Intelligent color matching method and system for color masterbatch

Abstract

The application discloses a color masterbatch intelligent color matching method and system, and relates to the technical field of intelligent manufacturing of high polymer materials.The method comprises the following steps: obtaining target sample color data; constructing a historical formula database storing historical color matching records; establishing a color and formula mapping relationship model based on the historical formula database; inputting the target color data into the mapping relationship model to match candidate formulas and calculate color difference values; if the minimum color difference value is less than a preset threshold, directly outputting a recommended formula; otherwise, selecting the closest candidate formula, iteratively correcting according to the color difference direction, and generating predicted color data through virtual color matching simulation on the corrected formula, and repeatedly iterating until the predicted color difference is less than the preset threshold, and then outputting the recommended formula.The application combines historical data driving and iterative correction, effectively reduces the dependence on artificial experience, improves the color matching efficiency and accuracy, and realizes the continuous accumulation and reuse of color matching knowledge.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology for polymer materials, and in particular to an intelligent color matching method and system for color masterbatch. Background Technology

[0002] Color masterbatch is a granular coloring material formed by uniformly loading pigments or dyes onto a carrier resin. It is widely used in the coloring and processing of plastic products. In the production and application of color masterbatch, color matching (i.e., determining the masterbatch formula based on the target color) is a crucial step in determining product quality and customer satisfaction.

[0003] Currently, color matching and prototyping in the color masterbatch industry mainly relies on manual experience. Specifically, after the customer provides a color sample, the color matching personnel use their personal experience to determine the type and ratio of the required pigment and conduct color matching trials; or they compare the sample color with a standard color card, estimate the pigment formula, and then conduct a sample verification.

[0004] Manual color matching often requires multiple cycles of trial and error, adjustment, and more trial and error. Color matching personnel estimate the formula based on experience and create a sample. Then, they visually inspect or measure the color difference using instruments. Based on the direction of the color difference and experience, they adjust the formula and create another sample, repeating this process until the color difference meets the requirements. Each sample requires a complete set of steps, including weighing, mixing, extrusion, cooling, pelletizing, and sample injection molding, which is time-consuming and severely restricts the speed of new product development and customer response.

[0005] Furthermore, the varying levels of experience and judgment among color matching personnel lead to inconsistent quality and efficiency in color matching. Experienced color matchers may achieve the desired result after 2-3 adjustments, while novices may require 5-8 or even more. This over-reliance on personal experience makes it difficult to guarantee the stability and repeatability of the color matching process, and hinders the effective transfer and replication of experience.

[0006] Successful color matching experiences often remain only in the memory of the colorists, lacking systematic recording and management. When similar color matching needs arise again, it's necessary to start from scratch, resulting in a waste of knowledge and experience. Even if companies maintain historical formula records, they lack effective retrieval and matching methods, making it difficult to achieve rapid color matching.

[0007] Every invalid proof means a waste of raw materials, energy, manpower, and time. In manual color matching mode, the proportion of invalid proofs is relatively high, especially for complex colors or special effect colors, where the cost of color matching is even more prominent.

[0008] To address the aforementioned issues, some solutions have proposed color matching methods based on particle swarm optimization, but these are mainly aimed at the architectural coatings field and do not involve the full utilization of historical formula data and virtual sampling simulation in masterbatch color matching.

[0009] Therefore, developing an intelligent color matching method and system for color masterbatches that can fully utilize historical color matching data, reduce reliance on human experience, and improve color matching efficiency and accuracy is of great significance for promoting the transformation of the color masterbatch industry from "experience-driven" to "data-driven". Summary of the Invention

[0010] The present invention aims to provide a color master intelligent color matching method and system to solve the technical problems of low color matching efficiency, strong dependence on human experience, difficulty in reusing color matching knowledge, and high color matching cost in the prior art.

[0011] To achieve the above objectives, the present invention provides the following technical solution: a color masterbatch intelligent color matching method, comprising the following steps: S1. Obtain the color data of the target sample. The color data includes the spectral reflectance data of the target color and the corresponding chromaticity value. S2. Construct a historical formula database. The historical formula database stores multiple historical color matching records. Each historical color matching record contains the color masterbatch formula and its corresponding measured color data. S3. Establish a color-formula mapping model based on the historical formula database. The mapping model is used to output the corresponding candidate color masterbatch formula based on the input target color data. S4. Input the target color data obtained in step S1 into the mapping relationship model, and calculate the color difference value between the target color data and the measured color data of each historical color matching record in the historical formula database. If the minimum color difference value is less than the preset color difference threshold, the candidate formula corresponding to the minimum color difference value is directly used as the recommended formula and the process jumps to step S8. If the minimum color difference value is greater than or equal to the preset color difference threshold, the candidate formula corresponding to the historical color matching record with the minimum color difference value is selected as the preliminary candidate formula, and the process proceeds to step S5. S5. Iteratively correct the candidate formulas based on the direction of color difference; S6. Perform virtual color matching simulation on the corrected candidate formula to generate predicted color data for the virtual color matching sample; S7. Calculate the color difference between the predicted color data and the target color data, and determine whether the color difference is less than the preset color difference threshold. If the color difference value is greater than or equal to the preset color difference threshold, jump back to step S5 until the color difference value is less than the preset color difference threshold. If the color difference value is less than the preset color difference threshold, proceed to step S8; S8, Recommended formula for output color difference values ​​less than the preset color difference threshold.

[0012] In step S5, "iteratively correcting the candidate formula according to the color difference direction" can be understood as: iteratively correcting the initial candidate formula or the candidate formula after the nth correction according to the color difference direction. That is, when moving from step S4 to step S5, step S5 at this time corresponds to: iteratively correcting the preliminary candidate formula according to the color difference direction (to obtain the candidate formula after the first correction). Subsequently, when steps 6 and 7 are passed for the first time, and the color difference value is less than the preset color difference threshold in step 7, the process jumps to step S8 and outputs a recommended formula with a color difference value less than the preset color difference threshold. If the first time steps 6 and 7 are passed, and the color difference value in step 7 is greater than or equal to the preset color difference threshold, then jump back to step S5. At this time, step 5 changes to: iteratively correcting the candidate formula after the first (n=1) correction according to the color difference direction. After that, when the process goes through steps 6 and 7 for the second time, and the color difference value is less than the preset color difference threshold in step 7, the process jumps to step S8 and outputs the recommended formula with a color difference value less than the preset color difference threshold. If the second time passes through steps 6 and 7, and the color difference value in step 7 is greater than or equal to the preset color difference threshold, then jump back to step S5. At this time, step 5 changes to: iteratively correcting the candidate formula after the second correction (where n increases to 2) according to the color difference direction. Following this pattern, when it is confirmed in step S7 that the color difference value is less than the preset color difference threshold, step S8 is executed.

[0013] In a further embodiment of the present invention, the historical color matching record in step S2 further includes at least one of the following information: Types and grades of carrier resins; The processing parameters for color masterbatch include processing temperature, screw speed, and residence time; Application scenario information for color masterbatch, including target product type and usage environment requirements; Color matching operator identification and corresponding color matching success rate; The data includes the initial mixing ratio, the color difference between the initial mixing ratio sample and the sample, and the color difference adjustment process data for obtaining the final mixing ratio.

[0014] The data obtained for the color difference adjustment process of the final formulation are as follows: Based on the initial formulation and the obtained "color difference value between the initial formulation sample and the target sample", record the adjustment data for each component each time, the color data measured after each adjustment, and the color difference value with the target sample. For example, it can be used as a reference, learning and determination for the subsequent "adjustment range", or as one of the reference, learning and determination elements.

[0015] In a further embodiment of the present invention, obtaining the color data of the target sample in step S1 includes: Spectral reflectance data of the target sample were collected using a spectrophotometer; Convert spectral reflectance data into CIE Lab chromaticity values ​​or CIE LCh chromaticity values; Use the chromaticity value as the target color data input.

[0016] In a further embodiment of the present invention, the virtual color matching simulation in step S6 includes: Based on the proportions of each component in the candidate formulation and their respective optical properties, the predicted color data is calculated through color mixing calculations. The predicted color data includes the predicted spectral reflectance data and the corresponding predicted chromaticity values.

[0017] In a further embodiment of the present invention, the iterative correction in step S5 includes: Calculate the direction and magnitude of the deviation between the candidate formula color data and the target color data in each color channel; Based on the direction and amount of deviation, determine the color masterbatch component that needs adjustment and the direction of adjustment; The magnitude of a single adjustment is determined based on similar color matching records in the historical formula database. The candidate formulations are modified according to the direction and magnitude of the adjustment to generate the modified candidate formulations.

[0018] In this step, “calculating the deviation direction and amount of the candidate formulation color data and the target color data in each color channel” can be understood as: calculating the deviation direction and amount of the measured color data of the preliminary candidate formulation or the predicted color data of the candidate formulation after the nth correction and the target color data in each color channel.

[0019] A further aspect of the present invention includes... Step S9: After weighing, mixing and sampling the pigments according to the recommended formula output in Step 8; Obtain the measured color data of the sample; Calculate the measured color difference between the measured color data and the target color data; If the measured color difference value is less than the preset color difference threshold, the recommended formula and its corresponding measured color data will be stored as a new historical color matching record in the historical formula database.

[0020] In a further aspect of this invention, in step S9, if the measured color difference value is greater than or equal to a preset color difference threshold, the process returns to step S5 and continues until the final measured color difference value is less than the preset color difference threshold. That is, by returning to step S5, the recommended formula from step S8 is automatically fine-tuned based on the direction of the measured color difference, and the fine-tuned formula and its corresponding measured color data are stored in a historical formula database as learning data for more accurate subsequent adjustments. For example, it can be used as a reference, learning, and determination of the subsequent "adjustment range," or as one of the elements for reference, learning, and determination.

[0021] A further aspect of the present invention involves correcting the predicted chromaticity value using measured color data from a sample, thereby updating the predicted color data calculated in the color mixing calculation.

[0022] A further aspect of this invention involves storing the recommended formula and its corresponding measured color data as new historical color matching records in the historical formula database, and triggering incremental updates to the mapping relationship model.

[0023] In a further aspect of the present invention, in step S9, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database, When at least one formula modification is performed, the initial candidate formula, the color difference value between the initial candidate formula and the target sample, and the color difference adjustment process data for obtaining the final ratio are recorded simultaneously.

[0024] In a further aspect of the present invention, in step S9, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database, When the process returns to step S5 at least once from step S9, all the corrected data returned from step S9 to step S5 is recorded in the "Color Difference Adjustment Process Data for Obtaining the Final Ratio" of the color matching task for the same target sample.

[0025] On the other hand, the present invention also provides the following technical solution: a color masterbatch intelligent color matching system, comprising: The target color acquisition module is used to acquire the color data of the target sample. The color data includes the spectral reflectance data of the target sample and the corresponding chromaticity value. The historical formula database is used to store historical color matching records. Each historical color matching record contains the color masterbatch formula and its corresponding measured color data. The mapping relationship model module is used to establish a mapping relationship model between colors and formulas based on a historical formula database; The recipe matching module is used to input the target color data into the mapping relationship model and match it to obtain candidate recipes; The color difference comparison module is used to calculate the color difference value between the measured color data and the target color data, and to calculate the color difference value between the predicted color data and the target color data, and to determine whether the color difference value is less than the preset color difference threshold. The formula correction module is used to iteratively correct candidate formulas according to the direction of color difference when the color difference value is greater than or equal to the preset color difference threshold. The virtual color matching simulation module is used to perform virtual color matching simulation on candidate formulations and generate predicted color data for virtual color matching samples. The recipe output module is used to output recommended recipes with color difference values ​​less than a preset color difference threshold.

[0026] The native color masterbatch intelligent color matching system also executes the aforementioned native color masterbatch intelligent color matching method, specifically: S1. Obtain the color data of the target sample through the target color acquisition module. The color data includes the spectral reflectance data of the target color and the corresponding chromaticity value. S2. Construct a historical formula database through a historical formula database. The historical formula database stores multiple historical color matching records. Each historical color matching record contains a color masterbatch formula and its corresponding measured color data. S3. Based on the historical formula database, establish a color-formula mapping model in the mapping model module. The mapping model is used to output the corresponding candidate color masterbatch formula based on the input target color data. S4. Using the formula matching module and the color difference comparison module, input the target color data obtained in step S1 into the mapping relationship model, and calculate the color difference value between the target color data and the measured color data of each historical color matching record in the historical formula database. If the minimum color difference value is less than the preset color difference threshold, the candidate formula corresponding to the minimum color difference value is directly used as the recommended formula and the process jumps to step S8. If the minimum color difference value is greater than or equal to the preset color difference threshold, the candidate formula corresponding to the historical color matching record with the minimum color difference value is selected as the preliminary candidate formula, and the process proceeds to step S5. S5. The candidate formula is iteratively corrected through the formula correction module based on the direction of color difference; S6. Using the virtual color matching simulation module, perform virtual color matching simulation on the corrected candidate formula to generate predicted color data for the virtual color matching sample. S7. Calculate the color difference value between the predicted color data and the target color data through the color difference comparison module, and determine whether the color difference value is less than the preset color difference threshold. If the color difference value is greater than or equal to the preset color difference threshold, jump back to step S5 until the color difference value is less than the preset color difference threshold. If the color difference value is less than the preset color difference threshold, proceed to step S8; S8. Through the formula output module, output a recommended formula with a color difference value less than the preset color difference threshold.

[0027] In a further embodiment of the present invention, the system also includes a data update module.

[0028] A further aspect of the present invention is that the intelligent color matching method for color masterbatch executed by the color masterbatch intelligent color matching system further includes: Step S9: After weighing, mixing and sampling the pigments according to the recommended formula output in Step 8 (for example, the weighing, mixing and sampling of the pigments can be completed by the operator according to the recommended formula). The measured color data of the sample is obtained through the target color acquisition module; The color difference comparison module calculates the measured color difference between the measured color data and the target color data. If the measured color difference value is less than the preset color difference threshold, the recommended formula and its corresponding measured color data will be stored as new historical color matching records in the historical formula database through the data update module.

[0029] In a further embodiment of the present invention, if the measured color difference value is greater than or equal to the preset color difference threshold in step S9, the process returns to step S5 and continues until the final measured color difference value is less than the preset color difference threshold.

[0030] In a further embodiment of the present invention, in step S9, after the data update module stores the data, the predicted chromaticity value is corrected by the measured color data of the sample, and the predicted color data calculated by the color mixing calculation is updated.

[0031] In a further embodiment of the present invention, the data update module stores the recommended formula and its corresponding measured color data as new historical color matching records in the historical formula database, and then triggers an incremental update of the mapping relationship model.

[0032] In a further aspect of the present invention, in step S9, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database, When at least one formula modification is performed, the initial candidate formula, the color difference value between the initial candidate formula and the target sample, and the color difference adjustment process data for obtaining the final ratio are recorded simultaneously.

[0033] In a further aspect of the present invention, in step S9, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database, When the process returns to step S5 at least once from step S9, all the corrected data returned from step S9 to step S5 is recorded in the "Color Difference Adjustment Process Data for Obtaining the Final Ratio" of the color matching task for the same target sample.

[0034] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention; Figure 3 This is another method flowchart of the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] Please see Figure 1-3 In this embodiment, a smart color matching method for color masterbatch is provided, including the following steps: S1. Obtain the color data of the target sample. The color data includes the spectral reflectance data of the target color and the corresponding chromaticity value. S2. Construct a historical formula database. The historical formula database stores multiple historical color matching records. Each historical color matching record contains the color masterbatch formula and its corresponding measured color data. S3. Establish a color-formula mapping model based on the historical formula database. The mapping model is used to output the corresponding candidate color masterbatch formula based on the input target color data. S4. Input the target color data obtained in step S1 into the mapping relationship model, and calculate the color difference value between the target color data and the measured color data of each historical color matching record in the historical formula database. If the minimum color difference value is less than the preset color difference threshold, the candidate formula corresponding to the minimum color difference value is directly used as the recommended formula and the process jumps to step S8. If the minimum color difference value is greater than or equal to the preset color difference threshold, the candidate formula corresponding to the historical color matching record with the minimum color difference value is selected as the preliminary candidate formula, and the process proceeds to step S5. S5. Iteratively correct the candidate formulas based on the direction of color difference; S6. Perform virtual color matching simulation on the corrected candidate formula to generate predicted color data for the virtual color matching sample; S7. Calculate the color difference between the predicted color data and the target color data, and determine whether the color difference is less than the preset color difference threshold. If the color difference value is greater than or equal to the preset color difference threshold, jump back to step S5 until the color difference value is less than the preset color difference threshold. If the color difference value is less than the preset color difference threshold, proceed to step S8; S8, Recommended formula for output color difference values ​​less than the preset color difference threshold.

[0039] In step S5, "iteratively correcting the candidate formula according to the color difference direction" can be understood as: iteratively correcting the initial candidate formula or the candidate formula after the nth correction according to the color difference direction. That is, when moving from step S4 to step S5, step S5 at this time corresponds to: iteratively correcting the preliminary candidate formula according to the color difference direction (to obtain the candidate formula after the first correction). Subsequently, when steps 6 and 7 are passed for the first time, and the color difference value is less than the preset color difference threshold in step 7, the process jumps to step S8 and outputs a recommended formula with a color difference value less than the preset color difference threshold. If the first time steps 6 and 7 are passed, and the color difference value in step 7 is greater than or equal to the preset color difference threshold, then jump back to step S5. At this time, step 5 changes to: iteratively correcting the candidate formula after the first (n=1) correction according to the color difference direction. After that, when the process goes through steps 6 and 7 for the second time, and the color difference value is less than the preset color difference threshold in step 7, the process jumps to step S8 and outputs the recommended formula with a color difference value less than the preset color difference threshold. If the second time passes through steps 6 and 7, and the color difference value in step 7 is greater than or equal to the preset color difference threshold, then jump back to step S5. At this time, step 5 changes to: iteratively correcting the candidate formula after the second correction (where n increases to 2) according to the color difference direction. Following this pattern, when it is confirmed in step S7 that the color difference value is less than the preset color difference threshold, step S8 is executed.

[0040] In some embodiments, the historical color scheme record in step S2 further includes at least one of the following information: Types and grades of carrier resins; The processing parameters for color masterbatch include processing temperature, screw speed, and residence time; Application scenario information for color masterbatch, including target product type and usage environment requirements; Color matching operator identification and corresponding color matching success rate; The data includes the initial mixing ratio, the color difference between the initial mixing ratio sample and the sample, and the color difference adjustment process data for obtaining the final mixing ratio.

[0041] The data obtained for the color difference adjustment process of the final formulation are as follows: Based on the initial formulation and the obtained "color difference value between the initial formulation sample and the target sample", record the adjustment data for each component each time, the color data measured after each adjustment, and the color difference value with the target sample. For example, it can be used as a reference, learning and determination for the subsequent "adjustment range", or as one of the reference, learning and determination elements.

[0042] In some embodiments, obtaining the color data of the target sample in step S1 includes: Spectral reflectance data of the target sample were collected using a spectrophotometer; Convert spectral reflectance data into CIE Lab chromaticity values ​​or CIE LCh chromaticity values; Use the chromaticity value as the target color data input.

[0043] In some embodiments, the virtual color scheme simulation in step S6 includes: Based on the proportions of each component in the candidate formulation and their respective optical properties, the predicted color data is calculated through color mixing calculations. The predicted color data includes the predicted spectral reflectance data and the corresponding predicted chromaticity values.

[0044] In some embodiments, the iterative correction in step S5 includes: Calculate the direction and magnitude of the deviation between the candidate formula color data and the target color data in each color channel; Based on the direction and amount of deviation, determine the color masterbatch component that needs adjustment and the direction of adjustment; The magnitude of a single adjustment is determined based on similar color matching records in the historical formula database. The candidate formulations are modified according to the direction and magnitude of the adjustment to generate the modified candidate formulations.

[0045] In this step, “calculating the deviation direction and amount of the candidate formulation color data and the target color data in each color channel” can be understood as: calculating the deviation direction and amount of the measured color data of the preliminary candidate formulation or the predicted color data of the candidate formulation after the nth correction and the target color data in each color channel.

[0046] In some embodiments, it also includes Step S9: After weighing, mixing and sampling the pigments according to the recommended formula output in Step 8; Obtain the measured color data of the sample; Calculate the measured color difference between the measured color data and the target color data; If the measured color difference value is less than the preset color difference threshold, the recommended formula and its corresponding measured color data will be stored as a new historical color matching record in the historical formula database.

[0047] In a further aspect of this invention, in step S9, if the measured color difference value is greater than or equal to a preset color difference threshold, the process returns to step S5 and continues until the final measured color difference value is less than the preset color difference threshold. That is, by returning to step S5, the recommended formula from step S8 is automatically fine-tuned based on the direction of the measured color difference, and the fine-tuned formula and its corresponding measured color data are stored in a historical formula database as learning data for more accurate subsequent adjustments. For example, it can be used as a reference, learning, and determination of the subsequent "adjustment range," or as one of the elements for reference, learning, and determination.

[0048] A further aspect of the present invention involves correcting the predicted chromaticity value using measured color data from a sample, thereby updating the predicted color data calculated in the color mixing calculation.

[0049] A further aspect of this invention involves storing the recommended formula and its corresponding measured color data as new historical color matching records in the historical formula database, and triggering incremental updates to the mapping relationship model.

[0050] In some embodiments, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database in step S9, When at least one formula modification is performed, the initial candidate formula, the color difference value between the initial candidate formula and the target sample, and the color difference adjustment process data for obtaining the final ratio are recorded simultaneously.

[0051] In some embodiments, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database in step S9, When the process returns to step S5 at least once from step S9, all the corrected data returned from step S9 to step S5 is recorded in the "Color Difference Adjustment Process Data for Obtaining the Final Ratio" of the color matching task for the same target sample.

[0052] In some embodiments, in step S9, if the measured color difference value is greater than or equal to a preset color difference threshold, the recommended formula is automatically fine-tuned according to the measured color difference direction, and the fine-tuned formula and its corresponding measured color data are stored in the historical formula database.

[0053] Unlike the above method of "continuing to return to step S5 for automatic fine-tuning", in this solution, automatic fine-tuning includes: Based on the mapping relationship model, starting with the recommended formula in step S8 and aiming to reduce the measured color difference value (based on the measured color data of the sample, the target color data, and the measured color difference value between the two), local optimization iteration is performed to output the fine-tuned formula.

[0054] In this embodiment, a color master intelligent color matching system is also provided, including: Target color acquisition module 1 is used to acquire the color data of the target sample. The color data includes the spectral reflectance data of the target sample and the corresponding chromaticity value. Historical formula database 2 is used to store historical color matching records. Each historical color matching record contains the color masterbatch formula and its corresponding measured color data. Mapping relationship model module 3 is used to establish a mapping relationship model between color and formula based on the historical formula database; Formula matching module 4 is used to input target color data into the mapping relationship model and match to obtain candidate formulas; The color difference comparison module 5 is used to calculate the color difference value between the measured color data and the target color data, and to calculate the color difference value between the predicted color data and the target color data, and to determine whether the color difference value is less than the preset color difference threshold. Formula correction module 6 is used to iteratively correct candidate formulas according to the direction of color difference when the color difference value is greater than or equal to the preset color difference threshold. Virtual color matching simulation module 7 is used to perform virtual color matching simulation on candidate formulations and generate predicted color data of virtual color matching samples; Formula output module 8 is used to output recommended formulas with color difference values ​​less than a preset color difference threshold.

[0055] The native color masterbatch intelligent color matching system also executes the aforementioned native color masterbatch intelligent color matching method, specifically: S1. Obtain the color data of the target sample through the target color acquisition module. The color data includes the spectral reflectance data of the target color and the corresponding chromaticity value. S2. Construct a historical formula database through a historical formula database. The historical formula database stores multiple historical color matching records. Each historical color matching record contains a color masterbatch formula and its corresponding measured color data. S3. Based on the historical formula database, establish a color-formula mapping model in the mapping model module. The mapping model is used to output the corresponding candidate color masterbatch formula based on the input target color data. S4. Using the formula matching module and the color difference comparison module, input the target color data obtained in step S1 into the mapping relationship model, and calculate the color difference value between the target color data and the measured color data of each historical color matching record in the historical formula database. If the minimum color difference value is less than the preset color difference threshold, the candidate formula corresponding to the minimum color difference value is directly used as the recommended formula and the process jumps to step S8. If the minimum color difference value is greater than or equal to the preset color difference threshold, the candidate formula corresponding to the historical color matching record with the minimum color difference value is selected as the preliminary candidate formula, and the process proceeds to step S5. S5. The candidate formula is iteratively corrected through the formula correction module based on the direction of color difference; S6. Using the virtual color matching simulation module, perform virtual color matching simulation on the corrected candidate formula to generate predicted color data for the virtual color matching sample. S7. Calculate the color difference value between the predicted color data and the target color data through the color difference comparison module, and determine whether the color difference value is less than the preset color difference threshold. If the color difference value is greater than or equal to the preset color difference threshold, jump back to step S5 until the color difference value is less than the preset color difference threshold. If the color difference value is less than the preset color difference threshold, proceed to step S8; S8. Through the formula output module, output a recommended formula with a color difference value less than the preset color difference threshold.

[0056] In a further embodiment of the present invention, the system also includes a data update module 9.

[0057] A further aspect of the present invention is that the intelligent color matching method for color masterbatch executed by the color masterbatch intelligent color matching system further includes: Step S9: After weighing, mixing and sampling the pigments according to the recommended formula output in Step 8 (for example, the weighing, mixing and sampling of the pigments can be completed by the operator according to the recommended formula). The measured color data of the sample is obtained through the target color acquisition module; The color difference comparison module calculates the measured color difference between the measured color data and the target color data. If the measured color difference value is less than the preset color difference threshold, the recommended formula and its corresponding measured color data will be stored as new historical color matching records in the historical formula database through the data update module.

[0058] In a further embodiment of the present invention, if the measured color difference value is greater than or equal to the preset color difference threshold in step S9, the process returns to step S5 and continues until the final measured color difference value is less than the preset color difference threshold.

[0059] In a further embodiment of the present invention, in step S9, after the data update module stores the data, the predicted chromaticity value is corrected by the measured color data of the sample, and the predicted color data calculated by the color mixing calculation is updated.

[0060] In a further embodiment of the present invention, the data update module stores the recommended formula and its corresponding measured color data as new historical color matching records in the historical formula database, and then triggers an incremental update of the mapping relationship model.

[0061] In some embodiments, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database in step S9, When at least one formula modification is performed, the initial candidate formula, the color difference value between the initial candidate formula and the target sample, and the color difference adjustment process data for obtaining the final ratio are recorded simultaneously.

[0062] In some embodiments, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database in step S9, When the process returns to step S5 at least once from step S9, all the corrected data returned from step S9 to step S5 is recorded in the "Color Difference Adjustment Process Data for Obtaining the Final Ratio" of the color matching task for the same target sample.

[0063] This embodiment provides a method for intelligent color matching of color masterbatches. First, system configuration is performed, including: Target color acquisition module: Equipped with an X-Rite Ci7800 spectrophotometer, measuring wavelength range 360-750nm, measuring aperture 25.4mm, and capable of outputting CIE Lab colorimetric values ​​(D65 light source, 10° viewing angle). The spectrophotometer constitutes the system's signal input terminal; for example, the spectrophotometer's signal terminal is connected to the signal input terminal of the system's main control unit.

[0064] Historical Formula Database: Stores 8,726 historical color matching records accumulated over the past three years. Each record includes: formula number, color masterbatch components and ratios (including carrier resin type and brand, type and amount of each color powder), measured Lab value, processing parameters (temperature, screw speed), application scenario, color matching personnel, color matching date, color difference compliance status, etc.

[0065] Mapping Relationship Model Module: Constructs the mapping relationship model. The mapping relationship model employs a three-layer artificial neural network (3 nodes in the input layer corresponding to L, a, and b* values, 64 nodes in the hidden layer, and the output layer corresponding to the allocation ratio of each color powder group). It uses 80% of the data from the historical database for training and 20% for validation.

[0066] Virtual color matching simulation module: Based on the Kubelka-Munk theory, it has a built-in database of the absorption coefficient K and scattering coefficient S of each color powder in the carrier resin.

[0067] It is equipped with weighing equipment or weighing system, high-speed mixer or high-speed mixing system, twin-screw extruder or twin-screw extrusion system, and pelletizer or pelletizing system.

[0068] When choosing colors: Step S1: Taking the standard sample of the color of the car interior parts provided by the customer as an example, the operator places the sample on a spectrophotometer and measures it three times under the conditions of D65 light source and 10° observation angle, and takes the average value to obtain the target color Lab value: L=45.3, a=-2.1, b*=8.7.

[0069] Steps S2-S4: Input the target Lab value into the mapping model, and the model searches and matches it in the historical formula database. Calculate the color difference ΔE*00 between the target color and the measured color of each record in the database, and select the formula with the smallest color difference value as the initial candidate formula.

[0070] The record with the smallest color difference value is: Formula number P-9521, Lab value: L=46.8, a=-1.5, b*=10.2, color difference value ΔE*00=2.35, the formula is 65 parts PP carrier, 18 parts titanium dioxide, 2 parts carbon black, 8 parts iron oxide yellow, and 0.5 parts phthalocyanine blue.

[0071] Since ΔE*00=2.35 is greater than the preset color difference threshold (set to ΔE≤0.5 in this embodiment), it is necessary to enter the iterative correction process and virtual color matching simulation.

[0072] Step S5: Calculate the color difference between the predicted color and the target color: ΔL = +1.5 (brighter), Δa = +0.6 (redder), Δb* = +1.5 (yellower).

[0073] Based on the direction of color difference: the target color is darker, greener, and bluer, so it is necessary to increase the amount of carbon black (reduce L), reduce the amount of iron oxide yellow (reduce b), and slightly increase the amount of phthalocyanine blue (reduce a and b).

[0074] Based on the adjustment experience of similar color matching records in the historical database (searching records with similar color difference directions and statistically analyzing their adjustment range), the adjustment range for a single instance was determined as follows: increase carbon black by 0.3 parts, decrease iron oxide yellow by 0.5 parts, and increase phthalocyanine blue by 0.2 parts.

[0075] The revised formula is as follows: 65 parts PP carrier, 18 parts titanium dioxide, 2.3 parts carbon black, 7.5 parts iron oxide yellow, and 0.7 parts phthalocyanine blue.

[0076] Steps S6-S7: Input the modified candidate formula P-9521-1 into the virtual color matching simulation module, calculate the predicted color based on the Kubelka-Munk theory, and obtain the predicted Lab values ​​as follows: L=45.5, a=-1.9, b*=8.9.

[0077] The color difference value ΔE*00 is calculated to be 0.42, which is less than the preset color difference threshold (in this embodiment, ΔE≤0.5), and the iteration converges.

[0078] Step S8: The system outputs a recommended formula: 65 parts PP carrier, 18 parts titanium dioxide, 2.3 parts carbon black, 7.5 parts iron oxide yellow, and 0.7 parts phthalocyanine blue, with a predicted color difference value ΔE*00 = 0.42. For example, the recommended formula is output via a display screen. For example, the signal terminal of the display screen is connected to the signal output terminal of the system's main control unit.

[0079] Step S9: The staff weighs the color powder according to the formula, puts it into a high-speed mixer and mixes it for 10 minutes (800 rpm). Then, it is melt-blended and extruded through a twin-screw extruder at 190-220℃, cooled and pelletized to obtain a sample.

[0080] The sample was measured using a spectrophotometer. The measured Lab values ​​were: L=45.6, a=-2.0, b*=8.8. The measured color difference value ΔE00 was calculated to be 0.38, which is less than the preset color difference threshold of 0.5, indicating that the color matching is qualified.

[0081] The system automatically stores the recommended formula and its corresponding measured Lab values ​​and process parameters as new historical color matching records in the database, and triggers incremental updates to the mapping relationship model.

[0082] In some embodiments, the system further includes an instruction input unit 10, such as a touch input unit or a key input unit that is signal-connected to the main control unit of the system.

[0083] For example, the color matching task for the same sample can be started, ended, or the color matching task for different samples can be switched through the instruction input unit.

[0084] For example, the target color acquisition module can be controlled via the command input unit to start, stop, or resume sample detection. This is suitable for color matching tasks involving the same sample or different samples.

[0085] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for intelligent color matching of color masterbatches, characterized in that, Includes the following steps: S1. Obtain the color data of the target sample. The color data includes the spectral reflectance data of the target color and the corresponding chromaticity value. S2. Construct a historical formula database. The historical formula database stores multiple historical color matching records. Each historical color matching record contains the color masterbatch formula and its corresponding measured color data. S3. Establish a color-formula mapping model based on the historical formula database. The mapping model is used to output the corresponding candidate color masterbatch formula based on the input target color data. S4. Input the target color data obtained in step S1 into the mapping relationship model, and calculate the color difference value between the target color data and the measured color data of each historical color matching record in the historical formula database. If the minimum color difference value is less than the preset color difference threshold, the candidate formula corresponding to the minimum color difference value is directly used as the recommended formula and the process jumps to step S8. If the minimum color difference value is greater than or equal to the preset color difference threshold, the candidate formula corresponding to the historical color matching record with the minimum color difference value is selected as the preliminary candidate formula, and the process proceeds to step S5. S5. Iteratively correct the candidate formulas based on the direction of color difference; S6. Perform virtual color matching simulation on the corrected candidate formula to generate predicted color data for the virtual color matching sample; S7. Calculate the color difference between the predicted color data and the target color data, and determine whether the color difference is less than the preset color difference threshold. If the color difference value is greater than or equal to the preset color difference threshold, jump back to step S5 until the color difference value is less than the preset color difference threshold. If the color difference value is less than the preset color difference threshold, proceed to step S8; S8, Recommended formula for output color difference values ​​less than the preset color difference threshold.

2. The intelligent color matching method for color masterbatch according to claim 1, characterized in that, Step S2 also includes at least one of the following information in the historical color scheme record: Types and grades of carrier resins; The processing parameters for color masterbatch include processing temperature, screw speed, and residence time; Application scenario information for color masterbatch, including target product type and usage environment requirements; Color matching operator identification and corresponding color matching success rate; The data includes the initial mixing ratio, the color difference between the initial mixing ratio sample and the sample, and the color difference adjustment process data for obtaining the final mixing ratio.

3. The intelligent color matching method for color masterbatch according to claim 1, characterized in that, Step S1 involves obtaining the color data of the target sample, including: Spectral reflectance data of the target sample were collected using a spectrophotometer; Convert spectral reflectance data into CIE Lab chromaticity values ​​or CIE LCh chromaticity values; Use the chromaticity value as the target color data input.

4. The intelligent color matching method for color masterbatch according to claim 1, characterized in that, Step S6, virtual color matching simulation, includes: Based on the proportions of each component in the candidate formulation and their respective optical properties, the predicted color data is calculated through color mixing calculations. The predicted color data includes the predicted spectral reflectance data and the corresponding predicted chromaticity values.

5. The intelligent color matching method for color masterbatch according to claim 1, characterized in that, The iterative correction in step S5 includes: Calculate the direction and magnitude of the deviation between the candidate formula color data and the target color data in each color channel; Based on the direction and amount of deviation, determine the color masterbatch component that needs adjustment and the direction of adjustment; The magnitude of a single adjustment is determined based on similar color matching records in the historical formula database. The candidate formulations are modified according to the direction and magnitude of the adjustment to generate the modified candidate formulations.

6. The intelligent color matching method for color masterbatch according to claim 1, characterized in that, Also includes Step S9: After weighing, mixing and sampling the pigments according to the recommended formula output in Step 8; Obtain the measured color data of the sample; Calculate the measured color difference between the measured color data and the target color data; If the measured color difference value is less than the preset color difference threshold, the recommended formula and its corresponding measured color data will be stored as a new historical color matching record in the historical formula database.

7. The intelligent color matching method for color masterbatch according to claim 6, characterized in that, In step S9, if the measured color difference value is greater than or equal to the preset color difference threshold, return to step S5 and continue execution until the final measured color difference value is less than the preset color difference threshold.

8. The intelligent color matching method for color masterbatch according to claim 6, characterized in that, In step S9, when the recommended formula and its corresponding measured color data are stored as new historical color matching records in the historical formula database, the following also applies: Trigger incremental updates to the mapping relationship model; and / or When the formula is revised at least once, the initial candidate formula, the color difference value between the initial candidate formula and the target sample, and the color difference adjustment process data for obtaining the final ratio are recorded simultaneously. and / or When the formula has been modified at least once, the predicted chromaticity value is corrected by the measured color data of the sample, and the predicted color data calculated by the color mixing calculation is updated.

9. A color masterbatch intelligent color matching system, executing the color masterbatch intelligent color matching method according to any one of claims 1-5, characterized in that, include: The target color acquisition module is used to acquire the color data of the target sample. The color data includes the spectral reflectance data of the target sample and the corresponding chromaticity value. The historical formula database is used to store historical color matching records. Each historical color matching record contains the color masterbatch formula and its corresponding measured color data. The mapping relationship model module is used to establish a mapping relationship model between colors and formulas based on a historical formula database; The recipe matching module is used to input the target color data into the mapping relationship model and match it to obtain candidate recipes; The color difference comparison module is used to calculate the color difference value between the measured color data and the target color data, and to calculate the color difference value between the predicted color data and the target color data, and to determine whether the color difference value is less than the preset color difference threshold. The formula correction module is used to iteratively correct candidate formulas according to the direction of color difference when the color difference value is greater than or equal to the preset color difference threshold. The virtual color matching simulation module is used to perform virtual color matching simulation on candidate formulations and generate predicted color data for virtual color matching samples. The recipe output module is used to output recommended recipes with color difference values ​​less than a preset color difference threshold.

10. A color masterbatch intelligent color matching system, executing the color masterbatch intelligent color matching method according to any one of claims 6-8, characterized in that, include: The target color acquisition module is used to acquire the color data of the target sample. The color data includes the spectral reflectance data of the target sample and the corresponding chromaticity value. The historical formula database is used to store historical color matching records. Each historical color matching record contains the color masterbatch formula and its corresponding measured color data. The mapping relationship model module is used to establish a mapping relationship model between colors and formulas based on a historical formula database; The recipe matching module is used to input the target color data into the mapping relationship model and match it to obtain candidate recipes; The color difference comparison module is used to calculate the color difference value between the measured color data and the target color data, and to calculate the color difference value between the predicted color data and the target color data, and to determine whether the color difference value is less than the preset color difference threshold. The formula correction module is used to iteratively correct candidate formulas according to the direction of color difference when the color difference value is greater than or equal to the preset color difference threshold. The virtual color matching simulation module is used to perform virtual color matching simulation on candidate formulations and generate predicted color data for virtual color matching samples. The formula output module is used to output recommended formulas with color difference values ​​less than a preset color difference threshold. Data update module.

Images