KR102631118B1 - Method for calculating nanoparticle amount and calculating apparatus performing the same - Google Patents

Abstract

A nanoparticle quantity calculation method and calculation device that can improve the working environment considering the health and safety of users by more accurately calculating the amount of harmful nanoparticles at low cost during the process in which nanoparticles are generated are proposed. The nanoparticle amount calculation method according to the present invention collects learning data including total volatile organic compound learning data and nanoparticle weight learning data in the process in which volatile organic compounds and nanoparticles are generated, and infers the amount of generated nanoparticles. Learning a nanoparticle mass calculation model, which is an artificial intelligence model; and inputting total volatile organic compound data into the learned nanoparticle amount calculation model to infer the amount of generated nanoparticles.

Description

Nanoparticle mass calculation method and calculation device {METHOD FOR CALCULATING NANOPARTICLE AMOUNT AND CALCULATING APPARATUS PERFORMING THE SAME}

The present invention relates to a method and device for calculating the amount of nanoparticles. Specifically, it is possible to more accurately calculate the amount of harmful nanoparticles at low cost during a process in which nanoparticles are generated, thereby improving the working environment in consideration of the health and safety of users. It relates to a method and device for calculating the amount of nanoparticles.

The basic principle of 3D printers, which are used in a wide variety of fields such as industry, life, or medicine, is to create 3D objects by stacking thin 2D layers. 3D printers can form 3D prints using plastic, photocurable resin, or metal, and particles of various components are generated during the process, which can have an adverse effect on the user's health.

For example, when using FDM (fused deposition modeling) equipment in a 3D printer working environment, volatile organic compounds and fine nanoparticles may be generated due to the nature of the heat printing process. These volatile organic compounds or fine nanoparticles have a detrimental effect on the human body of workers, and sensing is difficult due to the size of the nanoparticles.

In the case of volatile organic compounds, TVOC (Total Volatile Organic Compounds) sensors have been commercialized, making it possible to measure the amount generated. However, in the case of fine nanoparticles, it is possible to use a very expensive analysis device to measure the amount of nanoparticles. There is a problem, and in the case of fine nanoparticle sensors, they are still only in the laboratory stage and have not been commercialized, so the development of the technology is requested.

The present invention was created to solve the above problems, and the purpose of the present invention is to accurately calculate the amount of harmful nanoparticles at low cost during the process in which nanoparticles are generated, thereby improving the working environment in consideration of the health and safety of users. The aim is to provide a method and device for calculating the amount of nanoparticles possible.

In order to achieve the above object, the nanoparticle amount calculation method according to an embodiment of the present invention collects learning data including total volatile organic compound learning data and nanoparticle amount learning data in a process in which volatile organic compounds and nanoparticles are generated. Thus, learning a nanoparticle amount calculation model, which is an artificial intelligence model that infers the amount of generated nanoparticles; and inputting total volatile organic compound data into the learned nanoparticle amount calculation model to infer the amount of generated nanoparticles.

The total volatile organic compound learning data may be data obtained from a total volatile organic compound sensor.

Nanoparticle mass learning data may be data obtained from a nanoparticle analysis device.

The learning data may further include individual volatile organic compound learning data.

The volatile organic compound may be any one of benzene, toluene, xylene, methyl ethyl ketone, acetone, isopropyl alcohol, and glycol ether.

The learning data may further include temperature learning data, which is learning data about temperature.

The learning data may further include humidity learning data, which is learning data about humidity.

The learning data may further include fine particle learning data, which is learning data about the amount of fine particles.

The process in which volatile organic compounds and nanoparticles are generated may be an FDM-type 3D printing process.

According to another aspect of the present invention, artificial intelligence collects learning data including total volatile organic compound learning data and nanoparticle weight learning data in a process in which volatile organic compounds and nanoparticles are generated, and infers the amount of nanoparticles generated. A learning unit that trains a model to calculate the amount of nanoparticles; and a calculation unit that inputs total volatile organic compound data into the learned nanoparticle weight calculation model and infers the generated nanoparticle amount. A nanoparticle weight calculation device including a is provided.

According to another aspect of the present invention, in the FDM method 3D printing process, learning including total volatile organic compound learning data and nanoparticle amount learning data based on the amount of volatile organic compounds and nanoparticles generated from the plastic filament according to process conditions. Collecting data and learning a nanoparticle amount calculation model, which is an artificial intelligence model that infers the amount of generated nanoparticles; And inputting the total volatile organic compound data into the learned nanoparticle amount calculation model to infer the amount of nanoparticles generated in the 3D printing process. A method for calculating the amount of nanoparticles in the FDM-type 3D printing process is provided. .

According to another aspect of the present invention, a 3D printing unit that produces a 3D output by heating, stacking, and curing plastic filaments; and Nano, an artificial intelligence model that analyzes volatile organic compounds and nanoparticles generated from heated plastic filaments, collects learning data including total volatile organic compound learning data and nanoparticle weight learning data, and infers the amount of nanoparticles generated. A 3D printing system capable of process monitoring that includes a nanoparticle mass calculation unit that learns a particle mass calculation model, inputs total volatile organic compound data into the learned nanoparticle mass calculation model, and infers the generated nanoparticle mass. provided.

According to the present invention, when performing a process that generates particles that have an adverse effect on the human body, such as volatile organic compounds and nanoparticles, an artificial intelligence algorithm is used to learn a model to calculate the amount of nanoparticles generated during the process in various work environments at low cost. Since it is possible to infer the amount of nanoparticles, it is possible to carry out the process in a safer working environment.

Figure 1 is a block diagram of a nanoparticle weight calculation device according to an embodiment of the present invention.
Figure 2 is a flowchart provided to explain a method for calculating the amount of nanoparticles according to another embodiment of the present invention.
Figure 3 is a schematic diagram of how the nanoparticle amount calculation method is performed in the FDM method 3D printing process according to another embodiment of the present invention, and Figure 4 is a nanoparticle amount calculation method in the FDM method 3D printing process according to another embodiment of the present invention. This is a schematic diagram of how the calculation method is performed.

Hereinafter, embodiments of the present invention will be described with reference to the attached drawings. However, the embodiments of the present invention can be modified into various other forms, and the scope of the present invention is not limited to the embodiments described below. Embodiments of the present invention are provided to more completely explain the present invention to those skilled in the art. In the attached drawings, there may be components shown with a specific pattern or with a predetermined thickness, but this is for convenience of explanation or distinction, so even if they have a specific pattern and a predetermined thickness, the present invention does not describe the features of the components shown. It is not limited to just that.

Figure 1 is a block diagram of a nanoparticle weight calculation device according to an embodiment of the present invention. The nanoparticle weight calculation device 100 according to this embodiment collects learning data including total volatile organic compound learning data and nanoparticle weight learning data in the process in which volatile organic compounds and nanoparticles are generated, and generates nanoparticles. A learning unit 120 that trains a nanoparticle mass calculation model, which is an artificial intelligence model for inferring the quantity; and a calculation unit 140 that inputs total volatile organic compound data into the learned nanoparticle amount calculation model and infers the generated nanoparticle amount.

The learning unit 120 is an artificial intelligence model that collects learning data including total volatile organic compound learning data and nanoparticle amount learning data in the process in which volatile organic compounds and nanoparticles are generated, and infers the amount of generated nanoparticles. Train the nanoparticle mass calculation model. The learning DB 110 is a database that stores data for learning a nanoparticle weight calculation model, and stores, for example, total volatile organic compound learning data or nanoparticle weight learning data.

In processes where volatile organic compounds and nanoparticles are generated, volatile organic compounds can be measured using a TVOC (Total Volatile Organic Compounds) sensor that senses total volatile organic compounds, and the amount of nanoparticles can be measured quantitatively through an analysis device rather than a sensor. It can be analyzed. Total volatile organic compound learning data is obtained from a total volatile organic compound sensor, and the amount of nanoparticles is analyzed through a nanoparticle analysis device to obtain nanoparticle mass learning data that learns a nanoparticle mass calculation model.

In the calculation unit 140, the volatile organic compounds to be measured and the total volatile organic compounds data detected by the TVOC sensor unit 130 of the process in which nanoparticles are generated are input to the nanoparticle amount calculation model learned in the learning unit 120. do. The calculation unit 140 infers the amount of nanoparticles generated in the process in which volatile organic compounds and nanoparticles are generated from the total volatile organic compound data using a nanoparticle amount calculation model.

Figure 2 is a flowchart provided to explain a method for calculating the amount of nanoparticles according to another embodiment of the present invention. The nanoparticle weight calculation method using the nanoparticle weight calculation device according to this embodiment is as follows.

First, learning data including total volatile organic compound learning data and nanoparticle weight learning data are collected in a process in which volatile organic compounds and nanoparticles are generated (S210). With the collected learning data, a nanoparticle amount calculation model, which is an artificial intelligence model that infers the amount of generated nanoparticles, is trained (S220). The nanoparticle mass calculation model can be learned by applying an artificial intelligence algorithm, especially a regression algorithm.

Once the learning of the nanoparticle mass calculation model is completed, the total volatile organic compounds data obtained from the process in which volatile organic compounds and nanoparticles are generated are input into the learned nanoparticle mass calculation model (S230), and the total volatile organic compounds data obtained in the process are generated. Obtain the nanoparticle amount by inferring the nanoparticle amount (S240).

Figure 3 is a schematic diagram of how the nanoparticle amount calculation method is performed in the FDM method 3D printing process according to another embodiment of the present invention, and Figure 4 is a nanoparticle amount calculation method in the FDM method 3D printing process according to another embodiment of the present invention. This is a schematic diagram of how the calculation method is performed.

The method for calculating the amount of nanoparticles according to the present invention is applicable to all equipment that generates volatile organic compounds and nanoparticles. In particular, volatile organic compounds and nanoparticles are generated in the FDM method 3D printing process. Referring to FIG. 3, in the 3D printing process of the FDM (fused deposition modeling) method, thermoplastic plastic in the form of a thin filament is melted in a nozzle, printed in the form of a thin film, and 3D output 320 is obtained by stacking one layer at a time. The nozzle melts the plastic filament 310 with high heat, and the extracted filament is hardened at room temperature. Since the plastic filament 310 is melted by heating to a high temperature, volatile organic compounds and nanoparticles are generated through this process.

Nanoparticles generated during the FDM 3D printing process are correlated with the amount of volatile organic compounds generated. As the amount of volatile organic compounds increases, the amount of nanoparticles generated also increases.

The method of learning a nanoparticle weight calculation model using total volatile organic compound learning data and nanoparticle weight learning data in the FDM-type 3D printing process is as follows. To measure nanoparticles, the plastic filament that needs to be measured is first selected. A deposition process is performed on this plastic filament under various nozzle temperature conditions. At this time, measurement data is collected using a total volatile organic compound sensor for each deposition process condition, and air is collected at each process condition and analyzed through analysis equipment. Measure the amount of nanoparticles. The deposition process conditions and the data obtained from the total volatile organic compound sensor are used as independent variables, and the amount of nanoparticles measured using analysis equipment is used as the dependent variable. Then, a regression algorithm, one of the supervised learning algorithms of artificial intelligence, is used. It can be learned.

Afterwards, the total volatile organic compound sensor data measured in real time through the TVOC sensor 330 in the FDM process is applied to the nanoparticle amount calculation model, and the amount of nanoparticles in the process is inferred.

The learning data may further include individual volatile organic compound learning data in addition to the total volatile organic compound learning data and nanoparticle weight learning data. Individual volatile organic compound learning data refers to learning data for each volatile organic compound, not total volatile organic compounds. Volatile organic compounds are one of Benzene, Toluene, Xylene, Methyl ethyl ketone, Acetone, Isopropyl alcohol, and Glycol ethers. The learning DB 110 may store individual volatile organic compound learning data in addition to the total volatile organic compound learning data.

In order to increase the accuracy of the inferred nanoparticle amount, individual volatile organic compound learning data regarding the most major volatile organic compounds that may be generated in the process are used along with the total volatile organic compound learning data as an independent variable in the nanoparticle mass calculation model. You can add Individual volatile organic compound learning data can be obtained from the specific hazardous sensor 350 (FIG. 4).

For example, in the FDM-type 3D printing process, if a lot of toluene is generated among the volatile organic compounds from the plastic filament, toluene learning data using a toluene sensor can be added, and if a lot of formaldehyde is generated, formaldehyde learning data using a formaldehyde sensor can be added. You can.

In the case of a total volatile organic compound sensor, the measurement amount may change depending on temperature and humidity. To reflect the influence of temperature and humidity, a temperature sensor, a humidity sensor, or a temperature and humidity sensor 340 can be added to the total volatile organic compound sensor and applied to nanoparticle measurement. Accordingly, the learning DB 110 can further store temperature learning data, which is learning data about the amount of fine particles, and can further store humidity learning data, which is learning data about humidity.

Additionally, the learning DB 110 may further include fine particle learning data, which is learning data regarding the amount of fine particles. Currently, sensors for measuring the amount of nanoparticles are not commercially available, but sensors that measure the amount of micron particles are commercially available. Therefore, if fine particle learning data, which is learning data about the fine particle amount, is acquired using the fine particle sensor 360 and used to learn a nanoparticle weight calculation model, the influence of relatively large particle size on the nanoparticle amount can be determined. By removing it, more accurate inference of nanoparticle amount is possible.

In a nanoparticle mass calculation model using an artificial intelligence algorithm, the accuracy of the algorithm can be improved by increasing the number of independent variables. In the FDM process, a TVOC sensor, specific hazardous gas sensor, temperature and humidity sensor, and fine particle sensor are simultaneously applied to monitor the process, and this monitoring data is used to apply an artificial intelligence algorithm to implement a highly accurate nanoparticle calculation device.

According to another aspect of the present invention, in the FDM method 3D printing process, learning including total volatile organic compound learning data and nanoparticle amount learning data based on the amount of volatile organic compounds and nanoparticles generated from the plastic filament according to process conditions. Collecting data and learning a nanoparticle amount calculation model, which is an artificial intelligence model that infers the amount of generated nanoparticles; And inputting the total volatile organic compound data into the learned nanoparticle amount calculation model to infer the amount of nanoparticles generated in the 3D printing process. A method for calculating the amount of nanoparticles in the FDM-type 3D printing process is provided. .

According to another aspect of the present invention, a 3D printing unit that produces a 3D output by heating, stacking, and curing plastic filaments; and Nano, an artificial intelligence model that analyzes volatile organic compounds and nanoparticles generated from heated plastic filaments, collects learning data including total volatile organic compound learning data and nanoparticle weight learning data, and infers the amount of nanoparticles generated. A 3D printing system capable of process monitoring that includes a nanoparticle mass calculation unit that learns a particle mass calculation model, inputs total volatile organic compound data into the learned nanoparticle mass calculation model, and infers the generated nanoparticle mass. provided.

Meanwhile, of course, the technical idea of the present invention can be applied to a computer-readable recording medium containing a computer program that performs the functions of the device and method according to this embodiment. Additionally, the technical ideas according to various embodiments of the present invention may be implemented in the form of computer-readable code recorded on a computer-readable recording medium. A computer-readable recording medium can be any data storage device that can be read by a computer and store data. For example, of course, computer-readable recording media can be ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical disk, hard disk drive, etc. Additionally, computer-readable codes or programs stored on a computer-readable recording medium may be transmitted through a network connected between computers.

In addition, although preferred embodiments of the present invention have been shown and described above, the present invention is not limited to the specific embodiments described above, and the technical field to which the invention pertains without departing from the gist of the present invention as claimed in the claims. Of course, various modifications can be made by those of ordinary skill in the art, and these modifications should not be understood individually from the technical idea or perspective of the present invention.

100: Nanoparticle mass calculation device
110: Learning DB
120: Learning Department
130: TVOC sensor unit
140: Calculation unit
310: plastic filament
320: 3D output
330: TVOC sensor
340: Temperature and humidity sensor
350: Specific harmful sensor
360: Fine particle sensor

Claims (12)

Nanoparticle amount calculation model, an artificial intelligence model that infers the amount of generated nanoparticles by collecting learning data including total volatile organic compound learning data and nanoparticle amount learning data in the process in which volatile organic compounds and nanoparticles are generated, learning step; and
A nanoparticle mass calculation method including the step of inputting total volatile organic compound data into the learned nanoparticle mass calculation model and inferring the generated nanoparticle mass. In claim 1,
A nanoparticle mass calculation method characterized in that the total volatile organic compound learning data is data obtained from a total volatile organic compound sensor. In claim 1,
Nanoparticle weight calculation method characterized in that the nanoparticle weight learning data is data obtained from a nanoparticle analysis device. In claim 1,
A nanoparticle mass calculation method characterized in that the learning data further includes individual volatile organic compound learning data. In claim 4,
A method for calculating nanoparticle weight, wherein the volatile organic compound is any one of benzene, toluene, xylene, methyl ethyl ketone, acetone, isopropyl alcohol, and glycol ether. In claim 1,
A nanoparticle mass calculation method characterized in that the learning data further includes temperature learning data, which is learning data related to temperature. In claim 1,
A nanoparticle mass calculation method characterized in that the learning data further includes humidity learning data, which is learning data related to humidity. In claim 1,
A method for calculating the amount of nanoparticles, characterized in that the learning data further includes fine particle learning data, which is learning data about the amount of fine particles. In claim 1,
A method for calculating the amount of nanoparticles, characterized in that the process in which volatile organic compounds and nanoparticles are generated is an FDM-type 3D printing process. Nanoparticle amount calculation model, an artificial intelligence model that infers the amount of generated nanoparticles by collecting learning data including total volatile organic compound learning data and nanoparticle amount learning data in the process in which volatile organic compounds and nanoparticles are generated, A learning department that teaches; and
A nanoparticle mass calculation device including a calculation unit that inputs total volatile organic compound data into the learned nanoparticle mass calculation model and infers the generated nanoparticle mass. In the FDM method 3D printing process, learning data including total volatile organic compound learning data and nanoparticle amount learning data are collected based on the amount of volatile organic compounds and nanoparticles generated from the plastic filament according to process conditions, and the nanoparticles generated Learning a nanoparticle mass calculation model, which is an artificial intelligence model that infers the quantity; and
A method for calculating the amount of nanoparticles in an FDM-type 3D printing process, including the step of inferring the amount of nanoparticles generated in the 3D printing process by inputting total volatile organic compound data into the learned nanoparticle amount calculation model. A 3D printing unit that produces 3D prints by heating, stacking, and curing plastic filaments; and
Nanoparticle, an artificial intelligence model that analyzes volatile organic compounds and nanoparticles generated from heated plastic filaments and collects learning data including total volatile organic compound learning data and nanoparticle weight learning data to infer the amount of nanoparticles generated. A 3D printing system capable of process monitoring that includes a nanoparticle mass calculation unit that learns a quantity calculation model, inputs total volatile organic compound data into the learned nanoparticle weight calculation model, and infers the generated nanoparticle quantity.