KR20200134481A - Apparatus and method for detecting fault in plastic injection molding process - Google Patents

Abstract

The present invention relates to an apparatus and method for detecting defects in a plastic injection molding process, which collects process data for a plastic injection molding machine in process under predetermined molding conditions, detects defects from the process data based on a defect detection model, and determine the molding conditions to be proposed to solve the defects based on a molding condition proposal model.

Description

Plastic injection molding process defect detection device and method {APPARATUS AND METHOD FOR DETECTING FAULT IN PLASTIC INJECTION MOLDING PROCESS}

Various embodiments relate to a plastic injection molding process defect detection apparatus and method.

In general, plastic injection molding is a process in which a plastic resin is melted by a thermoplastic plastic processing method, then compressed, pushed into a mold, and cooled to produce a product. An injection molding machine is a machine that performs an injection molding process. When molding conditions such as injection temperature, injection input, and injection speed are input to the injection molding machine, the injection molding machine operates accordingly to produce a product.

Even if the molding conditions are input to the injection molding machine, the molding conditions may not be compliant with the surrounding environment and mechanical reasons, and thus, defects may occur. In addition to the cause of the injection molding machine, defects may occur in the mold temperature and the condition of raw materials. If a defect occurs during the injection molding process, the operator must stop the operation of the injection molding machine and change the molding conditions. However, when a defect occurs, it is difficult for the operator to immediately recognize the occurrence of the defect, and even if he or she recognizes the occurrence of the defect, trial and error are experienced to find an appropriate molding condition for normal production. For this reason, productivity decreases with delay in dealing with defects.

Various embodiments learn a defect detection model and a molding condition prediction model when a defect occurs using sensor values of an injection molding machine and peripheral devices, and use the learned models to detect injection molding defects and propose a molding condition to be changed. Systems and methods can be provided.

An operation method of an electronic device according to various embodiments includes an operation of collecting process data for a plastic injection molding machine in process under a predetermined molding condition, an operation of detecting a defect from the process data based on a defect detection model, and molding. It may include an operation of determining a molding condition to be proposed to solve the defect based on the condition proposal model.

An electronic device according to various embodiments is connected to an interface module for an interface with a plastic injection molding machine in process under predetermined molding conditions, and the interface module, and collects process data for the injection molding machine through the interface module. And, it may include a processor configured to detect a defect from the process data.

According to various embodiments, the processor may be configured to detect the defect from the process data based on a defect detection model and determine a molding condition to be proposed to solve the defect based on a molding condition suggestion model. .

According to various embodiments, the electronic device may efficiently detect and respond to a defect occurring in a plastic injection molding machine. That is, when a defect occurs in the injection molding machine, the electronic device can detect the defect within a short time, and a countermeasure for the defect can be applied. Through this, the productivity of the injection molding machine can be improved.

1 is a diagram illustrating a system according to various embodiments.
2 is a diagram illustrating an electronic device according to various embodiments.
3 is a diagram illustrating a method of operating an electronic device according to various embodiments.
4 is a diagram illustrating an operation of applying the failure detection model of FIG. 3.
FIG. 5 is a diagram illustrating a second possible defective data detection operation of FIG. 4.
6 is a diagram illustrating an operation of applying a molding condition proposal model of FIG. 3.

Hereinafter, various embodiments of the present document will be described with reference to the accompanying drawings.

1 is a diagram illustrating a system 100 according to various embodiments.

Referring to FIG. 1, a system 100 according to various embodiments may include a plastic injection molding machine 110 and an electronic device 130.

The plastic injection molding machine 110 may perform a plastic injection molding process. The injection molding machine 110 may injection-mold the raw material 111 to produce a product. To this end, the injection molding machine 110 may melt the raw material 111. In addition, the injection molding machine 110 applies pressure to the raw material 111 by controlling the extrusion cylinder 117 and the screw 118 by driving the hydraulic motor 115 while inputting the raw material 111 through the hopper 113. can do. Through this, the injection molding machine 110 may push the raw material 111 into the mold 119. In addition, the injection molding machine 110 can produce a product by cooling the raw material 111.

The injection molding machine 110 may perform an injection molding process under predetermined molding conditions. For example, the molding conditions may include a number of parameters as shown in Table 1 below. Here, in the injection molding machine 110, the molding conditions may be set by the operator 120. In addition, the injection molding machine 110 may include a plurality of sensors (not shown). The sensors may be distributed and disposed in the injection molding machine 110. Through this, the sensors can respectively detect sensor values at each location.

The electronic device 130 may detect a defect by monitoring the injection molding machine 110. To this end, while the injection molding machine 110 performs an injection molding process, the electronic device 130 may collect process data for the injection molding machine 110. For example, the process data may include a plurality of process values as shown in [Table 1] below. The electronic device 130 may receive sensor values of sensors from the injection molding machine 110 and detect process data based on the sensor values. In addition, the electronic device 130 may detect a defect from process data. In addition, the electronic device 130 may determine a molding condition to be proposed to solve the defect. To this end, the electronic device 130 may store at least one of a failure detection model or a molding condition proposal model. Through this, the electronic device 130 may propose the determined molding condition to the operator 120 or directly control the injection molding machine 110 based on the determined molding condition. For example, when a molding condition is proposed from the electronic device 130, the operator 120 may reset the linear condition of the injection molding machine 110 based on the proposed molding condition.

2 is a diagram illustrating an electronic device 130 according to various embodiments.

Referring to FIG. 2, an electronic device 130 according to various embodiments may include at least one of an interface module 210, an input module 220, an output module 230, a memory 240, or a processor 250. It may include. In some embodiments, at least one of the components of the electronic device 130 may be omitted, or one or more other components may be added to the electronic device 130.

The interface module 210 may interface with the plastic injection molding machine 110 in the electronic device 130. At this time, the interface module 210 may communicate with the injection molding machine 110 by a wired method or a wireless method. The interface module 210 may include at least one of a receiving device and a transmitting device. The receiving device may receive a signal from the injection molding machine 110, and the transmitting device may transmit a signal to the injection molding machine 110.

The input module 220 may input a command used for at least one component of the electronic device 130. The input module 220 is an input device configured to directly input a command or data to the electronic device 130 by the operator 120, a sensing device configured to detect a surrounding environment to generate data, or wired communication or wireless communication. It may include at least one of a communication device configured to receive a command or data from an external device through. For example, the input device may include at least one of a microphone, a mouse, a keyboard, and a camera.

The output module 230 may provide information to the outside of the electronic device 130. The output module 230 includes at least one of an audio output device configured to audibly output information, a display device configured to visually output information, or a communication device configured to transmit information to an external device through wired communication or wireless communication. It can contain either.

The memory 240 may store various types of data used by at least one component of the electronic device 130. The data may include input data or output data for a program or a command related thereto. For example, the memory 240 may include at least one of a volatile memory or a nonvolatile memory. The memory 240 may store at least one of a failure detection model or a molding condition proposal model.

The processor 250 may execute a program in the memory 240 to control components of the electronic device 130 and perform data processing or operation. The processor 250 may detect a defect by monitoring the injection molding machine 110. To this end, while the injection molding machine 110 performs an injection molding process, the processor 250 may collect process data for the injection molding machine 110 through the interface module 210. In addition, the processor 250 may detect a defect from process data based on the defect detection model. In addition, the processor 250 may determine a molding condition to be proposed to solve the defect based on the molding condition proposal model. According to an embodiment, the processor 250 may propose the determined molding condition to the operator 120 through the output module 230. According to another embodiment, the processor 250 may directly control the injection molding machine 110 through the interface module 210 based on the determined molding condition.

3 is a diagram illustrating a method of operating an electronic device 130 according to various embodiments.

Referring to FIG. 3, the electronic device 130 may collect process data for the plastic injection molding machine 110 in operation 310. While the injection molding machine 110 performs an injection molding process, the electronic device 130 may collect process data for the injection molding machine 110. In this case, the injection molding machine 110 may perform an injection molding process under predetermined molding conditions. For example, the molding conditions may include a plurality of parameters as shown in [Table 1]. The processor 250 may receive sensor values of sensors from the injection molding machine 110 through the interface module 210 and detect process data based on the sensor values. For example, the process data may include a plurality of process values as shown in [Table 1].

The electronic device 130 may apply a failure detection model in operation 320. The electronic device 130 may analyze process data based on a defect detection model. In this case, the processor 250 may reduce a range to be analyzed in the process data. In addition, the processor 250 may analyze the reduced range from the process data.

4 is a diagram illustrating an operation of applying the failure detection model of FIG. 3.

Referring to FIG. 4, in operation 410, the electronic device 130 may detect first possible defect data from process data. The processor 250 may remove good product data that are clearly distinguished from the process data. For example, based on a predetermined reference value, the processor 250 may classify process data from good product data and first possible defect data, and remove good product data. Through this, the processor 250 may detect the first possible defect data from the process data.

The electronic device 130 may classify data attributes based on the first possible defective data in operation 420. To this end, the processor 250 may remove a correlation between data attributes. For example, the processor 250 may remove a correlation between data attributes by performing a principal component analysis (PCA) transformation on the first possible defective data. The electronic device 130 may select at least one of the data attributes in operation 430. The processor 250 may select at least one of the data attributes based on a defect and a correlation for each of the data attributes. In this case, the processor 250 may select a data attribute having a high correlation with a defect. The electronic device 130 may derive a Gaussian distribution of the first possible defective data based on the data attribute selected in operation 440.

In operation 450, the electronic device 130 may detect the second possible defective data from the first possible defective data based on a Gaussian distribution. The processor 250 may divide the entire area of the Gaussian distribution into a first area and a second area by using the threshold value. In the Gaussian distribution, the processor 250 may detect the inner side of the threshold value as the first area and the outer side of the threshold value as the second area. In addition, the processor 250 may detect second possible defective data corresponding to the second area from the first possible defective data. Here, the processor 250 may detect the good product data corresponding to the first area from the first possible defect data and remove the good product data.

FIG. 5 is a diagram illustrating a second possible defective data detection operation of FIG. 4.

Referring to FIG. 5, in operation 510, the electronic device 130 may divide the entire area of the Gaussian distribution into a first area and a second area by using a candidate threshold. To this end, the memory 240 may store a plurality of candidate thresholds. The processor 250 may select any one of candidate thresholds. Further, the processor 250 may divide the entire area of the Gaussian distribution into a first area and a second area by using the selected candidate threshold. In the Gaussian distribution, the processor 250 may detect the inner side of the threshold value as the first area and the outer side of the threshold value as the second area.

The electronic device 130 may determine whether the candidate threshold meets a predetermined condition in operation 520. To this end, the processor 250 may check at least one of a ratio between the first region and the second region (recall) or the ratio between the entire region and the second region (precision). Further, the processor 250 may determine whether at least one of the ratio of the first region and the second region or the ratio of the entire region and the second region meets the condition. At this time, the condition may be set in advance by the operator 120.

If it is determined in operation 520 that the condition is not met, the electronic device 130 may change the candidate threshold in operation 525. The processor 250 may select another one of candidate thresholds. Further, the electronic device 130 may return to operation 510. Through this, the electronic device 130 may repeatedly perform at least some of operations 510 to 525.

If it is determined that the condition is satisfied in operation 520, the electronic device 130 may determine the candidate threshold as the final threshold in operation 530. Through this, the processor 250 may maintain the first region and the second region in the Gaussian distribution. In operation 540, the electronic device 130 may detect second possible defective data corresponding to the second area. The processor 250 may detect second possible defective data corresponding to the second area from the first possible defective data. Here, the processor 250 may detect the good product data corresponding to the first area from the first possible defect data and remove the good product data. After that, the electronic device 130 may return to FIG. 4.

Referring to FIG. 4, in operation 460, the electronic device 130 may analyze second possible defective data. In this case, the electronic device 130 may learn a failure detection model by using the second possible failure data. For example, the processor 250 may learn by applying a machine learning algorithm such as a multilayer perceptron (MLP) or a decision tree model to the second possible bad data. After that, the electronic device 130 may return to FIG. 3.

Referring to FIG. 3, the electronic device 130 may detect a defect while analyzing process data in operation 330. When a defect is detected in operation 330, the electronic device 130 may apply a molding condition proposal model in operation 340. Through this, the electronic device 130 may determine a molding condition to be proposed to solve the defect.

6 is a diagram illustrating an operation of applying a molding condition proposal model of FIG. 3.

Referring to FIG. 6, in operation 610, the electronic device 130 may analyze process data according to parameters of a linear condition. The processor 250 may analyze process data according to each of the parameters. At this time, the memory 240 may store a molding condition proposal model as shown in Table 2 below. Here, the molding condition proposal model may include records related to previously generated defects. Records related to previously occurring defects are process data of the injection molding machine 110 related to the previously generated defect, that is, sensor values detected by the injection molding machine 110, set values of parameters of molding conditions, and previously generated. It may include whether or not to change the setting value for each parameter in response to the defect. For example, in response to a previously occurring defect, whether or not to change at least one parameter changed by the operator 120, that is, reset, is stored as 1, and at least maintained unchanged by the operator 120 Whether to change one of the remaining parameters may be stored as 0. Through this, the processor 250 may compare the process values of the process data and the setting values for each parameter of the molding condition with a previously stored record.

The electronic device 130 may determine at least one parameter related to the defect from the parameters in operation 620. The processor 250 may determine a parameter to be changed among parameters. To this end, based on a result of comparing the process values of the process data and the setting value for each parameter of the molding condition with a previously stored record, the processor 250 may check whether the setting value for each parameter is changed from the previously stored record. For example, the processor 250 may detect a parameter whose change is set to 1 in a previously stored record.

The electronic device 130 may propose to the worker 120 to change the determined parameter in operation 630. The processor 250 may propose to the operator 120 to change the determined parameter through the output module 230. In this case, the processor 250 may provide a suggested value for the determined parameter. To this end, as shown in [Table 2], the previously stored record may further include a set value of at least one parameter changed, that is, reset by the operator 120 in response to a previously generated defect. The processor 250 may determine a setting value of a parameter determined from a previously stored record as a suggested value.

According to an embodiment, the processor 250 may provide a suggested value of the determined parameter to the operator 120 through the output module 230. For example, when a change of at least one of the parameters is proposed from the electronic device 130, the operator 120 may reset the proposed parameter. For example, if at least one suggested value among parameters is provided from the electronic device 130, the operator 120 may reset the proposed parameter to the suggested value. According to another embodiment, by controlling the injection molding machine 110 through the interface module 210, the processor 250 resets at least one of the parameters of the molding condition of the injection molding machine 110 to a suggested value. can do.

Meanwhile, if a defect is not detected in operation 330, the electronic device 130 may return to operation 310. Through this, the electronic device 130 may repeatedly perform at least some of operations 310 to 340.

An operation method of the electronic device 130 according to various embodiments includes an operation of collecting process data for a plastic injection molding machine 110 in process under a predetermined molding condition, and defects from the process data based on a defect detection model. It may include an operation of detecting and determining a molding condition to be proposed to solve the defect based on the molding condition proposal model.

According to various embodiments, the operation of detecting the defectiveness from the process data includes an operation of detecting first possible defect data by removing good product data from the process data, and detecting the defect from the first possible defective data. It may include an operation to do.

According to various embodiments, the operation of detecting the defect from the first possible defective data is an operation of deriving a Gaussian distribution of the first possible defective data based on at least one attribute of the first possible defective data. , An operation of dividing the entire area of the Gaussian distribution into a first area and a second area by using a threshold value, an operation of detecting second possible defective data corresponding to the second area from the first possible defective data, and And detecting the failure from the second possible failure data.

According to various embodiments, the operation of dividing the entire area into the first area and the second area is an operation of dividing the entire area into the first area and the second area by using a predetermined candidate threshold, The operation of determining whether at least one of the ratio of the first region and the second region or the ratio of the entire region and the second region meets a predetermined condition, and if the condition is met, the candidate threshold is It may include an operation of determining the threshold value and maintaining the first region and the second region.

According to various embodiments, the operation of dividing the entire region into the first region and the second region may further include an operation of changing the candidate threshold if the condition is not met.

According to various embodiments, the molding condition may include a plurality of parameters.

According to various embodiments, the determining of a molding condition to be proposed to solve the defect may include analyzing the process data according to each of the parameters, and determining at least one parameter related to the defect from the parameters. An operation and an operation of suggesting adjustment of the determined parameter may be included.

According to various embodiments, the determining of a molding condition to be proposed to solve the defect may include determining a suggested value for the determined parameter, outputting the determined suggested value, or based on the determined suggested value. Thus, at least one of the operations of controlling the injection molding machine 110 may be further included.

The electronic device 130 according to various embodiments is connected to an interface module 210 for an interface with a plastic injection molding machine 110 in process under a predetermined molding condition, and the interface module 210, and the interface module Through 210, a processor 250 configured to collect process data on the injection molding machine 110 and detect a defect from the process data may be included.

According to various embodiments, the processor 250 is configured to detect the defect from the process data based on a defect detection model, and determine a molding condition to be proposed to solve the defect based on a molding condition proposal model. Can be.

According to various embodiments, the processor 250 removes good product data from the process data, detects first possible defective data, and based on at least one attribute of the first possible defective data, the first 1 A Gaussian distribution of possible defective data is derived, the entire area of the Gaussian distribution is divided into a first area and a second area by using a threshold, and a second area corresponding to the second area from the first possible defective data It may be configured to detect possible defective data and detect the defectiveness from the second possible defective data.

According to various embodiments, the molding condition may include a plurality of parameters.

According to various embodiments, the processor 250 analyzes the process data according to each of the parameters, determines at least one parameter related to the failure from the parameters, and proposes to adjust the determined parameter. Can be configured to

According to various embodiments, the electronic device 130 may efficiently detect and respond to a defect occurring in the plastic injection molding machine 110. That is, when a defect occurs in the injection molding machine 110, the electronic device 130 may detect the defect within a short time, and a countermeasure for the defect may be applied. Through this, the productivity of the injection molding machine 110 may be improved.

Various embodiments of the present document and terms used therein are not intended to limit the technology described in this document to a specific embodiment, and should be understood to include various modifications, equivalents, and/or substitutes of the corresponding embodiment. In connection with the description of the drawings, similar reference numerals may be used for similar elements. Singular expressions may include plural expressions unless the context clearly indicates otherwise. In this document, expressions such as "A or B", "at least one of A and/or B", "A, B or C" or "at least one of A, B and/or C" are all of the items listed together. It can include possible combinations. Expressions such as "first", "second", "first" or "second" can modify the corresponding elements regardless of their order or importance, and are only used to distinguish one element from another. The components are not limited. When it is mentioned that a certain (eg, first) component is “(functionally or communicatively) connected” or “connected” to another (eg, second) component, the certain component is It may be directly connected to the component, or may be connected through another component (eg, a third component).

The term "module" used in this document includes a unit composed of hardware, software, or firmware, and may be used interchangeably with terms such as, for example, logic, logic blocks, parts, or circuits. A module may be an integrally configured component or a minimum unit or a part of one or more functions. For example, the module may be configured as an application-specific integrated circuit (ASIC).

Various embodiments of the present document are implemented as software including one or more instructions stored in a storage medium (eg, memory 240) readable by a machine (eg, electronic device 130). Can be. For example, the processor of the device (eg, the processor 250) may call at least one of the one or more instructions stored from the storage medium and execute it. This enables the device to be operated to perform at least one function according to the at least one command invoked. The one or more instructions may include code generated by a compiler or code that can be executed by an interpreter. A storage medium that can be read by a device may be provided in the form of a non-transitory storage medium. Here,'non-transient' only means that the storage medium is a tangible device and does not contain a signal (e.g., electromagnetic waves). It does not distinguish between temporary storage cases.

According to various embodiments, each component of the described components may include a singular number or a plurality of entities. According to various embodiments, one or more of the above-described corresponding components may be omitted, or one or more other components may be added. Alternatively or additionally, a plurality of components may be integrated into one component. In this case, the integrated component may perform one or more functions of each component of the plurality of components in the same or similar to that performed by the corresponding component among the plurality of components prior to integration.

Claims (10)

In the method of operating an electronic device,
Collecting process data for a plastic injection molding machine in process under predetermined molding conditions;
Detecting a defect from the process data based on a defect detection model; And
And determining a molding condition to be proposed to solve the defect based on a molding condition proposal model.
The method of claim 1, wherein detecting the defect from the process data comprises:
Removing the good data from the process data to detect first possible defect data;
And detecting the failure from the first possible failure data.
The method of claim 2, wherein the operation of detecting the defect from the first possible defect data comprises:
Deriving a Gaussian distribution of the first possible defective data based on at least one attribute of the first possible defective data;
Dividing the entire area of the Gaussian distribution into a first area and a second area by using a threshold value;
Detecting second possible defective data corresponding to the second area from the first possible defective data; And
And detecting the failure from the second possible failure data.
The method of claim 3, wherein the operation of dividing the entire area into the first area and the second area,
Dividing the entire area into the first area and the second area by using a predetermined candidate threshold;
Determining whether at least one of the ratio of the first region and the second region or the ratio of the entire region and the second region satisfies a predetermined condition; And
And if the condition is met, determining the candidate threshold as the threshold and maintaining the first region and the second region.
The method of claim 4, wherein the operation of dividing the entire area into the first area and the second area,
If the condition is not met, changing the candidate threshold.
The method of claim 1,
The molding conditions include a number of parameters,
The operation of determining a molding condition to be proposed to solve the defect,
Analyzing the process data according to each of the parameters, and determining at least one parameter related to the failure from the parameters; And
And suggesting adjustment of the determined parameter.
The method of claim 6, wherein the determining of a molding condition to be proposed to solve the defect comprises:
Determining a suggested value for the determined parameter;
Outputting the determined suggested value; or
The method further comprising at least one of controlling the injection molding machine based on the determined suggested value.
In the electronic device,
An interface module for interfacing with a plastic injection molding machine in process under predetermined molding conditions; And
A processor connected to the interface module and configured to collect process data for the injection molding machine through the interface module and detect a defect from the process data,
The processor,
Detecting the defect from the process data based on the defect detection model,
An electronic device configured to determine a molding condition to be proposed to solve the defect based on a molding condition proposal model.
The method of claim 8, wherein the processor,
By removing good product data from the process data, the first possible defect data is detected,
Based on at least one attribute of the first possible defective data, a Gaussian distribution of the first possible defective data is derived,
Using a threshold, the entire area of the Gaussian distribution is divided into a first area and a second area,
Detecting second possible defective data corresponding to the second area from the first possible defective data,
An electronic device configured to detect the failure from the second possible failure data.
The method of claim 8,
The molding conditions include a number of parameters,
The processor,
Analyzing the process data according to each of the parameters, determining at least one parameter related to the failure from the parameters,
An electronic device configured to propose adjustment of the determined parameter.

Images