JP2006289874A - Injection-molding backup device, injection-molding system and injection-molding backup control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an injection-molding backup device capable of presenting an appropriate and highly reliable molding failure remedy to a user. <P>SOLUTION: This injection-molding backup device has an input part 41 which receives molding information required for an injection-molding process by an injection-molding machine 2 and a restriction information storage part 45b which stores restriction information for operating the injection-molding machine 2. In addition, a counter-measure inferring means 100 infers a molding failure remedy from molding information received by the input part 41. Further, an inferred results control part 45 judges it from the restriction information whether the molding failure remedy inferred by the counter-measure inferring means 100 is appropriate and outputs the molding failure remedy judged as appropriate. Thus, an inappropriate molding failure remedy is no longer presented to the user and can be prevented from being put into practice by the user. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to, for example, an injection molding support device for optimizing molding conditions for resin injection molding, a system having the same, and the like.

Conventionally, when a molding defect occurs in a molded product obtained by injection molding, a skilled molding engineer skilled in injection molding work adjusts injection molding conditions based on the type and degree of molding defect. Yes. However, it takes a long time to train skilled molding engineers, and it is not easy to actually obtain skilled molding engineers.
For this reason, an expert system has been constructed that extracts and organizes the experience and know-how possessed by skilled molding engineers as described above to create a database and presents countermeasures for molding defects based on the database.
In Patent Document 1, in view of the fact that it is difficult to set the initial molding conditions for the target molded product even if the expert system is used, the initial molding conditions for the target molded product are selected from the molding conditions at the time of good product molding. Alternatively, the initial filling adjustment is performed based on the initial setting value according to the default setting rule and is derived and set in the injection molding machine. If a molding defect occurs in the molded product molded in accordance with the initial molding conditions, a plurality of molding defect countermeasures for eliminating the molding defect are sequentially derived for the molding defect. It has been proposed to cause the injection molding machine to take measures.

Conventionally, a method for statistically deriving countermeasures for molding defects has been implemented by an injection molding support apparatus based on an experimental design method. According to this method, there is an advantage that a defective molding countermeasure can be derived without being a skilled molding engineer.
Further, as disclosed in Patent Document 2, there is a proposal that derives a countermeasure for molding defects by combining an expert system and an experimental design method.

Japanese Patent Laid-Open No. 7-24894 JP-A-4-77219

In the expert system proposed up to now and the injection molding support device based on the experimental design method, if data using the same mold and resin exist in the database in the past, a certain result can be obtained, If such data does not exist, the proposed countermeasure against molding defects may not always be appropriate. If it is a skilled molding engineer, even if the countermeasure for improper molding failure is presented, it can be determined that it is not appropriate.
However, a user who derives countermeasures for molding defects using an injection molding support apparatus based on an expert system or an experimental design method is not necessarily a skilled molding engineer. For this reason, when a countermeasure for improper molding defects is presented, there are problems such as the possibility that a defective product is likely to be generated or the injection molding machine is easily damaged as a result of the user implementing the countermeasure for molding defects. Has occurred. As a cause of damage to injection molding machines, injection molding machines have limited data (for example, maximum injection pressure, maximum injection speed, etc.) based on machine specifications, but currently molding defects that do not take into account this limited data It is mentioned that measures are presented. For example, although the maximum injection pressure of a certain injection molding machine is 200 MPa, an injection pressure of 210 MPa may be presented as a countermeasure for molding defects. In addition, when an injection pressure of 200 MPa is presented as a countermeasure for defective molding, if the operation at the maximum injection pressure is continued in this injection molding machine, the durability of the injection molding machine is reduced or the injection molding machine is easily damaged. .
The present invention has been made based on such a technical problem, and an object thereof is to provide an injection molding support apparatus and the like that can present a user with appropriate and highly reliable countermeasures against molding defects.

For this purpose, in the present invention, the appropriateness of the molding defect countermeasure is output after determining in advance whether or not the countermeasure against the molding defect by the countermeasure reasoning unit is appropriate based on the constraint information for operating the injection molding machine. That is, the injection molding support device of the present invention includes an input unit that receives molding information necessary for performing injection molding with an injection molding machine, and a constraint information storage unit that stores constraint information for operating the injection molding machine. The countermeasure reasoning unit infers a molding defect countermeasure based on the molding information received by the input unit. Then, the inference control unit determines whether or not the molding defect countermeasure inferred by the countermeasure inference unit is based on the above-described constraint information, and outputs the molding defect countermeasure determined to be appropriate. Thereby, an inappropriate molding defect countermeasure is not presented to the user, and it can be avoided that the user implements an inappropriate molding defect countermeasure.
Restriction information for operating the injection molding machine includes restriction data based on the machine specifications of the injection molding machine (hereinafter referred to as “machine restriction data”). Injection molding machines have machine limit data such as maximum injection pressure and maximum injection speed for each model, but when the countermeasure for molding defects is outside the range of machine limit data, the reason control section Does not output such inappropriate measures for molding defects.

The countermeasure reasoning unit can be configured by an optimization method reasoning unit that infers a molding defect countermeasure using an optimization method based on the molding information received by the input unit. As an optimization method, neuro, genetic algorithm (GA), etc. can be used. Conventional expert systems do not have a learning function, and the transfer of know-how in producing user-specific molded products is insufficient. If provided, the know-how of injection molding can be effectively transferred.
When inference based on the optimization method is performed, it is preferable to perform a plurality of test shots in advance and use the test shot data obtained as a result.
In addition to the optimization method reasoning unit, a knowledge base reasoning unit and a statistical analysis unit can be further provided. The knowledge-based reasoning unit takes measures against molding defects based on experience information that associates molding defects that may occur in an injection molding machine with effective countermeasures when molding defects set based on the experience of a skilled molding engineer occur. Infer. In addition, the statistical analysis unit uses the molding history information that associates the molding conditions with the molding results when the molding conditions are applied, and infers molding defect countermeasures based on the experimental design method. The statistical analysis unit also has a learning function.

The present invention can also be an injection molding system including the above-described injection molding support device. That is, the injection molding system of the present invention stores an injection molding machine main body, an input unit for receiving molding information necessary for performing injection molding with the injection molding machine, and constraint information for operating the injection molding machine. The constraint information storage unit, the countermeasure reasoning unit that infers the molding defect countermeasure based on the molding information received by the input unit, and the suitability of the molding defect countermeasure inferred by the countermeasure reasoning unit based on the constraint information are judged to be appropriate. And an inference control unit that outputs countermeasures against molding defects.
In this system, the countermeasure reasoning unit can change the constraint information based on the result of actually performing injection molding by applying countermeasures against molding defects. That is, since the injection molding system of the present invention learns the constraint information, it is possible to present measures against molding defects to the user in a short time.

  Further, the present invention is an injection molding support control method executed by an injection molding support device that presents measures for molding defects based on a preset program, and molding information necessary for performing injection molding with an injection molding machine , Inferring molding defect countermeasures based on molding information, and determining the suitability of molding defect countermeasures based on constraint information for operating an injection molding machine, and outputting molding defect countermeasures determined to be appropriate It can also be grasped as an injection molding support control method characterized by comprising the steps of:

Furthermore, the present invention provides an injection molding support apparatus including a constraint information storage unit that stores constraint information classified into a plurality of stages based on any one of types, dimensions, applications, and grades of molded products. In this injection molding support device, the classification selection unit accepts molding information necessary for injection molding by the injection molding machine and selection of the above-described classification. Then, based on the molding information and the selected classification, the countermeasure reasoning unit infers a molding defect countermeasure.
For example, depending on the type of the molded product, there is no problem even if there is some color unevenness. Rather, there is a case where it is desired to shorten the molding cycle and produce a product in large quantities. In some cases, priority is given to the accuracy of injection molding over the molding cycle. In such a case, a user-desired classification is selected from the constraint information classified into a plurality of stages, and a molding defect countermeasure is performed accordingly.

  According to the present invention, even when there is no data using the same mold or resin in the past, it is possible to present a highly reliable countermeasure against molding defects to the user.

Hereinafter, embodiments of an injection molding support apparatus and an injection molding system according to the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a block diagram showing a configuration of an injection molding system 1 of the present embodiment. As shown in FIG. 1, the injection molding system 1 includes an injection molding machine 2 and an injection molding support device 3 that supports the operation of the injection molding machine 2.

First, the injection molding machine 2 related to the injection molding system 1 and the injection molding support apparatus 3 according to the present invention will be described with reference to FIG.
FIG. 2 is a schematic diagram showing a configuration example of an injection molding machine 2 whose operation should be supported by the injection molding support apparatus 3 according to the present embodiment. As shown in FIG. 2, the injection molding machine 2 includes molds 20 a and 20 b, a mold clamping device 21, and an injection device 22.
The mold clamping device 21 includes die plates 23 a and 23 b for attaching the molds 20 a and 20 b and a tie bar 24. The injection device 22 includes a nozzle 25 that injects molten resin, an injection cylinder 26 that supplies the molten resin to the nozzle 25, a screw 27 that can move back and forth in the injection cylinder 26, and a screw 27 that is rotatable. A hopper 28 for supplying the raw material resin, a motor 29 for rotating the screw 27 in the injection cylinder 26, an actuator 30 for moving the screw 27 back and forth in the injection cylinder 26, and a signal output from the injection molding support device 3 side. And a central control unit 31 for controlling the molding parameters. The molding parameters are each element constituting the molding conditions of the injection molding machine 2, and examples include injection speed, injection pressure, resin flow rate, mold clamping speed, holding pressure, mold and resin temperature, and the like.

  When injection molding is performed by the injection molding machine 2, first, raw material resin is supplied from the hopper 28 into the injection cylinder 26, and then the motor 27 is operated to rotate the screw 27. The raw material resin supplied into the injection cylinder 26 moves toward the nozzle 25 by the rotation of the screw 27. At this time, the raw material resin is melted by a heater (not shown) provided in the injection cylinder 26. By operating the actuator 30, the screw 27 is moved in the direction of the nozzle 25, so that the molten resin is injected from the nozzle 25. The molten resin injected from the nozzle 25 is filled into a cavity between the mold 20a and the mold 20b. When a predetermined amount of molten resin is injected, the screw 27 stops moving forward, holds the mold for a predetermined time, and molds an injection molded product. Thereafter, the die plate 23a is moved along the tie bar 24 to open the mold, and the injection molded product is removed from the mold to complete one molding.

Next, the injection molding support device 3 will be described with reference to FIG.
As shown in FIG. 1, the injection molding support device 3 presents countermeasures for molding defects by executing predetermined processing based on a preset program. The injection molding support device 3 includes an input unit (classification selection unit) 41, an evolutionary reasoning unit (countermeasure reasoning unit, optimization method reasoning unit) 42, an expert unit (countermeasure reasoning unit, knowledge base reasoning unit) 43, a statistical analysis. The unit 44, the inference result control unit (inference control unit) 45, the improvement condition output unit 46, the user database 50, and the knowledge database 60 are functionally provided.

The user database 50 includes a trial firing database 51 and a molding history information database 52.
In the trial firing database 51, trial firing data in which molding condition information during trial firing and trial molded product information are associated with each other is stored. The test shot data is stored in the test shot database 51 every time the user performs a test shot.
The molding condition information at the time of trial firing is information indicating the molding conditions used at the time of trial firing. The trial firing molding condition information includes information on the model name of the injection molding machine 2 to be used, the injection speed at the time of trial firing, the injection pressure, and the mold holding temperature. In addition to these, it is preferable that the material information at the time of trial firing, that is, information on the type, viscosity, temperature, etc. of the resin used is included in the molding condition information at the time of trial firing. Here, the test shot is an injection molding test that is performed to determine the optimum molding conditions for obtaining a target molded product. The trial-molded product is a prototype obtained as a result of the trial shot. The test shot molded product information is information indicating the properties of the test shot molded product. The trial molded product information includes the presence or absence of molding defects such as filling defects and cracks, the degree of molding defects, the dimensions and weight of the trial molded products.

  The molding history information database 52 stores molding history information. The molding history information is information indicating molding conditions used by the user in the past and molding results when the molding conditions are applied. The molding history information includes molding conditions under which good results were obtained when the user performed molding in the past (good molding conditions) and molding conditions under which poor results were obtained when the user performed molding in the past. (Defective molding conditions) are included.

  The knowledge database 60 stores experience information that associates “molding defects” with measures that are effective when molding defects occur. Effective measures when molding defects occur are set based on the experience (know-how) of skilled molding engineers. Molding defects include insufficient filling and cracking. For example, a countermeasure of “increasing the resin temperature” is associated with molding defects such as insufficient filling, and stored in the knowledge database 60 as experience information.

When a molding defect occurs, the injection molding support apparatus 3 outputs a highly reliable molding condition suitable for eliminating the defect and presents it to the user.
In the injection molding support device 3, molding information is received by the input unit 41. The molding information is information necessary for performing injection molding with the injection molding machine 2. The molding information includes information on molding conditions (hereinafter referred to as “molding condition information”) and information on molded products (hereinafter referred to as “molded product information”).

Information on molding conditions includes molding machine information (model name of injection molding machine 2, etc.), mold information (mold dimensions, gate depth, gate shape, etc.), material information (resin name, product name of resin material, etc.) )and so on.
The molded product information is a concept that includes both information on a target molded product (hereinafter referred to as “target molded product information”) and information on the above-described test-fired molded product. The target molded product information includes the size and weight of the target molded product. Note that the above-described trial firing data is also included in the molding information in the present invention.

  The input unit 41 receives at least one of molding condition information and molded product information as molding information. The input unit 41 stores the received information in the user database 50 as necessary. At that time, when both the molding condition information and the molded product information are received, they are stored in association with each other.

  Based on the molding information input by the input unit 41 and the experience information stored in the knowledge database 60, the expert unit 43 measures molding defects so that molding defects do not occur in the injection molding machine 2, that is, injection speed, injection Whether or not to adjust molding parameters such as pressure, resin flow rate, mold clamping speed, holding pressure, mold and resin temperature, and if so, molding parameters to be adjusted by the injection molding machine 2 and their adjustment amounts; And the result is output to the inference result control unit 45.

  The statistical analysis unit 44 uses the molding history information stored in the molding history information database 52 to infer a molding defect countermeasure based on the experimental design method, and outputs the result to the inference result control unit 45. In order for the statistical analysis unit 44 to perform inference based on the experiment design method, the statistical analysis unit 44 may store software for the experiment design method in advance.

The evolutionary reasoning unit 42 uses the information stored in the user database 50 to infer a molding defect countermeasure based on the optimization method, and outputs the result to the reasoning result control unit 45. Neuro, genetic algorithm (GA), simulated annealing, etc. can be used as optimization methods.
The evolutionary reasoning unit 42 is common to the statistical analysis unit 44 in that the information stored in the user database 50 is used to infer countermeasures against molding defects, but the reasoning methods used are different. The statistical analysis unit 44 uses molding history information stored in the molding history information database 52, while the evolutionary reasoning unit 42 mainly uses trial data stored in the trial database 51. The statistical analysis unit 44 extracts the molding history information highly relevant to the molding condition input by the input unit 41 and infers the countermeasure against molding defects after extracting from the molding history information database 52. Unless there is molding history information at the time of molding, the properties of the molded product assumed to be obtained when the input molding conditions are applied, and the properties of the molded product obtained as a result of actually applying the molding conditions Are often different. For this reason, when molding history information is used, it cannot be said that a reliable molding defect countermeasure can be derived. On the other hand, the information stored in the trial firing database 51 corresponds directly to actual trial molding conditions and trial molding product information, that is, actual information. A highly reliable countermeasure against molding defects can be derived.

  The evolutionary reasoning unit 42, the expert unit 43, and the statistical analysis unit 44 are common in that they infer countermeasures against molding defects, and in the present specification, these may be collectively referred to as countermeasure inference means 100. The above-described evolutionary reasoning unit 42, expert unit 43, and statistical analysis unit 44 need not all be activated. Therefore, when the user operates the injection molding support apparatus 3, one or a plurality of countermeasure reasoning means 100 to be activated can be arbitrarily selected. The input unit 41 accepts selection from the user.

The inference result control unit 45 includes a determination unit 45a and a constraint information storage unit 45b.
The determination unit 45a evaluates the suitability of the molding defect countermeasure output by the countermeasure reasoning unit 100 based on the constraint information stored in the constraint information storage unit 45b. Then, when it is determined that the countermeasure for molding defects is appropriate, that is, when the injection molding machine 2 performs the countermeasure for molding defects, the injection molding machine 2 does not cause any defects such as damage. The unit 45a sends the countermeasure for molding defects to the improvement condition output unit 46.
In the constraint information storage unit 45b, constraint information for operating the injection molding machine 2 is stored. The restriction information indicates whether there is a problem in the operation of the injection molding machine 2, specifically, the injection molding machine 2 is damaged, or the operation of the injection molding machine 2 must be stopped. This is information for the determination unit 45a to determine whether or not there is any. The constraint information can be arbitrarily set by the user based on the machine specifications of the injection molding machine 2 (for example, information such as maximum injection pressure, maximum holding pressure, maximum injection speed, maximum screw rotation speed, maximum clamping force). it can.
The constraint information is stored in the constraint information storage unit 45b in association with the model name of the injection molding machine 2, for example, as in the table of FIG.

In addition, the restriction information that has been classified into a plurality of stages is stored in the restriction information storage unit 45b in accordance with the type, size, application, grade, etc. of each molded article, and the user can arbitrarily classify the classification as necessary. You may make it selectable. For example, depending on the type of molded product, high accuracy may be required, while on the other hand, it may be desired to place importance on the molding cycle rather than accuracy. In such a case, the user can select the classification of the constraint information in accordance with requirements such as accuracy and molding cycle required for the target molded product. In this case as well, it is assumed that there is no problem in the operation of the injection molding machine 2.
For example, as shown in FIG. 4, it is assumed that there are a household bucket and an automobile interior panel as the types of molded products for injection molding. In many cases, the required precision of injection molding is different between a household bucket and an automobile interior panel, and it is not always necessary to use the same constraint information for both. Also, even if the type of molded product is common in terms of parts for automobiles, the parts for automobiles are like interior panels that are placed in a position where they can be seen by the user. It may be desired to vary the injection molding accuracy depending on whether the component is arranged. If the former, we would like to focus on the quality of grade A that satisfies a certain standard rather than the molding cycle, but if the latter, we would like to maintain the quality of grade B, which is a lower standard than grade A, but rather want to shorten the molding cycle. To do. In this case, the constraint information can be set for the latter so as to increase the upper limit of the injection speed than the former.

  The improvement condition output unit 46 outputs one or more molding defect countermeasures received from the determination unit 45a. The improvement condition output unit 46 can be configured by, for example, a display unit of a personal computer, a printer, or the like. The improvement condition output unit 46 may be configured to receive an instruction to perform molding defect countermeasures from the user. When the implementation instruction is received, the countermeasures for molding defects are sent to the central control unit 31 of the injection molding machine 2.

Hereinafter, the flow of processing performed by the countermeasure inference means 100, the inference result control unit 45, and the improvement condition output unit 46 will be described with reference to FIG.
First, in step S <b> 101, the input unit 41 receives input of molding information from the user, and outputs the received molding information to the countermeasure reasoning unit 100.
In step S102, the countermeasure reasoning means 100 selected by the user in advance is activated, infers a molding defect countermeasure for eliminating the molding defect, and outputs the defect countermeasure obtained as a result of the inference to the inference result control unit 45.
When the evolutionary reasoning unit 42 is activated as the countermeasure reasoning means 100, the evolutionary reasoning unit 42 reads the trial shooting data stored in the trial firing database 51, and infers a molding defect countermeasure based on the optimization method. When the statistical analysis unit 44 is activated, the statistical analysis unit 44 reads the molding history information stored in the molding history information database 52 and infers a molding defect countermeasure based on the experimental design method. When the expert unit 43 is activated, the expert unit 43 infers a molding defect countermeasure based on the knowledge data stored in the knowledge database 60. When one or more countermeasure reasoning means 100 are selected, each activates and infers a molding defect countermeasure.

The determination unit 45a of the inference result control unit 45 determines whether or not the molding defect countermeasure output by the countermeasure inference means 100 is appropriate (step S103). This determination is made based on whether the molding parameter based on the molding defect countermeasure is within the range of the machine restriction data stored in the constraint information storage unit 45b. When one or more countermeasure reasoning means 100 are selected, the determination unit 45a determines whether or not the molding defect countermeasures output by each are appropriate. In addition, when there is constraint information provided according to the type of a molded product to be injection-molded, the determination unit 45a also refers to this constraint information.
The inference result control unit 45 outputs only the molding defect countermeasures output by the countermeasure reasoning means 100 that are determined to be appropriate to the improvement condition output unit 46 (step S104). The output result is presented to the user by the improvement condition output unit 46 (step S105), and a series of processing for presenting a molding defect countermeasure is completed. In this way, by outputting only countermeasures for molding defects determined to be appropriate and not outputting countermeasures for improper molding defects, the user is likely to become defective products or conditions that damage the injection molding machine 2. Can be avoided in advance.

  When the user selects a certain molding defect countermeasure from among the one or more molding defect countermeasures presented, the molding defect countermeasure is sent to the central control unit 31 on the injection molding machine 2 side. Then, a test shot is performed with the molding parameters corresponding to the molding defect countermeasure. As a result of the trial shooting, the user determines whether or not there is a defect in the obtained molded product. If it is determined that the molded product obtained is defective as a result of the test shot, the process returns to step S101 in FIG. 5 and the result of applying the molding defect countermeasure is input by the user. Then, the input unit 41 receives this and sends the result of applying the molding defect countermeasure to the countermeasure inference means 100 and the inference result control unit 45. When the evolutionary reasoning unit 42 and / or the statistical analysis unit 44 are activated as the countermeasure reasoning means 100, these accumulate and learn the results of applying the molding defect countermeasures, and the reasoning result control unit 45 appropriately The constraint information is changed based on the result. Thereby, after the next time, the injection molding support apparatus 3 can present appropriate improved molding conditions in a shorter time.

FIG. 6 is a diagram showing a processing flow when the evolutionary reasoning unit 42 is selected as the countermeasure reasoning means 100. In order to cause the evolutionary reasoning unit 42 to infer countermeasures against molding defects using the trial shot data, a trial shot is performed prior to the start of the evolutionary reasoning unit 42.
First, the injection molding machine 2 actually performs a test shot under preset initial molding conditions (step S201).
In step S202, it is determined whether or not the obtained molded product has a defect. If it is determined here that the product is good without molding defects, the process is terminated.
If there is a molding defect, the trial data is stored in the trial database 51 via the input unit 41 (step S203). If desired by the user, the trial shooting data is also stored in the molding history information database 52.
In step S204, it is determined whether the number of trial shots has reached a preset value. If the set number of times has not been reached, the processes from step S201 to S203 are repeated until the set number of times is reached. When it is determined that the set number of times has been reached, the input unit 41 notifies the evolutionary reasoning unit 42 to that effect.
In response to this notification, the evolutionary reasoning unit 42 reads the trial shooting data stored in the trial shooting database 51 (step S205). And it progresses to the process after step S101 shown in FIG. The reason why the trial firing is repeated until the number of trial firings reaches a preset value is that the technique called the optimization method used by the evolutionary reasoning unit 42 requires data sufficient for inference. By collecting trial data sufficient for inference and performing inference, the evolutionary inference unit 42 can derive a realistic and highly reliable defect countermeasure.

As described above, the user can arbitrarily select which one of the evolutionary reasoning unit 42, the expert unit 43, and the statistical analysis unit 44 is used as the countermeasure reasoning means 100.
A combination of the evolutionary reasoning unit 42, the expert unit 43, and the statistical analysis unit 44 is shown in FIG.
In case 1, only the evolutionary reasoning unit 42 is used. Case 1 is effective when the type of mold or resin is significantly different from conventional molded products, and past accumulated information is not very useful in inferring countermeasures against molding defects. In this case, it is essential to perform a test shot.
In case 2, the evolutionary reasoning unit 42 and the expert unit 43 are used. In Case 3, the evolutionary reasoning unit 42, the expert unit 43, and the statistical analysis unit 44 are used in combination. In Case 4, the evolutionary reasoning unit 42 and the statistical analysis unit 44 are used. Case 2 is effective when the original data is not stored to the extent that correct statistical analysis can be performed. Case 3 is most effective in that optimum conditions can be examined from all directions. Note that Case 4 is preferably applied when the contents of the user database 50 are substantial. This is because the content of the knowledge database 60 is considered to be obsolete in such a case.
In case 5, the expert unit 43 is used. Case 5 is effective when initial molding conditions for trial shooting are determined or when the molding history information stored in the molding history information database 52 is small.
In case 6, the expert unit 43 and the statistical analysis unit 44 are used.
In case 7, only the statistical analysis unit 44 is used. Case 7 is effective when the molding history information stored in the molding history information database 52 is substantial.

As described above, according to the injection molding support apparatus 3 of the present invention and the injection molding system 1 equipped with the same, the inference result control unit 45 determines in advance whether or not the countermeasure against molding defects inferred by the countermeasure reasoning means 100 is appropriate. The countermeasures for molding defects that are judged to be appropriate are output. As a result, an inappropriate molding defect countermeasure is not presented to the user, and problems that occur when an inappropriate molding defect countermeasure is implemented can be avoided.
In addition, the conventional expert system has no learning function, and the transfer of know-how in producing a user-specific molded product was insufficient, but as shown in the above embodiment, the evolutionary reasoning with the learning function If the unit 42 and / or the statistical analysis unit 44 are provided, the know-how of injection molding can be effectively transmitted.

In the above description, the evolutionary reasoning unit 42 uses the trial data to infer the molding defect countermeasure based on the optimization method. However, in addition to the trial data, the knowledge data stored in the knowledge database 60 and / or the molding is used. The molding history information stored in the history information database 52 may be used to infer countermeasures for molding defects.
Further, as in the injection molding support device 3A of FIG. 8, based on the inferred result of the molding defect countermeasure performed by the expert unit 43 and / or the statistical analysis unit 44, the evolutionary reasoning unit 42 infers the molding defect countermeasure. It may be done. This allows the evolutionary reasoning unit 42 to learn past cases and know-how of skilled molding engineers. Table 1 shows the variations of the original data used by the evolutionary reasoning unit 42 for inference. Keep it. (1) in Table 1 corresponds to (1) in FIG. 1, (2) and (3) in Table 1 correspond to (2) and (3) in FIG. 8, respectively. It means the following cases.
(1): Use test firing data as original data.
(2): The inference result of the molding defect countermeasure performed by the expert unit 43 is used as original data.
(3): The inference result of the molding defect countermeasure performed by the statistical analysis unit 44 is used as original data.

  The more original data, that is, the data used as the basis of inference, the higher the possibility that a better countermeasure against molding defects can be inferred. On the other hand, the more data that is the basis of inference, the longer it takes to make an inference. Therefore, cases a to g may be selected by comparing the accuracy of inference and the time required for inference.

Further, in the injection molding support devices 3 and 3A, it is preferable that the knowledge data and / or molding history information to be used is close to the molding information of the target molded product among those stored in each database. When all the molding history information stored in the knowledge database 60 and / or the molding history information database 52 is used, it is considered that the accuracy of the molding defect countermeasure inference is improved, but the inference takes time.
As an application example, the evolutionary reasoning unit 42 may infer the molding defect countermeasure based on the optimization method using the knowledge data and / or the molding history information without using the trial shooting data. For example, when determining the initial molding conditions for the test shot, there is no test shot data. In this case, it is possible to set initial molding conditions in a short time by inferring molding defect countermeasures using the molding history information stored in the molding history information database 52.

Furthermore, the constraint information used by the determination unit 45a may be generated based on the inferred result of the molding defect countermeasure by the expert unit 43 and / or the statistical analysis unit 44. For example, for the model 100MXZ of the injection molding machine 2 shown in FIG. 3, it is assumed that the maximum injection pressure is limited to 220 MPa as a machine specification. At this time, for example, even if the constraint information is normally determined as “maximum injection pressure (MPa) −10 MPa”, the upper limit of the injection pressure is set to 215 MPa as a result of inference by the expert unit 43 and / or the statistical analysis unit 44. However, particularly when a result indicating that there is no problem in the injection molding machine 2 is obtained, restriction information on the injection pressure may be set based on this result. As a result, the upper limit of the constraint information can be widened, and the molding defect countermeasures that are determined to be unsuitable for the determination unit 45a are reduced, so that the effect of increasing the user's options can be expected.
Here, Table 2 shows variations of the constraint information used by the determination unit 45a. In addition, (4)-(6) in Table 2 means the following cases, respectively.
(4): The machine restriction data stored in the restriction information storage unit 45b is used as restriction information.
(5): Use the inference result that is likely to cause a molding defect made by the expert unit 43 as constraint information.
(6): Use the inference result performed by the statistical analysis unit 44 that is likely to cause molding defects as constraint information.

In the above embodiment, the example in which the injection molding machine 2 and the injection molding support device 3 are configured as separate devices has been described, but they may be configured as an integrated device.
Alternatively, the injection molding machine 2 and the injection molding support device 3 may be installed at a remote location and connected by an arbitrary communication line. In this case, a plurality of injection molding machines 2 can be connected to one injection molding support device 3.

It is a figure which shows the structure of the injection molding system in this Embodiment. It is a figure which shows the example of 1 structure of an injection molding machine. It is a figure which shows an example of the restriction information (machine restriction data) stored in the restriction information storage unit. It is a figure which shows an example of the constraint information set by the kind etc. of molded article. It is a figure which shows the flow of the process which a countermeasure reasoning means, an inference result control part, and an improvement condition output part perform. It is a figure which shows the flow of a process at the time of selecting the evolutionary reasoning engine part as a countermeasure reasoning means. It is a figure which shows the combination of an evolutionary reasoning part, an expert part, and a statistical analysis part. It is a figure which shows the other structure of the injection molding system in this Embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1, 1A ... Injection molding system, 2 ... Injection molding machine 3, 3A ... Injection molding assistance apparatus, 41 ... Input part (classification selection part), 42 ... Evolutionary reasoning part (Countermeasure reasoning part, Optimization method reasoning part) , 43 ... Expert section (measure inference section, knowledge base inference section), 44 ... statistical analysis section, 45 ... inference result control section (inference control section), 45a ... determination section, 45b ... constraint information storage section, 46 ... improvement conditions Output unit: 50 ... user database, 51 ... trial shooting database, 52 ... molding history information database, 60 ... knowledge database, 100 ... countermeasure reasoning means

Claims (9)

An input unit for receiving molding information necessary for injection molding with an injection molding machine;
A constraint information storage unit that stores constraint information for operating the injection molding machine;
A countermeasure reasoning unit for inferring a molding defect countermeasure based on the molding information received by the input unit;
Based on the constraint information, an inference control unit that determines whether or not the countermeasure for molding defects inferred by the countermeasure reasoning unit, and outputs a countermeasure for molding defects determined to be appropriate;
An injection molding support apparatus comprising:   2. The injection molding support apparatus according to claim 1, wherein the restriction information is restriction data based on a machine specification of the injection molding machine.   3. The injection molding support apparatus according to claim 1, wherein the countermeasure reasoning unit infers the molding defect countermeasure using an optimization method based on the molding information received by the input unit.   4. The injection molding support apparatus according to claim 3, wherein the countermeasure reasoning unit is an optimization method reasoning unit that performs a plurality of test shots in advance and infers based on the test shot data obtained as a result. The countermeasure reasoning unit
Knowledge to infer the molding defect countermeasures based on experience information that associates molding defects that may occur in the injection molding machine with effective countermeasures that are set based on the experience of a skilled molding engineer. The base reasoning section,
Using a molding history information that associates molding conditions and molding results when the molding conditions are applied, a statistical analysis unit that infers the molding defect countermeasures based on an experimental design method,
The injection molding support apparatus according to claim 4, further comprising: An injection molding machine body;
An input unit for receiving molding information necessary for injection molding with an injection molding machine;
A constraint information storage unit that stores constraint information for operating the injection molding machine;
A countermeasure reasoning unit for inferring a molding defect countermeasure based on the molding information received by the input unit;
Based on the constraint information, an inference control unit that determines whether or not the countermeasure for molding defects inferred by the countermeasure reasoning unit, and outputs a countermeasure for molding defects determined to be appropriate;
An injection molding system comprising:   The injection molding system according to claim 6, wherein the countermeasure reasoning unit changes the constraint information based on a result of actually performing injection molding by applying the molding defect countermeasure output from the inference control unit. . An injection molding support control method executed by an injection molding support device that presents measures for molding defects based on a preset program,
Receiving molding information necessary for performing injection molding on an injection molding machine;
Inferring the molding defect countermeasure based on the molding information;
Determining the suitability of the molding defect countermeasure based on constraint information on operating the injection molding machine, and outputting the molding defect countermeasure determined to be appropriate;
An injection molding support control method comprising: A constraint information storage unit for storing constraint information classified into a plurality of stages based on any of the types, dimensions, uses, and grades of the molded article;
A classification selection unit that accepts selection of the classification information and the classification necessary for performing injection molding with an injection molding machine;
A countermeasure reasoning unit for inferring a molding defect countermeasure based on the molding information received by the classification selection unit and the selection of the classification;
An injection molding support apparatus comprising:

Images