Physical constraint model for material formula optimization and construction method thereof

Abstract

The invention relates to the technical field of material informatics, and discloses a physical constraint model for material formula optimization, and the model comprises a component proportion constraint model which is used for limiting the proportion of each component in a thermodynamic stable interval; the process-structure constraint model is used for representing regulation and control of process parameters on a molecular chain structure and an aggregation state structure and limiting the structure parameters to be in a manufacturable interval; the performance prediction constraint model is used for limiting the prediction performance to be in a physical achievable range; a three-layer constraint framework of'component proportion-process-structure-performance prediction 'is constructed, thermodynamic stability, compatibility / phase separation criteria, reaction kinetics, rheology and other laws are embedded, and exclusive constraint modules are configured according to ten types of product characteristics, so that physical or process infeasible pseudo-optimal schemes are screened out in the optimization process, and the optimal performance of the product is optimized. And the possibility of formula recommendation is improved.

Description

Technical Field

[0001] This invention relates to the field of materials informatics, and in particular to a physical constraint model for materials formulation optimization and its construction method. Background Technology

[0002] Currently, material formulation design still largely relies on the "trial and error method," repeatedly adjusting between multi-component systems and complex process windows, resulting in long research and development cycles and high costs.

[0003] A key reason for this is the fragmentation of data regarding formulation components, process parameters, structural characterization, and performance indicators, lacking unified correlation modeling. This leads to a disconnect between the design scheme and actual processing / mass production requirements. Furthermore, some existing data-driven methods prioritize fitting correlations, failing to explicitly incorporate physicochemical constraints such as mass / equivalence conservation, compatibility and phase equilibrium boundaries, diffusion / curing kinetics, rheology, and process windows. This easily generates pseudo-optimal formulations that are "predicted to meet standards but physically infeasible or technologically unrealizable," resulting in a high verification failure rate. Simultaneously, the significant differences in multimodal data structures make fusion and alignment difficult, easily causing distortion in feature correlations.

[0004] Therefore, there is an urgent need to establish a material formulation optimization modeling method that integrates mechanisms and data, so as to enhance the physical rationality and process feasibility of the formulation while ensuring the accuracy of prediction, thereby improving R&D efficiency and the feasibility of the results. Summary of the Invention

[0005] To address the problem that existing material formulation designs often lack physicochemical mechanisms and manufacturability constraints, leading to discrepancies between predicted results and actual processes, this invention discloses a physical constraint model for material formulation optimization and its construction method. By constructing a three-layer constraint architecture of "component ratio - process - structure - performance prediction," it embeds thermodynamic stability, compatibility / phase separation criteria, reaction kinetics, rheology, and other laws, and configures dedicated constraint modules according to the characteristics of ten major product categories. This allows for the elimination of pseudo-optimal solutions that are physically or technologically infeasible during the optimization process, thereby improving the feasibility of formulation recommendations.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is as follows: A physical constraint model for material formulation optimization, applicable to formulation design and performance prediction of finely processed polymer products, employing a three-layer constraint architecture, including: (1) Component ratio constraint model: Based on the principles of chemical composition and thermodynamic stability, component molar ratio constraint equation or mass ratio constraint equation is established to limit the proportion of each component to be within the thermodynamically stable range. (2) Process-structure constraint model: The mapping relationship between processing parameters and material structure parameters is established by using reaction kinetic theory equations or rheological theory equations. This model is used to characterize the control of process parameters on molecular chain structure and aggregate structure, and to limit the structure parameters to the manufacturable range. (3) Performance prediction constraint model: Based on the classical theory of materials science and the hybrid law, the theoretical value range of the performance index is constructed to limit the predicted performance to be within the physically achievable range; The component ratio constraint model, process-structure constraint model and performance prediction constraint model work together to construct a quantitative mapping relationship of the entire link of component-process-structure-performance, and are used to eliminate candidate solutions that violate any constraint conditions, so as to avoid generating pseudo-optimal solutions that exceed the theoretical limit.

[0007] Optionally, the polymer fine-processed product is selected from at least one of plastics, rubber, adhesives, inks, coatings, oils, surface treatment agents, metalworking fluids, cleaning agents, and textile chemicals.

[0008] Optionally, the component ratio constraint model limits the range of component ratios from the perspective of thermodynamic stability; the process-structure constraint model describes the regulation law of process parameters on structure from the perspective of kinetics; and the performance prediction constraint model limits the range of performance index values ​​from the perspective of performance theoretical limits.

[0009] Optionally, the constraints of the component ratio constraint model are selected from at least one of component molar ratio constraints, component mass ratio constraints, compatibility constraints, and phase separation risk constraints.

[0010] Optionally, the component ratio constraint model is constrained based on the mechanism principle of the category to which the target product belongs. The mechanism principle is selected from at least one of the following: free radical polymerization reaction functional group activity principle, rubber vulcanization reaction mechanism, adhesive reaction functional group activity principle, pigment dispersion and film formation mechanism, film-forming resin crosslinking reaction mechanism, lubrication mechanism, surface modification reaction mechanism, cooling-lubrication-rust prevention synergistic mechanism, stain removal mechanism, and fabric modification mechanism.

[0011] Optionally, the phase separation risk constraint is established based on at least one of the interaction parameter, the difference in solubility parameter, and the ratio of pigment volume concentration to critical pigment volume concentration.

[0012] Optionally, the theoretical equations used in the process-structure constraint model are selected from at least one of the following: Arrhenius equation, vulcanization reaction kinetic equation, Mooney-Rivlin equation, curing reaction kinetic equation, viscosity-temperature equation, drying kinetic equation, coating kinetic equation, adsorption kinetic equation, film formation kinetic equation, cooling kinetic equation, diffusion kinetic equation, and adsorption-diffusion equation.

[0013] Optionally, the theoretical equations used in the performance prediction constraint model are selected from at least one of the following: the Mark-Howewen equation, the rubber elastic modulus equation, the Akron abrasion equation, the cohesive energy density equation, the interfacial tension equation, the wear resistance equation, the hardness equation, the oxidation kinetics equation, and the decontamination kinetics equation.

[0014] Another object of the present invention is to provide a method for constructing a physical constraint model for material formulation optimization as described above, comprising: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure; S2: Establish a physical constraint model, construct a quantitative mapping relationship across the entire chain of components, processes, structures, and performance, and output several candidate formulations that meet the physical constraints.

[0015] Optionally, the four-level standardized dataset structure is a data structure that classifies data hierarchically according to purpose, main performance, preparation process, and auxiliary performance.

[0016] The present invention has the following beneficial effects: By establishing a three-layer physical constraint framework applicable to polymer precision processing products, thermodynamic stability, compatibility / phase separation criteria, and physicochemical laws such as reaction kinetics and rheology are embedded in the "component ratio constraint - process constraint - structural constraint - performance prediction constraint". Dedicated constraint modules are set up based on the material characteristics of ten major product categories, eliminating pseudo-optimal solutions that only "predict and meet" data fitting but are physically infeasible or technologically unrealizable. Simultaneously, the model quantitatively correlates formulation components, preparation processes, structural evolution, and performance indicators across the entire chain, clarifying the constraint logic and feasible domain boundaries of each link, providing an interpretable theoretical basis for formulation optimization and process parameter adjustment. At the application level, the overall framework can be uniformly adapted to multiple product categories, while modules in each field can be configured according to specific mechanisms and theoretical equations, balancing universality and specificity. Furthermore, constraint thresholds and theoretical ranges can be aligned with industrial production standards and process windows, improving the credibility and operability of the recommended results. This effectively shortens the R&D iteration cycle, reduces trial-and-error costs, and increases the success rate of engineering implementation in intelligent formulation design and process optimization scenarios, driving the industry towards intelligent and precise development. Attached Figure Description

[0017] Figure 1 This is a structural diagram of the physical constraint model used for material formulation optimization in this invention. Detailed Implementation

[0018] The embodiments described below are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0019] To address the problem that existing material formulation design lacks physicochemical mechanisms and manufacturability constraints, leading to frequent discrepancies between predicted results and actual processes, this invention provides a physical constraint model for material formulation optimization. This physical constraint model is applicable to the formulation design and performance prediction of finely processed polymer products and employs a three-layer constraint architecture, including: (1) Component ratio constraint model: Based on the principles of chemical composition and thermodynamic stability, component molar ratio constraint equation or mass ratio constraint equation is established to limit the proportion of each component to be within the thermodynamically stable range. (2) Process-structure constraint model: The mapping relationship between processing parameters and material structure parameters is established by using reaction kinetic theory equations or rheological theory equations. This model is used to characterize the control of process parameters on molecular chain structure and aggregate structure, and to limit the structure parameters to the manufacturable range. (3) Performance prediction constraint model: Based on the classical theory of materials science and the hybrid law, the theoretical value range of the performance index is constructed to limit the predicted performance to be within the physically achievable range; Among them, the component ratio constraint model, the process-structure constraint model and the performance prediction constraint model work together to construct a quantitative mapping relationship of the entire link of component-process-structure-performance, and are used to eliminate candidate solutions that violate any constraint conditions, so as to avoid generating pseudo-optimal solutions that exceed the theoretical limit.

[0020] The present invention preferably uses a component ratio constraint model to limit the range of component ratios from the perspective of thermodynamic stability; a process-structure constraint model to describe the regulation law of process parameters on structure from the perspective of kinetics; and a performance prediction constraint model to limit the range of performance index values ​​from the perspective of performance theoretical limits.

[0021] Furthermore, the polymer fine processing products of the present invention are preferably selected from at least one of plastics, rubber, adhesives, inks, coatings, oils, surface treatment agents, metalworking fluids, cleaning agents and textile chemicals.

[0022] Specifically, when the target product is plastic, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes the key components such as "matrix resin-reinforcing phase-toughening phase" in the plastic formulation as the constraint objects. It comprehensively adopts the functional group activity mechanism of free radical polymerization reaction and the thermodynamic compatibility criterion of mixing to jointly limit the stoichiometric relationship and phase stability of the components.

[0023] S21: Establish a component ratio constraint model The preferred approach is to introduce the interaction parameter χ based on the Flory-Huggins theory, which is calculated as follows: Where ΔH is the enthalpy of mixing, R is the gas constant, T is the absolute temperature, and φ1 and φ2 are the volume fractions of the two components to be evaluated.

[0024] By setting a threshold boundary for χ, the compatibility between the polymer matrix and the toughening agent is limited to a feasible range. The preferred constraint is that the interaction parameter between the polymer (e.g., PP) and the toughening agent (e.g., POE) satisfies χ≤0.05, so as to reduce the risk of phase separation and ensure the stability of the system.

[0025] Based on the functional group activity and stoichiometric equilibrium relationship of free radical polymerization, molar ratio constraint equations or mass ratio constraint equations are established among monomers, reinforcing agents, and toughening agents to ensure that the formulation meets the basic conditions for molding and performance realization in terms of reaction stoichiometry and composition.

[0026] Preferably, the following boundaries are set for the key mass ratio and addition amount: the mass ratio of inorganic reinforcing phase (e.g., glass fiber) to organic matrix (e.g., PP) is limited to between 0.15 and 0.50 to avoid insufficient reinforcing phase leading to insignificant strengthening effect, or excessive reinforcing phase leading to poor interfacial bonding and decreased processing fluidity; the addition amount of toughening agent (e.g., POE) is not higher than 8% to suppress the risks of phase separation, decreased strength, or abnormal processing viscosity while improving toughness.

[0027] S22: Establish a process-structure constraint model Based on the Arrhenius equation, a correspondence is established between temperature parameters such as barrel temperature and die temperature and the molecular chain flowability or structural evolution rate. The Arrhenius equation is as follows: Where k is the reaction rate constant or structural evolution rate characterization quantity, A is the pre-exponential factor, and E a R is the activation energy, R is the gas constant, and T is the absolute temperature.

[0028] The effect of shear rate on melt viscosity is described using the Cross rheological model. The Cross model is as follows: Where η is the apparent viscosity, η0 is the zero-shear viscosity, λ is the characteristic time constant, λ is the shear rate, and n is the flow behavior exponent. By converting process parameters such as injection speed or screw speed into shear rate, the variation law of melt viscosity can be obtained, and a correspondence can be established with melt flow rate (MFR), thus forming a mapping link of "process parameters - flow behavior - structure formation".

[0029] S23: Establish a performance prediction constraint model Preferably, the theoretical range of mechanical properties such as elastic modulus is defined by the hybrid law of composite material mechanics. The hybrid law can be expressed as: Where E is the elastic modulus of the composite system, E1 and E2 are the elastic moduli of each component, and φ1 and φ2 are the volume fractions of each component; based on this theory, reasonable range boundaries for performance indicators such as tensile strength and elastic modulus can be further formed.

[0030] Preferably, the Mark-Houwink equation is used to establish the correspondence between molecular weight and viscosity characterization. The Mark-Houwink equation is as follows: Where [η] is the intrinsic viscosity, M is the molecular weight, and K and a are constants related to the system. This relationship allows for the establishment of a constrained correlation between molecular weight-related structural characterization and flow / viscosity characterization, and enables consistency verification with the reasonable range of MFR. When constraints need to be imposed on swelling and other related properties, the Flory-Rehner theory can be further used to limit the swelling performance to an achievable range, thereby avoiding predictions that are physically invalid.

[0031] When the target product is rubber, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes the raw rubber (matrix) - vulcanization system (vulcanizing agent / accelerator, etc.) - reinforcing filler (such as carbon black) in the rubber formulation as the key constraint objects. It establishes the full-link constraint relationship of composition, process, structure and performance around the "component reaction - process control - crosslinking network - elasticity and wear resistance" of the vulcanization process. By prioritizing the execution of composition constraints and process-structure constraints during the generation and screening of candidate solutions, and then executing performance constraints, the model can pre-eliminate physically infeasible or process-unfeasible solutions.

[0032] S21: Establish a component ratio constraint model Based on the stoichiometric relationship of the vulcanization reaction, a molar ratio constraint is established between the vulcanizing agent and the double bond of the raw rubber, which is preferably expressed as: n 硫化剂 : n 双键 =1:(50-100) Where, n 硫化剂 n is the amount of the vulcanizing agent. 双键 This represents the amount of reactant double bonds in raw rubber.

[0033] Furthermore, the above-mentioned measurement constraints can be equivalently converted into formulation dosage boundaries for rapid screening of candidate formulations. Preferably, the amount of vulcanizing agent is 1%-3% of the raw rubber mass to reduce the risk of performance degradation caused by insufficient or excessive vulcanization.

[0034] To avoid filler agglomeration, abnormal viscosity increase, and narrowing of the processing window, an upper limit boundary is set for the amount of reinforcing filler. Preferably, the amount of reinforcing fillers such as carbon black added should not exceed 50 phr (based on the mass of 100 parts of rubber), and this is one of the necessary inspection items for component ratio constraints.

[0035] S22: Establish a process-structure constraint model The evolution of the degree of sulfidation over time is described using a kinetic equation for the sulfidation reaction, and the correspondence between process parameters such as temperature and time and structural evolution is established. The preferred equation is: Where α is the degree of sulfidation (or conversion rate), k is the reaction rate constant, and n is the reaction order.

[0036] By inputting temperature and time, the evolution curve of α(t) can be obtained, which can be used to characterize the crosslinking network formation process, and manufacturability boundaries can be applied to structure-related parameters (such as crosslinking density, effective network segment length, etc.).

[0037] To characterize the effect of the cross-linked network structure on the elastic response, the Mooney-Rivlin constitutive relation was used to establish the mapping between stress and tensile ratio, and the preferred equation was: Where σ is stress, C1 and C2 are material constants, and λ is the stretch ratio.

[0038] By verifying the consistency between the degree of sulfurization / crosslinking state and the reasonable range of constitutive parameters, it is possible to identify structural anomalies such as "insufficient sulfurization leading to imperfect network" or "over-sulfurization leading to embrittlement".

[0039] S23: Establish a performance prediction constraint model Based on the theory of rubber elasticity, a correspondence between elastic modulus and cross-linked network is established to limit the physically realizable range of elastic modulus. The preferred equation is: Where E is the elastic modulus, ρ is the density, R is the gas constant, T is the absolute temperature, and M is the gas constant. c The molecular weight is the distance between crosslinking points. Based on this, a theoretical range boundary can be given for the predicted modulus / hardness of candidate solutions, which serves as one of the core criteria for performance prediction constraints.

[0040] Based on the Akron wear model to constrain the wear amount, the optimal equation is: Where V is the wear amount, K is the wear coefficient, F is the load, L is the sliding distance, E is the elastic modulus, and H is the hardness.

[0041] This model incorporates the effects of reinforcing fillers and cross-linked networks on wear resistance into theoretical consistency verification, avoiding pseudo-optimal results such as "significantly improved wear resistance but inconsistent with modulus / hardness". It is preferable to set target performance thresholds as constraint boundaries, for example: compression set not exceeding 20% ​​and tensile strength not less than 15 MPa; when the predicted performance of a candidate formulation exceeds the above physical range or does not meet the threshold boundaries, it is determined to violate the performance prediction constraints and is eliminated.

[0042] When the target product is an adhesive, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes the key components in the adhesive formulation, such as "resin-curing agent-diluent / additive", as the constraint objects. It comprehensively adopts the functional group activity and stoichiometric balance mechanism, curing kinetics and viscosity-temperature relationship to jointly limit the reaction stoichiometry, curing process and interfacial bonding feasibility, thereby forming a constraint link of "component reaction-process curing-crosslinking structure-adhesion performance".

[0043] S21: Establish a component ratio constraint model Preferably, the functional group molar ratio constraint between the curing agent and the resin is established based on the stoichiometric equilibrium of functional group reactions. This functional group molar ratio constraint can be expressed as: By setting boundaries for the functional group molar ratio, incomplete curing due to insufficient curing agent or network structure embrittlement and performance fluctuations due to excessive curing agent can be avoided. Furthermore, an optimal boundary can be set for the diluent: the amount of diluent added should not exceed 20%, in order to reduce the risk of insufficient strength, abnormal volatilization shrinkage, or interfacial wetting imbalance caused by excessive dilution of the system.

[0044] S22: Establish a process-structure constraint model Preferably, a curing reaction kinetic equation is used to describe the effect of temperature and time on the degree of curing. The curing reaction kinetic equation is as follows: Where α represents the degree of curing, A represents the pre-exponential factor, and E represents the degree of curing. a Let R be the activation energy, T be the absolute temperature, and n be the reaction order. This equation establishes a mapping relationship between structural parameters such as curing temperature / curing time, degree of curing, and crosslinking density, and limits these parameters to a manufacturable range.

[0045] Simultaneously, viscosity-temperature relationship is preferably used to constrain coating and flow manufacturability, wherein the viscosity-temperature equation is: Where η is viscosity, η0 is reference viscosity, β is viscosity-temperature coefficient, and T0 is reference temperature. This relationship constrains the coating temperature window and the application viscosity range to form a mapping link of "process temperature - flow behavior - film formation / curing structure".

[0046] S23: Establish a performance prediction constraint model Preferably, the theoretical range of the cohesive strength of the system is defined by the cohesive energy density equation, which is: Where, ΔH v For heat of vaporization, V m The volume is the molar volume.

[0047] Meanwhile, the interfacial tension relationship is preferably used to verify the consistency of compatibility and wetting behavior. The interfacial tension equation can be expressed as: Where γ is the interfacial tension, γ1 and γ2 are the surface tensions of the two phases, and γ1 d γ2 d This represents the dispersive component. Through the above theoretical range and consistency check, achievable boundaries can be set for properties such as shear strength (e.g., metal-to-metal shear strength not less than 5 MPa), and pseudo-optimal solutions that only meet the data fitting criteria but do not hold true at the interface mechanism can be eliminated.

[0048] When the target product is ink, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes the key components such as "resin-pigment-solvent / auxiliary agent" in the ink formulation as the constraint objects. It comprehensively adopts pigment dispersion and film formation mechanism, drying diffusion kinetics and interface wetting criteria to jointly limit the dispersion stability, film structure and printing performance feasibility, thereby forming a constraint link of "component dispersion-printing process-film structure-printing performance".

[0049] S21: Establish a component ratio constraint model Preferably, pigment volume concentration (PVC) is introduced to constrain dispersion and film stability, and the PVC equation is: Among them, V p V is the volume of the pigment. b This refers to the binder volume. By setting boundaries for PVC, placing it within a reasonable range relative to the critical pigment volume concentration (CPVC) (e.g., 60%-80% CPVC), the risk of increased porosity or insufficient film formation can be reduced.

[0050] Meanwhile, solvent / resin compatibility is preferably defined by a solubility parameter difference, wherein the difference is: Here, δ1 and δ2 are the solubility parameters of the two components. A threshold is set for Δδ (e.g., not higher than 1.5) to reduce the risk of phase separation and precipitation.

[0051] S22: Establish a process-structure constraint model Preferably, the film thickness relationship is used to describe the effect of printing speed and viscosity on film thickness, and the film thickness equation is: Where h is the film thickness, k is a constant, μ is the ink viscosity, v is the printing speed, and σ is the surface tension. This equation establishes a mapping relationship between process parameters and structural parameters such as film thickness and density.

[0052] Meanwhile, it is preferable to use a drying diffusion kinetic equation to describe the solvent evaporation and diffusion process, the drying kinetic equation being: Where M is the solvent mass, D is the diffusion coefficient, A is the area, ΔC is the concentration difference, and δ is the diffusion layer thickness. This forms a "printing process - drying process - film structure" chain, and limits the film structure to the manufacturable range.

[0053] S23: Establish a performance prediction constraint model Preferably, the adhesion performance is defined by an adhesion force relationship, and the adhesion force equation is: Where F is the adhesion force, A is the contact area, γ is the surface tension, and θ is the contact angle.

[0054] Meanwhile, a color difference equation is preferably used to constrain color stability, and the color difference equation is: By setting target ranges for adhesion and color difference, achievable boundaries can be set for indicators such as cross-cut adhesion (e.g., ≥4B) and abrasion resistance, thereby eliminating pseudo-optimal solutions that do not hold true in terms of physical meaning.

[0055] When the target product is paint, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes the key components in the coating formulation, such as "film-forming resin, curing agent, pigments / fillers / solvents", as the constraint objects. It comprehensively adopts cross-linking reaction stoichiometry and curing kinetics, coating kinetics and protection mechanism to jointly limit the formation of cross-linked structure, film thickness uniformity and the feasibility of protective performance.

[0056] S21: Establish a component ratio constraint model Preferably, the reaction equilibrium between the curing agent and the resin is constrained by functional group stoichiometry, and the molar ratio constraint of the crosslinking reaction is: Meanwhile, the volume fraction of pigments and fillers is preferably used to constrain the stability of film formation, and the volume fraction equation is as follows: Among them, V f V represents the volume of pigments and fillers. r This refers to the resin volume. A further upper limit for pigment and filler addition (e.g., no more than 60%) can be set to reduce the risk of pinholes, sagging, or insufficient crosslinking.

[0057] S22: Establish a process-structure constraint model Preferably, coating kinetics is used to describe the effects of pressure difference and viscosity on coating speed / film thickness formation. The coating kinetics equation is as follows: Where v is the coating speed, ΔP is the pressure difference, μ is the viscosity, h is the film thickness, L is the coating length, and k is a constant. A mapping link of "curing temperature / time - crosslinking density - film integrity" is established by combining curing kinetics (similar to the adhesive) to limit the film structure to the manufacturable range.

[0058] S23: Establish a performance prediction constraint model Preferably, a corrosion protection relationship is used to set the theoretical boundary for corrosion resistance performance, and the corrosion protection equation is: Where t is the protection time, d is the film thickness, ρ is the density, i is the corrosion current, and M is the molar mass.

[0059] Meanwhile, the relationship between hardness and crosslinking density can be used to verify performance consistency, for example: Where H is hardness, ρ c The crosslinking density is defined. By setting achievable ranges for indicators such as salt spray resistance and hardness, candidate solutions that only fit the criteria but whose mechanisms are not valid are eliminated.

[0060] When the target product is oil, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes key components such as the "base oil-additive package" in the oil formulation as the constraint objects, and comprehensively adopts the lubrication mechanism and oxidation kinetics to jointly limit the viscosity-temperature behavior, lubricating film formation and antioxidant stability.

[0061] S21: Establish a component ratio constraint model Preferably, the synergistic effect of additives is used to constrain the viscosity contribution of the system, and the relationship is as follows: Where, η total For total viscosity, η base Based on base oil viscosity, η i φ contributes to the viscosity characterization of the additive. i This refers to the mass fraction. Further, an upper limit can be set for the total amount of additives (e.g., ≤15%) and a window for the amount of anti-wear agent (e.g., 0.5%-2%) to avoid formulation instability or amplified side effects.

[0062] S22: Establish a process-structure constraint model Preferably, the viscosity-temperature relationship is used to constrain the effect of temperature on viscosity, and the viscosity-temperature equation is: A mapping between operating parameters and lubrication state is established using the lubrication film thickness relationship. The lubrication film thickness equation is as follows: Where h is the film thickness, v is the sliding speed, and F is the load. This link limits the "operating condition-viscosity-film thickness" to within an achievable range, avoiding candidate solutions that theoretically cannot form an effective lubricating film.

[0063] S23: Establish a performance prediction constraint model The boundaries between friction and oxidation resistance can be set using tribological relationships and oxidation kinetics, such as the relationship between friction coefficients: And the expression of oxidation kinetics for acid value increase: By setting achievable ranges for factors such as friction coefficient and acid value growth rate, pseudo-optimal solutions that are "predicted well but cannot be maintained under operating conditions" can be eliminated.

[0064] When the target product is a surface treatment agent, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes "active component-solvent / carrier-auxiliary agent" as key objects and comprehensively adopts surface adsorption mechanism and film formation thermodynamics to jointly limit the coverage, adsorption rate and film structure stability.

[0065] S21: Establish a component ratio constraint model Preferably, the relationship between coverage and concentration is described using an adsorption equilibrium relationship, and the adsorption equilibrium equation is: Where θ is the coverage, K is the adsorption constant, and c is the concentration of the active component. The adsorption is limited to a feasible range by setting a boundary for c (e.g., 2%-10%).

[0066] S22: Establish a process-structure constraint model Preferably, an adsorption kinetic equation is used to describe the effect of treatment temperature and time on the adsorption rate. The adsorption kinetic equation is as follows: By establishing a mapping link of "processing conditions - adsorption amount - film thickness / uniformity", the film structure is limited to the manufacturable range.

[0067] S23: Establish a performance prediction constraint model The performance consistency can be verified by using the relationship between interfacial bonding strength and film thickness, and the expression is: Where σ represents the bonding strength, γ represents the interface-related characterization quantity, and δ represents the film thickness. Achievable boundaries can be set in conjunction with performance targets such as corrosion resistance and adhesion, thereby eliminating physically invalid predictions.

[0068] When the target product is a metalworking fluid, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes key components such as "emulsion system - rust prevention system - lubrication system" as constraint objects, and comprehensively adopts the cooling heat transfer mechanism and the boundary lubrication film formation mechanism to jointly limit the emulsion stability, cooling capacity and rust prevention capacity.

[0069] S21: Establish a component ratio constraint model Preferably, the emulsifier system is constrained by the HLB demand relationship related to emulsification stability, wherein the relationship is as follows: HLB req For the desired hydrophilic-lipophilic balance value, HLB i For component HLB value, φ i It is a mass fraction. The dosage windows for emulsifiers (e.g., 5%-15%) and rust inhibitors (e.g., 1%-5%) can be set to maintain emulsion stability and performance balance.

[0070] S22: Establish a process-structure constraint model Preferably, a cooling heat transfer equation is used to establish the mapping between the operating conditions and the cooling capacity. The heat transfer equation is as follows: Where Q is the heat transfer, h is the heat transfer coefficient, A is the heat exchange area, and ΔT is the temperature difference.

[0071] A mapping between cutting speed / load and lubrication film thickness is established using the boundary lubrication film formation relationship, and the equation is as follows: This limits cooling and lubrication to a manufacturable and maintainable operating window.

[0072] S23: Establish a performance prediction constraint model The performance boundary can be defined by the relationship between rust prevention time and rust inhibitor concentration / liquid film thickness, and the equation is as follows: Where t is the rust prevention time, c is the rust inhibitor concentration, and δ is the liquid film thickness. Target boundaries are set in conjunction with indicators such as tool wear to eliminate unattainable candidate solutions.

[0073] When the target product is a cleaning agent, the physical constraint model can be constructed and constraints applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes key components such as surfactants, detergent builders, solvents / auxiliaries as constraints and comprehensively adopts detergency kinetics and interfacial chemical mechanisms to jointly limit the detergency rate, detergency upper limit, and impact on the substrate.

[0074] S21: Establish a component ratio constraint model Preferably, a detergency synergy relationship is used to describe the synergistic contribution of surfactants and builders, and the detergency synergy equation is as follows: Where D is the detergency rating, and c s c is the surfactant concentration. a This refers to the concentration of the builder. The surfactant concentration (e.g., 1%-5%) and the builder / surfactant mass ratio (e.g., 0.5-1.0) can also be set to ensure synergistic effectiveness and system stability.

[0075] S22: Establish a process-structure constraint model Preferably, a decontamination kinetic equation is used to describe the effects of temperature and time on the decontamination rate. The decontamination kinetic equation is as follows: Where α is the temperature coefficient.

[0076] When using ultrasonic enhancement methods, an ultrasonic cavitation-related expression can be introduced to establish a mapping between "ultrasonic power and particle removal efficiency," and the equation is as follows: This limits the process window to avoid candidate solutions that are "theoretically high in decontamination but unattainable by the process".

[0077] S23: Establish a performance prediction constraint model Preferably, the performance boundary is checked for consistency using a decontamination rate calculation formula, wherein the decontamination rate equation is: Furthermore, the influence of corrosion rate relationship constraints on the substrate can be introduced, and the corrosion rate equation is as follows: Candidate solutions that are unacceptable to the substrate are eliminated by setting boundaries for D (e.g., ≥95%) and the upper limit of corrosion rate.

[0078] When the target product is a textile chemical, the physical constraint model can be constructed and constraints can be applied as follows: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure that is hierarchically classified according to purpose, main performance, preparation process, and auxiliary performance; S2: Establish a physical constraint model. This model takes key components such as "functional components-fixation system-carrier / auxiliary agent" as constraint objects, and comprehensively adopts the fabric adsorption-diffusion mechanism and curing reaction mechanism to jointly limit the dyeing / finishing accessibility, fixation reaction quantification and wash resistance stability.

[0079] S21: Establish a component ratio constraint model Preferably, the adsorption-diffusion equilibrium relationship is used to describe the correspondence between solution concentration and fabric loading, and the equilibrium equation is: Among them, c f c represents the concentration of functional components in the fabric. s Where K is the solution concentration and K is the partition coefficient. A functional component concentration window (e.g., 0.5%-3%) can be set to ensure adsorption without overloading.

[0080] Simultaneously, it is preferable to establish a molar ratio constraint between the fixing agent and the dye using a stoichiometric relationship of the fixing reaction, wherein the constraint is: n 固色剂 :n 染料 =1.0-1.5 This is to avoid problems such as hardening of the hand and fluctuations in color fastness caused by insufficient or excessive color fixation.

[0081] S22: Establish a process-structure constraint model The adsorption-diffusion kinetic equation is preferably used to describe the effects of finishing temperature, roll-off rate, etc., on the concentration distribution inside the fabric. The equation is as follows: Where D is the diffusion coefficient and x is the positional variable in the thickness direction. A mapping link of "baking temperature / time - degree of crosslinking - functional film stability" is established by combining curing kinetics (referencing the curing kinetics of adhesives) to limit the structural parameters to within an achievable range.

[0082] S23: Establish a performance prediction constraint model Preferably, a softness evaluation relationship is used to set a theoretical range for the feel performance, and the equation is: Where S is a measure of softness, E is the elastic modulus, and ρ is the density.

[0083] Meanwhile, the wash fastness is preferably defined by the relationship between the load amount and the film thickness, and the equation is: Where L is the characterization of wash fastness grade, and δ is the thickness of the functional film. This allows setting achievable boundaries for feel ratings (e.g., ≥4 grades) and colorfastness (e.g., ≥90%) to eliminate physically invalid candidate solutions.

[0084] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A physical constraint model for material formulation optimization, characterized in that, The physical constraint model is applicable to the formulation design and performance prediction of finely processed polymer products. It adopts a three-layer constraint architecture, including: (1) Component ratio constraint model: Based on the principles of chemical composition and thermodynamic stability, component molar ratio constraint equation or mass ratio constraint equation is established to limit the proportion of each component to be within the thermodynamically stable range. (2) Process-structure constraint model: The mapping relationship between processing parameters and material structure parameters is established by using reaction kinetic theory equations or rheological theory equations. This model is used to characterize the control of process parameters on molecular chain structure and aggregate structure, and to limit the structure parameters to the manufacturable range. (3) Performance prediction constraint model: Based on the classical theory of materials science and the hybrid law, the theoretical value range of the performance index is constructed to limit the predicted performance to be within the physically achievable range; The component ratio constraint model, process-structure constraint model and performance prediction constraint model work together to construct a quantitative mapping relationship of the entire link of component-process-structure-performance, and are used to eliminate candidate solutions that violate any constraint conditions, so as to avoid generating pseudo-optimal solutions that exceed the theoretical limit.

2. The physical constraint model for material formulation optimization according to claim 1, characterized in that, The polymer fine processing products are selected from at least one of plastics, rubber, adhesives, inks, coatings, oils, surface treatment agents, metalworking fluids, cleaning agents, and textile chemicals.

3. The physical constraint model for material formulation optimization according to claim 1, characterized in that, The component ratio constraint model limits the range of component ratios from the perspective of thermodynamic stability; the process-structure constraint model describes the regulation law of process parameters on structure from the perspective of kinetics; and the performance prediction constraint model limits the range of performance index values ​​from the perspective of performance theoretical limits.

4. The physical constraint model for material formulation optimization according to claim 3, characterized in that, The constraints of the component ratio constraint model are selected from at least one of the following: component molar ratio constraint, component mass ratio constraint, compatibility constraint, and phase separation risk constraint.

5. The physical constraint model for material formulation optimization according to claim 2, characterized in that, The component ratio constraint model is constrained based on the mechanism principle of the target product category. The mechanism principle is selected from at least one of the following: free radical polymerization reaction functional group activity principle, rubber vulcanization reaction mechanism, adhesive reaction functional group activity principle, pigment dispersion and film formation mechanism, film-forming resin crosslinking reaction mechanism, lubrication mechanism, surface modification reaction mechanism, cooling-lubrication-rust prevention synergistic mechanism, stain removal mechanism, and fabric modification mechanism.

6. The physical constraint model for material formulation optimization according to claim 4, characterized in that, The phase separation risk constraint is established based on at least one of the following: interaction parameter, solubility parameter difference, and ratio of pigment volume concentration to critical pigment volume concentration.

7. The physical constraint model for material formulation optimization according to claim 1, characterized in that, The theoretical equations used in the process-structure constraint model are selected from at least one of the following: Arrhenius equation, vulcanization reaction kinetic equation, Mooney-Rivlin equation, curing reaction kinetic equation, viscosity-temperature equation, drying kinetic equation, coating kinetic equation, adsorption kinetic equation, film formation kinetic equation, cooling kinetic equation, diffusion kinetic equation, and adsorption-diffusion equation.

8. The physical constraint model for material formulation optimization according to claim 1, characterized in that, The theoretical equations used in the performance prediction constraint model are selected from at least one of the following: the Mark-Howewen equation, the rubber elastic modulus equation, the Akron abrasion equation, the cohesive energy density equation, the interfacial tension equation, the abrasion resistance equation, the hardness equation, the oxidation kinetics equation, and the decontamination kinetics equation.

9. A method for constructing a physical constraint model for material formulation optimization according to any one of claims 1-8, characterized in that, include: S1: Collect multi-source basic data and perform standardization processing, then construct a four-level standardized dataset structure; S2: Establish a physical constraint model, construct a quantitative mapping relationship across the entire chain of components, processes, structures, and performance, and output several candidate formulations that meet the physical constraints.

10. The method for constructing a physical constraint model for material formulation optimization according to claim 9, characterized in that, The four-level standardized dataset structure is a hierarchical classification data structure based on purpose, main performance, preparation process, and auxiliary performance.

Images