JP7634284B2 - Paint manufacturing method - Google Patents

Description

The present invention relates to a method for producing paint.

Adjustment of paint properties (one or more of color, gloss, and viscosity; hereinafter, these are sometimes collectively referred to as color, etc.) of paints is mainly done by adjusting paints for automobile repairs on-site and adjusting industrial paints mass-produced in paint factories. As automobiles age while being driven outdoors, they acquire colors, etc. that are unique to the vehicle body. Therefore, when carrying out repairs, it is necessary to calculate a paint mix that is suitable for such unique colors, etc. Furthermore, strictly speaking, since there are differences between lots, it is difficult to adjust to the desired color, etc. in one go, and adjustments are made multiple times to match the target color, etc. (such as the color of the normal parts of the incoming car).

On the other hand, industrial paints that are mass-produced in paint factories are adjusted to match each lot to a reference plate. Unlike automotive repair paints, which blur the boundary with the normal part, for example, when adjusting the paint properties of coil coating paint, painted steel plates painted with different paint lots are brought together during product assembly, resulting in a linear paint boundary, making slight differences in the paint properties visible. This requires extremely high adjustment precision. Also, because there are lot differences in the raw materials used to adjust the paint properties, it is difficult to adjust the paint to the desired properties in one go, and adjustments are made multiple times to match the target paint properties (the paint properties of the reference plate).

For example, one example of a lot difference in raw materials used to adjust paint properties is the lot difference in coloring strength of a colorant. Since the coloring strength of a lot when the primary data or the fluctuation response curve is created differs from the coloring strength of a lot when the paint is actually manufactured, it is very difficult to mix the color in one go exactly as calculated for paints that require extremely high adjustment accuracy. As a countermeasure, a method can be considered in which the coloring strength of each colorant lot is managed and the lot difference is taken into account in the calculation, but since the coloring strength fluctuates slightly even during the time from the production of the colorant to the production of the paint, it is necessary to measure the coloring strength immediately before production rather than for each lot, which results in enormous management costs. For this reason, the conventional method was to mix the color multiple times, repeatedly varying between shortfall and excess relative to the target paint properties, gradually reducing the fluctuation range, and stopping when it is within the target range.

However, in both the preparation of automotive repair paints and the preparation of industrial paints mass-produced in paint factories, if too much of the preparation ingredients are added, it is necessary to add more ingredients to restore the paint properties. For example, in the case of color, a colorant of the opposite hue (e.g., if too much black colorant is added, a white colorant) is added to restore the color, and in the case of gloss, if the gloss is too low, the original paint that does not contain a matting agent is added, and in the case of viscosity, if the viscosity becomes too low, the original high-viscosity paint is added, etc. In such cases, the amount of paint produced increases, often resulting in waste. In particular, in the production of industrial paints mass-produced in paint factories, several hundred kg to several tons (tons) may be produced per run, resulting in large waste of materials and resources. In addition, there were also disadvantages such as the need to transfer the paint to a larger manufacturing container to prevent it from overflowing, which required a lot of additional labor, so adjustments were made by relying on the skills of experienced workers.However, adjustments by experienced workers had the disadvantages of being costly and requiring a lot of labor, as the accuracy of adjustments varies depending on the level of skill, making it difficult to maintain quality.
In addition, a particular colorant having a maximum blending amount is often blended close to the target value when the paint is blended before adjustment, and even a small amount of a colorant having a minimum blending amount can result in a large excess being added, resulting in a noticeable problem of overproduction, in which an amount far exceeding the planned amount is produced in order to return to the original amount. For example, if the correct blending amount is 99 kg of white colorant and 1 kg of black colorant, even if each colorant is blended in the correct amount, if the coloring strength of the black colorant is 10% higher, 9.9 kg of white colorant must be added to return to the original amount, resulting in at least 9.9 kg of overproduction.

To address this issue of overproduction, Patent Literature 1 proposes applying a gap target value to a machine learning method.

Patent No. 7116266

However, machine learning methods have the problem of requiring large amounts of training data, and if there is insufficient training data for the gap target value, the prediction accuracy decreases, and there is a risk that the problem of overproduction will not be resolved, such as by mistakenly predicting the need to add excessive amounts of the adjustment mix.

The present invention aims to provide a method for producing paint that can prevent excessive amounts of additives from being added when adjusting the paint properties.

The gist and configuration of the present invention are as follows.
(1) A method for producing a paint having a desired paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint property is color,
A pre-adjustment paint property information acquisition process for acquiring information on the paint property before adjustment;
a gap target value setting process for setting a gap target value of a paint property;
A gap target value corresponding blend composition prediction step of predicting a blend composition of raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value by theoretical calculation using a computer,
In the gap target value setting step, an absolute value of a difference between the acquired paint property before adjustment and the gap target value is smaller than an absolute value of a difference between the acquired paint property before adjustment and a target paint property,
A method for producing a paint, characterized in that the gap target value setting step and the gap target value corresponding blend composition predicting step are repeated until a difference between the paint properties obtained by adding the predicted blend composition and the target paint properties becomes equal to or less than a predetermined reference value.

(2) A method for producing a paint having a desired paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint properties are one or more of color, gloss, and viscosity,
A pre-adjustment paint property information acquisition process for acquiring information on the paint property before adjustment;
a fluctuation response curve data acquisition step for obtaining information on the paint properties when only one of the parameters of the addition ratio of one or more colorants, the addition ratio of one or more gloss modifiers, and the addition ratio of one or more viscosity modifiers is varied based on the pre-variation paint composition, and obtaining fluctuation response curve data showing the relationship between the amount of variation of each of the parameters and the amount of variation of the paint property information;
a gap target value setting process for setting a gap target value of a paint property;
A gap target value corresponding blend composition prediction step, which predicts a suitable amount of fluctuation in the addition ratio of the raw material for adjusting the paint properties by using the gap target value and the fluctuation amount response curve data by a computer,
In the gap target value setting step, an absolute value of a difference between the acquired paint property before adjustment and the gap target value is smaller than an absolute value of a difference between the acquired paint property before adjustment and a target paint property,
A method for producing a paint, characterized in that the gap target value setting step and the gap target value corresponding blend composition predicting step are repeated until a difference between the paint properties obtained by adding the predicted blend composition and the target paint properties becomes equal to or less than a predetermined reference value.

(3) In the manufacturing method described in (1) or (2), in the gap target value setting step, the gap target value is set so that the smaller the absolute value of the difference between the acquired pre-adjustment paint properties and the target paint properties, the closer the gap target value becomes to the target paint properties, either continuously or stepwise.

(4) The paint properties consist of a plurality of properties,
The manufacturing method according to (1) or (2), wherein in the step of setting the target gap value, the target gap value is set independently for each of the properties.

(5) A method for producing a paint having a target paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint property is color,
A pre-adjustment paint property information acquisition process for acquiring information on the paint property before adjustment;
and a step of predicting a paint property adjustment composition with limitations, which is a paint property adjustment composition with limitations, by theoretical calculation under a condition in which each blend amount in the paint property adjustment composition of the raw materials is individually limited to a solution candidate generated from a finite numerical range,
The solution candidates include a case where the blend amount is 0,
A method for producing a paint, characterized in that the step of predicting a blend composition for adjusting limited paint properties is repeated until a difference between the paint properties obtained by adding the predicted blend composition and the target paint properties falls below a predetermined reference value.

(6) The manufacturing method described in (5) above, further comprising, in the step of predicting the composition of the raw materials for adjusting the paint properties with restrictions, a step of releasing some or all of the conditions and re-predicting the composition of the raw materials for adjusting the paint properties with theoretical calculations if the composition of the raw materials for adjusting the paint properties cannot be calculated by theoretical calculations in the step of predicting the composition of the raw materials for adjusting the paint properties with restrictions.

(7) In the gap target value corresponding mixture composition prediction step, the mixture composition of the raw material for adjusting the paint properties is predicted by theoretical calculation under the condition that each mixture amount in the mixture composition of the raw material for adjusting the paint properties is individually limited to a solution candidate generated from a finite numerical range,
The manufacturing method according to (1) above, wherein the candidate solutions include a case where the blending amount is 0.

(8) The manufacturing method described in (7) above, further comprising a step of releasing some or all of the conditions and re-predicting the composition of the raw materials for adjusting the paint properties by theoretical calculations in the gap target value corresponding composition prediction step, if the composition of the raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value cannot be calculated by theoretical calculations.

(9) In the gap target value corresponding mixture composition prediction step, the mixture amount of the raw material for adjusting the paint properties is individually limited to a candidate solution generated from a finite numerical range, and the mixture composition of the raw material for adjusting the paint properties is predicted by calculation based on the fluctuation response curve data,
The manufacturing method according to (2) above, wherein the candidate solutions include a case where the blending amount is 0.

(10) The manufacturing method described in (9) above, further comprising, in the gap target value corresponding composition prediction step, if the composition of the raw material for adjusting the paint properties suitable for obtaining the paint properties of the gap target value is not calculated based on the fluctuation response curve data, releasing some or all of the conditions and re-predicting the composition of the raw material for adjusting the paint properties by calculation based on the fluctuation response curve data.

(11) The method further includes a main colorant specifying step of specifying a colorant having the largest amount of colorant in the colorant composition of the paint before the adjustment,
The manufacturing method according to (7) or (9), wherein in the gap target value corresponding blend composition prediction process, a prediction is made for at least the main colorant using conditions limited to solution candidates generated from the finite numerical range.

(12) A method for producing a paint having a target paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint property is color,
A pre-adjustment paint property information acquisition process for acquiring information on the paint property before adjustment;
A target value corresponding blend composition prediction step of predicting a blend composition of raw materials for adjusting the paint properties suitable for obtaining the target paint properties by theoretical calculation using a computer;
A gap composition setting step of setting a gap composition of the raw material for adjusting the paint properties,
In the gap composition setting step, the set gap composition is smaller than the predicted target value corresponding composition,
A method for producing a paint, characterized in that the target value corresponding composition prediction step and the gap composition prediction step are repeated until a difference between the paint properties obtained by adding the predicted gap composition and the target paint properties becomes equal to or less than a predetermined reference value.

(13) A method for producing a paint having a target paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint properties are one or more of color, gloss, and viscosity,
a fluctuation response curve data acquisition step for obtaining information on the paint properties when only one of the parameters of the addition ratio of one or more colorants, the addition ratio of one or more gloss modifiers, and the addition ratio of one or more viscosity modifiers is varied based on the pre-variation paint composition, and obtaining fluctuation response curve data showing the relationship between the amount of variation of each of the parameters and the amount of variation of the paint property information;
A target value corresponding blend composition prediction step of predicting a suitable amount of fluctuation in the addition ratio of the raw material for adjusting the paint properties by using the target paint properties and the fluctuation amount response curve data by a computer;
A gap composition setting step of setting a gap composition of the raw material for adjusting the paint properties,
In the gap composition setting step, the set gap composition is smaller than the predicted target value corresponding composition,
A method for producing a paint, characterized in that the target value corresponding composition prediction step and the gap composition calculation step are repeated until a difference between the paint properties obtained by adding the gap composition and the target paint properties becomes equal to or less than a predetermined reference value.

(14) Prior to the gap blend composition calculation step, information on the paint properties before adjustment is acquired, and a pre-adjustment paint property information acquisition step is further included;
The manufacturing method according to (12) or (13), wherein in the gap composition calculation step, the gap composition is calculated so that the smaller the absolute value of the difference between the acquired paint properties before adjustment and the target paint properties is, the closer the gap composition becomes to the predicted composition continuously or stepwise.

(15) The manufacturing method described in (12) above, in which the gap composition calculation step calculates the gap composition for each colorant independently.

(16) In the manufacturing method described in (1), in the gap target value corresponding blend composition prediction step, when calculating the blend composition of the raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value, one of the methods of brute force search method, mathematical optimization method, and random search is used.

(17) The manufacturing method described in (2) above, in which in the gap target value corresponding blend composition prediction step, when predicting the suitable amount of variation in the addition ratio of the raw material for adjusting the paint properties and/or the suitable amount of variation in the manufacturing conditions, uses any one of a brute force search method, a mathematical optimization method, and a random search method.

(18) The manufacturing method described in (12) above, in which in the target value corresponding blend composition prediction step, one of the methods of brute force search method, mathematical optimization method, and random search is used to predict the blend composition of the raw materials for adjusting the paint properties suitable for obtaining the paint properties of the target value.

(19) The manufacturing method described in (13) above, in which in the target value corresponding blend composition prediction step, when predicting the suitable amount of variation in the addition ratio of the raw material for adjusting the paint properties and/or the suitable amount of variation in the manufacturing conditions, a brute force search method, a mathematical optimization method, or a random search method is used.

The present invention provides a method for producing paint that can prevent excessive amounts of additives from being added when adjusting the paint properties.

1 is a flowchart of a method for producing a paint according to a first embodiment of the present invention. 5 is a flowchart of a method for producing a paint according to a modified example of the first embodiment of the present invention. FIG. 13 is a diagram for explaining a fluctuation response curve data acquisition step. FIG. 1 shows the names of fluctuation response curves. FIG. 13 is a diagram illustrating an example of a predictive calculation. FIG. 13 is a diagram illustrating an example of a predictive calculation. 5 is a flowchart of a method for producing a paint according to a second embodiment of the present invention. 10 is a flowchart of a method for producing a paint according to a modified example of the second embodiment of the present invention. 10 is a flowchart of a method for producing a paint according to a third embodiment of the present invention.

The following describes in detail an embodiment of the present invention with reference to the drawings.

First Embodiment
The paint manufacturing method according to the first embodiment of the present invention is a method for manufacturing a paint having a target paint property by adding raw materials for adjusting the paint properties to an unadjusted paint.

Here, the term "paint before adjustment" refers to a paint in which, among the materials constituting the target paint composition, the raw materials having the same function as the raw materials for adjusting the paint properties are mixed in a predetermined amount of 0 to 100%, and the raw materials other than the raw materials having the same function as the raw materials for adjusting the paint properties are mixed in a predetermined amount of 100%, and the properties are approximately similar to the target properties. It may or may not contain raw materials for adjusting the paint properties. In addition, the amount of each material may be unknown. Here, the paint is, for example, a colorant prepared by dispersing a resin, a solvent, and one type of coloring pigment in an SG mill or the like, and multiple colorants of various colors are prepared, and in order to adjust to an arbitrary color, resin, solvent, and additives are added to one or more colorants and mixed. The paint is not particularly limited, but the form can be, for example, a water-based paint, a solvent-based paint, a powder paint, a solvent-free paint, etc. The applications include, for example, coil paints, general industrial paints, automotive paints, automotive repair paints, architectural paints, heavy-duty anticorrosive paints, and marine paints; the coating methods include spray coating, roller coating, brush coating, roll coating (including natural and reverse rotation), curtain flow coating, die coating, electrodeposition coating, powder coating, and electrostatic coating; the drying methods include baking drying, forced drying, natural drying, and ultraviolet curing; and the blending composition includes resin raw materials, pigments, design raw materials, solvents (including water), and additives. The resin raw materials can be acrylic resins, polyester resins, epoxy resins, alkyd resins, fluororesins, urethane resins, aminomelamine resins, isocyanate resins, blocked isocyanate resins, urethane-modified polyester resins, and other modified resins thereof; the pigments include inorganic pigments, organic pigments, coloring pigments, and extender pigments; the design raw materials include metallic, pearlescent materials, aggregates, silica, resin beads, wax, and the like; and the additives include viscosity modifiers, silicone-based additives, rust inhibitors, catalysts, and defoamers.

In addition, "raw materials for adjusting paint properties" refers to a general term for materials used for adjusting paint properties, such as colorants, gloss regulators, viscosity regulators, and design raw materials. The raw materials for adjusting paint properties include at least one of a plurality of types of colorants, one or more types of gloss regulators, and one or more types of viscosity regulators. It is preferable that the raw materials for adjusting paint properties include at least a plurality of types of colorants. The colorants include color pigments, and examples of the color pigments include black pigments, white pigments, yellow pigments, green pigments, red pigments, blue pigments, titanium oxide, and red iron oxide. Examples of gloss regulators include aggregates (sand, etc.), silica, reflective materials, and alumina as components. Examples of viscosity regulators include synthetic resin-based viscosity regulators, natural product-based viscosity regulators, and inorganic viscosity regulators. Furthermore, examples of synthetic resin-based viscosity regulators include polymer-type viscosity regulators and association-type viscosity regulators. Solvents such as water and organic solvents can also be used as viscosity regulators. Colorants, gloss adjusters, viscosity adjusters, and design ingredients may be used as they are, or may be dispersed in a solvent such as water or an organic solvent using a dispersant, etc.

In addition, the "color" can be, for example, the L ^* value, ^{a* value} , and b ^* value in the L*a ^* b ^* color space (JIS Z 8781-4 (2013)), the XYZ color system, the RGB color system, the Yxy color system, the Hunter Lab color system, the L ^* C ^* h ^* color system, the Munsell color system, and other color systems. The color can be measured using a known color measurement method. As an example, a spectrophotometer CM-M6 (manufactured by Konica Minolta) can be used to measure the L ^* value, a* value, and b ^* value measured by irradiating a light source from angles of 25°, 45°, and 75° when the light receiving unit perpendicular to the coating film formed on the test plate is set to 0 ^° . Alternatively, it can be measured using a spectrophotometer X-Rite MA68II (manufactured by X-Rite). The measurement angle can be appropriately adjusted depending on the purpose or the equipment used. Any other index can be used. Furthermore, any index can be used, such as reflection spectrum data, which is an index of color representing the reflection spectrum intensity at 5 nm intervals from 380 nm to 780 nm. The color of the resulting coating composition may be directly evaluated. As an example, the coating composition can be measured in a state in which it is placed in a quartz glass cell or a plastic cell using a spectrophotometer SE7700 (manufactured by Nippon Denshoku Industries Co., Ltd.). Here, adjusting any one of the properties of color, gloss, and viscosity may affect the other properties.

Figure 1 is a flow chart of a paint manufacturing method according to a first embodiment of the present invention. The example in Figure 1 involves theoretical calculations by computer. In the example in Figure 1, the paint property to be adjusted is color. As shown in Figure 1, in the first embodiment, first, primary data is acquired (step S101). The primary data is actual measurement data of the spectral reflectance at wavelengths of 400 to 700 nm for each paint composition for calculation. By acquiring such data, the relationship between the composition of the paint composition for calculation in any paint formulation and the change in color can be obtained, and the composition of the raw materials for adjusting the paint properties suitable for obtaining the paint property of the gap target value described below can be used to predict by theoretical calculations by computer.

Next, in the first embodiment, the paint before adjustment is charged (step S102). At this time, the main colorant, which is the colorant with the largest amount mixed, may be multiplied by a rate of 100% so that the other colorants are mainly adjusted.

Next, information on the paint properties (color in this example) before adjustment is obtained (step S103: process of obtaining information on paint properties before adjustment). As an example, the paint before adjustment prepared in step S102 is applied to a substrate, such as art paper or a metal plate, to prepare a sample plate, and a color difference meter or the like is used to measure the color difference between a standard plate (which has the target paint properties and is prepared in advance) and the sample plate.

Next, a gap target value of the paint property (color in this example) is set (step S104: gap target value setting step). Here, the difference in absolute value between the pre-adjustment paint property (color) acquired in step S103 and the gap target value is smaller than the difference in absolute value between the pre-adjustment paint property (color) acquired and the target paint property. For example, if the pre-adjustment paint property (color) is 25 and the target paint property is 30, the absolute value of the difference between the pre-adjustment paint property (color) and the target paint property is 5 (=30-25). Therefore, the value obtained by multiplying this by a rate of, for example, 0.5 to 0.9 (hereinafter also referred to as approach rate) is 3.5 (=5×0.7) if the approach rate is, for example, 0.7. In this case, the gap target value can be set to 28.5 (=25+3.5). In this case, the absolute value of the difference between the paint properties (color) before adjustment and the gap target value is 1.5 (= 30 - 28.5), which is smaller than the absolute value of the difference between the paint properties (color) before adjustment and the target paint properties, which is 5 (= 30 - 25).
In the same way, when managing the color difference, if the difference (color difference) between the paint properties before adjustment and the target paint properties is -5 and the approach rate is 0.7, the initial gap target value can be a color difference of -1.5. In this case, the absolute value of the color difference of the gap target value is "1.5", which is smaller than the absolute value of the color difference before multiplication by the approach rate, which is "5".

Next, the composition of the raw materials for adjusting the paint properties suitable for obtaining the paint properties of the target gap value is predicted by theoretical calculation using a computer (step S105: gap target value corresponding composition prediction process). As described above, the theoretical calculation can be performed using the primary data acquired in step S101, using known methods such as the Kubelka-Munk equation and mathematical optimization.

Next, from the predicted formulations, one formulation is selected in which the difference between the color difference of the gap target value and the calculated color difference is equal to or less than a predetermined reference value (step S106). When multiple candidates are predicted, the one in which the difference between the color difference of the gap target value and the calculated color difference is the smallest can be selected, although this is not particularly limited. In a calculation method in which only one formulation can be selected, the one in which the difference between the color difference of the gap target value and the calculated color difference is the smallest is calculated and used as the formulation of the raw material for adjusting the paint properties.
In the gap target value corresponding blend composition prediction process (step S105), when calculating the blend composition of raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, and random search.

Next, the paint adjusted to the selected composition is applied to a plate to prepare a sample plate, and the color difference between the standard plate and the sample plate is measured using a color difference meter or the like (step S107).

Next, it is determined whether the color difference measured in step S107 is equal to or less than a predetermined reference value (step S108). If the measured color difference exceeds the predetermined reference value, the process returns to step S104, and steps S104 to S108 are repeated. However, the target gap value set in step S104 is set based on the color difference obtained in step S107. For example, in the first step S104, when the paint properties at the time of preparation are 25, the target paint properties are 30, and the approach rate is 0.7, the initial target gap value is 28.5 as described above. As a result, if the measurement result in the first step S107 indicates that the paint properties of the paint adjusted by adding the blending composition selected in step S106 are 28.4, then in the second step S104, based on the current value of "28.4", the absolute value of the difference between the current paint properties and the target paint properties is 1.6 (=30-28.4), and therefore this can be multiplied by, for example, an approach rate of 0.9 to set the gap target value to 29.84 (=28.4+1.6×0.9).
In the case of a method of management using color difference, for example, when the difference (color difference) between the paint properties before adjustment and the target paint properties is -5 and the approach rate is 0.7, the initial gap target value is a color difference of -1.5 as described above. As a result, if the result of the measurement in the first step S107 shows that the difference in the paint adjusted by adding the blending composition selected in step S106 is -1.6, which is 0.1 lower than the gap target value (0.1 larger in absolute value) due to the influence of raw material lot differences, etc., in the second step S104, this can be multiplied by, for example, an approach rate of 0.9 to set the gap target value to a color difference of -0.16 (=-1.6+1.6×0.9).
On the other hand, if the measured color difference is equal to or less than the predetermined reference value, the color matching is completed (step S109).
The effects of the coating material manufacturing method according to the first embodiment will be described below.

In the paint manufacturing method according to the first embodiment of the present invention, information on the paint properties (color in this example) before adjustment is acquired (step S103: before-adjustment paint property information acquisition step), and a gap target value of the paint property (color in this example) is set (step S104: gap target value setting step). In the gap target value setting step, the absolute value of the difference between the acquired before-adjustment paint property (color in this example) and the gap target value is smaller than the absolute value of the difference between the acquired before-adjustment paint property and the target paint property. In the first embodiment, a blending composition of raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value is predicted by theoretical calculation using a computer (step S105: gap target value corresponding blending composition prediction step).
In this way, in the first embodiment, a gap target value that is lower than the original target value (the difference from the current paint properties is smaller) is set, and the composition of the raw materials for adjusting the paint properties that corresponds to that gap target value is predicted. From the predicted compositions, one composition is selected such that the difference between the color difference of the gap target value and the calculated color difference is equal to or less than a predetermined reference value (in this example, the color difference is 0.1 or less) to adjust the paint properties (step S106), thereby preventing the amount of material used for adjustment from becoming excessive when adjusting the paint properties of the paint.
Then, the paint adjusted by adding the selected compounding composition is applied to a plate to prepare a sample plate, and the color difference between the standard plate and the sample plate is measured using a color difference meter or the like (step S107). It is then determined whether the color difference is below a predetermined reference value (step S108). As a result, steps S104 to S108 are repeated as necessary to ultimately produce a paint having the desired paint properties.
Here, for color, the blend amount and the amount of color change are not linear, and L ^* , a ^* , and b ^* behave separately. In contrast, in the first embodiment, the gap target value is set by multiplying the approach rate based on the paint properties (color), so that it is possible to more accurately prevent the blend amount for adjustment from becoming excessive, compared to, for example, a method of multiplying a predicted blend amount by the approach rate.
For example, the colorant addition amount and the color variation are not in a linear relationship, and the more colorant is added, the smaller the color change becomes. In other words, when the predicted blending amount is multiplied by 90% of the approach rate, the color difference will be closer than when the color difference is multiplied by 90% of the approach rate. For example, when the L ^* value of a white colorant is adjusted to a black paint at 90% of the color difference approach rate to a color difference of -0.5 to -0.05, when the approach rate of the amount of addition is 90%, the color difference will be -0.5 to -0.02, which is too close. When there is no colorant lot difference, the number of color adjustments may be reduced by getting closer, but when there is a colorant lot difference and its coloring power is 20% stronger, the color difference will be -0.5 to 0.04 (=-0.5 + 0.5 x 0.9 x 1.2) at 90% of the color difference approach rate, which is only a slight over, and the color adjustment is completed if the control width is ±0.05. However, if the approach rate of the amount added is 90%, it will go beyond the opposite side, from -0.5 to 0.076 (=-0.5 + (0.5-0.02) x 1.2), and it will be necessary to add a lot of black colorant to restore the color, which will lead to the overproduction mentioned above.
It may be possible to anticipate this and lower the approach rate of the amount of addition, but this would require checking the fluctuation curve of the amount of addition and the color, which would be cumbersome and prone to errors.
On the other hand, the color difference approach rate is efficient and easy to set because it is faithful to the intended approach to the target property.

<Modification of the first embodiment>
Next, a modified example using a fluctuation response curve will be described. In this modified example, any one or more of color, gloss, and viscosity are targeted for adjustment.
Here, the color is as described above, and the gloss is not particularly limited, but gloss can be used as an index. The gloss can be measured using a known gloss measurement method, and as an example, the 60° gloss of a coating film formed on a test plate can be measured using a specular gloss meter (VG 7000 gloss meter (manufactured by Nippon Denshoku Industries Co., Ltd.)) in accordance with JIS K 5600-4-7 (specular gloss).
The viscosity can be measured using a known viscosity measuring method, for example, in accordance with JIS K 5600-2-2 (flow cup method).
The "pre-variation coating composition" may consist of only a resin and a solvent, or may contain one or more coating property adjusting materials (such as colorants, gloss adjusters, and viscosity adjusters) in addition to the resin and solvent.

Figure 2 is a flow chart of a method for producing a paint according to a modified example of the first embodiment of the present invention. As shown in Figure 2, in this modified example, first, based on the pre-variation paint composition, information on the paint properties when only one of the parameters of the addition ratio of one or more types of colorants, the addition ratio of one or more types of gloss modifiers, and the addition ratio of one or more types of viscosity modifiers is variously changed is obtained in advance for each parameter, and variation response curve data showing the relationship between the variation of each parameter and the variation of the paint property information is obtained (step S201: variation response curve data acquisition process).

In the example shown in FIG. 3, in variation 1, the gloss regulator addition ratio is varied from 0 to 0.03%, and data on the relationship between the variation in color (variation in L ^* value (ML1), variation in a ^* value (Ma1), variation in b ^* value (Mb1)), variation in gloss (Mg1), and variation in viscosity (Mv1) at that time are obtained, as well as data on the relationship between the variation in parameters and the variation in paint property information, which is similarly obtained for other variations 2 to 6. Although one pre-variation paint composition may be prepared, in order to accommodate various adjustments, it is preferable to prepare multiple pre-variation paint compositions and obtain data on the relationship between the variation in parameters and the variation in paint property information for the properties of each pre-variation paint composition. For example, multiple pre-variation paint compositions are preferably prepared in increments of 5 for color (L ^* value, a ^* value, b ^* value), and more preferably in increments of 1.

In the method of this modification, fluctuation response curve data is obtained from the data obtained as described above. FIG. 4 shows an example of the names of the fluctuation response curves.
The term "obtain" as used herein includes not only the case of obtaining by calculation using a calculation unit (processor) of a computer, but also the case of receiving the above-mentioned fluctuation response curve data using a communication unit of a computer, or reading the above-mentioned fluctuation response curve data recorded in a memory using a reading unit, etc. Note that even when the addition rate of the gloss regulator is changed as only one parameter, for example, if the target paint properties include not only the target gloss but also the target color and target viscosity, it is preferable to obtain not only the gloss but also the fluctuation amount of the color and viscosity. This is because a change in the addition rate of the gloss regulator can affect not only the gloss but also the color and viscosity.
Furthermore, even when adjusting gloss, not only the gloss modifier but also the colorant, viscosity modifier, and manufacturing conditions can affect the gloss, so fine adjustments may be used, and it is also possible to obtain a fluctuation response curve when the colorant and viscosity modifier are varied, as shown in Figure 4.

Next, in this modified example, the unadjusted paint is prepared (step S202).

Next, information on the paint properties before adjustment (in this example, one or more of color, gloss, and viscosity) is obtained (step S203: pre-adjustment paint property information acquisition process). As an example, the pre-adjustment paint prepared in step S202 is applied to a substrate, such as art paper or a metal plate, to prepare a sample plate, and a gloss meter or the like is used to measure the difference in gloss between a standard plate (which has the target paint properties and is prepared in advance) and the sample plate. In the case of viscosity, the paint can be directly measured with a viscometer or the like to obtain the difference in viscosity from a standard product. In the case of color, the color difference between the standard plate and the sample plate can be measured using a color difference meter or the like, as in step S103.

Next, a gap target value of the paint property (in this example, one or more of color, gloss, and viscosity) is set (step S204: gap target value setting step). Here, the absolute value of the difference between the pre-adjustment paint property acquired in step S203 and the gap target value is smaller than the absolute value of the difference between the pre-adjustment paint property acquired and the target paint property. For example, if the pre-adjustment paint property is 25 and the target paint property is 30, the difference between the pre-adjustment paint property and the target paint property is 5 (=30-25), and the value obtained by multiplying this by an approach rate of, for example, 0.5 to 0.9 is 3.5 (=5×0.7) if the approach rate is 0.7. In this case, the gap target value can be 28.5 (=25+3.5).
Also, similarly to the explanation regarding step S104, it is possible to manage the paint based on differences in paint properties such as color differences.

Next, the computer uses the target gap value and the fluctuation response curve data to predict the fluctuation amount of the addition ratio of the raw material for adjusting the suitable paint properties (step S205: target gap value corresponding blend composition prediction step).
In addition, when two types of additives are added simultaneously, the amount of change in the paint properties when the first type of additive is added is calculated using a change response curve, and when it is assumed that the paint changes from "paint properties before adjustment" to "paint properties 1," for example, the amount of change when the second type of additive is added is calculated as the amount of change from "paint properties before adjustment," which can be calculated as "change amount 2." In the case of simultaneous addition in this case, the amount of change after addition can be calculated as "paint properties before adjustment" + "change amount 1" + "change amount 2."
5 and 6 are diagrams for explaining the prediction calculation.
A specific calculation method in the second method will be described. FIG. 5 shows the calculation results obtained by substituting the fluctuation amount of each coating film property into the fluctuation amount response curve when the gloss modifier is added at 0.01%, 0.02%, and 0.03%. Similarly, FIG. 6 shows the calculation results in the case of a viscosity modifier. The fluctuation amount of the g value when the gloss modifier is added at 0.01% and the viscosity modifier is added at 0.02% is Mg (0.01) and Sg (0.02), respectively. In addition, the fluctuation amount when both are added simultaneously can be calculated by adding up the fluctuation amount of the g value when each colorant is added alone [Mg (0.01) + Sg (0.02)]. Although not shown in particular, when the addition ratio of the colorant is variously changed, the same method as shown in FIG. 5 and FIG. 6 can be used.
It is also preferable to create a fluctuation response curve as an approximation curve (e.g., by the least squares method) based on the fluctuation response curve data, and in the predictive calculation, use the fluctuation response curve to calculate the amount of fluctuation in one or more of the color, gloss, and viscosity when at least one of the addition ratios of multiple types of colorants, the addition ratios of one or more types of gloss modifiers, and the addition ratios of one or more types of viscosity modifiers is varied.

Next, from among the predicted mixture compositions, one mixture composition is selected in which the difference between the gap target value property and the calculated property is equal to or less than a predetermined reference value (step S206). When multiple candidates are predicted, the one in which the difference between the gap target value property and the calculated property is the smallest can be selected, although this is not particularly limited. In a calculation method in which only one mixture composition can be selected, the one in which the difference between the gap target value color difference and the calculated color difference is the smallest is calculated, and this is used as the mixture composition of the raw material for adjusting the paint properties.
In the gap target value corresponding blend composition prediction process (step S205), when calculating the blend composition of raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, and random search.

Next, the paint adjusted by adding the selected compounding composition is applied to a plate to prepare a sample plate, and the difference in gloss between the standard plate and the sample plate is measured using a gloss meter or the like (step S207). In the case of viscosity, liquid measurement can be performed as described above. In the case of color, the color difference between the standard plate and the sample plate can be measured using a color difference meter or the like as described above. Color can also be measured liquid.

Next, it is determined whether the property difference measured in step S207 is equal to or smaller than a predetermined reference value (step S208). If the property difference measured exceeds the predetermined reference value, the process returns to step S204, and steps S204 to S208 are repeated. However, the target gap value set in step S204 is set based on the property difference obtained in step S207. For example, in the first step S204, when the paint property at the time of preparation is 25, the target paint property is 30, and the approach rate is 0.7, the initial target gap value is 28.5 as described above. As a result, if the measurement result in the first step S207 indicates that the paint properties of the paint adjusted by adding the blending composition selected in step S206 are 28.4, then in the second step S204, based on the current value of "28.4", the absolute value of the difference between the current paint properties and the target paint properties is 1.6 (=30-28.4), and therefore this can be multiplied by, for example, an approach rate of 0.9 to set the gap target value to 29.84 (=28.4+1.6×0.9).
On the other hand, if the measured property difference is equal to or smaller than the predetermined reference value, the adjustment is completed (step S109).
The effects of the coating material manufacturing method according to the modified example of the first embodiment will be described below.

In the paint manufacturing method according to the modified example of the first embodiment of the present invention, information on the paint properties when only one of the parameters of the addition ratio of one or more colorants, the addition ratio of one or more gloss adjusters, and the addition ratio of one or more viscosity adjusters is changed based on the paint composition before the change, and the information on the paint properties is obtained in advance for each parameter, and the variation response curve data showing the relationship between the variation of each parameter and the variation of the paint property information is obtained (step S201: variation response curve data acquisition step), and the gap target value of the paint properties is set (step S204: gap target value setting step). In the gap target value setting step, the absolute value of the difference between the acquired paint properties before the adjustment and the gap target value is smaller than the absolute value of the difference between the acquired paint properties before the adjustment and the target paint properties. In this modified example, the computer predicts the variation of the addition ratio of the raw material for adjusting the paint properties (step S205: gap target value corresponding blending composition prediction step).
In this manner, in this modified example, a gap target value lower than the original target value (small difference from the current paint properties) is set, the composition of the raw materials for adjusting the paint properties corresponding to that gap target value is predicted, and from the predicted compositions, one composition is selected such that the difference between the properties of the gap target value and the calculated properties is equal to or less than a predetermined reference value (step S206), and the paint properties are adjusted. Therefore, when adjusting the paint properties of the paint, it is possible to prevent the amount of material used for adjustment from becoming excessive.
For example, in the case of adjusting gloss, a paint adjusted by adding a selected compounding composition is applied to a plate to prepare a sample plate, and the difference in gloss between the standard plate and the sample plate is measured using a gloss meter or the like (step S207), and then it is determined whether the difference in gloss is equal to or less than a predetermined reference value (step S208). As a result, steps S204 to S208 are repeated as necessary to produce a paint having the final target paint properties. In the case of viscosity, liquid measurement can be performed as described above. In the case of color, the color difference between the standard plate and the sample plate can be measured using a color difference meter or the like. Color can also be liquid measured.

The color adjustment shown in FIG. 1 and the gloss and/or viscosity adjustment shown in FIG. 2 to FIG. 6 can be performed simultaneously. In other words, the color can be adjusted by the method illustrated in FIG. 1, and the gloss and/or viscosity can be adjusted by the method illustrated in FIG. 2 to FIG. 6, and these can be performed simultaneously to adjust the color, gloss and/or viscosity. That is, as a preliminary preparation for prediction, steps S101 and S201 are performed, then the paint before adjustment is prepared (steps S102, S202), and then each paint property (color, gloss and/or viscosity) before adjustment is measured and obtained (steps S103, S203). Next, the gap target value for each paint property (color, gloss and/or viscosity) is set (steps S104, S204). Then, for gloss and/or viscosity, the amount of variation in the appropriate ratio of raw materials for adjusting the paint properties is predicted by calculation using the above-mentioned variation response curve data, and the color variation when the raw materials for adjusting the paint properties necessary for adjusting the gloss and/or viscosity are added is also calculated (step S205), and for color, the appropriate blend composition is predicted by the above-mentioned theoretical calculation after taking into account the color variation due to the gloss and/or viscosity adjustment (step S105). Then, each property is measured, and depending on whether it is below a predetermined reference value, repeat or end the adjustment (steps S104 to S109, steps S204 to S209).

Here, in the first embodiment and its modified examples, in the gap target value setting process (steps S104 and S204), it is preferable to set the gap target value so that the smaller the absolute value of the difference between the acquired pre-adjustment paint properties and the target paint properties, the closer the gap target value becomes to the target paint properties, either continuously or stepwise.
As described above, the first embodiment and its modified examples include repeating steps S104 to S108 (steps S204 to S208), and because the paint properties of the paint gradually approach the target paint properties, the amount of blended composition added for adjustment gradually decreases, and therefore the excess amount in the case of adding too much of the blend also becomes smaller, so the risk gradually decreases.
Therefore, by repeating steps S104 to S108 (steps S204 to S208), the smaller the absolute value of the difference between the acquired pre-adjustment paint properties and the target paint properties, the closer the gap target value is to the target paint properties, either continuously or stepwise (for example, by adopting an approach rate closer to "1"), thereby reducing the number of repetitions required to produce paint as much as possible, and improving production efficiency without compromising the effect of suppressing overproduction.

In the first embodiment and its modified examples, the paint properties are composed of a plurality of properties, and in the gap target value setting step (step S104, step S204), it is preferable to set a gap target value independently for each property. Specifically, the paint properties include color, gloss and/or viscosity, and the color further includes L ^* , a ^* , and b ^* . If a common gap target value is used, a problem of overproduction will occur if any property exceeds the target value due to overfilling. In addition, if the common gap target value is set excessively lower than the target paint property, productivity will decrease. In contrast, if a gap target value is set independently for each property, overproduction can be further suppressed while maintaining production efficiency.

For example, when ΔL ^* is more than 1, the approach rate can be set to 50%, when ΔL ^* is 0.5 to 1.0, the approach rate can be set to 70%, and when ΔL ^* is 0.5 or less, the approach rate can be set to 100%, etc. This can also be applied independently to each property (Δa ^* , Δb ^* , Δg, Δv). In the case of gloss, for example, when Δg is more than 1 in the case of a gloss control width of ±0.3, the approach rate can be set to 50%, and when Δg is 1 or less, the approach rate can be set to 100%. In the case of viscosity, for example, when Δv is 30 seconds or more, the approach rate can be set to 50%, and when Δv is less than 30 seconds, the approach rate can be set to 100%.

Here, in the gap target value corresponding blend composition prediction process (step S104) in the first embodiment, when calculating the blend composition of raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, and random search.
Similarly, in the gap target value corresponding compounding composition prediction process (step S204) in the modified example of the first embodiment, when predicting the suitable amount of variation in the addition ratio of raw materials for adjusting the paint properties and/or the suitable amount of variation in the manufacturing conditions, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, random search, and Bayesian optimization.

The brute force search method is a method of calculating the variation values of color, gloss, and viscosity by theoretical calculation and/or calculation using variation response curve data for all patterns of the extracted variation amounts (for example, 100,000 to 1 billion) by specifying a specified range and a numerical interval of a parameter, and finding various variation amounts when the difference from the target value is equal to or smaller than a certain value in a brute force search method. Here, it is synonymous with grid search.
The mathematical optimization method is a technique for finding x1, x2, x3, and x4 as an optimization problem of functions of the amount of variation x1 in the addition ratio of colorant 1, the amount of variation x2 in the addition ratio of colorant 2, the amount of variation x3 in the addition ratio of gloss modifier, and the amount of variation x4 in the addition ratio of viscosity modifier.
As the mathematical optimization method, it is preferable to use the simplex method, the interior point method, the gradient descent method, the quadratic programming method, the continuous linear programming method, the sequential quadratic programming method, the Lagrange undetermined multiplier method, the barrier function method, or the like.
Random search is a method of determining the amount of variation of various colors, gloss, and viscosity when the difference from the target value is equal to or less than a certain value by specifying the number of trials and calculating the variation values of color, gloss, and viscosity by theoretical calculations and/or calculations using variation response curve data for a variation pattern randomly extracted from within the specified value range of a parameter.

Second Embodiment
Next, a description will be given of a method for producing a paint according to a second embodiment of the present invention. The method for producing a paint having a desired paint property is also performed by adding raw materials for adjusting the paint property to the paint before adjustment.

Figure 7 is a flow chart of a paint manufacturing method according to a second embodiment of the present invention. The example of Figure 7 involves theoretical calculations by computer (as in the example of Figure 1). In the example of Figure 7, the paint property is color (as in the example of Figure 1). As shown in Figure 7, in the second embodiment, first, primary data is acquired (step S301). Next, the paint before adjustment is prepared (step S302), and then information on the paint property before adjustment (color in this example) is acquired (step S303: paint property information acquisition process before adjustment). These steps are the same as those in the embodiment shown in Figure 1, so a detailed explanation will be omitted.

Next, in the example of FIG. 7, a compounding composition of raw materials for adjusting paint properties suitable for obtaining paint properties of a target value is predicted by theoretical calculation by a computer (step S304: target value corresponding compounding composition prediction step). For the theoretical calculation, the primary data acquired in step S301 can be used. Next, in the example of FIG. 7, one compounding composition is selected from the predicted compounding compositions, in which the difference between the color difference of the target value and the calculated color difference is equal to or less than a predetermined reference value (in this example, the color difference is 0.1 or less) (step S305). When multiple candidates are predicted, the one with the smallest difference between the color difference of the target value and the calculated color difference can be selected, although this is not particularly limited. In a calculation method in which only one compounding composition can be selected, the one with the smallest difference between the color difference of the gap target value and the calculated color difference is calculated and used as the compounding composition of raw materials for adjusting paint properties.
In the target value corresponding blend composition prediction process (step S304), when calculating the blend composition of raw materials for adjusting the paint properties suitable for obtaining the target paint properties, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, and random search.

Next, in the example of FIG. 7, the gap composition of the raw materials for adjusting the paint properties is set (step S306: gap composition setting process). Here, in the gap composition setting process (step S306), the set gap composition is smaller than the predicted target value corresponding composition. When the predicted target value corresponding composition is 1.5 kg of red colorant, 0.2 kg of black colorant, and 0.6 kg of gloss adjuster, for example, by multiplying by an approach rate of 0.7, the gap composition can be set to 1.05 (= 1.5 x 0.7) kg of red colorant, 0.14 (= 0.2 x 0.7) kg of black colorant, and 0.42 (= 0.6 x 0.7) kg of gloss adjuster.

Next, the paint adjusted by adding the set composition is applied to a plate to prepare a sample plate, and the color difference between the standard plate and the sample plate is measured using a color difference meter or the like (step S307).

Next, it is determined whether the color difference measured in step S307 is equal to or less than a predetermined reference value (step S308). If the measured color difference is greater than the predetermined reference value, the process returns to step S304, and steps S304 to S308 are repeated. However, in step S304, a prediction is made based on the difference (color difference) between the paint property (color) of the paint adjusted by adding the set blending composition and the target paint property (color). For example, in the first step S304, a prediction is made based on the case where the paint property at the time of preparation is 25 and the target paint property is 30, and as a result of the measurement in step S307, the paint property of the paint adjusted by adding the gap blending composition set in step S306 is 28.4, in the second step S304, a prediction is made based on the current value of "28.4" and the target value of "30".
On the other hand, if the measured color difference is equal to or less than the predetermined reference value, the color matching is completed (step S309).
The effects of the coating material manufacturing method according to the second embodiment will be described below.

In the paint manufacturing method according to the second embodiment of the present invention, information on the paint properties before adjustment (color in this example) is acquired (step S301: paint property information before adjustment acquisition step), the mixture composition of the raw materials for adjusting the paint properties suitable for obtaining the paint properties of the target value is predicted by theoretical calculation by a computer (step S304: target value corresponding mixture prediction step), and the gap mixture composition of the raw materials for adjusting the paint properties is set (step S306: gap mixture setting step). Then, in the gap mixture setting step (step S306), the set gap mixture composition is smaller than the predicted target value corresponding mixture composition.
In this way, in the second embodiment, the paint properties are adjusted by setting a gap composition smaller than the predicted composition, so that when adjusting the paint properties of the paint, it is possible to prevent the amount of the paint used for adjustment from becoming excessive.
Then, the paint adjusted by adding the set gap compounding composition is applied to a plate to prepare a sample plate, and the color difference between the standard plate and the sample plate is measured using a color difference meter or the like (step S307). Next, it is determined whether the color difference is equal to or less than a predetermined reference value (step S308). As a result, by repeating steps S304 to S308 as necessary, a paint having the final target paint properties can be produced.

<Modification of the second embodiment>
Next, a modified example of the second embodiment using a fluctuation response curve will be described. In this modified example, one or more of color, gloss, and viscosity are targeted for adjustment.
8 is a flow chart of a method for producing a paint according to a modified example of the second embodiment of the present invention. As shown in FIG. 8, in this modified example, first, based on the pre-variation paint composition, information on the paint properties when only one of the parameters of the addition ratio of one or more colorants, the addition ratio of one or more gloss modifiers, and the addition ratio of one or more viscosity modifiers is changed in various ways is obtained for each parameter, and fluctuation response curve data showing the relationship between the fluctuation amount of each parameter and the fluctuation amount of the paint property information is obtained (step S401: fluctuation response curve data acquisition step). This step is the same as step S201 in the example of FIG. 2, which is a modified example of the first embodiment, so a detailed description will be omitted again.

Next, in this modified example, the paint before adjustment is prepared (step S402). Next, information on the paint properties before adjustment (in this example, one or more of color, gloss, and viscosity) is acquired (step S403: paint property information acquisition process before adjustment). These steps are similar to steps S202 and S203 in the example of FIG. 2, which is a modified example of the first embodiment, so a detailed explanation will be omitted.

Next, the computer uses the target value and the fluctuation response curve data to predict the amount of fluctuation in the addition ratio of the raw material for adjusting the appropriate paint properties (step S404: target value corresponding composition prediction process). Next, from the predicted compositions, one composition is selected in which the difference between the target value properties and the calculated properties is equal to or less than a predetermined reference value (step S405). These steps can be performed in the same way by replacing the "gap target value" in steps S205 and S206 of the example in Figure 2, which is a modified example of the first embodiment, with the "original target value" (without the process of multiplying the approach rate), so a detailed explanation will be omitted.

Next, as shown in FIG. 8, the gap composition of the raw materials for adjusting the paint properties is set (step S406: gap composition setting process). In the gap composition setting process, the set gap composition is smaller than the predicted target value corresponding composition. When the predicted target value corresponding composition is 1.5 kg of red colorant, 0.2 kg of black colorant, and 0.6 kg of gloss adjuster, the gap composition can be set, for example, by multiplying the approach rate of 0.7 to 1.05 (= 1.5 x 0.7) kg of red colorant, 0.14 (= 0.2 x 0.7) kg of black colorant, and 0.42 (= 0.6 x 0.7) kg of gloss adjuster.

Next, the paint adjusted by adding the set composition is applied to a plate to prepare a sample plate, and the difference in properties between the standard plate and the sample plate is measured (step S407).

Next, it is determined whether the property difference measured in step S407 is equal to or less than a predetermined reference value (step S408). If the property difference measured exceeds the predetermined reference value, the process returns to step S404, and steps S404 to S408 are repeated. However, in step S404, prediction is made based on the difference between the paint property of the paint adjusted by adding the set blending composition and the target paint property. For example, in the first step S404, a prediction is made based on the paint property at the time of preparation being 25 and the target paint property being 30, and as a result of the measurement in step S407, if the paint property of the paint adjusted by adding the gap blending composition set in step S406 is 28.4, in the second step S404, a prediction is made based on the current value of "28.4" and the target value of "30".
On the other hand, if the measured property difference is equal to or smaller than the predetermined reference value, the adjustment is completed (step S309).
The effects of the coating material manufacturing method according to the modified example of the second embodiment will be described below.

In the modified example of the second embodiment, the fluctuation response curve data is acquired as described above (step S401: fluctuation response curve data acquisition step), and the computer predicts the fluctuation of the addition ratio of the raw material for adjusting the paint properties using the target paint properties and the fluctuation response curve data (step S404: target value corresponding composition prediction step), and sets the gap composition of the raw material for adjusting the paint properties (step S406: gap composition setting step). In the gap composition setting step, the set gap composition is smaller than the predicted target value corresponding composition.
In this way, in the modified example of the second embodiment, the paint properties are adjusted by setting a gap composition smaller than the predicted composition, so that when adjusting the paint properties of the paint, it is possible to prevent the amount of the paint used for adjustment from becoming excessive.
Then, the paint adjusted by adding the set gap compounding composition is applied to a plate to prepare a sample plate, and the property difference between the standard plate and the sample plate is measured (step S407). Next, it is determined whether the property difference is equal to or less than a predetermined reference value (step S408). As a result, by repeating steps S404 to S408 as necessary, a paint having the final target paint properties can be produced.

The color adjustment shown in FIG. 7 and the gloss and/or viscosity adjustment shown in FIG. 8 can be performed simultaneously. In other words, the color can be adjusted by the method illustrated in FIG. 7, and the gloss and/or viscosity can be adjusted by the method illustrated in FIG. 8, and these can be performed simultaneously to adjust the color, gloss and/or viscosity. That is, as a preliminary preparation for prediction, steps S301 and S401 are performed, then the paint before adjustment is prepared (steps S302, S402), and then each paint property before adjustment (color, gloss and/or viscosity) is measured and obtained (steps S303, S403). Next, for color, the appropriate composition is predicted by the theoretical calculation, and for gloss and/or viscosity, the color fluctuation when the raw materials for adjusting the paint properties necessary for adjusting the gloss and/or viscosity are added is calculated by calculation using the fluctuation response curve data (steps S404, S405), and for color, the appropriate composition is predicted by the theoretical calculation after taking into account the color fluctuation due to the adjustment of the gloss and/or viscosity (steps S304, S305). Next, the gap composition for each paint property (color, gloss and/or viscosity) is set (steps S306, S406). Then, each property is measured, and depending on whether it is below a predetermined reference value, it is repeated or the adjustment is terminated (steps S304 to S309, S404 to S409).

Here, in the second embodiment and its modified example, a pre-adjustment paint property information acquisition step is further included, which acquires information on the paint properties before adjustment, prior to the gap composition calculation step (step S306, step S406). In the gap composition calculation step (step S306, step S406), it is preferable to calculate the gap composition continuously or stepwise so that the gap composition approaches the predicted composition as the absolute value of the difference between the acquired pre-adjustment paint properties and the target paint properties becomes smaller.
As described above, the second embodiment and its modified examples include repeating steps S304 to S308 (steps S404 to S408), and because the paint properties of the paint gradually approach the target paint properties, the amount of blended composition added for adjustment gradually decreases, and therefore the excess amount in the case of adding too much of the blend also becomes smaller, so the risk gradually decreases.
Therefore, as steps S304 to S308 (steps S404 to S408) are repeated, the smaller the absolute value of the difference between the acquired pre-adjustment paint properties and the target paint properties, the smaller the gap composition is set so that the gap composition approaches the target paint properties continuously or stepwise (for example, by adopting an approach rate closer to "1").This reduces the number of repetitions required to produce paint, and makes it possible to improve production efficiency without compromising the effect of suppressing overproduction.

In the second embodiment, in the gap composition calculation step (step S306), it is preferable to calculate the gap composition for each colorant independently. Specifically, the color further includes L ^* , a ^* , and b ^* . If a common gap composition is used, if any property exceeds the target value due to excessive addition, a problem of overproduction will occur. In addition, if the common gap composition is set too low, productivity will decrease. In contrast, if the gap composition is set independently for each colorant, overproduction can be further suppressed while maintaining manufacturing efficiency.

For example, when ΔL ^* is greater than 1, the approach rate can be 50%, when ΔL ^* is 0.5 to 1.0, the approach rate can be 70%, when ΔL ^* is 0.5 or less, the approach rate can be 100%, etc. This can also be applied independently to each colorant.

Here, in the target value corresponding blend composition prediction process (step S304) in the second embodiment, when predicting the blend composition of raw materials for adjusting the paint properties suitable for obtaining the target value of the paint properties, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, and random search.
Similarly, in the target value corresponding blend composition prediction step (step S404) in the modified example of the second embodiment, when predicting the suitable amount of variation in the addition ratio of raw materials for adjusting the paint properties and/or the suitable amount of variation in the manufacturing conditions, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, and random search.

Third Embodiment
The paint manufacturing method according to the third embodiment of the present invention is a method of manufacturing a paint having a target paint property by adding raw materials for adjusting the paint property to the paint before adjustment. FIG. 9 is a flow chart of the paint manufacturing method according to the third embodiment of the present invention. The example of FIG. 9 involves theoretical calculation by a computer. In the example of FIG. 9, the paint property is color. As shown in FIG. 9, in the third embodiment, first, primary data is acquired (step S501), then the primary color of the paint before adjustment is prepared (step S502), and then information on the paint property before adjustment (color in this example) is acquired (step S503: paint property information acquisition process before adjustment). These steps S501 to S503 are the same as steps S101 to S103 in the example of FIG. 1, so a detailed description will be omitted.

Next, in the third embodiment, the main colorant of the paint before adjustment is identified (step S504). Here, the "main colorant" is the colorant that is mixed in the largest amount in the paint before adjustment. Although not particularly limited, the main colorant may be identified by visually inspecting the paint before adjustment, or the main colorant may be identified by estimation based on the paint properties acquired in step S503.

Next, a condition is set that limits each amount in the composition of the raw materials for adjusting the paint properties of the colorant to a candidate solution (including the case where the amount is 0) generated from a finite numerical range. In this example, a condition is set that fixes the composition of the main colorant to 0 (step S505). In other words, a set of solutions in which the main colorant is not added is set as a prerequisite. Note that the target for limiting the candidate solutions is the main colorant in this example, but it is not limited to the main colorant as long as it is a colorant, and multiple colorants may be targeted. Also, in this example, the amount of the main colorant is fixed to a single solution of 0, but it is sufficient to limit it to a finite numerical range, and it is preferable to limit it to a numerical range close to 0, including the case of 0. Such a limit can be achieved by multiplying the normal numerical range by a rate close to 0, etc.

Next, a gap target value for the paint properties (color in this example) is set (step S506: gap target value setting process). Here, the difference between the pre-adjustment paint properties (color) acquired in step S503 and the gap target value is smaller than the difference between the pre-adjustment paint properties (color) acquired and the target paint properties. Step S503 is similar to step S103 in FIG. 1, so a detailed explanation will be omitted.

Next, a suitable composition is predicted by theoretical calculation (step S507: restricted paint property adjustment composition prediction process). That is, a restricted paint property adjustment composition, which is a composition of raw materials for adjusting paint properties, is predicted by theoretical calculation under conditions in which the amounts of each ingredient in the composition for adjusting the paint properties of the colorant are individually limited to solution candidates generated from a finite numerical range. In this example, a composition suitable for obtaining the target gap value is predicted under conditions limited to a set of solutions in which the amount of the main colorant is 0.

Next, it is determined whether or not there is a composition among the predicted compositions in which the difference between the color difference of the gap target value and the calculated color difference is equal to or less than a predetermined reference value (step S508). In the limited paint property adjustment composition prediction step (step S507), if the composition of the raw material for adjusting the paint property is not calculated by theoretical calculation, some or all of the conditions are released, and the composition of the raw material for adjusting the paint property is predicted again by theoretical calculation (step S509). On the other hand, in the limited paint property adjustment composition prediction step (step S507), if the composition of the raw material for adjusting the paint property is calculated by theoretical calculation, one composition is selected from the predicted compositions in which the difference between the color difference of the gap target value and the calculated color difference is equal to or less than a predetermined reference value (in this example, the color difference is 0.1 or less) (if multiple candidates are predicted, the one with the smallest difference between the color difference of the gap target value and the calculated color difference can be selected, although this is not particularly limited). In addition, in a calculation method in which only one blend composition can be selected, the blend composition for the raw material for adjusting the paint properties is calculated to have the smallest difference between the color difference of the gap target value and the calculated color difference.
In the step of predicting the blend composition for adjusting the paint properties with restrictions (step S507), when calculating the blend composition of raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value under the above-mentioned limited conditions, it is preferable to use any one of the methods of brute force search method, mathematical optimization method, and random search.

Next, the paint adjusted by adding the selected composition is applied to a plate to prepare a sample plate, and the color difference between the standard plate and the sample plate is measured using a color difference meter or the like (step S510). Next, it is determined whether the color difference measured in step S510 is equal to or less than a predetermined reference value (step S511). If the measured color difference exceeds the predetermined reference value, the process returns to step S505, and steps S505 to S511 are repeated. However, the gap target value set in step S506 is set based on the color difference obtained in step S510.
On the other hand, if the measured color difference is equal to or less than the predetermined reference value, the color matching is completed (step S512).
The effects of the coating material manufacturing method according to the third embodiment will be described below.

As mentioned above, it is often the case that a particular colorant having a maximum blending amount is blended close to the target value when the paint before adjustment is blended, and for example, if the correct blending amount is 99 kg of white colorant and 1 kg of black colorant, even if each colorant is blended in the correct amount, if the coloring strength of the black colorant is 10% higher, the original amount cannot be restored unless 9.9 kg of white colorant is added, resulting in at least 9.9 kg of overproduction. In such cases, the problem of overproduction has become evident.
In contrast, in the third embodiment, under conditions in which the amounts of each component in the composition of raw materials for adjusting the paint properties of the colorant are individually limited to solution candidates generated from a finite numerical range, a restricted composition for adjusting the paint properties, which is a composition of raw materials for adjusting the paint properties, is predicted by theoretical calculation (step S507: restricted composition prediction process for adjusting the paint properties). The solution candidates include cases in which the amount is 0, and the restricted composition prediction process for adjusting the paint properties is repeated until the difference between the paint properties obtained by adding the predicted composition and the target paint properties becomes equal to or less than a predetermined reference value.
According to the third embodiment, for example, it is possible to obtain a solution under the precondition that the blend composition that is originally close to the target value as described above is not excessively added (when the blend amount is set to 0 as a solution candidate, the blend amount is not added), so that it is possible to suppress overproduction, particularly in a situation where a specific colorant having a particularly maximum blend amount as described above is blended close to the target value at the time of intended charging.
Furthermore, in the third embodiment, if a solution does not exist, prediction is performed again with some or all of the conditions removed, so that there is no risk of being unable to calculate a predicted value due to limiting the conditions.
Furthermore, in the third embodiment, a gap target value that is lower than the original target value (the difference from the current paint properties is small) is set, and the composition of the raw materials for adjusting the paint properties that corresponds to that gap target value is predicted. From the predicted compositions, one composition is selected such that the difference between the color difference of the gap target value and the calculated color difference is equal to or less than a predetermined reference value, and the paint properties are adjusted. Therefore, when adjusting the paint properties of the paint, it is possible to prevent the amount of material used for adjustment from becoming excessive.
Then, the paint adjusted by adding the selected compounding composition is applied to a plate to prepare a sample plate, and the color difference between the standard plate and the sample plate is measured using a color difference meter or the like. It is then determined whether the color difference is below a predetermined reference value. As a result, steps S505 to S511 are repeated as necessary to ultimately produce a paint having the target paint properties.
Here, for color, the blend amount and the amount of color change are not linear, and L ^* , a ^* , and b ^* behave separately. In contrast, in the third embodiment, the gap target value is set by multiplying the approach rate based on the paint properties (color), so that it is possible to more accurately prevent the blend amount for adjustment from becoming excessive, compared to, for example, a method of multiplying a predicted blend amount by the approach rate.

In the example shown in FIG. 9, in the gap target value corresponding composition prediction process (step S105) in the example of FIG. 1, the composition of the raw materials for adjusting the paint properties of the colorant is predicted by theoretical calculation under conditions in which each amount in the composition of the raw materials for adjusting the paint properties of the colorant is individually limited to solution candidates generated from a finite numerical range, and the solution candidates include cases in which the amount is 0.
On the other hand, the method of Fig. 9 can also be applied to the method of Fig. 2, and in the gap target value corresponding mixture composition prediction step (step S205) in the example of Fig. 2, the mixture composition of the raw material for adjusting the paint properties of the colorant is predicted by calculation based on the fluctuation response curve data under conditions in which the amount of the raw material for adjusting the paint properties of the colorant in the mixture composition is individually limited to solution candidates generated from a finite numerical range, and the solution candidates can also include cases in which the amount of the raw material is 0. In this case, too, it is possible to obtain the effect of suppressing overproduction in a situation in which a specific colorant having a particularly maximum amount as described above is added close to the target value at the time of intended preparation.

Here, the computer can include a calculation unit and an acquisition unit. The calculation unit performs prediction in the gap target value corresponding mixture composition prediction process (step S105, step S205), prediction in the target value corresponding mixture composition prediction process (step S304, step S404), and prediction in the limited paint property adjustment mixture composition prediction process (step S507). The calculation unit can be a processor or the like. The acquisition unit can be configured to acquire fluctuation response curve data. The acquisition unit can be a processor (in this case, the calculation unit may have such a function, or may be a processor separate from the calculation unit), or may be, for example, a communication unit capable of receiving the above-mentioned fluctuation response curve data, or may be a reading unit having a function of reading the above-mentioned fluctuation response curve data recorded in a memory.
It should be noted that the color adjustment shown in Fig. 9 and the gloss and/or viscosity adjustment shown in Fig. 2 or 8 can be performed simultaneously. In other words, the color can be adjusted by the method illustrated and explained in Fig. 9, and the gloss and/or viscosity can be adjusted by the method illustrated and explained in Fig. 2 or 8, and these can be performed simultaneously to adjust the color, gloss and/or viscosity.

The present invention will be explained in more detail with reference to the following examples, but the present invention is not limited to these.

The production rate (average of 20 runs) was compared when paint was produced using the methods of Examples 1 to 4 and Comparative Examples 1 and 2.

First, the methods for preparing the colorants used in Examples 1 to 4 and Comparative Examples 1 and 2 will be described.
<Preparation Example of White Colorant>
20 parts by mass of acrylic resin as a resin, 35 parts by mass of isophorone as an organic solvent, and 46 parts by mass of titanium oxide as a white pigment were mixed, and dispersed using a sand mill (dispersion medium: glass beads) until the maximum particle diameter of the pigment coarse particles became 10 μm or less, to prepare White Colorant-10.
<Adjustment of colorant to reproduce lot variation in coloring strength>
In order to reproduce the lot variation in coloring strength, the white colorant-8 was prepared in the same manner as the white colorant-10, except that the pigment coarse particles were dispersed until the maximum particle size was 8 μm or less.
Similarly, a white colorant-12 was prepared in the same manner as in white colorant-10, except that the pigment coarse particles were dispersed until the maximum particle size was 12 μm or less.
<Preparation examples of other colorants>
Black colorant 1-10, black colorant 2-10, yellow colorant-10, red colorant-10, black colorant 1-8, black colorant 2-8, yellow colorant-8, red colorant-8, black colorant 1-12, black colorant 2-12, yellow colorant-12, and red colorant-12 were prepared in the same manner as in the preparation example of the white colorant, except that the type and amount of each material was changed as shown in Table 1 below. The formulation of each colorant is shown in Table 1.
Which of the lots with different coloring strengths, colorant-8, -10, and -12, was used was selected randomly using a random number table. Below, unless a lot with different coloring strength was specifically indicated, all were carried out in the same manner. For example, if "white colorant" was indicated, which of white colorant-8, white colorant-10, and white colorant-12 was used was selected randomly, which was intended to reproduce the fact that the coloring strength of the colorant lot used when actually manufacturing paint in a factory is unknown.

<Gloss adjuster>
The gloss adjusting agent used was commercially available silica (silicon dioxide) used as a matting agent.
<Viscosity modifier>
The viscosity modifier used was isophorone, which is also used as a solvent.

<Preparation Example of Calculation Coating Composition W>
5 parts by mass of acrylic resin, 25 parts by mass of fluororesin, 35 parts by mass of isophorone, and 35 parts by mass of cyclohexanone were added, and the resins were uniformly dissolved using a disper, and then 100 parts by mass of the white colorant was further added and mixed uniformly using a disper to prepare a calculation coating composition W (solid content concentration: 48% by mass).
<Preparation examples of other coating compositions for calculation>
Except for changing the type and amount of each material as shown in Table 2 below, the calculation coating compositions K1, K2, Y and R were prepared in the same manner as in the preparation example of the calculation coating composition W. The formulations of each calculation coating composition are shown in Table 2.

<Preparation example of paint 1 before fluctuation>
5 parts by mass of acrylic resin, 25 parts by mass of fluororesin, 35 parts by mass of isophorone, and 35 parts by mass of cyclohexanone were added, and the resins were uniformly dissolved using a disper, and then 25 parts by mass of white colorant, 2 parts by mass of black colorant 1, 69 parts by mass of yellow colorant, 4 parts by mass of red colorant, and 2 parts by mass of gloss adjuster 1 were added and mixed uniformly using a disper to prepare paint 1 before dispersion (solids concentration: 47% by mass).

<Preparation examples of paints 2 to 4 before fluctuation>
Except for changing the type and amount of each material as shown in Table 3 below, paints 2 to 4 before the change were prepared in the same manner as in the preparation example of paint 1 before the change. The formulation of each paint before the change is shown in Table 3.

Details of the materials used in preparing the coating compositions are as follows:
Acrylic resin: Paraloid B44 (manufactured by Rohm & Haas), solid content: 100% by mass
Fluorine resin: KYNAR500 (manufactured by ARKEMA), solid content: 100% by mass
White pigment: TI-PURE R-706 (titanium oxide, manufactured by DuPont)
Black pigment 1: Mitsubishi Carbon Black MA-100 (carbon black, manufactured by Mitsubishi Chemical Corporation)
Black pigment 2: SUNBLACK X15 (carbon black, manufactured by Shiraishi Calcium Co., Ltd.)
Yellow pigment: TAROX synthetic iron oxide LL-XLO (yellow iron oxide, manufactured by Titan Kogyo Co., Ltd.)
Red pigment: TODA COLOR 140ED (iron oxide, manufactured by Toda Kogyo Co., Ltd.)
Gloss adjuster 1: GASIL HP395 (synthetic silica, manufactured by INEOS SILICAS)
Gloss adjuster 2: Silica 435 (silicon dioxide, manufactured by Fuji Silysia Chemical Ltd.)
Organic solvent: Isophorone (manufactured by ARKEMA)
・Organic solvent: Cyclohexanone (manufactured by Ube Industries)

<Method of preparing coating film>
Fine Tough G Primer (epoxy resin-based primer: manufactured by Nippon Paint Industrial Coatings Co., Ltd.) was applied as an undercoat paint to the surface of a base material (zinc-aluminum alloy plated steel plate: 1,800 × 300 × 0.35 mm) using a roll coater (standard film thickness: 5 μm), and then baked for 60 seconds under conditions that resulted in a maximum temperature reached by the base material being 210° C., to form an undercoat paint film.
Next, a paint was applied onto the undercoat coating film using a roll coater (standard film thickness: 15 μm), and then baked for 60 seconds under conditions that resulted in a maximum material temperature of 250° C., followed by immediate cooling to obtain a coating film of the paint composition.

Next, we will explain the predictions for each example and comparative example. First, we will explain how to create a fluctuation response curve.

<Example of acquiring paint properties>
The paint properties required in each of the Examples and Comparative Examples were measured by the following methods.
1) Measurement of color data The color (L ^* , a ^* , b* values, or L*, a*, ^{b *} value differences from the target standard plate) of the various coating films obtained by the ^coating film preparation method described above was measured using a spectrophotometer LabScan ^XE (manufactured by HunterLab).
2) Measurement of Gloss Data For each of the coating films obtained by the above-mentioned coating film manufacturing method, the 60° gloss was measured using a gloss meter VG7000 (manufactured by Nippon Denshoku Industries Co., Ltd.).
3) Measurement of Viscosity Data For each of the coating compositions obtained in the above coating composition preparation examples, the number of seconds measured at 25°C using a Ford Cup No. 4 was taken as the viscosity value.

<Example of creating a fluctuation response curve>
A coating film of the pre-variation paint composition 1 was prepared, and color, gloss, and viscosity data of the pre-variation paint properties were obtained. Next, a white colorant was added at 0.03%, and a coating film was prepared in the same manner, and data of the color L ^* value of the post-variation paint properties 1 was obtained. Similarly, 0.1%, 0.3%, 1%, 3%, and 10% of the white colorant were added, and data of the color L ^* value of the post-variation paint properties 2 to 7 were obtained. A variation response curve WL(x)-1 was obtained as an approximation curve by the least squares method or the like from these pre-variation paint properties and post-variation paint properties 1 to 7.
In the same manner as above, other paint property data was obtained for black colorant 1, black colorant 2, yellow colorant, red colorant, gloss modifier 1, gloss modifier 2, and viscosity modifier, and fluctuation response curves were obtained for each. The results are shown in Table 4.

For the pre-variation coating compositions 2 to 4 (paint properties before adjustment: L ^* value: 29 to 90, a ^* value: -1 to 13, b ^* value: 2 to 36, gloss: 7 to 29, viscosity: 230 to 260 seconds), a fluctuation response curve was obtained in the same manner as above. The results are shown in Table 4.

<Example of creating primary data>
<Example 1 of Acquiring Absorption Coefficient and Scattering Count>
The three types of coating compositions were applied, namely, the calculation coating composition W, the calculation coating composition K1, and a coating composition obtained by mixing the calculation coating composition W and the calculation coating composition K1 in an 80:20 ratio, and the spectral reflectance of each was measured. The absorption coefficient K of the calculation coating composition W and the absorption coefficient K and scattering coefficient S of the calculation coating composition K1 were calculated.

<Acquisition Example 2 of Absorption Coefficient and Scattering Count>
Except for changing the calculation coating composition K1 to the calculation coating compositions K2, Y, and R, the absorption coefficient K and scattering coefficient S of the calculation coating composition K2, the calculation coating composition Y, and the calculation coating composition R were calculated in the same manner as in Example 1 for obtaining the absorption coefficient and scattering coefficient.

<Example of obtaining the theoretical calculation formula for hue fluctuation>
The theoretical calculation formula for hue variation was obtained using these absorption coefficients and scattering coefficients, the refractive indexes of pigments and coatings, and the Kubelka-Munk theory, Duncan theory, and Sanderson correction formula.

<Criteria for passing paint adjustment>
The criteria for pass were that all of the following three conditions were satisfied: The differences between the target values and the actually measured values of L ^* , a ^* , and b ^* were designated ΔL ^* , Δa ^* , and Δb ^* , respectively.
1) The value of ΔE=√(ΔL ^{* 2} + Δa ^{* 2} + Δb ^{* 2} ) is 0.05 or less, and each of ΔL ^* , Δa ^* , and Δb ^* is 0.05 or less.
2) The difference in gloss between the target value and the actual measured value is 0.3 or less from the standard plate when the gloss value is less than 10, 0.5 or less when it is 10 or more but less than 20, and 2.0 or less when it is 20 or more.
3) The difference in viscosity between the target value and the actual measured value is 5 seconds or less. <Criteria for passing the adjustment count>
Among the 20 adjustments, an average number of adjustments of less than 5 was judged as pass, and an average number of adjustments of 5 or more was judged as fail.
<Production rate pass criteria>
In the 20 adjustments, a production rate of less than 105% was judged as pass, and a production rate of 105% or more was judged as fail.

Example 1
In Example 1, as shown in Fig. 2, a pre-adjustment paint property information acquisition step of acquiring information on the paint property before adjustment, a gap target value setting step of setting a gap target value of the paint property, and a gap target value corresponding composition prediction step of predicting a composition of raw materials for adjusting the paint property suitable for obtaining the paint property of the gap target value using a fluctuation response curve were performed, and the gap target value setting step and the gap target value corresponding composition prediction step were repeated until the difference between the paint property obtained by adding the predicted composition and the target paint property became equal to or less than a predetermined reference value. In the gap target value setting step, the difference between the acquired paint property before adjustment and the gap target value was made smaller than the difference between the acquired paint property before adjustment and the target paint property.
<Pre-adjustment paint property information acquisition process>
Raw paint 1 was prepared. Colorant lots were selected randomly.
The coating film properties of Paint 1 before preparation were obtained, and were ΔL ^* value = 4.5, Δa ^* value = 0.3, and Δb ^* value = -1.4. In the same manner, the coating film properties of Paints 2 to 4 before preparation were obtained, and information on the properties of the paints before preparation was obtained. The preparation of the paints before preparation is summarized in Table 5 below.
<Setting the target gap value for paint properties>
The target gap values for paint 1 before adjustment were an approach rate of 0.7, ΔL ^* value = 1.35, Δa ^* value = 0.09, and Δb ^* value = -0.42. Similarly, the target gap values for paints 2 to 4 before adjustment were set.
<Prediction process of blend composition of raw materials for adjusting paint properties>
- Data input for corrected blend prediction The lower and upper limits of the corrected blend variation range for each colorant and gloss modifier, which are raw materials for adjusting the paint properties, were set to any amount between 0-2% of the total amount of paint before adjustment, and the viscosity modifier was similarly set to a range of 0-10%.Based on the set variation range of the corrected blend, data for 10 million possible combinations of the corrected blend composition of the raw materials for adjusting the paint properties (combinations of the amount of each colorant, gloss modifier, and viscosity modifier added) was created.

<Mixture prediction using variation response curves and brute force search>
The candidate data of 10 million different amounts of addition was substituted into the fluctuation response curves WL(x)-1 to Sg(x)-1 shown in Table 4, and the predicted values of the paint properties of the adjusted paint composition 1 were obtained. Specifically, the fluctuation amount of the L ^* value when the amount of white colorant added in the first candidate formulation of 10 million combinations was substituted into WL(x)-1, and the sum of all the fluctuation amounts of the L* value when the black colorant 1, the yellow colorant, the red colorant, the gloss modifier 1, the viscosity modifier, the nip pressure, and the maximum temperature were substituted into KL(x)-1, YL(x)-1, RL(x)-1, ML(x)-1, and SL(x)-1, respectively, is the fluctuation amount of ^{the L* value} of the first candidate formulation. In the same manner, the fluctuation amounts of the a ^* and b ^* values, gloss, and viscosity of each candidate were obtained. This was repeated 10 million times.
From the candidate data for corrected formulation compositions in which the difference between the L ^* , a ^* , b ^* values, gloss, and viscosity fluctuations of the gap target value of adjusted paint composition 1 at an approach rate of 0.7 and the predicted E value was within 0.05, the difference in gloss was within 0.3, and the difference in viscosity fluctuation was within 5 seconds, one with the smallest total amount of paint property adjusting material added was extracted, and predicted values for the corrected addition amount of raw materials for adjusting the paint properties and the corrected fluctuation amount of the painting conditions were obtained.

<Acquisition of paint properties of adjusted paint composition and acquisition of difference from target properties>
A corrective blend of raw materials for adjusting the predicted paint properties was added to the paint 1 before adjustment, and the adjusted paint composition 1 was prepared to obtain the paint properties. The differences from the target properties were obtained, and the ΔL ^* value was 1.1, Δa ^* value was 0.09, and Δb ^* value was -0.4.

<Re-adjustment of the prepared coating composition>
The paint properties of the prepared paint composition 1 did not satisfy the pass criteria for the target values of the paint properties, and therefore it was judged to be unacceptable.
The paint composition before the readjustment and its properties before the readjustment were the paint composition after the previous adjustment that had failed and its properties after the adjustment.
As in the previous study, 10 million candidate data for the corrected blend composition of raw materials for adjusting the paint properties were created, and the 10 million candidate data for the corrected blend composition of raw materials for adjusting the paint properties and coating conditions were substituted into the fluctuation response curves WL(x)-1, KL(x)-1, YL(x)-1, RL(x)-1, ML(x)-1, SL(x)-1 to obtain predicted paint properties of the adjusted paint composition.
In the same manner as for the unadjusted paint 1, the predicted corrective blend of paint property adjusting materials was added to the unadjusted paint to adjust the adjusted paint composition and obtain the paint properties. This readjustment operation was repeated, and it was confirmed that the paint properties of the adjusted paint composition met the target pass criteria, and the preparation of the paint was completed. (Adjustment times: 2)
Repeat operation 1
The process from preparation of Post-Prepared Coating Composition 1 through to the end of the paint preparation, including the step of randomly selecting a colorant lot, was repeated five times.
・Repeat operation 2
The same operation was carried out for pre-prepared paint compositions 2, 3, and 4. Including pre-prepared paint 1, four types of paint were prepared, each repeated five times for a total of 20 preparations. The average number of preparations was 3.8, and the production rate was 101%, which was acceptable.

Example 2
In Example 2, as shown in Figures 1 and 2, a pre-adjustment paint property information acquisition step of acquiring information on the paint property before adjustment, a gap target value setting step of setting a gap target value of the paint property, and a gap target value corresponding composition prediction step of predicting a composition of raw materials for adjusting the paint property suitable for obtaining the paint property of the gap target value, the hue being predicted by theoretical calculation by computer, and the gloss and viscosity being predicted by a fluctuation response curve, were performed, and the gap target value setting step and the gap target value corresponding composition prediction step were repeated until the difference between the paint property obtained by adding the predicted composition and the target paint property became equal to or less than a predetermined reference value. In the gap target value setting step, the difference between the acquired paint property before adjustment and the gap target value was made smaller than the difference between the acquired paint property before adjustment and the target paint property.
<Pre-adjustment paint property information acquisition process>
The same procedure as in Example 1 was carried out.
<Setting the target gap value for paint properties>
The same procedure as in Example 1 was carried out.
<Prediction process of blend composition of raw materials for adjusting paint properties>
Acquisition of candidate combination data for corrected blend composition As in Example 1, 10 million candidate combination data (combinations of the amounts of each colorant, gloss modifier, and viscosity modifier added) for the corrected blend composition of raw materials for adjusting paint properties was created.

<Prediction of gloss and viscosity modifier blending using variation response curve and brute force search>
The candidate data for 10 million different amounts of addition were substituted into the fluctuation response curves ML(x)-1 to Sg(x)-1 shown in Table 4, and the predicted values of the paint properties of the adjusted paint composition 1 were obtained. Specifically, the sum of the amount of change in the L ^* value when the amount of gloss modifier 1 added in the first candidate formulation of 10 million combinations was substituted into ML(x)-1 and the amount of change in the L ^* value when the viscosity modifier was substituted into SL(x)-1 was the amount of change in the L ^* value of the first candidate formulation. In the same manner, the amount of change in the a ^* and b ^* values, gloss, and viscosity of each candidate was obtained. This was repeated 10 million times.
From the candidate data for the corrected formulation composition in which the difference between the gloss of the gap target value of the adjusted paint composition 1 and the predicted gloss was within 0.3 and the difference in viscosity was within 5 seconds, the one with the smallest difference was extracted, and predicted values for the corrected addition amount and corrected variation amount of the gloss adjuster and viscosity adjuster were obtained.

<Colorant correction blend composition prediction using theoretical calculations and mathematical optimization>
A theoretical formula was obtained to calculate color variation from the corrected blend composition.
The colorant blend amount for the gap target value of the adjusted coating composition 1 was calculated by mathematical optimization using a theoretical formula to determine the combination of the gap target value and the corrected blend composition of the colorant with the smallest predicted value, taking into account the fluctuations in the L ^* , a ^* , and b ^* values that fluctuate due to the predicted values of the corrected addition amounts of the gloss modifier 1 and the viscosity modifier obtained from the fluctuation response curve. Note that the colorant was assumed to have no effect on gloss and viscosity.

<Obtaining paint properties of adjusted paint composition>
The same procedure as in Example 1 was carried out.

<Re-adjustment of the prepared coating composition>
The same procedure as in Example 1 was carried out.
The average number of times the coating composition was prepared was 3.6, and the production rate was 101%, which was acceptable.

Example 3
In Example 3, as shown in Figures 7 and 8, a pre-adjustment paint property information acquisition step for acquiring information on the paint property before adjustment, a target value corresponding composition prediction step for predicting the composition of the raw material for adjusting the paint property suitable for obtaining the paint property of the target value, the hue is predicted by theoretical calculation by computer, and the gloss and viscosity are predicted by the fluctuation response curve, and a gap composition setting step for setting the gap composition of the raw material for adjusting the paint property are performed, and the target value corresponding composition prediction step and the gap composition prediction step are repeated until the difference between the paint property obtained by adding the predicted gap composition and the target paint property becomes equal to or less than a predetermined reference value. In the gap composition setting step, the set gap composition is made smaller than the predicted target value corresponding composition.
The same procedure as in Example 2 was carried out except that a gap compounding composition was used instead of setting a target gap value. The average number of times the coating composition was adjusted was 4.1 times, and the production rate was 102%, which was acceptable.

Example 4
In Example 4, as shown in Fig. 9, a pre-adjustment paint property information acquisition step for acquiring information on the paint property before adjustment, and a restricted paint property adjustment composition prediction step for predicting a restricted paint property adjustment composition, which is a composition of raw materials for adjusting the paint property, by theoretical calculation under conditions in which each blend amount in the blend composition of the raw materials for adjusting the paint property of the primary color is individually limited to a solution candidate generated from a finite numerical range, are performed, and the restricted paint property adjustment composition prediction step is repeated until the difference between the paint property obtained by adding the predicted blend composition and the target paint property becomes equal to or less than a predetermined reference value. The solution candidates include the case where the blend amount is 0.
In <Colorant Corrected Blend Composition Prediction by Theoretical Calculation and Mathematical Optimization>, the same procedure as in Example 2 was carried out except that the blend correction amount of the main colorant was fixed at zero and no target gap value was set.

<Comparative Example 1>
In Comparative Example 1, the composition of raw materials for adjusting the paint properties suitable for obtaining the target paint properties (the original paint properties, not the gap target value) was predicted by theoretical calculation using a computer, and the above prediction was repeated until the difference between the paint properties obtained by adding the predicted composition and the target paint properties became equal to or less than a predetermined reference value.
The same procedure as in Example 2 was carried out, except that no target gap value was set and the blending composition of raw materials for adjusting the paint properties was predicted to obtain the target paint properties.
The average number of times the coating composition was adjusted was 3.5, which was acceptable, but the production rate was 118%, which was unacceptable.

<Comparative Example 2>
In Comparative Example 2, the color, gloss, and viscosity variations of 27 paint types (paint properties before adjustment: L ^* value 24-94, a ^* value -3-10, b ^* value -2-37, gloss 3-55, viscosity 130-280 seconds) were learned using a neural network from the results of approximately 1,000 variations in paint property adjustment materials, painting conditions, and paint properties, and artificial intelligence for color prediction was created, and the predicted value of the correction variation amount of the correction addition amount was obtained by the brute force search method. The same procedure as in Example 1 using the gap target value was carried out. The average number of times the paint composition was adjusted was 5.7, and the production rate was 109%, which did not meet the pass standard.

In this disclosure, each term has the following meaning.
The "number of adjustments" refers to the average number of adjustments required to pass a test when a total of 20 paint compositions, five for each of the four coating colors, were prepared and adjusted. A smaller number of adjustments means higher prediction accuracy.
"Production rate" refers to the proportion of the adjusted production rate when the planned production rate of the target paint composition is taken as 1. Here, the fact that the production rate does not greatly exceed 100% means that the paint property prediction system has high prediction accuracy and the calculation method of the corrected blend composition is highly accurate, and that it is an efficient system that does not produce more paint composition than necessary, while allowing for raw materials and errors.
The evaluation results are shown in Table 6 below.

As shown in Table 6, according to Examples 1 to 4, the production rate was lower than that of Comparative Examples 1 and 2, and it can be seen that when adjusting the paint properties of the paint, it was possible to prevent the amount of adjustment compounded from becoming excessive, and that the number of repeated adjustments did not increase significantly. Furthermore, a comparison of Examples 2 and 4 shows that in Example 4, in which a limited paint property adjustment compounding composition prediction process was performed, the production rate was even lower than in Example 2, in which this process was not performed. Furthermore, a comparison of Examples 1 and 2 with Example 3 shows that the production rate was particularly low in the method in which a gap target value setting process was performed.

Claims (16)

A method for producing a paint having a desired paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint property is color,
A pre-adjustment paint property information acquisition process for acquiring information on the paint property before adjustment;
a gap target value setting process for setting a gap target value of a paint property;
a gap target value corresponding blend composition prediction step, which is carried out after the gap target value setting step, and predicts a blend composition of raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value by theoretical calculation using a computer,
In the gap target value setting step, an absolute value of a difference between the acquired pre-adjustment paint property and the gap target value is smaller than an absolute value of a difference between the acquired pre-adjustment paint property and a target paint property,
A method for manufacturing a paint, characterized in that the gap target value setting step and the gap target value corresponding blend composition predicting step are repeated until a difference between the paint properties obtained by adding the predicted blend composition and the target paint properties becomes equal to or less than a predetermined reference value. A method for producing a paint having a desired paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint properties are one or more of color, gloss, and viscosity,
A pre-adjustment paint property information acquisition process for acquiring information on the paint property before adjustment;
a fluctuation response curve data acquisition step for obtaining information on the paint properties when only one of the parameters of the addition ratio of one or more colorants, the addition ratio of one or more gloss modifiers, and the addition ratio of one or more viscosity modifiers is varied based on the pre-variation paint composition, and obtaining fluctuation response curve data showing the relationship between the amount of variation of each of the parameters and the amount of variation of the paint property information;
a gap target value setting process for setting a gap target value of a paint property;
A gap target value corresponding blend composition prediction step, which is performed after the gap target value setting step , predicts a suitable amount of fluctuation in the addition ratio of the raw material for adjusting the paint properties by using the gap target value and the fluctuation amount response curve data by a computer,
In the gap target value setting step, an absolute value of a difference between the acquired paint property before adjustment and the gap target value is smaller than an absolute value of a difference between the acquired paint property before adjustment and a target paint property,
A method for producing a paint, characterized in that the gap target value setting step and the gap target value corresponding blend composition predicting step are repeated until a difference between the paint properties obtained by adding the predicted blend composition and the target paint properties becomes equal to or less than a predetermined reference value. The manufacturing method according to claim 1 or 2, wherein in the gap target value setting step, the gap target value is set so that the smaller the absolute value of the difference between the acquired pre-adjustment paint properties and the target paint properties, the closer the gap target value is to the target paint properties, either continuously or stepwise. The paint properties include a plurality of properties,
3. The manufacturing method according to claim 1, wherein in the step of setting the target gap value, the target gap value is set independently for each of the properties. In the gap target value corresponding blend composition prediction step, the blend composition of the raw material for adjusting the paint properties is predicted by theoretical calculation under conditions in which each blend amount in the blend composition of the raw material for adjusting the paint properties is individually limited to a solution candidate generated from a finite numerical range,
The manufacturing method according to claim 1 , wherein the solution candidates include a case where the blending amount is 0. The manufacturing method according to claim 5, further comprising a step of canceling some or all of the conditions and re-predicting the composition of the raw material for adjusting the paint properties by theoretical calculation when the composition of the raw material for adjusting the paint properties suitable for obtaining the paint properties of the target gap value is not calculated by theoretical calculation in the gap target value corresponding composition prediction step. In the gap target value corresponding mixture composition prediction step, the mixture composition of the raw material for adjusting the paint properties is predicted by calculation based on the fluctuation response curve data under conditions in which the mixture amount in the mixture composition of the raw material for adjusting the paint properties is individually limited to a candidate solution generated from a finite numerical range,
The manufacturing method according to claim 2 , wherein the solution candidates include a case where the blending amount is 0. The manufacturing method according to claim 7, further comprising a step of canceling a part or all of the conditions and re-predicting the composition of the raw material for adjusting the paint properties by calculation based on the fluctuation response curve data, when the composition of the raw material for adjusting the paint properties suitable for obtaining the paint properties of the gap target value is not calculated based on the fluctuation response curve data, in the gap target value corresponding composition prediction step. The method further includes a main colorant specifying step of specifying a main colorant that is blended in the largest amount in the colorant blend composition of the paint before the adjustment,
8. The manufacturing method according to claim 5, wherein in the gap target value corresponding blend composition prediction step, a prediction is made for at least the main colorant using conditions limited to solution candidates generated from the finite numerical range. A method for producing a paint having a desired paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint property is color,
A pre-adjustment paint property information acquisition process for acquiring information on the paint property before adjustment;
A target value corresponding blend composition prediction step of predicting a blend composition of raw materials for adjusting the paint properties suitable for obtaining the target paint properties by theoretical calculation using a computer;
A gap composition setting step of independently setting the gap composition of the raw material for adjusting the paint properties for each colorant,
In the gap composition setting step, the set gap composition is smaller than the predicted target value corresponding composition,
A method for producing a paint, characterized in that the target value corresponding composition prediction step and the gap composition setting step are repeated until a difference between the paint properties obtained by adding the predicted gap composition and the target paint properties becomes equal to or less than a predetermined reference value. A method for producing a paint having a desired paint property by adding a raw material for adjusting the paint property to a paint before adjustment,
The paint properties are at least one of gloss and viscosity,
a fluctuation response curve data acquisition step for obtaining information on the paint properties when only one of the parameters of the addition ratio of one or more colorants, the addition ratio of one or more gloss modifiers, and the addition ratio of one or more viscosity modifiers is varied based on the pre-variation paint composition, and obtaining fluctuation response curve data showing the relationship between the amount of variation of each of the parameters and the amount of variation of the paint property information;
A target value corresponding blend composition prediction step of predicting a suitable amount of fluctuation in the addition ratio of the raw material for adjusting the paint properties by using the target paint properties and the fluctuation amount response curve data by a computer;
A gap composition setting step of setting a gap composition of the raw material for adjusting the paint properties,
In the gap composition setting step, the set gap composition is smaller than the predicted target value corresponding composition,
A method for producing a paint, characterized in that the target value corresponding composition prediction step and the gap composition setting step are repeated until a difference between the paint properties obtained by adding the gap composition and the target paint properties becomes equal to or less than a predetermined reference value. Prior to the gap blending composition setting step, information on the paint properties before adjustment is acquired, and
12. The manufacturing method according to claim 10 or 11, wherein in the gap composition setting step, the gap composition is calculated so that the smaller the absolute value of the difference between the acquired paint properties before adjustment and the target paint properties, the closer the gap composition is to the predicted composition continuously or stepwise. The manufacturing method according to claim 1, wherein in the gap target value corresponding blend composition prediction step, one of the methods of brute force search method, mathematical optimization method, and random search is used to calculate the blend composition of the raw materials for adjusting the paint properties suitable for obtaining the paint properties of the gap target value. The manufacturing method according to claim 2, wherein in the gap target value corresponding blend composition prediction step, when predicting the amount of variation in the suitable addition ratio of the raw material for adjusting the paint properties, one of the methods of brute force search method, mathematical optimization method, and random search is used. The manufacturing method according to claim 10, wherein in the target value corresponding blend composition prediction step, when predicting the blend composition of the raw materials for adjusting the paint properties suitable for obtaining the paint properties of the target value, one of the methods of brute force search method, mathematical optimization method, and random search is used. 12. The manufacturing method according to claim 11, wherein in the target value corresponding blend composition prediction step, when predicting the suitable amount of fluctuation in the addition ratio of the raw material for adjusting the paint properties, any one of a brute force search method, a mathematical optimization method, and a random search method is used.

Images