WO2021049848A1 - 인공지능 기반의 사출성형시스템 및 사출성형시스템에서의 성형조건 생성방법 - Google Patents
인공지능 기반의 사출성형시스템 및 사출성형시스템에서의 성형조건 생성방법 Download PDFInfo
- Publication number
- WO2021049848A1 WO2021049848A1 PCT/KR2020/012107 KR2020012107W WO2021049848A1 WO 2021049848 A1 WO2021049848 A1 WO 2021049848A1 KR 2020012107 W KR2020012107 W KR 2020012107W WO 2021049848 A1 WO2021049848 A1 WO 2021049848A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- molding
- product
- standard data
- molding condition
- mold
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
- B29C45/17—Component parts, details or accessories; Auxiliary operations
- B29C45/76—Measuring, controlling or regulating
- B29C45/766—Measuring, controlling or regulating the setting or resetting of moulding conditions, e.g. before starting a cycle
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
- B29C45/17—Component parts, details or accessories; Auxiliary operations
- B29C45/76—Measuring, controlling or regulating
- B29C45/7686—Measuring, controlling or regulating the ejected articles, e.g. weight control
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76003—Measured parameter
- B29C2945/761—Dimensions, e.g. thickness
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76003—Measured parameter
- B29C2945/7613—Weight
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76177—Location of measurement
- B29C2945/7629—Moulded articles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76451—Measurement means
- B29C2945/76461—Optical, e.g. laser
- B29C2945/76464—Optical, e.g. laser cameras
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76929—Controlling method
- B29C2945/76939—Using stored or historical data sets
- B29C2945/76949—Using stored or historical data sets using a learning system, i.e. the system accumulates experience from previous occurrences, e.g. adaptive control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76929—Controlling method
- B29C2945/76979—Using a neural network
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
Definitions
- the present invention relates to an injection molding system.
- Injection molding is the most widely used manufacturing method in manufacturing plastic products.
- products such as TVs, mobile phones, PDAs, etc.
- various parts including covers and cases may be manufactured through injection molding.
- manufacturing of a product through injection molding is carried out through the following processes. First, a molding material to which pigments, stabilizers, plasticizers, fillers, and the like are added is put into a hopper to make it melt. Next, the molten molding material is injected into the mold and then cooled to solidify. Next, after extracting the solidified molding material from the mold, unnecessary parts are removed. Products having various types and sizes are manufactured through these processes.
- An injection molding machine is used as an equipment for performing such injection molding.
- the injection molding machine includes an injection device for supplying a molding material in a molten state, and a clamping device for solidifying the molding material in a molten state through cooling.
- the present invention is to solve the above problems, and to provide an artificial intelligence-based injection molding system capable of providing molding conditions with high accuracy in a short time and a method of generating molding conditions in the injection molding system. It should be.
- An object of the present invention is to provide an artificial intelligence-based injection molding system capable of generating molding conditions using a deep learning-based molding condition generation model and a molding condition generation method in the injection molding system.
- the present invention is to provide an artificial intelligence-based injection molding system and a method of generating molding conditions in the injection molding system, which can provide optimal molding conditions by additionally learning molding conditions that are incorrectly output from the molding condition generation model. Make it an assignment.
- the artificial intelligence-based injection molding system for achieving the above object extracts target specification data of the product produced by the mold from mold information about the mold to which the molten first molding material is supplied.
- a molding condition output unit 220 for inputting the extracted target standard data into a pre-learned molding condition generation model 230 and outputting a molding condition;
- An injection molding machine (100) for producing the product by supplying the first molding material to the mold according to the molding condition;
- a determination unit 250 that compares the production standard data of the produced product with the target standard data to determine whether the molding condition is suitable, and the molding condition is determined to be inadequate by the determination unit 250 Then, the molding condition output unit 220 is characterized in that the production standard data and the molding condition are generated as one feedback data set, and the molding condition generation model 230 is learned from the feedback data set. .
- Extracting target standard data which is a product standard, from mold information about a mold to which a molding material in a molten state is supplied according to another aspect of the present invention for achieving the above object; Inputting the extracted target standard data into a pre-learned molding condition generation model and outputting a molding condition; Producing a product by supplying the molding material to the mold according to the molding condition; Measuring production standard data of the produced product; Comparing the measured production standard data with the target standard data to determine whether a molding condition is suitable; And if the molding condition is determined to be inadequate as a result of determining whether the molding condition is suitable, learning the molding condition generation model by using the inappropriate molding condition and the production standard data as one feedback data set. do.
- the present invention it is possible to generate molding conditions using a deep learning-based molding condition generation model, thereby ensuring the performance of the molding condition generation model.
- the performance of the molding condition generation model can be gradually improved, thereby generating an optimum molding condition.
- FIG. 1 is a view showing an artificial intelligence-based injection molding system according to an embodiment of the present invention.
- FIG. 2 is a view showing the configuration of an injection molding machine according to an embodiment of the present invention.
- 3 is a view showing that the fixed mold and the moving mold are opened.
- FIG. 4 is a view showing that a fixed mold and a moving mold are molded and closed by a moving part.
- FIG. 5 is a view showing the configuration of a molding condition generating apparatus according to an embodiment of the present invention.
- FIG. 6 is a flowchart showing a method of generating molding conditions in an injection molding system according to an embodiment of the present invention.
- FIG. 1 is a view showing an artificial intelligence-based injection molding system according to an embodiment of the present invention.
- the artificial intelligence-based injection molding system (10, hereinafter referred to as'injection molding system') according to the present invention uses molding materials to produce products according to optimal molding conditions.
- the injection molding system 10 includes an injection molding machine 100 and a molding condition generating device 200.
- the injection molding machine 100 performs injection molding to manufacture a product.
- FIGS. 1 and 2 are views showing the configuration of an injection molding machine 100 according to an embodiment of the present invention.
- the injection molding machine 100 will be described in more detail with reference to FIGS. 1 and 2.
- the injection molding machine 100 includes an injection device 102 and a clamping device 103.
- the injection device 102 supplies the molded material in a molten state to the clamping device 103.
- the injection device 102 may include a barrel 121, an injection screw 122 disposed inside the barrel 121, and an injection drive unit 123 for driving the injection screw 122.
- the barrel 121 may be disposed parallel to the first axis direction (X axis direction).
- the first axial direction (X-axis direction) may be a direction parallel to a direction in which the injection device 102 and the clamping device 103 are spaced apart from each other.
- the injection drive unit 123 rotates the injection screw 122 to move the molding material supplied into the barrel 121 in the first direction (FD arrow direction). I can.
- the molding material can be melted by friction and heating.
- the first direction (FD arrow direction) is a direction from the injection device 102 toward the clamping device 103 and may be a direction parallel to the first axis direction (X-axis direction).
- the injection drive unit 123 can move the injection screw 122 in the first direction (FD arrow direction). . Accordingly, the molten molding material can be supplied from the barrel 121 to the clamping device 103.
- the clamping device 103 solidifies the molten molding material through cooling.
- the clamping device 103 is a fixed mold plate 131 to which the fixed mold 150 is coupled, a moving mold plate 132 to which the moving mold 160 is coupled, and the moving mold plate 132 in the first axis direction (X-axis direction). It may include a moving unit 133 to move along.
- 3 and 4 are views showing that a moving part molds and closes a fixed mold and a moving mold.
- the moving part 133 moves the moving platen 132 in the second direction (the direction of the SD arrow) to mold-close the moving mold 160 and the fixed mold 150, the injection device 102 A molten molding material is supplied into the fixed mold 150.
- the second direction (SD arrow direction) is a direction parallel to the first axis direction (X axis direction) and opposite to the first direction (FD arrow direction).
- the clamping device 103 may include a tie bar 134.
- the tie bar 134 guides the movement of the moving platen 132.
- the movable platen 132 may be movably coupled to the tie bar 134.
- the moving platen 132 may move along the tie bar 134 in the first axis direction (X axis direction).
- the tie bar 134 may be disposed parallel to the first axis direction (X axis direction).
- the tie bar 134 may be coupled to be inserted into each of the fixed plate 131 and the movable plate 132.
- the injection molding machine 100 produces a product by supplying the molding material to the molded moving mold 160 and the fixed mold 150 according to the molding conditions generated by the molding condition generating device 200.
- the moving mold 160 and the fixed mold 150 will be described as molds.
- the molding condition generating device 200 generates a molding condition and transmits it to the injection molding machine 100. At this time, the molding condition generating device 200 determines whether or not the molding conditions are suitable by using the product produced according to the molding conditions in order to generate the optimum molding conditions.
- the molding condition generating apparatus 200 includes a standard data extracting unit 210, a molding condition outputting unit 220, a molding condition generating model 230, and a determining unit 250.
- the standard data extraction unit 210 extracts target standard data, which is the standard of a product, from the mold information. Specifically, the standard data extraction unit 210 extracts target standard data of a product produced from a corresponding mold from mold information about a mold to which the first molding material in a molten state is supplied. At this time, the first molding material means a molding material used in the product to be produced.
- the standard data includes at least one of shape information and product weight information.
- the shape information is the total volume of the product produced in the mold, the volume of the cavity of the mold, the number of cavities, the number of gates of the mold, the surface area of the product, the surface area of the cavity, the first projection of the product.
- the first to third projection areas mean the areas vertically projected from the respective axial planes (XY, YZ, ZX) of the product.
- the diameter of the gate means a circular diameter or a hydraulic diameter.
- the standard data extraction unit 210 may generate mold information by scanning a mold for producing a corresponding product, and extract shape information of the product from the mold information. Unlike this embodiment, the standard data extraction unit 210 may generate mold information by receiving a mold drawing of a product, and extract shape information of the product therefrom.
- the standard data extraction unit 210 extracts the solid state density of the first molding material from among the plurality of molding materials from the material property database 215. In addition, the standard data extraction unit 210 may extract the weight of the product using the solid density of the extracted first molding material and the total volume of the product.
- the material property database 215 stores solid state densities of a plurality of molding materials.
- the molding condition generating device 200 is shown to include a material property database 215, but this is only an example, and the material property database 215 is separate from the molding condition generating device 200. It may be composed of the composition of.
- the molding condition output unit 220 inputs the target standard data extracted by the standard data extraction unit 210 into the pre-learned molding condition generation model 230 to output the molding conditions.
- the molding condition is at least one of the temperature of the mold, the temperature of the barrel 121, the injection speed of the injection molding machine 100, the holding time of the injection molding machine 100, and the holding pressure of the injection molding machine 100 It may include.
- the molding condition output unit 220 transmits the output molding condition to the injection molding machine 100. Accordingly, the injection molding machine 100 produces a product by supplying the first molding material to the mold according to the molding conditions.
- the molding condition output unit 220 when the molding condition output unit 220 produces the product using the output molding condition and it is determined that the molding condition is inappropriate, the production standard data of the product produced under the inappropriate molding condition and the corresponding molding The condition is created as one set of feedback data. Then, the molding condition output unit 220 learns the molding condition generation model 230 from the feedback data set.
- the molding condition output unit 220 may transfer learning to a feedback data set when learning the molding condition generation model 230.
- the molding condition output unit 220 may input target standard data to the molding condition generation model 230 to output the modified molding condition.
- the present invention generates production standard data of products produced under inappropriate molding conditions and corresponding molding conditions as a single data set to learn the molding condition generation model 230, thereby gradually improving the performance of the molding condition generation model 230. Not only can it be improved, but because it can automatically find the optimal molding conditions, it is possible to produce the highest quality products even without skilled experts.
- the molding condition generation model 230 When the target standard data is input through the molding condition output unit 220, the molding condition generation model 230 generates molding conditions according to the data.
- the molding condition generation model 230 may be learned by the molding condition output unit 220.
- the molding condition generation model 230 according to the present invention produces a product using the molding condition output by the molding condition output unit 220 and it is determined that the molding condition is inappropriate, It can be additionally learned by using the production standard data of the produced product and the corresponding molding conditions as one feedback data set.
- the shaping condition generation model 230 may be a neural network network that enables shaping conditions to be output according to target standard data based on a plurality of weights and a plurality of biases.
- the molding condition generation model 230 may be implemented with an artificial neural network (ANN) algorithm.
- ANN artificial neural network
- the determination unit 250 compares the production standard data of the product produced using the molding conditions output by the molding condition output unit 220 and the target standard data extracted by the standard data extraction unit 210 to determine the molding conditions. Judging whether it is suitable or not. Specifically, when the production standard data is out of a predetermined reference range from the target standard data, the determination unit 250 determines that the corresponding molding condition is inappropriate. In addition, if the production standard data is within a predetermined reference range from the target standard data, the determination unit 250 determines that the corresponding molding condition is suitable.
- the determination unit 250 measures the weight of the product included in the production standard data as 100g, the weight of the product included in the target standard data is extracted as 90g, and if the reference range is 5g, the weight of the production standard data is As it is out of the standard range from the weight of the target standard data, it is judged that the molding condition is inappropriate.
- the determination unit 250 When determining that the molding condition is inappropriate, the determination unit 250 transmits a stop command to the injection molding machine 100. Accordingly, the injection molding machine 100 stops production of the product. In addition, if the determination unit 250 determines that the molding condition is inappropriate, it transmits a feedback learning command to the molding condition output unit 220. Accordingly, the molding condition output unit 220 learns the molding condition generation model 230 by using inappropriate molding conditions and production standard data as one feedback data set.
- the molding condition generating apparatus 200 may further include a standard data measuring unit 240 and a model generating unit 260 as shown in FIG. 5.
- the standard data measuring unit 240 measures production standard data of a product produced from the injection molding machine 100. To this end, as shown in FIG. 5, the standard data measurement unit 240 includes a takeout unit 242 and a standard data generation unit 244.
- the take-out unit 242 takes out the product produced from the mold.
- the take-out unit 242 may be a multi-joint take-out robot.
- the standard data generation unit 244 generates production standard data from the extracted product. Specifically, the standard data generation unit 244 generates first shape information by photographing a product, generates first weight information by measuring the weight of the product, and produces the first shape information and the second weight information. Create standard data. In this case, the standard data generation unit 244 may be implemented as a vision system (not shown) to generate the first shape information.
- the first shape information of the product is the total volume of the product, the volume of the part corresponding to the cavity of the mold, the number of parts corresponding to the cavity, the number of parts corresponding to the gate of the mold, and the product Surface area of the product, the first projected area of the product (XY), the second projected area of the product (YZ), the third projected area of the product (ZX), the maximum thickness of the product, the average thickness of the product, the standard deviation of the thickness of the product, It may include at least one of a diameter of a portion corresponding to the gate, a maximum flow distance from the portion corresponding to the gate to an end of the product, and a ratio of the maximum flow distance to the average thickness of the product.
- the standard data generation unit 244 transmits the generated production standard data to the determination unit 250.
- the model generation unit 260 generates a molding condition generation model 230. Specifically, the model generation unit 260 may generate the forming condition generation model 230 by learning a neural network network using a plurality of training data sets.
- the model generation unit 260 generates a plurality of learning data sets by integrating a plurality of learning molding conditions collected in advance and learning standard data of a product produced according to each learning molding condition.
- the learning molding condition may include at least one of the temperature of the mold, the temperature of the barrel 121, the injection speed of the injection molding machine 100, the holding time of the injection molding machine 100, and the holding pressure of the injection molding machine 100.
- the learning standard data may include at least one of shape information and product weight information.
- the model generation unit 260 generates a forming condition generation model 230 by learning a neural network network using a plurality of generated training data sets.
- the model generator 260 builds a weight prediction system by training a neural network network having a predetermined layer structure as a training data set, and converts the training data set to the same value domain, with a minimum-maximum normalization (Min-Max). normalization).
- the learning data set may be divided into n-dimensional input data consisting of shape information and molding conditions, and one-dimensional output data consisting of weight information of a product.
- n may mean the number of information included in the shape information and molding conditions. For example, if the shape information includes 15 pieces of information and the molding condition includes 5 pieces of information, n is 20.
- model generation unit 260 distributes the input data and the output data according to a predetermined ratio for training, verification, and testing.
- the model generation unit 260 extracts shape information related to the weight information of the product from among the shape information in order to increase the accuracy of the molding condition generation model 230, and generates a weight prediction system of the product using this.
- the model generator 260 may perform sensitivity analysis to extract shape information related to weight information of a product from among shape information.
- the model generator 260 may perform a grid search or a random search to determine a hyper parameter of a neural network.
- the grid search may be applied to the activation function, the optimization method, and the initialization method, and the random search may be applied to the remaining hyper parameters.
- the model generation unit 260 generates a molding condition generating model 230 for deriving a molding condition corresponding to the weight when the weight is presented in reverse using the generated weight prediction system. Accordingly, when the shape information and the weight information are input to the molding condition generation model 230, weight information according to the shape information is input, and a molding condition according to the shape information may be input.
- the model generation unit 260 may generate the molding condition generation model 260 from the weight prediction system using particle swarm optimization or random search.
- the user can guide the user to find the process conditions without specialized knowledge about injection molding, so that the dependence of the expert is lowered, and the corresponding molding Since the condition generation model 230 can be improved through additional learning using feedback data, there is an effect that it is possible to establish a smart factory in the injection field based on an unmanned injection molding system by securing higher accuracy.
- FIG. 6 is a flowchart showing a method of generating molding conditions in an injection molding system according to an embodiment of the present invention.
- the injection molding system 10 extracts target standard data, which is the product standard, from the mold information (S600). Specifically, the injection molding system 10 extracts target standard data of a product produced from the corresponding mold from mold information about a mold to which the first molding material in a molten state is supplied. At this time, the first molding material means a molding material used in the product to be produced.
- the standard data includes at least one of shape information and product weight information.
- the shape information of the product is the total volume of the product produced in the mold, the volume of the cavity of the mold, the number of cavities, the number of gates of the mold, the surface area of the product, the surface area of the cavity, the number of the product.
- the first to third projection areas mean the areas vertically projected from the respective axial planes (XY, YZ, ZX) of the product.
- the diameter of the gate means a circular diameter or a hydraulic diameter.
- the injection molding system 10 extracts the solid state density of the first molding material from among the plurality of molding materials from the material property database 215. In addition, the injection molding system 10 may extract the weight of the product using the solid density of the extracted first molding material and the total volume of the product.
- the injection molding system 10 inputs the extracted target standard data into the pre-learned molding condition generation model 230 and outputs the molding conditions (S610).
- the molding condition is at least one of the temperature of the mold, the temperature of the barrel 121, the injection speed of the injection molding machine 100, the holding time of the injection molding machine 100, and the holding pressure of the injection molding machine 100 It may include.
- the injection molding system 10 produces a product by supplying the first molding material to the mold according to the molding condition (S620).
- the injection molding system 10 measures production standard data of the produced product (S630).
- the injection molding system 10 compares the measured production standard data with the target standard data to determine whether the molding conditions are suitable (S640). Specifically, in addition, if the production standard data is within a predetermined reference range from the target standard data, the injection molding system 10 determines that the corresponding molding condition is suitable (S650). The injection molding system 10 determines that the molding condition is inappropriate if the production standard data is out of a predetermined reference range from the target standard data (S660).
- the injection molding system 10 determines that the molding conditions are inappropriate, it stops production of the product.
- the injection molding system 10 may transfer the molding condition generation model 230 to a feedback data set.
- the injection molding system 10 inputs the target standard data into the molding condition generation model 230 and outputs the modified molding conditions.
- This component may be provided as a series of computer directives through a computer-readable medium or a machine-readable medium including volatile and non-volatile memory.
- the directives may be provided as software or firmware, and may, in whole or in part, be implemented in hardware configurations such as ASICs, FPGAs, DSPs, or other similar devices.
- the directives may be configured to be executed by one or more processors or other hardware configurations, wherein the processor or other hardware configurations perform or perform all or part of the methods and procedures disclosed herein when executing the series of computer directives. To be able to do it.
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Mechanical Engineering (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Injection Moulding Of Plastics Or The Like (AREA)
Abstract
Description
Claims (11)
- 용융상태의 제1 성형재료가 공급되는 금형에 대한 금형정보로부터 상기 금형으로 생산되는 제품의 타겟 규격데이터를 추출하는 규격데이터 추출부(210);추출된 상기 타겟 규격데이터를 미리 학습된 성형조건 생성모델(230)에 입력하여 성형조건을 출력하는 성형조건 출력부(220);상기 성형조건에 따라 상기 제1 성형재료를 상기 금형으로 공급하여 상기 제품을 생산하는 사출성형기(100); 및생산된 상기 제품의 생산 규격데이터와 상기 타겟 규격데이터를 비교하여 상기 성형조건의 적합여부를 판단하는 판단부(250)를 포함하고,상기 판단부(250)에 의해 상기 성형조건이 부적합한 것으로 판단되면, 상기 성형조건 출력부(220)는, 상기 생산 규격데이터와 상기 성형조건을 하나의 피드백데이터 세트로 생성하고, 상기 피드백데이터 세트로 상기 성형조건 생성모델(230)을 학습시키는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제1항에 있어서,상기 성형조건 출력부(220)는,상기 성형조건 생성모델(230)을 상기 피드백데이터 세트로 전이학습(Transfer Learning) 시키는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제1항에 있어서,상기 성형조건 출력부(220)는,상기 성형조건 생성모델(230)의 학습이 완료되면, 상기 타겟 규격데이터를 상기 성형조건 생성모델(230)에 입력하여 수정된 성형조건을 출력하는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제1항에 있어서,상기 성형조건 생성모델(230)은 복수개의 가중치 및 복수개의 바이어스(bias)를 기초로 상기 타겟 규격데이터에 따라 상기 성형조건이 출력될 수 있게 하는 신경망 네트워크로 구성된 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제1항에 있어서,상기 성형조건 생성모델(230)을 생성하는 모델 생성부(260)를 더 포함하고,상기 모델 생성부(260)는,복수개의 학습 성형조건과 각 학습 성형조건에 따라 생산되는 제품의 학습 규격데이터를 통합하여 복수개의 학습데이터 세트를 생성하고, 상기 복수개의 학습데이터 세트로 신경망 네트워크를 학습시켜 상기 성형조건 생성모델(230)을 생성하는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제1항에 있어서,상기 성형조건은상기 금형의 온도, 배럴의 온도, 상기 사출성형기의 사출속도, 상기 사출성형기의 보압시간, 및 상기 사출성형기의 보압압력 중 적어도 하나를 포함하는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제1항에 있어서,상기 규격데이터는 형상정보 및 상기 제품의 무게정보 중 적어도 하나를 포함하는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제7항에 있어서,상기 형상정보는,상기 금형에서 생산되는 상기 제품의 전체부피, 상기 금형의 캐비티(cavitiy)의 부피, 상기 캐비티의 개수, 상기 금형의 게이트의 개수, 상기 제품의 표면적, 상기 캐비티의 표면적, 상기 제품의 제1 투영면적(XY), 상기 제품의 제2 투영면적(YZ), 상기 제품의 제3 투영면적(ZX), 상기 제품의 최대두께, 상기 제품의 평균두께, 상기 제품의 두께의 표준편차, 상기 게이트의 직경, 상기 게이트에서 상기 제품의 말단까지의 최대 유동거리, 상기 최대 유동거리 대 상기 제품의 평균두께의 비 중 적어도 하나를 포함하는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제7항에 있어서,복수개의 성형재료의 고상밀도가 저장된 재료물성 데이터 베이스(215)를 더 포함하고,상기 규격데이터 추출부(210)는,상기 재료물성 데이터 베이스(215)로부터 복수개의 성형재료 중 상기 제1 성형재료의 고상밀도를 추출하고, 상기 제품의 전체부피와 상기 제1 성형재료의 고상밀도를 이용하여 상기 제품의 무게를 산출하는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 제1항에 있어서,생산된 상기 제품의 상기 생산 규격데이터를 측정하는 규격데이터 측정부(240)를 더 포함하고,상기 규격데이터 측정부(240)는,상기 금형으로부터 생산된 상기 제품을 취출하는 취출유닛(242);취출된 상기 제품을 촬영하여 생성되는 제1 형상정보 및 상기 제품의 무게를 측정하여 생성되는 제1 무게정보를 포함하는 상기 생산 규격데이터를 생성하는 규격데이터 생성유닛(244)를 포함하는 것을 특징으로 하는 인공지능 기반의 사출성형시스템.
- 용융상태의 성형재료가 공급되는 금형에 대한 금형정보로부터 제품의 규격이 되는 타겟 규격데이터를 추출하는 단계;추출된 상기 타겟 규격데이터를 미리 학습된 성형조건 생성모델에 입력하여 성형조건을 출력하는 단계;상기 성형조건에 따라 상기 성형재료를 상기 금형으로 공급하여 제품을 생산하는 단계;생산된 상기 제품의 생산 규격데이터를 측정하는 단계;측정된 상기 생산 규격데이터와 상기 타겟 규격데이터를 비교하여 성형조건의 적합여부를 판단하는 단계; 및상기 성형조건의 적합여부 판단결과, 상기 성형조건이 부적합한 것으로 판단하면, 부적합한 성형조건과 상기 생산 규격데이터를 하나의 피드백데이터 세트로 하여 상기 성형조건 생성모델을 학습시키는 단계를 포함하는 것을 특징으로 하는 사출성형시스템에서의 성형조건 생성방법.
Priority Applications (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202080062899.3A CN114364503B (zh) | 2019-09-11 | 2020-09-08 | 基于人工智能的注塑成型系统以及注塑成型系统中的成型条件生成方法 |
| EP20862082.3A EP4029672A4 (en) | 2019-09-11 | 2020-09-08 | INJECTION MOLDING SYSTEM BASED ON ARTIFICIAL INTELLIGENCE AND METHOD FOR PRODUCING A MOLDING CONDITION IN AN INJECTION MOLDING SYSTEM |
| JP2022514961A JP7411786B2 (ja) | 2019-09-11 | 2020-09-08 | 人工知能基盤の射出成形システムおよび射出成形システムでの成形条件生成方法 |
| US17/641,596 US12208552B2 (en) | 2019-09-11 | 2020-09-08 | Artificial intelligence-based injection molding system and method for generating molding condition in injection molding system |
| CA3149727A CA3149727C (en) | 2019-09-11 | 2020-09-08 | Artificial intelligence-based injection molding system and method for generating molding condition in injection molding system |
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR10-2019-0113137 | 2019-09-11 | ||
| KR20190113137 | 2019-09-11 | ||
| KR10-2020-0105270 | 2020-08-21 | ||
| KR1020200105270A KR20210031373A (ko) | 2019-09-11 | 2020-08-21 | 인공지능 기반의 사출성형시스템 및 사출성형시스템에서의 성형조건 생성방법 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2021049848A1 true WO2021049848A1 (ko) | 2021-03-18 |
Family
ID=74867294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/KR2020/012107 Ceased WO2021049848A1 (ko) | 2019-09-11 | 2020-09-08 | 인공지능 기반의 사출성형시스템 및 사출성형시스템에서의 성형조건 생성방법 |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US12208552B2 (ko) |
| EP (1) | EP4029672A4 (ko) |
| JP (1) | JP7411786B2 (ko) |
| CN (1) | CN114364503B (ko) |
| CA (1) | CA3149727C (ko) |
| WO (1) | WO2021049848A1 (ko) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7484610B2 (ja) * | 2020-09-23 | 2024-05-16 | コニカミノルタ株式会社 | 情報処理装置、学習装置、情報処理システム、情報処理方法、プログラム、および、記録媒体 |
| CN115230106B (zh) * | 2022-06-30 | 2025-05-02 | 苏州缔微致智能科技有限公司 | 一种注塑件加工工艺控制方法及系统 |
| CN115171825B (zh) * | 2022-07-26 | 2025-10-21 | 成都星云智联科技有限公司 | 型材产品规格参数控制方法、装置、电子设备及介质 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170101607A (ko) * | 2016-02-29 | 2017-09-06 | 엘에스엠트론 주식회사 | 사출성형장치를 이용하여 제품을 생산하는 방법 |
| KR20180116610A (ko) * | 2017-04-17 | 2018-10-25 | 영남대학교 산학협력단 | 형상과 무게 및 강도가 정해진 공극을 가진 사출성형물의 사출조건을 찾기위한 2차 수학모델을 이용한 서출성형 조건의 탐색방법 및 그 방법을 이용한 컴퓨터 프로그램 |
| KR20190022258A (ko) * | 2017-08-24 | 2019-03-06 | 엘에스엠트론 주식회사 | 보압 전환 위치 자동 설정 기능을 구비하는 사출성형기 및 제어 방법 |
| KR20190065834A (ko) * | 2017-12-04 | 2019-06-12 | 엘에스엠트론 주식회사 | 사출성형기 및 사출 제어 데이터 생성 방법 |
Family Cites Families (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0653380B2 (ja) * | 1987-02-27 | 1994-07-20 | 日精樹脂工業株式会社 | 射出成形機の成形条件設定方法 |
| JPH0477219A (ja) | 1990-07-19 | 1992-03-11 | Toshiba Corp | 成形支援エキスパートシステム |
| JPH04209004A (ja) * | 1990-12-03 | 1992-07-30 | Toyo Mach & Metal Co Ltd | 射出成形機の制御方法 |
| US6275741B1 (en) | 1998-10-05 | 2001-08-14 | Husky Injection Molding Systems Ltd. | Integrated control platform for injection molding system |
| JP2000258151A (ja) * | 1999-03-05 | 2000-09-22 | Canon Inc | 三次元形状測定装置および金型構造体 |
| US7580771B2 (en) * | 2004-10-19 | 2009-08-25 | Husky Injection Molding Systems Ltd. | Intelligent molding environment and method of configuring a molding system |
| JP4499601B2 (ja) * | 2005-04-01 | 2010-07-07 | 日精樹脂工業株式会社 | 射出成形機の制御装置 |
| JP4789138B2 (ja) * | 2005-04-14 | 2011-10-12 | 三菱重工プラスチックテクノロジー株式会社 | 射出成形支援装置、射出成形システム、射出成形支援制御方法 |
| JP2007313852A (ja) * | 2006-05-29 | 2007-12-06 | Mitsubishi Electric Engineering Co Ltd | 金型修正システム |
| US20130053995A1 (en) * | 2011-08-25 | 2013-02-28 | Konica Minolta Business Technologies, Inc. | Three-dimensional object molding apparatus and control program |
| DE102013008245A1 (de) * | 2013-05-15 | 2014-11-20 | Arburg Gmbh + Co. Kg | Verfahren zum Betreiben einer Maschine zur Verarbeitung von Kunststoffen |
| JP6054991B2 (ja) * | 2015-01-07 | 2016-12-27 | ファナック株式会社 | 射出成形システム |
| JP6346128B2 (ja) * | 2015-07-28 | 2018-06-20 | ファナック株式会社 | 最適な操作条件を算出できる射出成形システム及び機械学習器 |
| AT519096B1 (de) * | 2016-12-23 | 2018-04-15 | Engel Austria Gmbh | Verfahren zum Einstellen einer Formgebungsmaschine |
| JP6573647B2 (ja) * | 2017-07-10 | 2019-09-11 | 株式会社日本製鋼所 | 計算機と複数台の射出成形機とからなる射出成形機システム |
| JP2019119172A (ja) * | 2018-01-10 | 2019-07-22 | 池上金型工業株式会社 | プログラム及び情報処理装置 |
-
2020
- 2020-09-08 JP JP2022514961A patent/JP7411786B2/ja active Active
- 2020-09-08 CN CN202080062899.3A patent/CN114364503B/zh active Active
- 2020-09-08 EP EP20862082.3A patent/EP4029672A4/en active Pending
- 2020-09-08 WO PCT/KR2020/012107 patent/WO2021049848A1/ko not_active Ceased
- 2020-09-08 CA CA3149727A patent/CA3149727C/en active Active
- 2020-09-08 US US17/641,596 patent/US12208552B2/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170101607A (ko) * | 2016-02-29 | 2017-09-06 | 엘에스엠트론 주식회사 | 사출성형장치를 이용하여 제품을 생산하는 방법 |
| KR20180116610A (ko) * | 2017-04-17 | 2018-10-25 | 영남대학교 산학협력단 | 형상과 무게 및 강도가 정해진 공극을 가진 사출성형물의 사출조건을 찾기위한 2차 수학모델을 이용한 서출성형 조건의 탐색방법 및 그 방법을 이용한 컴퓨터 프로그램 |
| KR20190022258A (ko) * | 2017-08-24 | 2019-03-06 | 엘에스엠트론 주식회사 | 보압 전환 위치 자동 설정 기능을 구비하는 사출성형기 및 제어 방법 |
| KR20190065834A (ko) * | 2017-12-04 | 2019-06-12 | 엘에스엠트론 주식회사 | 사출성형기 및 사출 제어 데이터 생성 방법 |
Non-Patent Citations (2)
| Title |
|---|
| See also references of EP4029672A4 * |
| TERCAN HASAN, GUAJARDO ALEXANDRO, HEINISCH JULIAN, THIELE THOMAS, HOPMANN CHRISTIAN, MEISEN TOBIAS: "Transfer-Learning: Bridging the Gap between Real and Simulation Data for Machine Learning in Injection Molding", PROCEDIA CIRP, ELSEVIER, NL, vol. 72, 1 January 2018 (2018-01-01), NL, pages 185 - 190, XP055791813, ISSN: 2212-8271, DOI: 10.1016/j.procir.2018.03.087 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114364503B (zh) | 2025-06-06 |
| EP4029672A4 (en) | 2023-10-18 |
| JP7411786B2 (ja) | 2024-01-11 |
| CN114364503A (zh) | 2022-04-15 |
| US12208552B2 (en) | 2025-01-28 |
| EP4029672A1 (en) | 2022-07-20 |
| CA3149727C (en) | 2024-01-09 |
| JP2022547118A (ja) | 2022-11-10 |
| CA3149727A1 (en) | 2021-03-18 |
| US20220297362A1 (en) | 2022-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021049848A1 (ko) | 인공지능 기반의 사출성형시스템 및 사출성형시스템에서의 성형조건 생성방법 | |
| WO2021091191A1 (ko) | 인공지능 기반의 사출성형시스템 및 성형조건 생성방법 | |
| KR20210056235A (ko) | 인공지능 기반의 사출성형시스템 및 성형조건 생성방법 | |
| CN113733505B (zh) | 注射成型系统、成型条件校正系统以及注射成型方法 | |
| WO2013065931A1 (en) | Injection molding apparatus | |
| JP6573647B2 (ja) | 計算機と複数台の射出成形機とからなる射出成形機システム | |
| WO2016032078A1 (ko) | 사출압축성형이 적용된 필름인서트성형 장치 및 그 방법 | |
| CN107246927B (zh) | 一种塑料脱模力测试方法 | |
| CN119526655A (zh) | 塑胶模具防变形的脱模控制方法及系统 | |
| WO2025135535A1 (ko) | 사출 충전 예측 시뮬레이션 서비스를 운영하는 방법과 시스템 및 이 서비스를 운영하는 프로그램을 기록한 컴퓨터로 읽을 수 있는 기록매체 | |
| KR20210031373A (ko) | 인공지능 기반의 사출성형시스템 및 사출성형시스템에서의 성형조건 생성방법 | |
| US20230241826A1 (en) | Method and device for reducing the amount of reworking required on mold cavities prior to their use in series production | |
| CN117124568A (zh) | 一种电缆保护管生产用挤出机温度控制系统 | |
| KR102314599B1 (ko) | 가스 사출 성형 제품의 품질 검사 장치 및 그 방법 | |
| CN212446195U (zh) | 注塑模具注塑监控设备 | |
| CN213166722U (zh) | 一种模具温压监测装置 | |
| CN118544552B (zh) | 一种注塑机智能能耗管理系统 | |
| WO2016099069A1 (ko) | 고분자의 동일성 평가 방법 및 이를 이용한 시스템 | |
| CN115195037B (zh) | 一种tpv波纹软管的成型装置及生产方法 | |
| Petrova et al. | Incorporation of phenomenological models in a hybrid neural network for quality control of injection molding | |
| CN116945544A (zh) | 一种基于用途的塑料母粒生产过程调节控制方法及系统 | |
| Davoudinejad et al. | Assessment of sub-mm features replication capability in injection moulding using a multi-cavity tool produced by additive manufacturing | |
| Yang et al. | A study on the comparison of the predicting performance of quality of injection molded product according to the structure of artificial neural network | |
| WO2019135465A1 (ko) | 금형 | |
| CN105034277B (zh) | 多芯插芯注塑模具 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20862082 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 3149727 Country of ref document: CA |
|
| ENP | Entry into the national phase |
Ref document number: 2022514961 Country of ref document: JP Kind code of ref document: A |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| ENP | Entry into the national phase |
Ref document number: 2020862082 Country of ref document: EP Effective date: 20220411 |
|
| WWG | Wipo information: grant in national office |
Ref document number: 202080062899.3 Country of ref document: CN |