KR20250003102A - AI-based cold-pad batch dyeing system and dyeing process using the same - Google Patents

Abstract

The present invention relates to an AI-based cold-pad batch dyeing system capable of collecting sensing information in a cold pad batch process as big data and deriving optimal dyeing conditions through a learning process using machine learning, and a cold-pad batch dyeing method using the same.

Description

AI-based cold-pad batch dyeing system and cold-pad batch dyeing method using the same { AI-based cold-pad batch dyeing system and dyeing process using the same }

The present invention relates to an AI-based cold-pad batch dyeing system and a cold-pad batch dyeing method using the same. The present invention relates to an AI-based cold-pad batch dyeing system and a cold-pad batch dyeing method using the same, which collects cold-pad batch dyeing-related information into a big data server, learns the same through machine learning, and derives optimal dyeing conditions applicable to the field.

Cold-pad batch (CPB) dyeing is a method in which fabric is padded with a dye solution consisting of a mixture of reactive dye and an alkaline solution, aged for several hours, washed, and dried before dyeing. The characteristics of CPB dyeing are that it minimizes environmental pollution because it uses less water and generates less wastewater, dyes without friction on the surface of the fabric, so there is less pilling, and since it is a wide-range continuous dyeing method, there is less wrinkle generation, so it can completely solve the wrinkle defect of fabrics that contain spandex or have high density.

Recently, there have been various attempts to efficiently improve the dyeing process by incorporating digital technologies such as convolutional neural networks (CNNs) or deep learning. Specifically, Korean Patent Publication No. 10-2022-0165147 discloses a system for providing an assistant solution and a digital twin-based defect prediction module for collecting sensing information and work information reflecting the know-how of skilled workers in the textile dyeing process as big data, diagnosing work abnormalities through machine learning, and an AI-based optimal work support system for field workers that can apply this to the field to derive optimal dyeing work conditions. Korean Patent Publication No. 10-1995297 discloses a smart dyeing device that automatically operates to control the pH value, dyeing temperature, nozzle pressure, reel speed, cloth speed, and various valves of the dyeing machine to reduce color differences between dyeing lots and increase reproducibility, and a smart dyeing method using the same.

The dyeing method according to the above-mentioned conventional technology is mainly targeted at the immersion dyeing process, and thus fails to overcome the direct cost problems that the CPB dyeing industry is currently facing, such as a shortage of manpower due to rising labor costs and a lack of skilled workers, and an increase in costs due to increases in dye, preparation, and energy prices. In particular, the CPB dyeing method frequently causes defects due to color differences in the dyed material depending on the process conditions such as uneven pick-up rate by lot and temperature and tension during the process, and therefore continuous management through constant manpower is required during the operation of the equipment.

In the CPB dyeing industry, various improvements are being attempted using physical and chemical methods, but the reality is that a solution to resolve this problem has not yet been found.

In addition, due to the lack of a skilled workforce, unskilled and foreign workers are being used, which is resulting in frequent dyeing defects and the resulting increase in manufacturing costs, which is deteriorating corporate competitiveness.

The present invention has been made to solve the above-mentioned problems, and aims to provide an AI-based cold-pad batch dyeing system capable of collecting various sensing information generated during the CPB dyeing process among textile dyeing processes as big data and deriving optimal dyeing conditions through a learning process using machine learning, and a cold-pad batch dyeing method using the same.

The AI-based cold-pad batch dyeing method of the present invention for solving the above problem comprises: an information input step of inputting information on a material and structure of a dyeing material sample; a color matching step of analyzing a color of a dyeing material sample and transmitting the analyzed color information to a big data server; a dyeing solution preparation step of deriving a dyeing solution recipe from a big data server using the color information analyzed in the color matching step and preparing a dyeing solution; a transfer step of transferring the dyeing solution prepared in the dyeing solution preparation step to a padding machine; a padding step of immersing and compressing a fabric in the dyeing solution transferred to the padding machine to pad it; a batching step of rolling the padded fabric into a ROLL and batching it; a maturing step of fixing the dye picked up in the padding step to the fabric; a washing step of washing the dyeing material to remove unattached dye and impurities present on the surface; a drying step of drying the washed dyeing material; a fabric color measurement step of color-measuring the dried fabric to derive color information and transmitting the color information to a big data server; Including a transmission step for measuring data in real time in the above padding step and maturing, washing, and drying steps and transmitting the sensing information in real time to a big data server through a communication module; a process analysis step for comparing the sensing information with input variables and performing an analysis on the correlation between them; and a learning step for converting the information analyzed in the analysis step into big data and performing learning through machine learning; wherein the sensing information is the concentration and pickup rate of the dye solution in the padding step and the tension during padding as the control variables, and the temperature, humidity, and maturing time in the maturing step, and the tension and temperature applied to the fabric during washing and drying, and the input variables are the information on the material and structure of the dyeing material sample input in the information input step, the color information of the dyeing material sample analyzed in the color matching step, and the color information of the dried fabric derived in the fabric color measurement step, and it is preferable that the dyeing solution preparation step further includes a packaging step for mixing a dye and a preparation agent according to the dyeing solution recipe derived from the big data server to prepare the dyeing solution and packaging the dyeing material after the fabric color measurement step.

In addition, the AI-based cold-pad batch dyeing system according to the present invention comprises: a colorimeter that analyzes the color of a dye sample and transmits the color information to a big data server; a liquid dye solution preparation device that prepares a dye solution based on the analysis results of the color transmitted to the big data server; a big data server; a padding device that performs dyeing by immersing the dye sample in the dye solution and then pressing it; a maturing space that fixes the dye to the dye sample; And a communication module which transmits information sensed in the colorimeter, the liquid dye solution preparation device, the padding device, and the maturing space to a big data server, and transmits a control command produced by the big data server; wherein the big data server derives a dye solution recipe based on the color information measured by the colorimeter, and analyzes the correlation between the control variables and the input variables by comparing them, wherein the control variables are the concentration and pickup rate of the dye solution in the padding step and the tension during padding, the temperature and humidity and the maturing time in the maturing step, and the tension and temperature applied to the fabric during washing and drying, and the input variables are information on the material and structure of the dyeing material sample input in the information input step, the color information of the dyeing material sample analyzed in the color matching step, and the color information of the fabric after washing and drying derived in the fabric colorimeter step, and converts the correlation analyzed as described above into big data, and performs learning through machine learning, and the AI-based cold-pad batch dyeing system comprises: a washing unit which washes the dyeing material; The method may further include a drying unit for drying a dyed material after the washing process has been completed; and a packaging unit for packaging the dried dyed material; and in particular, the big data server preferably includes a model manager unit for determining, in a module form, an AI analysis model suitable for analyzing information transmitted from a colorimeter, a liquid dye solution preparation device, a padding device, and a maturing space, and for determining a customized AI analysis model suitable for the characteristics of the colorimeter, the liquid dye solution preparation device, the padding device, and the maturing space; a hybrid analysis unit for analyzing control variables and input variables based on the AI analysis model determined by the model manager unit to generate a control command; and a storage unit for storing the AI analysis model determined by the model manager unit and learning data learned by the hybrid analysis unit.

According to the present invention, sensing information for a CPB dyeing process is collected as big data, and the same is learned through machine learning, thereby providing a digital-based CPB dyeing method, and applying the same to a field has the effect of deriving optimal dyeing work conditions.

In addition, the present invention has the additional effect of reducing color differences between dyeing lots and increasing reproducibility by controlling dyeing solution, dyeing process conditions, and process conditions by deriving optimal dyeing operation conditions for CPB dyeing, thereby realizing improved productivity and reduced costs.

In addition, the present invention accelerates environmental friendliness by contributing to carbon reduction by digitalizing the CPB dyeing method, and in particular, has the effect of solving tailing, a defect in which a color difference occurs between the end and beginning of a dyed material during the CPB dyeing process.

Figure 1 is a flow chart of an AI-based cold-pad batch dyeing method according to the present invention.
Figure 2 is a conceptual diagram showing the correlation between sensing information in an AI-based cold-pad batch dyeing system according to the present invention.
Figure 3 is a schematic diagram of an AI-based cold-pad batch dyeing system according to the present invention.
Figure 4 is a block diagram of an AI-based cold-pad batch dyeing system according to the present invention.

It should be understood that the terms “include,” “have,” or “equip” in this application are intended to specify the presence of a feature, number, step, component, part, or a combination thereof described in the specification, and do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof. In addition, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by a person of ordinary skill in the art to which the present invention belongs. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning they have in the context of the relevant art, and shall not be interpreted in an ideal or excessively formal sense, unless explicitly defined in this application.

Hereinafter, the AI-based cold-pad batch dyeing system (100) according to the present invention and the cold-pad batch dyeing method using the same will be described in detail. FIG. 1 attached to the present invention is a flow chart for the AI-based cold-pad batch dyeing method according to the present invention, FIG. 2 is a conceptual diagram showing the correlation between sensing information in the AI-based cold-pad batch dyeing system (100) according to the present invention, FIG. 3 is a schematic diagram of the AI-based cold-pad batch dyeing system (100) according to the present invention, and FIG. 4 is a block diagram of the AI-based cold-pad batch dyeing system (100) according to the present invention.

The cold-pad batch dyeing method based on AI according to the present invention, as illustrated in FIG. 1, comprises: an information input step for inputting information on the material and structure of a dyeing material sample; a color matching step for analyzing the color of a dyeing material sample and transmitting the analyzed color information to a big data server; a dyeing solution preparation step for deriving a dyeing solution recipe from a big data server using the color information analyzed in the color matching step and preparing a dyeing solution; a transfer step for transferring the dyeing solution prepared in the dyeing solution preparation step to a padding machine; a padding step for padding by immersing and compressing a fabric in the dyeing solution transferred to the padding machine; a batching step for rolling the padded fabric into a ROLL and batching it; a maturing step for fixing the dye picked up in the padding step to the fabric; a washing step for washing the dyeing material to remove unattached dye and impurities present on the surface; a drying step for drying the washed dyeing material; a fabric color measurement step for colorimetrically measuring the dried fabric to derive color information and transmitting the color information to a big data server; It is preferable to include a transmission step for measuring data in real time in the above padding step and the maturing, washing, and drying steps and transmitting the sensing information in real time to a big data server via a communication module; a process analysis step for comparing the sensing information with input variables and performing an analysis on the correlation between them; and a learning step for converting the information analyzed in the above analysis step into big data and performing learning through machine learning.

Looking at this in detail, the above information input step is a step for inputting information on the material, structure, and density of the dye sample for which dyeing has been requested, and refers to a step for inputting information on the dye sample to be subjected to CPB dyeing. The above information input step can be input using a PC, etc.

The material of the dyeing material that is the target of CPB dyeing can be diverse. That is, the dyeing material can be a material that can be dyed with a reactive dye, and the main material is cellulose. The cellulose fibers refer to cotton fibers, flax fibers, rayon, tencel, and acetate fibers. Synthetic fibers and protein fibers that can be dyed with a reactive dye are also possible. In addition, the cellulose fibers and other material fibers such as nylon can be interwoven and interwoven.

In addition, in the information input step, information on the organization, density, yarn thickness, etc. of the dyed material sample can be input together. That is, the dyed material sample can be manufactured in a woven form such as a plain weave, a twill weave, or a satin weave, and a knitted form such as a single or double-sided weave, and the density of the fabric, that is, the number of warp and weft yarns or wales and course directions per inch of width and length, etc., is input. The yarn thickness is input in units such as count, denier, and tex.

The material, structure, and density of the dyeing material as described above are important factors in the CPB dyeing process, and they need to be clearly input in order to learn the color characteristics after dyeing.

After completing the information input step of inputting the material, structure, density, etc. of the dye sample as described above, the color matching step is performed. The color matching step refers to the step of analyzing the color of the dye sample using a colorimeter (Computer Color Matching System, hereinafter referred to as CCM) and deriving the analyzed color information and transmitting it to a big data server.

The above color matching step is a step of obtaining a color coordinate system by measuring the color of a dye sample using a CCM (10), and refers to a step of measuring color using a CCM (10) to derive color information of a dye sample. The color values measured using the CCM (10) can be expressed as color coordinate values.

Color coordinates are information for expressing colors in a color space, and were established as a standard by the International Commission on Illumination (CIE) in 1931. After several revisions, they were developed into the CIE 1976 L ^* u ^* v ^* colorimetric system.

In the above color matching step, color coordinates, i.e. color information, of the dye sample are derived by measuring color using CCM (10).

The above CCM (10) is a type of optical device and is widely used in industrial sites and research institutes for measuring and evaluating the colors of dyes, paints, foods, plastics, etc. The color information of the dye sample derived using the CCM (10) as described above can be transmitted to a big data server thereafter.

After the color matching step is performed as described above, a dyeing solution preparation step is performed to prepare a dyeing solution by deriving a dyeing solution recipe for dyeing the fabric in the same color as the dyeing material sample using the color information derived from the color matching step and transmitted to the big data server. According to the present invention, the dyeing solution preparation step can be performed using a liquid dyeing solution preparation device (computer Color Kitchen, hereinafter, CCK).

The above CCK (20) refers to a device that selects only the necessary dye among dyes of various colors based on the color information derived from the CCM (10) and transmitted to the big data server to prepare a dye solution. The CCM (10) and CCK (20) are connected to a network to exchange information, and each is connected to a big data server to transmit information, and based on this, a dye solution recipe can be derived and then the dye solution can be prepared.

The above CCK (20) may include a plurality of salt solution storage tanks and preparation storage tanks, valves for starting or stopping the supply of the salt solution and preparation, pumps mounted on the valves for transporting the salt solution and preparation, a flow meter for measuring the flow rates of the salt solution and preparation, and a control unit for controlling the operation of the salt solution storage tanks and preparation storage tanks and the valves and pumps based on a derived salt solution recipe.

The above CCK (20) is a device that automatically measures dye and preparations based on the dye requirement data from the colorimetric results of the CCM (10) to prepare a dye solution, and refers to a device that converts and measures the amount of dye solution dissolved at a certain dye concentration into the amount to be injected. The above dye solution preparation step can be performed using the above CCK (20).

After performing the above-mentioned dye solution preparation step, a transfer step is performed to transfer the dye solution prepared in the above-mentioned dye solution preparation step to a padding machine (30).

After the dye solution is transferred to the padding machine (30) through the above transfer step, a padding step is performed in which the fabric is immersed and pressed in the dye solution transferred to the padding machine (30) to perform padding.

The above padding step transfers the fabric to the prepared dye solution to impart the dye solution to the fabric and pressurizes the fabric so that the fabric has a consistent pick-up rate overall.

That is, the fabric is immersed in the dye solution prepared as described above to allow sufficient penetration and absorption of the dye solution, and then passes through a padding process to remove excess dye solution by passing it between rotating press rollers. The pressure of the press rollers is not limited, but is preferably 1 to 5 kg/cm2.

The above padding step can also be repeated several times through the above immersion and pressing process until the desired color is obtained.

In particular, according to the present invention, the temperature of the dye solution in the padding step is preferably 10 to 20°C, and is preferably managed so as not to exceed 30°C. That is, when the temperature of the dye solution is 10 to 20°C, the most stable color development is exhibited, and when the pick-up rate of the dye solution in the padding step is generally uniform, a result without uneven dyeing can be obtained.

In addition, the process conditions in the padding step, such as the concentration of the dye solution, the pickup rate, and the tension applied to the fabric during padding, can be sensed by various sensors, and the process conditions during padding sensed as described above can be transmitted in real time to a big data server.

According to the present invention, in order to prevent tailing and listing, which are dyeing defects that occur during CPB dyeing, it is necessary to maintain the tension applied to the fabric evenly during the padding step. That is, tailing, which is the most frequently occurring defect during the CPB dyeing process, refers to a phenomenon in which a difference occurs in dyeing concentration and/or color in the longitudinal direction of the dyed material. Therefore, in order to prevent the above-mentioned defect, tailing, it is very important to maintain the tension applied to the initial and final parts of the fabric evenly during padding.

After the padding step is performed as described above, a batching step is performed to roll the fabric and an ageing step is performed to fix the dye applied to the fabric in the padding step to the fabric.

The above batching step refers to the process of winding the fabric that has gone through the padding step into a roll in order to smoothly perform the subsequent maturing step.

In addition, the above maturing step refers to a step of fixing the dye to the fabric by placing the fabric padded with the dye solution in a maturing space (40) where temperature and humidity are controlled. The above maturing step may generally take 4 to 24 hours, and in order to prevent color differences between lots during the maturing step, it is necessary to maintain the temperature, humidity, and maturing time constant during the maturing step.

CPB dyeing has poor reproducibility, so it is necessary to control temperature, humidity, and maturation time during maturation to improve this.

The process conditions in the above maturation step, such as temperature, humidity, and maturation time, can be sensed by sensors, and the process conditions during maturation sensed as above can be transmitted in real time to a big data server.

After the maturing step as described above, a washing step is performed to wash the dyed material to remove any unattached dye and impurities present on the surface, and a drying step is performed to dry the washed dyed material.

The above washing step is a step for removing unfixed dye that is not fixed to the dyed object when the dyed object is immersed in the dye solution during the padding step, and the unfixed dye is removed using water at 90 to 100°C.

The above drying step refers to a step of removing the water used in the above washing step from the dyeing material or treating it with various processing agents and reacting it at a temperature of 130 to 220°C, and can be performed using a hot air dryer, etc.

CPB dyeing is a continuous dyeing process, and weight changes may occur depending on the tension applied to the fabric, and if the tension is high, the fabric may tear. Therefore, the tension applied to the fabric should be adjusted during the washing and drying stages to control tearing and weight of the fabric.

After the above drying step is performed, a fabric color measurement step is performed to measure the color of the fabric to which the dye is fixed, derive color information, and transmit it to a big data server. The fabric color measurement step can be performed using CCM (10), and the derived color information is transmitted to the big data server through a communication module.

As described above, various sensing information sensed during the padding step, maturing step, washing step, and drying step is transmitted to a big data server via a communication module.

The above transmission step transmits information sensed in each of the padding step, maturation step, washing step, and drying step to a big data server through various sensors installed in the padding machine (30) and the devices or spaces where the maturation step, washing step, and drying step are performed. The above transmission step can be performed through a communication module.

The information sensed as described above is transmitted to a big data server via a communication module as shown in Fig. 3, and the sensed information transmitted is input into the big data server as a control variable as shown in Fig. 2. That is, the control variables may include the concentration and pickup rate of the dye solution in the padding step, the tension during padding, the temperature and humidity and the maturing time in the maturing step, and the tension and temperature applied to the fabric during washing and drying.

After performing the above transmission step, a process analysis step is performed to compare the sensing information, that is, the control variables and input variables transmitted to the big data server, and analyze the correlation between them.

According to the present invention, the sensing information may be, as control variables, the concentration and pickup rate of the dye solution in the padding step, the tension during padding, the temperature and humidity and the maturing time in the maturing step, the tension and temperature applied to the fabric during washing and drying, and the like, and the input variables may be, information on the material and structure of the dyeing sample input in the information input step, the color information of the dyeing sample analyzed in the color matching step, and the color information of the washed and dried fabric derived in the fabric colorimetry step.

In order to perform the above process analysis step, the process is analyzed using the control variables and input variables transmitted and input to the big data server as shown in Fig. 2.

The control variables and input variables set as above are transmitted to the big data server through the communication module. As shown in Fig. 3, the big data server analyzes the correlation between the input variables and control variables transmitted as above through various analysis techniques, such as Pareto Analysis, Time Series Analysis, Multiple Linear Regression Analysis, Control Chart Analysis, or Predictive Analysis.

That is, the correlation between the process elements and the color of the dyed fabric is analyzed through analysis of the process conditions such as the concentration of the dye solution during the padding process, the pick-up rate of the dye solution, the tension during padding, the temperature and humidity during maturing, the maturing time, the tension and temperature applied to the fabric during washing and drying, the color information of the dyed sample, and the color information of the dyed fabric.

In the above process analysis step, the range where the color difference ( ΔE ) is 1.0 or less in the CPB dyeing process conditions, the color information of the dyed material sample, and the color information of the dyed fabric is analyzed through correlation analysis between the control variables and input variables, and data on the normal error range is established.

After performing the above process analysis step, the information analyzed in the above process analysis step is converted into big data, and a learning step is performed to perform learning through machine learning.

That is, the correlation between the control variables and input variables analyzed in the above process analysis step is transmitted to the big data server, and the big data server performs learning through big data and machine learning through an AI algorithm. These learning results can be implemented to enable work instructions or monitoring to field workers through a monitor.

FIG. 2 is a conceptual diagram showing the correlation between control variables and input variables in the AI-based cold pad batch dyeing system (100) according to the present invention. As shown in FIG. 2, the sensing information, which is the control variable, may be the concentration and pickup rate of the dye solution in the padding step, the tension during padding, the temperature and humidity and the maturing time in the maturing step, the tension and temperature applied to the fabric during washing and drying, etc., and the input variables may be the information on the material and structure of the dyeing material sample input in the information input step, the color information of the dyeing material sample analyzed in the color matching step, and the color information of the fabric on which washing and drying are completed derived in the fabric colorimetry step.

By deriving the correlation between the above control variables and input variables, it becomes possible to derive the dye recipe more accurately.

That is, the derivation of the correlation between the control variables and input variables can be performed through time series data analysis methods such as the Pearson Correlation Coefficient, Kendall Tau Rank Correlation Coefficient, Time Lagged Cross Correlation (TLCC), and Dynamic Time Warping (DTW).

The above big data server converts the information analyzed in the process analysis step into big data using the control variables and input variables set as above, and performs a learning step in which learning is performed through machine learning.

Through such a learning step, it is possible to derive a dyeing recipe based on the material, structure, and density of the dyeing material sample, and to predict the color of the dyed dyeing material based on the CPB dyeing process conditions. Based on the color information of the dyeing material predicted as described above, a big data server generates and transmits control commands for the padding machine (30) and the maturing space (40), respectively.

That is, a dyeing recipe can be derived through the correlation variables described above, and the color of a dyed dyeing material can be predicted. Based on the predicted result, work instructions can be given to field workers through the system (100) so that a CPB dyeing process can be performed under optimal work conditions.

The learning step for performing learning through the above machine learning can be performed through a machine learning learning processing module, and the machine learning learning processing module learns the sensing information transmitted from the padding machine (30) and the maturing space (40) and stores the learning results for which the learning process is completed so that the necessary process conditions can be set.

The learning results saved as above are used to derive a dyeing process according to the color information and process conditions of the dyeing material sample during the next dyeing process, and the dyeing process is performed by the dyeing process derived as above.

In the washing step for removing unfixed dye after the above dyeing process, the tension applied to the fabric is sensed and adjusted so that the tension does not exceed a certain value. In addition, the drying step for drying the fabric or treating it with a finishing agent can also adjust the tension by controlling the overfeed so that the weight required for the fabric can be implemented.

In addition, the packaging step after the drying step is a step of packaging the dyed material for sale, and can be performed using a packaging machine, etc.

The AI-based cold-pad batch dyeing method according to the present invention collects dyeing information and work information as big data in the cold pad batch dyeing process, learns this through machine learning, thereby providing digital-based dyeing process conditions, and applying this to the field to minimize the defect rate, improve productivity, reduce color differences between dyeing lots, and increase reproducibility. In addition, by digitalizing the CPB dyeing method as described above and contributing to carbon reduction, it accelerates environmental friendliness, and has the effect of solving tailing and listing, which are defects in which color differences occur between the end and the beginning of the dyed material during the CPB dyeing process.

Below, an AI-based cold-pad batch dyeing system (100) according to the present invention is described.

The AI-based cold-pad batch dyeing system (100) according to the present invention is equipped with a CCM (10) that analyzes the color of a dye sample to be dyed by CPB and transmits the color information to a big data server.

The above CCM (10) refers to equipment that measures the intensity of each wavelength of reflected light and calculates color coordinates using a CIE standard illuminant and standard observer.

The above CCM (10) uses a standard light source to analyze the color of a dye sample and transmits the analyzed color information to a big data server via a communication module.

In addition, the AI-based cold-pad batch dyeing system (100) is equipped with a CCK (20) that prepares a dye solution based on color information on a dye sample transmitted from the CCM (10) to a big data server.

The above CCK (20) refers to a device that selects only the necessary dyes among various colored dyes and prepares a dye solution based on the color information for the dye sample derived from the above CCM (10) and transmitted to the big data server.

The above CCK (20) may include a storage device having a plurality of dye storage tanks, a water level sensor in the tank, and a supply control panel; a metering and transfer device having a valve for starting or stopping the supply of dye, a pump for transferring the dye through a pipe by opening and closing the valve, and a flow meter for measuring the flow rate of the dye by a flow meter; a dyeing machine for treating the transferred dye on fabric; a program for detecting an electric signal from a dyeing machine, a processing machine, or a controller of these machines, processing it in a computer, and automatically supplying a desired amount of dye at a desired time; a control device for selecting a storage tank of the storage device according to a pre-input dyeing recipe, and receiving a signal provided from the dye level sensor and the supply control panel to control the storage device, and controlling the metering and transfer device by monitoring whether the valves and pumps are operating and the flow rate of the dye from the flow meter. In the above, the control device may include a computer capable of converting signals generated from the storage device, the metering and transfer device, and the controllers of various other dyeing machines into electric signals, and receiving and processing the converted electric signals.

The AI-based cold-pad batch dyeing system (100) according to the present invention is equipped with a big data server. Big data refers to a collection of a large amount of structured or unstructured data that is beyond the level of collection, storage, management, and analysis by the existing database management system (100), and a technology for extracting value from such data and analyzing the results.

Recently, research and development based on big data are actively being conducted. Big data refers to a large amount of data accumulated over a certain period of time, and a data set that is difficult to handle with ordinary software or computing systems (100). There is no regulation regarding the size of big data, but it is usually more than terabytes, and can range from exabytes to zettabytes. In addition, big data can be utilized as an Internet paradigm that collects and analyzes data in an Internet environment to find new value. In particular, a big data server is a server with a CPU with a high clock speed rather than multiple cores, and can efficiently process more data.

The big data server according to the present invention derives a dye recipe based on the color information measured by the colorimeter (10), compares control variables and input variables to analyze the correlation between them, converts the analyzed correlation into big data, and performs learning through machine learning.

According to the present invention, the control variables may be the concentration and pickup rate of the dye solution in the padding step, the tension during padding, the temperature and humidity and the maturing time in the maturing step, the tension and temperature applied to the fabric during washing and drying, and the input variables may be information on the material and structure of the dyeing sample input in the information input step, the color information of the dyeing sample analyzed in the color matching step, and the color information of the fabric on which drying and washing are completed derived in the fabric color measurement step.

In particular, a big data server according to the present invention may include a model manager unit, a hybrid analysis unit, and a storage unit.

The above model manager determines an AI analysis model suitable for analyzing information transmitted from the CCM (10), CCK (20), padding machine (30), and maturing space (40) in a module form, and determines a customized AI analysis model suited to the characteristics of the CCM (10), CCK (20), padding machine (30), and maturing space (40).

The above hybrid analysis unit analyzes control variables and input variables based on the AI analysis model determined as above to generate control commands, and stores learning data performed through machine learning.

Looking at this in detail, the model manager determines an AI analysis model suitable for analyzing information sensed and transmitted from the CCM (10), CCK (20), padding machine (30), and maturing space (40) in a module form, and determines and manages a customized AI analysis model that suits the characteristics of the CCM (10), CCK (20), padding machine (30), and maturing space (40).

The above model manager determines an AI analysis model suitable for analyzing the information in a module form based on at least one of the types and formats of information sensed and transmitted from the CCM (10), CCK (20), padding machine (30), and maturing space (40). Here, the suitable AI analysis model is an analysis model that is most optimized for analyzing the information, and refers to a model that performs calculations in a simple process to increase speed while accurately producing results.

Here, the AI analysis model can be any one of various artificial intelligence models such as artificial neural networks (ANN), machine learning, and deep learning.

The above model manager loads the determined AI analysis model from the storage unit and transfers the AI analysis model to the hybrid analysis unit.

The above model manager receives a control command generated through an AI analysis model transmitted to the hybrid analysis unit from the hybrid analysis unit, and controls the received control command to be transmitted to the CCM (10), CCK (20), padding machine (30), or maturing space (40) that transmitted the information, as shown in Fig. 4. At this time, the control command refers to a command that controls the operation of the CCM (10), CCK (20), padding machine (30), and maturing space (40).

The hybrid analysis unit receives the AI analysis model determined from the model manager unit, and analyzes the information transmitted from the CCM (10), CCK (20), padding machine (30), and maturing space (40), i.e., the control variables and input variables, using the received AI analysis model. At this time, the hybrid analysis unit analyzes the information based on the AI analysis model, but generates a control command for the device, which is an optimal result value, in a hybrid manner that complements the statistical analysis model. In addition, the hybrid analysis unit analyzes the information to generate learning data. At this time, it is preferable that the learning data is learning data that learns the control commands transmitted to the CCM (10), CCK (20), padding machine (30), and maturing space (40) by analyzing the control variables and input variables based on the AI analysis model determined by the model manager unit.

In addition, the storage unit stores the AI analysis model determined by the model manager unit in a module form and stores learning data learned by the hybrid analysis unit. The storage unit may include at least one storage medium among a flash memory type, a multimedia card micro type, a card type memory, a random access memory (RAM), and a static random access memory (SRAM).

In addition, the AI-based cold-pad batch dyeing system (100) according to the present invention may include a communication module that transmits information transmitted from the CCM (10), CCK (20), padding machine (30), maturing space (40), washing machine, and dryer to a big data server, and transmits a control command produced by the big data server.

The above communication module can be installed at any location in the CCM (10), CCK (20), padding machine (30), and maturing space (40), and the communication module can be a WiFi wireless communication module or a VLC (Visible Light Communication) communication module depending on the communication sensitivity and communication status.

In addition, the AI-based cold-pad batch dyeing system (100) according to the present invention may additionally include a packaging machine for packaging dried fabric.

The above washing unit is for removing unfixed dye from the CPB dyed fabric, and it is preferable that the washing unit washes in a washing bath at 90 to 100° C. In addition, it may be preferable to include an acid for neutralizing the alkali added in the soaping agent and dyeing padding step.

The above drying unit may be preferably a hot air dryer capable of drying or processing the CPB-dyed fabric washed in the washing unit at 130 to 220°C and capable of controlling the tension to control the weight of the fabric, and the packaging unit refers to a device that folds or rolls the CPB-dyed fabric and uses vinyl to package it.

The AI-based cold-pad batch dyeing system (100) configured as described above can reduce color differences between dyeing lots, lower the defect rate, and increase reproducibility. In addition, it can prevent tailing and listing phenomena, which are the most frequently occurring defects in the CPB dyeing process.

Although the present invention has been described with reference to the experimental examples illustrated in the drawings, these are merely exemplary, and those skilled in the art will understand that various modifications and equivalent other experimental examples are possible from the present invention. In addition, those skilled in the art to which the present invention pertains will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential characteristics of the present invention. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive, and the true technical protection scope of the present invention should be determined by the technical idea of the appended claims.

10 : Colorimeter
20: Liquid dye preparation device
30 : Padding machine
40: Aging space
100: AI-based cold-pad batch dyeing system

Claims (5)

Information entry step for entering information on the material and structure of the contaminated water sample;
A color matching step for analyzing the color of a dye sample and transmitting the analyzed color information to a big data server;
A dye solution preparation step for preparing a dye solution by deriving a dye solution recipe from a big data server using the color information analyzed in the color matching step above;
A transfer step for transferring the dye solution prepared in the above dye solution preparation step to a padding machine;
A padding step in which the fabric is immersed and pressed in a dye solution transferred to a padding machine to make padding;
Batching step of winding padded fabric into a ROLL;
A maturing step that fixes the dye picked up in the padding step to the fabric;
A washing step to wash the dye material to remove any unattached dye and impurities present on the surface;
A drying step for drying the washed dye;
Fabric color measurement step that measures the color of dried fabric to derive color information and transmits it to a big data server;
A transmission step that measures data in real time during the above padding step and maturing, washing, and drying steps and transmits the sensing information in real time to a big data server through a communication module;
A process analysis step for comparing the above sensing information and input variables and performing an analysis on the correlation between them; and
An AI-based cold-pad batch dyeing method, comprising a learning step of converting the information analyzed in the above analysis step into big data and performing learning through machine learning.
In claim 1,
The above-mentioned dye preparation step is,
AI-based cold-pad batch dyeing method characterized by preparing a dye solution by mixing dye and preparation according to a dye solution recipe derived from a big data server
In claim 1,
A washing step for washing the dyed material after the above maturing step to remove any unattached dye and impurities present on the surface;
A drying step for drying the washed dye; and
AI-based cold-pad batch dyeing method characterized by further comprising a packaging step of packaging the above dyeing agent;
A colorimeter that analyzes the color of a dye sample and transmits the color information to a big data server;
A liquid dye solution preparation device that prepares a dye solution based on the analysis results of the color transmitted to the above big data server;
big data server;
A padding machine that performs dyeing by immersing the dye material in a dye solution and then pressing it;
A maturing space for fixing the dye to the dyeing material; and
A communication module for transmitting information sensed from the colorimeter, the liquid dye solution preparation device, the padding device, the maturing space, the water purifier, and the dryer to a big data server, and transmitting a control command produced from the big data server; including,
The above big data server
Based on the color information measured by the above colorimeter, a dye recipe is derived,
By comparing the control variables and input variables, an analysis is conducted on the correlation between them.
An AI-based cold-pad batch dyeing system characterized by converting the correlations analyzed as above into big data and performing learning through machine learning.
In claim 4,
A washing unit for washing contaminated water;
A drying section for drying the washed dyed material; and
An AI-based cold-pad batch dyeing system further comprising a packaging machine for packaging dried dyeing material.