Characterization of phase separation of coating compositions

Abstract

The invention relates to a method of detecting phase separation of an aqueous or solvent borne or solventless coating composition, comprising: providing the coating composition in a container; providing a measuring instrument for receiving the container, the measuring instrument comprising a measuring probe; controlling the measuring instrument; a) moving the measuring probe through the coating composition along a predetermined measurement path at a predetermined velocity profile, the predetermined measurement path extending along a length axis of the container; obtaining a force-displacement profile by measuring a force exerted on the measuring probe while the probe is moving along the predetermined measurement path at the predetermined velocity profile; processing the force-displacement profile to detect at least one phase separation of the coating composition; and outputting a detection result.

Description

Technical Field

[0001] This invention relates to phase separation detection and characterization of aqueous, solvent-based, or solvent-free coating compositions, particularly aqueous, solvent-based, or solvent-free coating compositions. background

[0002] From production through storage and processing to drying or curing after use, coating compositions face numerous different requirements, such as their flow properties. These affect leveling, settling / precipitation (important for storage stability), and film formation.

[0003] Coating compositions contain components that tend to settle over time, thus forming different phases. This can lead to separation into a near-solid phase (precipitated phase) and a liquid phase, several liquid phases, or a combination of these phases. Phase separation can adversely affect further processing of the coating composition because redispersibility of the components can be difficult or incomplete, requiring methods and therefore time and money. Furthermore, since multiple phases can coexist, they may influence each other. To assess quality in terms of storage stability, the proportions of the different phases in the composition are therefore determined. To investigate and avoid these problems, the formulation is stored and its phases are examined during formulation development. Settling and dehydration shrinkage are both consequences of insufficient stability between two or more components, such as pigments and their solvents. What occurs is that one or more components are driven away (or detached) from another component, thus resulting in settling or dehydration shrinkage, depending on the specific gravity of the components.

[0004] Since the top liquid phase after dehydration and shrinkage is visually identifiable because it typically contains no pigment or has a low pigment concentration, it can be quantitatively assessed. However, the phase transitions of the colored and precipitated phases are usually not visually identifiable, making it difficult to easily quantify their amounts. Because the precipitated phase is generally not visually distinguishable from the colored phase, this method cannot be used to determine their amounts. Therefore, an attempt is made to detect the sediment phase using a spatula. To determine if a precipitated phase has formed, the spatula is scraped across the bottom of the sample glass (container). However, this only allows for very coarse characterization, such as no sediment phase, small or large amounts of precipitated phase, and / or soft or hard precipitated phase.

[0005] Due to the number of interdependent process parameters, the wide variety of coating compositions, pretreatment methods, and the surface type to which the composition is to be applied, it is currently impossible to predict whether a particular coating composition will provide a coating with acceptable storage stability quality. Therefore, storage stability quality can only be determined retrospectively at present.

[0006] Currently, storage stability is assessed visually by humans, such as employees. This purely visual assessment is often very coarse-grained, highly subjective, and difficult to reproduce. As a result, phase identification and surface coating quality assessment may require extensive employee experience, but can be extremely individualized, making it difficult to compare results. Furthermore, manual assessment of coating compositions is time-consuming and therefore expensive.

[0007] Overview

[0008] The object of this invention is to provide an improved method and corresponding system for characterizing phase separation of coating compositions, particularly aqueous, solvent-based, or solvent-free coating compositions, as expressly stated in the independent claims, and the resulting procedures and information for use in the environment of coating composition production. Examples are given in the dependent claims. The embodiments of this invention can be freely combined with each other, and are assumed not to be mutually exclusive.

[0009] In one aspect, the present invention relates to a method for detecting phase separation in aqueous, solvent-based, or solvent-free coating compositions. The method includes:

[0010] - The coating composition is provided in a container;

[0011] - Provide a measuring instrument for receiving the container, the measuring instrument including a measuring probe;

[0012] - Control the measuring instrument to

[0013] a) The measuring probe is moved through the coating composition along a predetermined measuring path at a predetermined speed curve, the predetermined measuring path extending along the length axis of the container.

[0014] b) Obtain a force-displacement curve by measuring the force applied to the measuring probe while the probe is being moved along a predetermined measurement path at a predetermined speed curve;

[0015] - Process the force-displacement curve to detect at least one phase separation of the coating composition; and

[0016] - Output the detection results.

[0017] The embodiments of the present invention have the following advantages: phase separation of the coating composition can be detected by obtaining the force-displacement curve.

[0018] The provided coating composition may be an aqueous, solvent-based, or solvent-free coating composition, which has been stored in a container at a specified temperature for a specified period of time. Storage conditions for the coating composition may include storage at 50°C for two weeks.

[0019] Force-displacement curves can be obtained in a reproducible, objective, and rapid manner because the measuring probe of the measuring instrument used to receive the container with the coating composition is moved through the coating composition along a predetermined measuring path at a predetermined speed curve.

[0020] The container is a container in which the coating composition is stored. The container may have a removable lid that can be removed (shortly) before the force-displacement curve is obtained.

[0021] Forces, particularly those proportional to dynamic pressure, are measured by moving a measuring probe through the coating composition along a predetermined measuring path. The predetermined measuring path is the route the measuring probe takes through the coating composition. This path is advantageously a straight path, in which the measuring probe is introduced from above and passes through the coating composition. The force measured is generated by the dynamic pressure exerted by the coating composition on the measuring probe.

[0022] The predetermined velocity profile for measuring probe displacement may include constant velocity, positive acceleration, and / or negative acceleration, or one or more portions of the velocity profile may include constant velocity and one or more portions of the velocity profile may include an acceleration. The force applied to the measuring probe is recorded as the probe moves along a predetermined measurement path at the predetermined velocity profile.

[0023] Force-displacement curves are obtained by measuring multiple forces applied to the measuring probe while the probe is being displaced along a predetermined measurement path at a predetermined velocity curve. The multiple measured forces can be visualized as curves.

[0024] The processing of force-displacement curves for detecting at least one phase separation of a coating composition can be performed manually or automatically by detecting changes in the recorded force-displacement curves.

[0025] For example, immersing a probe from ambient air into the coating composition causes a sudden increase in dynamic pressure, which in turn leads to a sudden increase in the recorded force-displacement curve. Therefore, a phase transition between the gas and liquid phases can be detected by the sudden increase in dynamic pressure. If no further increase in the force-displacement curve of the coating composition is detected, the output may indicate that there is no further phase separation in the coating composition, other than phase separation between the gas and liquid phases. Similarly, different liquid phases with different densities and therefore different dynamic pressures can be detected by the increase in the force-displacement curve. Furthermore, since the container, the amount of coating composition, the predetermined measurement path, and the predetermined velocity curve are likely known, the increase can be used to calculate which phase has which volume.

[0026] Embodiments of the present invention may be particularly advantageous in the context of the production of aqueous, solvent-based, or solvent-free coating compositions, such as paints, varnishes, printing inks, abrasive resins, pigment concentrates, and other coatings, since it was previously impossible to perform reproducible and objective characterization of the phases within the respective coating compositions.

[0027] Due to the large number of components and their interactions, the quality of a coating composition, including its storage stability, cannot be predicted. For example, dispersants (also known as dispersing aids or dispersing agents) are commonly used to disperse solids (such as pigments, fillers, or dyes) in liquid media to achieve efficient dispersion of the solids, reduce the mechanical shear forces required for dispersion, and simultaneously achieve the highest possible filler level. Dispersants support the breaking down of agglomerates, act as surfactants to wet and / or coat the surface of the solids or particles to be dispersed, and stabilize them to prevent unwanted re-agglomeration. In the manufacture of paints, coatings, printing inks, abrasive resins, pigment concentrates, and other coatings, dispersants facilitate the addition of solids such as pigments, dyes, and fillers, which, as important components, primarily determine the visual appearance and physicochemical properties of such systems.

[0028] Furthermore, in solvent-based formulations, for example, flowability can be adjusted by dissolving the molecular weight of the adhesive. In aqueous formulations, the adhesive is in the form of dispersed polymer particles, so the flow behavior cannot be adjusted by changing the molecular weight. Therefore, rheology modifiers must be used to adjust the flow properties of the aqueous coating composition.

[0029] The performance of a coating depends on different shear rates. Low viscosity is required during stirring (dispersion), but it is best to store it at a high viscosity to prevent pigment precipitation. For spray coating, the coating viscosity should be as low as possible, but once the coating is applied to the surface, the viscosity should be increased to prevent sagging on vertical substrates.

[0030] For optimal utilization, the solids must be uniformly dispersed in the composition, and once dispersed, they must stabilize. Currently, a wide variety of substances are used as dispersants, rheology modifiers, binders, and solvents, significantly influencing the optical properties, quality, and storage stability of coating compositions.

[0031] According to an embodiment, the processing of the force-displacement curve includes an evaluation of the force-displacement curve representing a specific characteristic of at least one phase separation.

[0032] The embodiments of the present invention may have the following advantages: the processing of force-displacement curves includes threshold calculation, change point detection, inflection point detection and / or application of the Ramer-Douglas-Peucker algorithm (RDP) or isolated forest for phase separation detection.

[0033] To detect phase separation, a threshold calculation can be used, where the threshold can be set manually, automatically assigned to dynamic pressure increases, or visually determined by analyzing force-displacement curves and detecting dynamic pressure increases. The threshold can be set based on the solvent used and / or other components of the coating composition.

[0034] Furthermore, change point detection can be used to determine phase separation. By detecting the locations of change points on a graph / line, the probability distribution of changes in a random process or time series can be identified. Additionally, Isolation Forest can be used to determine phase separation because it is an unsupervised learning algorithm for anomaly detection that works by isolating anomalies (in this case, changes). Furthermore, inflection point detection can be used to detect inflection points in the graph that bend inwards or outwards, referred to as knees or elbows. The Ramer-Douglas-Peucker algorithm (RDP) can also be used to determine phase separation because RDP is an algorithm that provides segment approximations, constructs approximate trajectories, and finds “valuable” inflection points in the force-displacement curve. RDP can be used to reduce the number of points in the force-displacement curve, which is approximated by a series of points.

[0035] The applicant has observed that thresholding, RDP, and change point detection are best suited for phase separation detection.

[0036] For example, phase separation of the coating composition can be detected from a force-displacement curve, which can be displayed in a graph where the height of the measuring probe in the container is shown on the X-axis and the measured force is shown on the Y-axis, as described below:

[0037] - The average force value is calculated for the first part of the measurement. The first part includes the recorded force measurements from the start of the measurement to the first increase in the first force-displacement pattern or force-displacement curve. In the first part, the measuring probe is not yet immersed in the sample, and the measurement is performed in air. Therefore, only background noise is measured.

[0038] - Move the recorded force-displacement curve (curve) by the previously calculated average value. Therefore, the noise may be near zero on the Y-axis.

[0039] - Determine the end of the first phase. The end of the first phase is defined as the first measurement point where the force exceeds a specified first threshold (e.g., 0.03 N).

[0040] - Starting from the end of the first phase and working backwards to the beginning of the measurement, the start of the first phase is determined using the recorded force-displacement curve (curve). The start of the first phase is defined as the first measurement point in that direction where the force is below a specified second threshold (e.g., 0.002 N).

[0041] The phase length is calculated based on the difference in X values ​​between these two thresholds (force-displacement modes).

[0042] Optionally or additionally, the determination of possible precipitate phases can be performed similarly to phase separation identification, as described in the preceding steps. The beginning (boundary) of the sedimentary phase can be detected when a defined threshold is exceeded. If the height of the bottom of the container is known, the X value (height) of the measurement point can be calculated as the difference in the amount of sedimentary phase.

[0043] Embodiments of the present invention may specify that the phase separation of the coating composition includes at least two liquid phases, namely a first phase and a second phase, wherein the second phase contains more filler and / or pigment (pigment phase) than the first phase (referred to herein as dehydration shrinkage).

[0044] Optionally or additionally, the determination of the possible colored phase (second liquid phase) can be performed in a manner similar to phase separation identification as described above. The start (boundary) of the colored phase can be defined as the end of the first phase (the starting point of the colored phase). The end of the colored phase can be defined as the start (if any) of the precipitate phase or the bottom of the container (the ending point of the colored phase). The length of the colored phase is calculated from the difference between these two points.

[0045] The ratio of the length of each phase to the total length of the coating composition can provide information about the stability of the coating composition. For example, the length of the first phase is 8 mm, and the total length of the coating composition is 20 mm. The proportion of the first phase is calculated to be 40% by volume, which indicates low storage stability of the coating composition.

[0046] Suitable fillers are, for example, those based on kaolin, talc, mica, other silicates, quartz, cristobalite, wollastonite, perlite, diatomaceous earth, fiber fillers, aluminum hydroxide, barium sulfate, glass, or calcium carbonate.

[0047] Preferably, the pigment is an organic or inorganic pigment or a carbon black pigment. Examples of inorganic pigments include iron oxide, chromium oxide, or titanium oxide. Suitable organic pigments are, for example, azo pigments, metal complex pigments, anthraquinone pigments, phthalocyanine pigments, and polycyclic pigments, especially thioindole, quinacridone, dioxazine, pyrrolopyrrole, naphthalenetetracarboxylic acid, perylene, isoaminefluorene, flavanone, pyrantrone, or isoviolanthrone series. The carbon black used can be gas black, lamp black, or furnace black. These carbon blacks can also be secondary oxidized and / or converted into beads.

[0048] Because the second phase contains more fillers and / or pigments, the dynamic pressure in the second phase is higher than that in the first phase, which may contain fewer fillers and / or pigments.

[0049] According to some embodiments, the second phase includes a liquid subphase and a precipitated subphase containing fillers and / or pigments.

[0050] The method may then include: stopping the acquisition of the force-displacement curve when the measuring force reaches or exceeds a predetermined limit, and the displacement of the probe when the predetermined limit is reached indicating the start of the precipitate subphase.

[0051] This method advantageously ensures that the measuring probe does not come into contact with or penetrate the precipitate phase. This reduces or eliminates the risk of damaging the measuring probe or measuring instrument.

[0052] In another embodiment, the method may include:

[0053] - Calculate the measure of the identifying phase used to provide qualitative and / or quantitative characterization of the coating composition, in particular the measure is

[0054] o Select a quantitative metric from the group including: the length of the measurement path between two detected phase boundaries, the travel time of the measurement probe between the two detected phase boundaries, the relative size of the detected phases, the number of detected phases; and / or

[0055] Qualitative metrics, particularly phase types selected from the group consisting of a gas phase, a first phase, and a second phase, wherein optionally the second phase contains more filler and / or pigment than the first phase, and / or optionally, the second phase comprises a liquid subphase and a precipitate subphase containing filler and / or pigment; and

[0056] - Qualitative and / or quantitative characterization of the output coating composition.

[0057] In another aspect, the present invention relates to a method for detecting phase separation in aqueous, solvent-based, or solvent-free coating compositions. The method includes:

[0058] - The force-displacement curves are processed by a phase separation identification procedure configured to identify force-displacement patterns, each of which is assigned to the boundary of a phase type; and

[0059] - Provides detection results for one or more phases identified by the phase separation identification program.

[0060] Embodiments of the present invention offer the advantage of providing phase characterization of coating compositions in a reproducible, objective, and rapid manner through a phase separation identification procedure. Individual phases in the coating composition are detected fully automatically and used to automatically calculate the coating composition characterization based on the type and / or extent (quantity) of one or more phases identified in a force-displacement curve analysis procedure. Therefore, a large number of force-displacement curves can be evaluated and labeled fully automatically using the automatically calculated characterization of the separately described coating compositions. This can be particularly useful in the context of high-throughput equipment used for testing and / or manufacturing coating compositions. The automated determination of coating composition characteristics improves the transparency and reproducibility of the evaluation of coating composition quality and other properties.

[0061] Embodiments of the present invention may have other advantages, as the automatically identified coating composition phases and the resulting characterization of the coating composition can be used as a database for performing many different forms of data analysis. In particular, computational characterization can be used as a qualitative and / or quantitative indicator of coating composition quality, coating composition storage stability, and / or the coating composition production process used to produce the coating composition.

[0062] Automated calculation of qualitative and / or quantitative coating composition characteristics allows for the automated analysis of large amounts of data and ensures the comparability of quality characteristics and / or coating production process parameters between different coating compositions. This is particularly useful when producing and testing many different types of coating compositions to identify those that are optimal for good storage quality.

[0063] Embodiments of the present invention may be particularly advantageous in the preparation of aqueous, solvent-based, or solvent-free coating compositions, such as paints, varnishes, printing inks, abrasive resins, pigment concentrates, and other coatings, because previously it was impossible to obtain reproducible objective characterization of the possible phase formations in the respective coating compositions. Due to the large number of components and their interactions, the quality of the coating composition could not be predicted.

[0064] According to some embodiments, the coating composition characterization provided by the phase separation identification program includes fine-grained quantitative characterization, such as values ​​within a continuous range or a set of at least 10 different predetermined values ​​or value ranges. The phase separation identification program is configured to convert the fine-grained quantitative characterization into a coarse-grained quantitative characterization so that the automatically calculated characterization is comparable to an existing coarse-grained dataset. For example, the existing coarse-grained dataset may be manually created. The coarse-grained quantitative characterization may be values ​​within a set of fewer than 10 different predetermined values ​​or categorical values ​​or value ranges. By mixing manually labeled and automatically labeled force-displacement curves of the coating composition that demonstrate comparability with each other, the automatic conversion can increase the data foundation for various data analyses or machine learning.

[0065] According to an embodiment, the method includes calculating a measure of the identification phase used to provide qualitative and / or quantitative characterization of the coating composition through a phase separation identification procedure; and outputting the qualitative and / or quantitative characterization of the coating composition.

[0066] For example, a coating composition may contain one or more phases of different phase types and in varying degrees. Qualitative and / or quantitative measures, such as those calculated for each phase, can be used to obtain qualitative and / or quantitative characterization of the coating composition, which integrates automatically obtained measures for automatically identifying phases in the coating composition.

[0067] The automatic identification and measurement of individual phases offer the advantage of objectifying the number, type, quantity, and other characteristics of each individual phase in the coating composition. Embodiments of the present invention are far less subjective than prior art methods based on manual / visual assessments of individual phases and / or the quality of the coating composition by employees. Visual assessments are not quantitative and can only provide a very rough classification of results based on coarse-grained quality grades or coarse-grained grading systems.

[0068] For phase separations where different phases can be visually identified, current methods allow for quantitative or semi-quantitative analysis of the amount of each phase by measuring its height. However, this is a time-consuming and manual process. Furthermore, phase boundaries, especially for stained or filled phases, are not always clearly measurable, which reduces detection accuracy and may even lead to situations where phase separation cannot be visually detected.

[0069] According to an embodiment, the measurement of a phase is or includes quantitative measurements selected from the group consisting of: the length of a measurement path between two adjacent detected force-displacement patterns, the travel time of a measurement probe between two adjacent detected force-displacements, or computational measurements such as the relative height of different phases or the number of detected force-displacement patterns.

[0070] Additionally or alternatively, the measure is or includes a qualitative measure. A qualitative measure may in particular identify the type of phase. For example, the phase may be selected from the group consisting of a gas phase, a first liquid phase, and a second liquid phase. The second liquid phase may contain more filler and / or pigment than the first liquid phase. The second liquid phase may include a liquid subphase and a precipitate subphase containing filler and / or pigment.

[0071] According to an embodiment, the method further includes:

[0072] - Provides aqueous, solvent-based, or solvent-free coating compositions in containers;

[0073] - Provides a measuring instrument for receiving containers, the measuring instrument including a measuring probe;

[0074] - A positioning measurement probe relative to the coating composition in the container, especially above it;

[0075] - Control the measuring instrument to move the measuring probe through the coating composition at a predetermined speed along a predetermined measuring path, which extends along the length axis of the container.

[0076] - The force-displacement profile is obtained by moving the measuring probe through the coating composition by measuring the force applied to the measuring probe while moving the probe along a predetermined measurement path at a predetermined speed profile.

[0077] The features of the aqueous, solvent-based, or solvent-free coating composition, container, measuring instrument, measuring probe, predetermined measuring path, predetermined velocity profile, and force-displacement profile may be the same as those disclosed above for the method for detecting phase separation of aqueous, solvent-based, or solvent-free coating compositions.

[0078] The embodiments of the present invention that control the predetermined measurement path and predetermined velocity curve of the measurement probe have the following advantages: the force-displacement curve can be provided as input to the phase separation identification program. In the case of obtaining the phase separation identification program based on a machine learning method, the embodiments of the present invention that control the predetermined measurement path and predetermined velocity curve of the measurement probe can ensure that the conditions used to obtain the force-displacement curve are similar to the conditions used to obtain the training force-displacement curve from which the phase separation identification program is derived. By controlling the force-displacement curve acquisition process, the embodiments can ensure that the acquired force-displacement curves are comparable and can be reproducibly and accurately processed by the phase separation identification program.

[0079] According to the embodiment, the processing of the force-displacement curve further includes:

[0080] - Through a phase separation identification procedure, force-displacement curves are processed using methods selected from thresholding, change point detection, isolated forest, knee / elbow detection, and / or the Ramer-Douglas-Peucker algorithm (RDP), thereby automatically labeling the force-displacement pattern with the phase type and instances of that phase in the force-displacement curve; and

[0081] - Output one or more identified phase instances.

[0082] For example, the identified force-displacement pattern can be output as coordinates.

[0083] Outputting coordinates (e.g., in the form of x and y coordinates of force-displacement patterns forming phase boundaries) may have the advantage that these coordinates can be easily further processed by a phase separation identification program, for example, for calculating the fraction of phases in the force-displacement curve. On the other hand, providing a graphical representation of pattern instances may have the advantage that the identified phases can be easily recognized by humans. For example, a phase separation identification program can generate a graphical user interface (GUI) configured to display image fragments that have been identified as representing specific types of phases through their respective force-displacement patterns. Providing a combination of coordinate information and graphical representation may have the advantage that the output of the phase separation identification program can be easily processed and understood by both software and humans.

[0084] According to an embodiment, the method further includes installing and / or instantiating a phase separation identification program on a data processing system including a graphical user interface (GUI), the data processing system being effectively coupled to a measuring probe of a measuring instrument for receiving a container. The phase separation identification program can be configured to generate the GUI, which is displayed to the user via the screen of the data processing system. In response to user actions via the GUI, the phase separation identification program acquires a force-displacement curve of the coating composition via the measuring probe. The acquired force-displacement curve is processed by the phase separation identification program to automatically identify the phase type, calculate a measure of the phase type, and calculate a qualitative and / or quantitative characterization of the coating composition. The phase separation identification program then executes the output of the calculated characterization via the GUI or another output interface of the data processing device.

[0085] According to an embodiment, the phase separation identification procedure is selected from the group consisting of:

[0086] - Applications, portable data processing devices, or stationary devices specifically designed for detecting phase separation; for example, specially designed quality control devices can include additional components such as measuring probes whose position relative to the coating composition can be checked by a phase separation identification procedure; embodiments of the present invention can ensure that these devices can automatically identify and characterize the coating composition phase and the quality of the coating composition under test in a reproducible and accurate manner.

[0087] - The application and data processing equipment are high-throughput (HT) devices (also known as HTE) for automated or semi-automated coating production; in particular, high-throughput devices can be devices that include automated force-displacement curve acquisition units as described herein with reference to embodiments of the invention; the use of phase separation identification programs in an HT device environment may be particularly advantageous because HT devices are capable of automatically producing and testing many different coating compositions, thereby generating a large amount of data, which can then be used to train machine learning programs to identify and / or predict coating compositions with desired coating quality characteristics; the automated generation and storage of qualitative and / or quantitative coating characterizations in an HT device environment allows coating phase and coating quality characterization to be considered in various big data applications, particularly in machine learning-based predictions; this is not possible based on subjective and inconsistent quality labels generated manually;

[0088] - Web applications that are downloaded and / or instantiated permanently or temporarily over a network; for example, a server may provide a Java application that implements a phase separation identification program via the Internet; or

[0089] - Programs executed within a browser, such as JavaScript programs. Regarding web applications and browser programs, the phase separation identification program can be implemented as a client-server system, wherein the client portion, instantiated on a data processing system, is responsible for acquiring force-displacement curves with sufficient quality and under appropriate conditions, while the server portion, instantiated on a remote server computer, is responsible for performing force-displacement curve analysis to identify the phase and provide characterization of the coating composition.

[0090] According to an embodiment, the phase separation identification procedure includes a predictive model that has learned to identify predetermined patterns from training data in a training step employing a machine learning procedure. In particular, the machine learning procedure may be a neural network.

[0091] According to an embodiment, the training data includes training force-displacement curves for multiple labeled coating compositions. For example, the training force-displacement curves may include force-displacement curves for many different coating compositions, where different coating compositions have been obtained by combining different types and / or amounts of components and / or by combining said components according to different production process parameters (such as mixing duration, mixing temperature, mixing speed, etc.). Therefore, the training data can cover a vast multidimensional data space, encompassing many different coating compositions and coating composition production parameters. The method also includes storing the training data in a database. Initially, the labels on the training data will be manually annotated. In subsequent training steps, the training data can be expanded using additional force-displacement curves of the coating compositions, which have been automatically labeled and preferably checked or corrected by an annotator. The labels preferably include phase boundaries, phase types, and / or one or more other characteristics of the coating composition, which in some cases may be the same measure as one or more phases included in the respective force-displacement curves.

[0092] The steps described above illustrate how an existing and / or trained phase separation identification program can be applied to a new (test) force-displacement curve that does not contain labels indicating the presence, type, or degree / quantity of an indicative phase. Embodiments of methods for generating and training predictive models for phase recognition software are described below. The test phase and training phase can be executed on the same or different data processing systems. For example, a model M1 can be trained on a first data processing system, the trained model M1 integrated into a phase separation identification program that may include some additional functions or modules, for example, to interact with a user and / or equipment used for producing or testing coatings, and the phase separation identification program can be transferred to a second data processing system.

[0093] Training phase of the phase separation recognition procedure model (M1)

[0094] According to an embodiment, the method includes performing a training step on training data, which includes a set of labeled digital training force-displacement curves of the coating composition, the labels identifying the position / location and type of the phases of the training force-displacement curves. A predictive model is trained using backpropagation through the labeled training force-displacement curves to identify patterns.

[0095] Providing a phase separation identification procedure during machine learning may have the following advantages: the generated predictive model will learn multiple highly complex interrelationships between several different factors, including the type and / or quantity of coating components and / or coating composition production parameters.

[0096] According to an embodiment, each training force-displacement curve has been assigned additional data processed during the training step to allow the predictive model to correlate the additional data with force-displacement patterns (and with quantification of the phase and / or qualitative and / or quantitative characterization of the coating composition). The additional data includes contextual data, which includes:

[0097] - One or more components of a coating composition for generating a coating composition to which a trained force-displacement profile has been obtained; the specifications of one or more components of the coating composition may include the type and / or amount of the component; for example, component-related information provided as contextual data may include the type and / or amount of a dispersant and / or the type or amount of a rheology modifier and / or the type or amount of one or more pigments and / or the type and amount of a solvent; and / or

[0098] - One or more production process parameters, which characterize the process of producing the coating composition, including, for example, the mixing rate, mixing temperature, and / or mixing duration of the coating composition; and / or

[0099] - System parameters for a pressure measurement system used to obtain training force-displacement curves. The system parameters are selected from the following group, which includes the temperature type of the coating composition, the measuring probe, the measuring probe sensitivity, the measuring path length, the measuring probe velocity when the probe moves along the measuring path, and the measuring probe velocity curve when the probe moves along the measuring path.

[0100] Training the predictive model to be integrated into the phase separation identification procedure based on the aforementioned contextual data may be advantageous, as the applicant has observed that the aforementioned contextual parameters can all affect the quantity and type of phases to be observed in the coating composition, thus affecting the quality of the coating composition. Labeling the training force-displacement curves with the aforementioned contextual data and / or phase quantification measures ensures that the trained predictive model can account for any factors that may affect the phase type and extent, as well as the characterization of the coating composition quality.

[0101] Other embodiments

[0102] In another aspect, the present invention relates to a computer-implemented method for providing coating composition-related prediction programs, such as composition quality prediction programs and / or coating composition specification prediction programs. The method includes:

[0103] - Provide a database that includes qualitative and / or quantitative characterization of coating compositions and their association with parameters selected from one or more of the following groups: one or more components of the coating composition, the relative and / or absolute amounts of the one or more components, and / or process parameters of the coating composition.

[0104] - A machine learning model is trained based on the association between coating composition characterizations in a database and one or more parameters to provide a predictive model (M2, M3) that has learned to correlate qualitative and / or quantitative characterizations of one or more coating compositions with their respective coating components and / or production process parameters used to generate the coating compositions; and

[0105] o provides a composition quality prediction program that includes a prediction model (M2), configured to predict coating composition properties from one or more input parameters selected from the group consisting of one or more components of the coating composition, relative and / or absolute amounts of one or more said components, and / or production process parameters, including the detection of phase separation; and / or

[0106] o provides a composition specification prediction program including a prediction model (M3) configured to use the prediction model (M3) to predict based on inputs that at least explicitly specify the required storage stability characteristics and perhaps one or more additional parameters related to the desired coating composition (components, process parameters, application parameters), and output one or more parameters related to the predicted coating composition having the storage characterization of the inputs and optionally satisfying the additional parameters as inputs, the one or more parameters being selected from the group consisting of one or more components of the coating composition, the relative and / or absolute amounts of one or more of the components and / or the production process parameters to be used to prepare the coating composition.

[0107] Embodiments of the present invention offer the advantage of providing a composition quality prediction program that can automatically predict the quality characterization of a specific coating composition based on the type and / or amount of the composition components and perhaps other parameters. This can significantly accelerate the testing and identification process of coating compositions that provide the desired quality characterization. Unlike prior art methods that rely on human experience and typically involve the manufacture and testing of large quantities of coating compositions on a workbench, embodiments of the present invention can allow for the precise prediction of whether a particular coating composition will possess characteristics of other coating compositions. This can significantly accelerate the process of identifying suitable coating compositions and reduce the costs associated with the reagents, machinery, and consumables required to perform this identification process.

[0108] The predictive model M1 of the phase separation identification procedure is preferably obtained by performing a machine learning step on manually labeled training-force-displacement curves, and has learned to correlate force-displacement patterns with phase type characterization metrics and coating composition characterization. The predictive model M2 of the composition quality prediction procedure can be trained on training data, which may or may not include force-displacement curves. The purpose of predictive model M2 is to predict the coating composition characterization, particularly quality-related characterization, in relation to the contextual parameters of the one or more components and perhaps the coating composition.

[0109] According to an embodiment, the method includes providing a plurality of force-displacement profiles, each relating to a coating composition. The coating compositions each have one or more phases of various different types. The method includes applying a phase separation identification procedure to the force-displacement profiles to identify force-displacement patterns in the force-displacement profiles, obtaining a measure of the phase represented by the identified force-displacement patterns, and calculating a qualitative and / or quantitative characterization of the coating composition represented by the force-displacement profiles. The method also includes storing, in a database, the qualitative and / or quantitative characterization of the phases (optionally, also the qualitative and / or quantitative measures of individual phases) in association with one or more parameters designed to produce coating compositions containing these phases, in order to provide training data for predictive models (M2, M3). For example, these parameters may be selected from the group consisting of: one or more components of the coating composition, the relative and / or absolute amounts of one or more of said components, and / or process parameters for the production of the coating composition.

[0110] This can be advantageous because automated phase identification and coating composition characterization can automatically allow for the annotation of a large number of force-displacement curves in a reproducible and comparable manner. Providing a large set of impartial training data ensures that the model M2 trained on this data set can accurately predict the performance of the coating composition.

[0111] According to the embodiments, the use of the composition quality prediction procedure includes:

[0112] - Provides each of many specifications of numerous different candidate coating compositions as input to the composition quality prediction procedure;

[0113] -Predict coating quality for each candidate coating composition using a composition quality prediction procedure;

[0114] - Select a candidate composition based on the respective predicted measures; and

[0115] - As a specification for predicting the recommended coating composition with the highest quality output among the selected candidate coating compositions; and / or

[0116] - Input the selected candidate coating composition specifications to the processor, which controls the equipment for producing and / or testing the composition for the coating composition, wherein the processor drives the equipment to produce the input coating composition.

[0117] For example, the prediction is generated by a computer system connected to equipment used for producing and / or testing coating compositions. This equipment may be an HT device.

[0118] According to an embodiment, the method further includes:

[0119] - Receive incomplete specifications of the coating composition;

[0120] - A set of candidate coating composition specifications is generated manually or automatically. This set of candidate compositions consists of different versions of the received incomplete coating compositions. The generation of these candidate coating composition specifications includes:

[0121] a) Supplement the specifications of the incomplete coating composition with one or more additional components; and / or

[0122] b) Supplement the specifications of the incomplete coating composition with different absolute or relative amounts of one or more of the other components and / or with different absolute or relative amounts of the explicitly specified components; and / or

[0123] c) Changing the amount of one or more components given in the incomplete coating composition; and / or

[0124] d) Supplement the specifications of the incomplete coating composition with one or more applied process parameters that characterize the process of producing the candidate coating composition on the substrate.

[0125] For example, the received incomplete specifications may clearly specify that the (desired) coating composition should include a specific dispersant and a specific pigment, but not their absolute or relative amounts. Candidate compositions can be generated by using different ratios of dispersant and pigment.

[0126] As another example, an incomplete specification may explicitly state that a particular pigment or combination of pigments should be used, but does not mention the amount of pigment or only provides a range of pigment amounts. Candidate compositions may be coating compositions that differ from each other in terms of pigment amount.

[0127] According to another example, an incomplete specification may explicitly specify all or at least most of the components of the coating composition and their respective amounts, but fails to mention contextual parameters, particularly manufacturing process parameters such as mixing time, mixing duration, mixing temperature, etc. Candidate compositions may be coating compositions that differ from each other in terms of the values ​​of one or more of the aforementioned contextual parameters.

[0128] According to another example, an incomplete specification may clearly specify all or at least most of the components of the coating composition and their respective amounts, but does not mention its phase separation properties. Candidate compositions may be coating compositions that differ from one another in terms of different amounts or different types of rheology modifiers.

[0129] According to the embodiments, in each of the following cases a), b), c) and / or d), additional data used to supplement the candidate coating composition specifications are used as input to the coating composition quality prediction program to perform predictions.

[0130] The above-described embodiments, based on providing multiple (hypothetical) candidate coating composition specifications, predicting coating quality, and selecting one candidate coating composition, appear to offer the most desirable predictive performance. However, an alternative approach could employ learned correlations between coating storage stability characteristics and coating composition-related parameters, which have been more directly integrated into the predictive model M3 to identify promising coating compositions.

[0131] According to an embodiment, the use of a coating composition specification prediction program includes providing a composition specification prediction program comprising a prediction model (M3) configured to use the prediction model (M3) to predict and output one or more parameters relating to a coating composition for producing a coating composition having an input storage characterization and optionally satisfying the additional parameters as input, based on at least explicitly defined desired storage stability characterization and optionally one or more additional parameters relating to the desired coating composition (components, process parameters, application parameters).

[0132] According to a preferred embodiment, the method further includes outputting a predicted specification of the coating composition to a person and / or inputting the specification of a selected candidate coating composition to a processor, the processor controlling an apparatus for producing and / or testing a composition for the coating composition, wherein the processor drives the apparatus to produce the input coating composition.

[0133] In addition to the desired storage stability, additional constraints can be provided as input. These constraints may consist of an incomplete, rough coating composition specification that indicates certain components or component material categories and certain absolute or relative amounts. Constraints may include requiring any suggested substitute components to belong to the same material category or ensuring that any substitution amount does not deviate from the amount specified in the constraints by more than a maximum threshold. Production process parameters and / or applied process parameters may also be provided as constraints.

[0134] According to an embodiment, the method is executed on a computer system effectively connected to an automated container receiving and force acquisition unit (RRFMA unit). The RRFMA unit includes a measuring probe. The RRFMA unit is part of an apparatus for producing and / or testing coating compositions, or is effectively connected to the apparatus via, for example, an automated transport device for automatically transporting coated surface samples back and forth relative to the RRFMA unit. The method also includes sending one or more control commands to the apparatus. The control commands cause the apparatus;

[0135] - Transport the produced coating composition to the RRFMA unit;

[0136] o Position the measuring probe relative to the coating composition inside the container, especially above it;

[0137] o Control the measuring instrument to move the measuring probe along a predetermined measuring path at a predetermined speed curve through the coating composition; the predetermined measuring path extends along the length axis of the container.

[0138] o To obtain a force-displacement curve as the measuring probe moves through the coating composition, the force applied to the measuring probe is measured while the probe is moving through the coating composition along a predetermined measurement path at a predetermined speed curve.

[0139] The acquired force-displacement curve is returned to the computer system.

[0140] This could be advantageous because it provides a fully automated system for predicting, producing, and testing compositions in automated production equipment, particularly HT equipment. Data obtained from force-displacement curve analysis based on test steps can be used to extend the training data of the composition quality prediction program, and the model M2 of the composition quality prediction program can be retrained based on the extended training data to obtain an improved version of model M2.

[0141] According to an embodiment, active learning is used in the training of a predictive model M1 to identify force-displacement patterns in force-displacement curves of a coating composition and thus a phase. In this case, active learning identifies a subset of one or more unlabeled test force-displacement curves of the coating composition for which manual annotation will provide the highest learning performance. The active learning module prompts the user to manually annotate (assign labels) each force-displacement curve in the set, where the labels indicate the type and location of the phase depicted. The one or more additionally annotated force-displacement curves are added to the training force-displacement curves, thereby expanding the training data. The predictive model M1 is retrained based on the expanded training data, thereby providing a more accurate improved version of the predictive model M1. Thus, the outdated predictive model in the phase separation identification process is replaced by the improved model.

[0142] In another aspect, the present invention relates to a system comprising:

[0143] - Equipment for producing and testing compositions for paints, varnishes, printing inks, abrasive resins, pigment concentrates, or other coatings, wherein the equipment comprises at least two processing stations, wherein the at least two processed transport systems are connected to each other, and self-propelled transport vehicles are capable of moving on the transport systems to transport components of the composition and / or the produced composition between the processing stations, and

[0144] - A computer system configured to perform a method as described in one of the embodiments herein.

[0145] In another aspect, the present invention relates to a computer program configured to perform the methods of any of the embodiments described herein. This computer program may be a computer program product.

[0146] In another aspect, the present invention relates to a phase separation identification procedure provided by executing a method for providing and using a phase separation identification procedure as described herein with respect to embodiments of the invention.

[0147] In another aspect, the present invention relates to a composition quality prediction program provided by performing a method for providing and using a composition quality prediction program as described herein with reference to embodiments of the invention.

[0148] In another aspect, the present invention relates to a composition specification prediction program, which is provided by performing a method for providing and using a composition specification prediction program as described herein with respect to embodiments of the invention.

[0149] On the other hand, the present invention relates to a coating composition produced according to a composition specification provided by performing any of the methods described herein for providing the coating composition specification. In particular, the specification can be calculated using a trained model (such as M2 or M3) based on a desired storage stability characterization provided as input by a coating composition specification prediction program or a composition quality prediction program according to embodiments of the invention.

[0150] In another aspect, the present invention relates to a volatile or non-volatile data storage medium comprising computer-readable instructions for implementing a phase separation identification procedure, a composition quality prediction procedure, and / or a composition specification prediction procedure.

[0151] In another aspect, the present invention relates to a volatile or non-volatile data storage medium comprising the above-described specifications, the data storage medium being effectively coupled to a device via an interface, the device being configured to produce a coating composition according to one or more specifications stored in the data storage medium.

[0152] In another aspect, the present invention relates to a measuring instrument for obtaining a force-displacement curve by measuring the force applied to a measuring probe as the probe is moved along a predetermined measuring path at a predetermined speed profile through a coating composition. The measuring instrument is configured to perform the method as described herein with respect to embodiments of the invention.

[0153] As used herein, the term "composition" or "coating composition" refers to a composition comprising two or more raw materials (components) formed from said raw materials. When references are made in the context of this application to the production or testing of a composition by automated equipment, it is understood that the product is produced based on information regarding the properties and / or amounts of the components.

[0154] As used herein, “coated surface” refers to a substrate surface that has been coated once or multiple times with a coating composition. For example, the coating composition can be applied by spreading, spraying, or coating the substrate, by immersing at least one surface of the substrate in the coating composition, or by other coating methods.

[0155] As used in this article, “program” is a type of software such as an application or a module or function of an application, or a script, or any other type of software code that can be executed by one or more processors such as a CPU or GPU.

[0156] The “phase separation identification program” used in this article is a software program or software module configured to analyze force-displacement curves to automatically identify one or more phases and calculate coating composition characteristics based on the identified phases plotted in the force-displacement curves.

[0157] The "force-displacement curve" depicts multiple recorded forces applied to the measuring probe as it moves along a predetermined measurement path at a predetermined speed through the coating composition. These multiple recorded measuring forces can be visualized as curves in a graph.

[0158] The “composition quality prediction program” used herein is a software program or module configured to receive a (complete or incomplete) specification of a coating composition and to predict one or more properties of the coating composition based on data detailed in that specification. The properties of the coating composition may include coating composition quality indicators, such as indicators of the coating surface quality obtained by applying the coating composition to a substrate. Quality indicators may, for example, be the likelihood of certain types of coating defects occurring.

[0159] The “composition specification prediction program” used herein is a software program or module configured to receive a desired coating surface characterization as input parameters. Optionally, the composition specification prediction program may be configured to receive one or more other input parameters, such as an incomplete coating composition specification to limit the solution space of the prediction. The composition specification prediction program is configured to predict one or more of the following output parameters based on the received input data: one or more components, one or more absolute or relative amounts of components, one or more production process parameters, and / or one or more application process parameters.

[0160] As used herein, a “composition specification” is a set of data containing parameters related to a coating composition. For example, these parameters may specify the identity and / or substance class of some or all of the components to be combined to produce the coating composition. Optionally, the specification may include additional parameters, such as relative or absolute amounts or ranges of the components, coating composition manufacturing process parameters, coating application process parameters, etc. These parameters may explicitly specify how the components must be handled and / or mixed to obtain the composition and / or how the composition should be applied to a substrate to obtain a specific coated surface. A composition specification may be complete or incomplete. For example, some specifications may only specify the type of component (e.g., a solvator based on an organic solvent) without specifying the exact identity and / or amount of the component. A composition “specification” may be provided in various forms, such as as a printout, a document such as an XML file, an object in an object-oriented programming language such as a JSON file, etc.

[0161] As used herein, “coating composition manufacturing process parameters” or “manufacturing process parameters” are parameters that indicate the characteristics of the process of processing and / or combining components to form a coating composition. Examples include mixing duration, mixing speed, mixing temperature, component mixing sequence, and equipment used for mixing or otherwise preparing the coating composition.

[0162] As used herein, a “known composition” is a composition whose properties (e.g., coating surface characteristics, rheology, elasticity, shelf life, etc.) of a specified product are known to the person or organization performing the neural network training. For example, a known composition may have been manufactured for a customer months or years ago, and the product’s performance has been empirically determined. Measurements do not necessarily have to be performed by the laboratory operator who now determines the predictive composition, but may be performed and published by other laboratories; therefore, in this case, the performance is derived from professional literature. Since compositions as defined above also include formulations as a subset, a “known composition” according to embodiments of the invention may also include a “known formulation” or simply a “known formula.”

[0163] As used in this article, "database" refers to any volatile or non-volatile data storage medium that stores data, especially structured data. A database can be one or more text files, spreadsheet files, directories in a directory tree, or a database of a relational database management system (DBMS) such as MySQL or PostgreSQL.

[0164] The "loss function" used in this paper for the prediction problem is the following function, which is used in the training of predictive models (such as models of neural networks) using machine learning procedures to train and improve the model. The loss function outputs a value whose magnitude indicates the quality of the predictive model. Here, the loss function is minimized during training because the magnitude of this value indicates the inaccuracy of the predictive model's predictions.

[0165] As used herein, "equipment" for producing and testing compositions refers to a device or system comprising multiple laboratory instruments and transport units capable of coordinated joint control of the laboratory instruments and transport units to perform automated or semi-automated workflows. Workflows may be, for example, coating composition preparation workflows (e.g., combination and mixing workflows), analytical workflows, or combinations of two or more of these workflows. Workflows may include automated preparation and / or automated storage of coating compositions and / or automated recording of force-displacement profiles and / or application of the composition to one or more substrates. The equipment may, for example, be a high-throughput device (HT device), also known as a "high-throughput equipment" (HTE).

[0166] "Testing" or "analyzing" a coating composition using automated production and / or testing equipment is the process of analyzing the chemical, physical, mechanical, optical, or other empirically measurable properties of the composition using one or more analytical modules. For example, testing may include acquiring and analyzing the force-displacement profile of the coating composition and calculating a quality measure of the coating based on one or more phases detected in the composition. The analysis may also include measuring other object properties such as opacity, elasticity, rheology, color, etc.

[0167] The “active learning module” used in this paper is a software program or a module of a software program designed to select a (relatively small) small set of candidate compositions from a set of candidate candidates, thereby exhibiting a strong learning effect after the properties of the selected candidate compositions are prepared and empirically measured, because these data are taken into account in training the predictive model.

[0168] As used herein, "model" or "predictive model" refers to a data structure or executable software program or program module configured to generate predictions based on input data. For example, the model may be one obtained during machine learning by training the model on manually labeled and / or automatically labeled training data. Predictive models may include, for example, neural network models, support vector models, random forests, decision trees, etc. According to embodiments of the invention, a predictive model suitable for calculating the characterization of a coating composition based on obtained force-displacement curves regarding the presence, location, and / or extent of one or more phases is also referred to as an "M1" model. A predictive model suitable for predicting the properties of a coating composition based on one or more input parameters, such as those relating to the components, component amounts, production process parameters, and / or applied process parameters of the composition, is also referred to as an "M2" model. A predictive model suitable for predicting one or more parameters relating to, such as, the components, component amounts, and / or production process parameters of the coating composition, based on input data that explicitly specifies the desired performance of the coating composition, is also referred to as an "M3" model. Attached Figures

[0169] The embodiments of the present invention will be explained in more detail below, with reference to the figures including the embodiments, wherein:

[0170] Figure 1A The force-displacement curves (curves) of the coating composition are shown;

[0171] Figure 1B Show Figure 1A The force-displacement curve shows the threshold (boundary) of the phase;

[0172] Figure 2 A flowchart illustrating a method for automated characterization of coating compositions;

[0173] Figure 3 A flowchart of a method for automated characterization of coating compositions is shown in more detail;

[0174] Figure 4 A flowchart illustrating the method for obtaining force-displacement curves is shown.

[0175] Figure 5 A block diagram of a data processing system for automated surface coating characterization is shown.

[0176] Figure 6AA data processing system in the form of a smartphone containing web applications is shown;

[0177] Figure 6B This illustrates a data processing system in the form of a customized phase identification quality inspection device;

[0178] Figure 6C A data processing system in the form of a computer is shown, which is connected to an apparatus for producing a coating composition;

[0179] Figure 7 Force-displacement curves of coating compositions containing automatically identified and labeled force-displacement patterns are shown, with each force-displacement pattern assigned to the boundary of the phase type. Detailed description

[0180] Figure 1A The subregion of the force-displacement curve 1202 obtained for the coating composition is shown.

[0181] For example, the force-displacement curve may have been obtained using a measuring probe in a force-displacement curve acquisition unit of equipment used for automated production and / or testing of coating compositions. Alternatively, the force-displacement curve may have been obtained, for example, by referencing Figures 6B-6C The data is obtained using the measurement probes of the aforementioned data processing system.

[0182] Figure 1A The force-displacement curve 1202 shown was obtained using a measuring head DSR301 with a measuring probe PP25. The sample was fixed in a 100 mL glass container on a shuttle stage. The fill level of the composition within the container (glass) could be varied. The measuring probe was moved into the sample at the center, and the force required to move the sample was measured. The parameters initial height [mm], final height [mm], and speed [mm / s] could be changed. A common parameter set was an initial height of 400 mm, a final height of 5 mm, and a speed of 1 mm / s.

[0183] The recorded force-displacement curve 1202 can be used to illustrate phase separation detection. Since the measuring probe is moved into the sample from above, the measurement process in the curve is from right to left (decreasing height).

[0184] To generate a sufficiently large training dataset, force-displacement curves of many different coating compositions containing many different types of phases were obtained.

[0185] Preferably, a large number (e.g., thousands) of force-displacement curves showing different phase types are obtained, which can be manually labeled (marked).

[0186] Figure 1B Show Figure 1AThe force-displacement curve 1202 includes the determined phase boundaries (force-displacement patterns) 1212 and 1214. Force-displacement patterns 1212 and 1214 have been automatically labeled with triangles using thresholding operations, each indicating an identified force-displacement pattern.

[0187] For example, phase separation of the coating composition can be detected from the force-displacement curve shown. The height of the measuring probe inside the container is displayed on the X-axis, thereby causing the measuring probe to move through the coating composition at a predetermined speed along a predetermined measurement path, and the measuring / recording force is displayed on the Y-axis. Since the height of the measuring probe is reduced during the measurement process, the measurement process is shown from right to left in the graph. The detection is performed using the following steps:

[0188] - Calculate the average value of the recorded forces used in the first measurement section. The first section includes the measured force values ​​recorded from the start of the measurement to the first increase in the first force-displacement pattern 1214 or force-displacement curve. In the first section, the measuring probe has not yet been immersed in the coating composition sample and the measurement is performed in air. Therefore, only background noise is measured.

[0189] - The recorded force-displacement curve (curve) is shifted by using the previously calculated average value. Therefore, the noise may be near zero on the Y-axis.

[0190] - Determine the end of the first phase (force-displacement pattern 1212 in the figure). The end of the first phase is defined as the first measurement point where the force exceeds a determined first threshold (e.g., 0.03 N).

[0191] - Starting from the end of the first phase and rewinding to the beginning of the measurement, the start of the first phase is determined using the recorded force-displacement curve (curve). The start of the first phase (force-displacement pattern 1214 in the figure) is defined as the first measurement point in that direction, at which the force decreases to below a determined second threshold (e.g., 0.002 N).

[0192] - Calculate the phase length from the difference of the X value (1216) of these two thresholds (force-displacement modes 1212, 1214).

[0193] Figure 2 A flowchart illustrating a method for detecting phase separation in aqueous, solvent-based, or solvent-free coating compositions is shown. In a first step 102, a phase separation identification procedure processes the force-displacement curves recorded for the coating composition. The phase separation identification procedure identifies one or more phases in step 102 and provides characteristics of the identified phases. For example, the procedure may determine that the coating composition comprises two phases. The characteristics of the identified phases may include the type, location, and extent of the identified phases. The data obtained in step 104 may be output to a user and / or may be used internally by the phase separation identification procedure to calculate derived data values, such as for characterizing polymeric coating compositions.

[0194] Figure 3 A flowchart of a method for automatically characterizing coating compositions is shown in more detail. Following steps 102 and 104, a phase separation identification procedure calculates values ​​for individual phases, such as volume and location, in step 106. These measurements are used in steps 106 and 108 to calculate and provide qualitative and / or quantitative characterization of the coating composition.

[0195] Figure 4 A flowchart illustrating a method for obtaining force-displacement profiles of aqueous, solvent-based, or solvent-free coating compositions is shown. Prior to step 110, various coating compositions can be produced by mixing multiple components with each other according to a mixing and production protocol. In step 110, the coating composition is provided in a container. Furthermore, a measuring instrument, including a measuring probe, is provided for receiving the container. The sample is automatically or manually delivered to a force acquisition unit. In step 112, the measuring probe is positioned relative to the coating composition within the container. In steps 114 and 116, the measuring probe is moved through the coating composition in a prescribed manner so that force-displacement profile analysis software 124 can correctly analyze the force-displacement profile. Thus, one or more force-displacement profiles describing the phase are obtained.

[0196] Figure 5 A block diagram of a data processing system 120 for automated coating characterization is shown. The data processing system includes one or more processors 126 and volatile or non-volatile storage medium 122. The storage medium may include force-displacement curves 125, such as training force-displacement curves for training a phase separation identification program 124 for a model M1, or test force-displacement curves 125 to be input into a trained predictive model M1. Additionally or alternatively, the storage medium may include training data for training a composition quality prediction program for a model M2 and / or may include a composition quality prediction program containing a trained predictive model M2.

[0197] The data processing system 120 can be implemented in many different ways. For example, the data processing system can be a monolithic computer system such as a desktop computer system, portable telecommunications equipment, smartphone, dedicated coating composition quality control equipment, or a computer system effectively connected to or incorporated into equipment used for the automated production and / or testing of coating compositions. Alternatively, the data processing system 120 can be a distributed computer system, such as a client / server computer system, optionally connected to one or more devices used for the automated production and / or testing of coating compositions. Components of the distributed computer system can communicate with each other via a network such as the Internet or an organizational intranet. Figure 5 A-5D illustrates some embodiments of the data processing system 120.

[0198] Figure 6A A data processing system in the form of a smartphone 130 is shown, which includes a phase separation identification program 124 in the form of a web application.

[0199] According to one example, the phase separation identification program is implemented as a script, which runs in a smartphone's browser and is downloaded by a user accessing a specific website, such as a company web portal generated by server 144 and provided via the internet or intranet. For example, program 124 can be implemented as a JavaScript program.

[0200] According to another embodiment, the phase separation identification program is implemented as a program that runs outside of a browser, such as a Java program.

[0201] The phase separation identification program can be implemented as a two-part program comprising a client component and a server component, which are interoperable and configured to exchange data via network communication 142. For example, the program component device 130 (“client application”), installed on a portable telecommunications device, can be configured to control the force-displacement curve acquisition process and output phase separation identification results to the user. The program component installed on the server (“server application”) can be configured to receive force-displacement curves from the client component via the network, analyze the force-displacement curves to detect phases, determine the magnitude of the identified phases, and calculate the qualitative and / or quantitative characterization of the coating composition. The server component returns the characterization to the client component and preferably also returns an indication of the type and magnitude of the identified phases.

[0202] Figure 6B A data processing system 150 in the form of a customized surface coating quality control device is shown, namely, dedicated hardware designed to control and objectify the quality of the coating composition and implicitly control and objectify the quality of the coating formation process. The device includes a storage medium with a phase separation identification program 124, an interface 152 allowing the user to control the quality inspection and testing process, and preferably several hardware components for testing the composition properties of sample compositions. For example, the device may include a measuring probe 134 engaged to the device via a robotic arm 158 or other connectors that allow changing the relative position of the container with the coating composition and the measuring probe, or vice versa. The measuring probe 134 includes a contact surface 134a through which the dynamic pressure of the coating composition through which the measuring probe moves is applied, thereby exerting a force on a support of the contact surface 134a, wherein the force acting on the support of the contact surface 134a is proportional to the dynamic pressure of the coating composition.

[0203] The control device 150 can be implemented as a portable or stationary device. For example, it can be implemented as part of an apparatus for the automated production and / or testing of coating compositions. The apparatus includes a conveyor belt 154 for feeding multiple composition samples 162, 164, 166, 168 to the control device 150, thereby allowing fully automated, rapid, and repeatable quality control of numerous coating compositions. As indicated, composition samples 162, 164, 166, 168 can comprise various coating compositions, which may thus result in different phase separations. As shown, composition samples 162 and 168 do not exhibit any phase separation, composition sample 164 comprises two phases, and composition sample 166 comprises three phases.

[0204] Figure 6C A data processing system in the form of a computer 170 is shown, which is connected to an apparatus 244 for manufacturing a coating composition.

[0205] Device 244 includes a main control computer 246 for controlling, monitoring, and / or scheduling tasks related to the production of coating compositions, the application of coating compositions to various surfaces, and / or testing of the coated surfaces or coating compositions (e.g., for determining rheological, chemical, physical, or other parameters of the coating composition). Each task is performed by several different units included in device 244. For example, the device may include one or more analyzers 257 for performing chemical, physical, mechanical, optical, or other forms of testing and analysis. The device may include one or more mixing units 256 configured for manufacturing various coating compositions, such as mixing components of a composition based on a specific manufacturing and mixing protocol. According to some embodiments, the device also includes a force-displacement curve acquisition unit 252, which includes measuring probes and mechanisms for positioning the sample and measuring probes relative to each other, such that the acquired force-displacement curves can be used as input by a phase separation identification procedure 124. One or more transport units 258, such as conveyor belts, connect the different units and transport components, mixtures, and coating compositions from one unit to another.

[0206] The control computer 246 includes a control unit 248 configured to transmit the force-displacement curve of the composition sample obtained from the force-displacement curve acquisition unit 252 to the phase separation identification program 124 of the computer system 170. Preferably, additional data, such as complete or incomplete specifications of the coating composition components and optionally information about the coating composition production process, are provided to the phase separation identification program along with the force-displacement distribution data. The phase separation identification program is configured to use the received force-displacement curve and optionally the additional data as input to automatically identify the phase plotted in the force-displacement curve, to calculate quantifications, and to calculate coating composition characterizations related to the phase quantifications. The results calculated by the phase separation identification program can be output to a user via a GUI and / or stored in a database 204.

[0207] Preferably, some data obtained by other units such as analyzer 257 or mixing unit 256 may be stored directly in the database in association with the identifier of a specific coating composition and / or coated sample, or may be sent to computer system 170 so that computer system 170 stores the data in the database.

[0208] Using a phase separation identification procedure in the context of device 244 can be particularly advantageous because, after obtaining the force-displacement profiles of the coating composition, they can be automatically analyzed for the phase to be tested. The results obtained can be correlated with formulation data and / or analytical data, and can therefore be used to optimize the composition.

[0209] Figure 7 Force-displacement curve 1302 is shown. The force-displacement curve includes labels 1312 and 1314 automatically created by the phase separation identification program 124 according to an embodiment of the present invention. The labels indicate force-displacement patterns 1312 and 1314 that have been automatically detected by the phase separation identification program. Furthermore, for example, the length of the phase (and thus the amount / volume of the phase) can be calculated from the difference of the X value (1316) of the two force-displacement patterns (1312, 1314).

[0210] In addition to the visual representation of the detected force-displacement patterns, the phase separation identification program is also configured to temporarily or permanently store the type and location of the identified phase in a structured form. For example, the location can be stored as graphical coordinates representing two force-displacement patterns of a phase. Storing the phase location and its position in a structured form allows the phase separation identification program to process the structured data to calculate the polymerization characteristics of the coating composition.

[0211] List of reference numerals

[0212] Steps 102-116

[0213] 120 Data Processing System

[0214] 122 Data storage media

[0215] 124 Phase Separation Identification Program

[0216] 125 Force-Displacement Curve

[0217] 126 processor

[0218] 130 Portable Telecommunications Equipment

[0219] 134 Measurement probe

[0220] 134a Contact surface of the measuring probe

[0221] 140 Browser

[0222] 142 Network

[0223] 144 Server Computer

[0224] 146 Network Server

[0225] 150 Coating Quality Control Equipment

[0226] 152 Control Panel

[0227] 154 Carrier / Conveyor Belt

[0228] 158 robotic arm

[0229] Samples 162-168

[0230] 170 Computer System

[0231] 204 Database

[0232] 244 Equipment for producing and / or testing coating compositions

[0233] 246 Main Control Computer

[0234] 248 Control Unit

[0235] 252 Force-Displacement Curve Acquisition Unit

[0236] 256 Hybrid Units

[0237] 257 Analyzer

[0238] 258 transport units

[0239] Force-displacement curves of the 1202 coating composition

[0240] Force-displacement curve marked 1210 1202

[0241] 1212 Add tags

[0242] 1214 Add tags

[0243] 1216 The length difference of the X-values ​​of labels 1212 and 1214

[0244] Force-displacement curves of the 1302 coating composition

[0245] 1312 Automatically generated labels based on calculated and recognized patterns

[0246] 1314 Automatically generate labels based on calculated and recognized patterns

[0247] 1316 Calculate the length difference of the X-value of the automatically generated labels for recognizing patterns 1312 and 1314.

Claims

1. A method for detecting phase separation in an aqueous, solvent-based, or solvent-free coating composition, comprising: The coating composition is provided in a container; A measuring instrument for receiving the container is provided, the measuring instrument including a measuring probe; Control the measuring instrument to a) The measuring probe is moved through the coating composition along a predetermined measuring path at a predetermined speed curve, the predetermined measuring path extending along the length axis of the container. b) Obtain a force-displacement curve by measuring the force applied to the measuring probe while the probe is moving along a predetermined measurement path at a predetermined speed curve; The force-displacement curve is processed to detect at least one phase separation of the coating composition; and Output the detection results.

2. The method as described in claim 1, wherein, The processing of the force-displacement curve includes thresholding, change point detection, inflection point detection, and / or the application of the Ramer-Douglas-Peucker algorithm (RDP) or isolated forest for phase separation detection.

3. The method of claim 2, wherein the phase separation comprises at least two phases, namely a first phase and a second phase, wherein the second phase contains more filler and / or pigment than the first phase.

4. The method of claim 3, wherein the second phase comprises a precipitated subphase and a liquid subphase containing fillers and / or pigments, the method further comprising: The acquisition of the force-displacement curve is stopped when the measured force reaches or exceeds the predetermined limit. The probe displacement at the predetermined limit indicates the start of the precipitation subphase.

5. The method according to any one of claims 2 to 4, further comprising: Calculate the measure used for the identified phase to provide qualitative and / or quantitative characterization of the coating composition; this measure is    Quantitative measures selected from the following group, which includes: the length of the measurement path between two detected phase boundaries, the travel time of the measurement probe between two detected phase boundaries, the relative size of the detected phases, the number of detected phases; and / or A qualitative measure is the type of phase selected from the group consisting of a gas phase, a first phase, and a second phase, wherein the second phase contains more filler and / or pigment than the first phase and / or the second phase contains a liquid subphase and a precipitate subphase containing filler and / or pigment. Output the qualitative and / or quantitative characterization of the coating composition.

6. A method for detecting phase separation in an aqueous, solvent-based, or solvent-free coating composition, the method comprising: The force-displacement curves (102) are processed by a phase separation identification procedure (124), which is configured to identify force-displacement patterns, each of which is assigned to a boundary of a phase type. and Provide (104) the detection results of one or more phases confirmed by the phase separation and identification procedure, The force-displacement curve depicts multiple recorded forces applied to the measuring probe as it is moved along a predetermined measuring path at a predetermined speed through the coating composition.

7. The method of claim 6, comprising: The quantification of the identified phase is calculated (106) using a phase separation identification procedure to provide qualitative and / or quantitative characterization of the coating composition; and Output (108) Qualitative and / or quantitative characterization of the coating composition.

8. The method as described in claim 7, This metric is a quantitative metric selected from the following group: the length of the measurement path between two detected force-displacement modes, the travel time of the measurement probe between the two detected force-displacement modes, the relative size of the detected phases, the number of detected force-displacement modes; and / or This measurement is a qualitative measurement, selected from the following group: gas phase, first liquid phase, and second liquid phase. The second liquid phase contains more filler and / or pigments than the first liquid phase, and the second liquid phase contains a liquid subphase and a precipitate subphase containing filler and / or pigments.

9. The method of any one of claims 6 to 8, further comprising: Provides aqueous, solvent-based, or solvent-free coating compositions in containers; A measuring instrument for receiving the container is provided, the measuring instrument including a measuring probe; Position the measuring probe relative to the coating composition in the container; The measuring instrument is controlled to move the measuring probe along a predetermined measuring path at a predetermined speed curve through the coating composition, the predetermined measuring path extending along the length axis of the container. A force-displacement curve is obtained by using a measuring probe to measure the force applied to the measuring probe while the measuring probe is moving through the coating composition at a predetermined speed along a predetermined measuring path.

10. The method of any one of claims 6 to 8, wherein the processing of the force-displacement curve further comprises: The force-displacement pattern is automatically labeled with the phase type and instances of that phase type in the force-displacement curve by performing a phase separation identification procedure on the force-displacement curve, including thresholding, change point detection, isolated forest, inflection point detection and / or the Ramer-Douglas-Peucker algorithm (RDP). and Output one or more identified phase instances.

11. The method according to any one of claims 6 to 8, wherein the phase separation identification procedure includes a predictive model M1 learned from training data in a training step, the training step using a machine learning procedure to identify a predetermined pattern, the method further comprising performing the training step on the training data, the training data including a set of labeled digital training force-displacement curves of the coating composition, the labels identifying the location / position and type of the phase in the training force-displacement curves, the predictive model being trained using backpropagation through the labeled training force-displacement curves to identify the pattern.

12. The method of claim 11, wherein each of the training force-displacement curves has been assigned supplementary data, which is processed in the training step to allow the prediction model to associate the supplementary data with the curve, the supplementary data including contextual data, the contextual data including: The coating comprises one or more components used to generate a coating composition to which a trained force-displacement curve has been obtained; and / or One or more production process parameters, which characterize the process of producing the coating composition; and / or System parameters for a pressure measurement system used to obtain training force-displacement curves are selected from the following group: temperature type of coating composition, measuring probe, measuring probe sensitivity, measuring path length, measuring probe speed as the probe moves along the measuring path, and measuring probe speed curve as the probe moves along the measuring path.

13. The method of claim 9, wherein, Positioning the measuring probe relative to the coating composition in the container includes positioning the measuring probe above the coating composition in the container.

14. The method of claim 11, wherein, The machine learning program is a neural network.

15. The method of claim 12, wherein, The specifications of the one or more components include the type and / or amount of dispersant and / or the type or amount of rheology modifier and / or the type or amount of one or more pigments and / or the type and amount of solvent.

16. The method of claim 12, wherein, The process parameters include the mixing rate and / or mixing duration of the coating composition.

17. A computer-implemented method for providing a coating composition-related prediction program, the method comprising: Provide a database (204) that includes qualitative and / or quantitative characterization of coating compositions and correlations with one or more parameters selected from the group consisting of one or more components of the coating composition, relative and / or absolute amounts of one or more of the components and / or process parameters of the coating composition. A machine learning program is trained on the correlation between the characterization of the coating composition and one or more parameters in a database to provide predictive models M2 and M3, which have learned to correlate the qualitative and / or quantitative characterization of one or more coating compositions with one or more parameters stored in association with the respective coating components and / or production process parameters used to produce the coating compositions. and A composition quality prediction program including a prediction model M2 is provided, which is configured to use the prediction model M2 to predict the performance of a coating composition from one or more input parameters selected from the group consisting of one or more components of the coating composition, relative and / or absolute amounts of one or more said components and / or production process parameters, said performance including the detection of phase separation by processing force-displacement curves; and / or A composition specification prediction program including a prediction model M3 is provided. This program is configured to use the prediction model M3 to predict and output one or more parameters related to the desired coating composition, based on inputs that at least explicitly specify the required storage stability characterization and one or more additional parameters associated with the desired coating composition. These parameters are selected from the group consisting of one or more components of the coating composition, relative and / or absolute amounts of one or more of the components, and / or manufacturing process parameters used to prepare the coating composition. The force-displacement curve depicts multiple recorded forces applied to the measuring probe as it is moved along a predetermined measuring path at a predetermined speed through the coating composition.

18. The method of claim 17, wherein, The additional parameters include components, process parameters, and application parameters.

19. The method of claim 17, wherein the method comprises: Multiple force-displacement profiles are provided for coating compositions made of a variety of different coating components, wherein at least some coating compositions have a variety of different types of one or more phases; A phase separation identification procedure is applied to a force-displacement curve to identify force-displacement patterns in the force-displacement curve, obtain a measure of the phase represented by the force-displacement patterns identified in the force-displacement curve, and calculate the qualitative and / or quantitative characterization of the coating composition represented by the force-displacement curve, wherein the phase separation identification procedure is the phase separation identification procedure described in the method for detecting phase separation of aqueous, solvent-based, or solvent-free coating compositions as described in any one of claims 6-12. The qualitative and / or quantitative characterization of the phases is stored in a database in association with one or more parameters related to the coating and / or process parameters used to produce coating compositions containing these phases.

20. A system comprising: Equipment (244) for producing and testing compositions for paints, varnishes, printing inks, abrasive resins, pigment concentrates, or other coatings, wherein the equipment comprises at least two processing stations connected to each other by a transport system, wherein self-propelled transport vehicles are capable of moving on the transport system to transport components of the composition and / or the produced composition between the processing stations, and A computer system (170) configured to perform the method described in any one of claims 6-19.

Images