding pressure P and the screw position Q at this time.

  Based on the method of the present invention, the user can first obtain the sensitivity between each characteristic parameter and the manufacturing parameter. Furthermore, an appropriate weight function can be set according to the sensitivity relationship. In addition, by adjusting an appropriate manufacturing parameter effectively and quickly by the error function, the user can produce a high-quality product with low error based on the manufacturing parameter. The supplementary explanation is that the step S1 in the above-described embodiment has been described by selecting two manufacturing parameters that affect the quality of the gear as an example, but the user can select parameters that are selected as necessary. May be increased to two or more.

  Finally, it should be noted that the components described in the embodiments of the present invention are merely illustrative and will not limit the scope of the present invention in any way. Replacements and changes by other steps belong to the scope of the claims of the present invention.

S1 to S10 steps

Claims (4)

Selecting at least two manufacturing parameters that affect the product quality to be measured and a plurality of feature parameters used to define the product quality to be measured, and setting a standard value for each of the feature parameters When,
An error function is created, the error function satisfies Equation 1, Err is the error function, Q _j is an error of the jth feature value, and the jth feature value The error is directly proportional to the numerical value obtained by subtracting the j-th standard value from the j-th feature parameter, and a _j corresponds to the error weight function of the j-th feature value, and j Is a positive integer step;
By obtaining a plurality of sets of setting values by pairing a plurality of settable numerical values of at least two of the manufacturing parameters, the setting values are actually set in the injection molding machine in sequence. Performing production and obtaining actual feature values for each of the feature parameters;
Calculating an error of each feature value using each actual feature value, and calculating a sensitivity relationship between each of the set values and each of the feature parameters described above;
Select one of the set values to obtain the corresponding actual feature value, set each weight function of the error function based on the sensitivity relationship, and calculate the total error by the error function Steps,
It is determined whether or not the total error is lower than a threshold value. When the total error is lower than the threshold value, all steps are completed. When the total error is higher than the threshold value, the setting value selection and the weight function are set. Overlapping steps, and
The manufacturing parameter acquisition method used for injection molding characterized by including these.

  The method for obtaining manufacturing parameters used in injection molding according to claim 1, wherein the at least two manufacturing parameters are a holding pressure and a screw position of the injection molding machine, respectively.   A drawing is created with at least two of the manufacturing parameters and the actual feature values, and the sensitivity relationship is obtained by obtaining an ascending numerical value of each of the manufacturing parameters and a corresponding change rate of the actual feature values. The method for obtaining manufacturing parameters used for injection molding according to claim 1, wherein the manufacturing parameters are obtained.   2. The injection according to claim 1, wherein the error of the feature value is obtained by subtracting the j-th standard value from the j-th feature parameter and then dividing the j-th standard value by the j-th standard value. A method for obtaining manufacturing parameters used for molding.

Images