A method and system for predicting the degree of clustering of magnetic particles in a magnetorheological fluid

Abstract

The application discloses a kind of magnetic particle cluster degree prediction method and system in mixed magnetic liquid, it is related to magnetic functional liquid characterization, particle dispersion system modeling and manufacturing process quality monitoring technical field.The method includes: obtaining at least one of the concentration parameter of the magnetic particle of mixed magnetic liquid, carrier medium viscosity parameter, magnetization input parameter, particle size parameter and action time window parameter;According to the equivalent magnetization intensity or equivalent magnetic dipole moment of the magnetic particle of magnetization input parameter determination;Based on the magnetic dipole interaction between the magnetic particle, carrier medium viscous resistance and action time window, calculate the capture range or capture ability parameter;Further calculate potential cluster descriptor, and obtain the cluster degree prediction value of magnetic particle cluster by saturation mapping;According to the cluster degree prediction value, output cluster risk evaluation result or process parameter recommendation result.The application can be used for formula screening, magnetization parameter selection, process window determination and manufacturing process quality evaluation of mixed magnetic liquid.

Description

Technical Field

[0001] This invention belongs to the technical field of magnetic functional liquid characterization, particle dispersion system modeling, and manufacturing process quality monitoring. Specifically, it relates to a method and system for predicting the degree of magnetic particle agglomeration in a magnetically mixed liquid. The magnetically mixed liquid refers to a functional liquid system formed by dispersing magnetic particles in a liquid carrier medium, including but not limited to magnetically mixed inks, magnetically mixed resins, photocurable magnetically mixed precursor liquids, magnetically mixed hydrogel precursor liquids, magnetically mixed elastomer precursor liquids, magnetically mixed coating liquids, and other liquid dispersion systems containing magnetic fillers. This invention establishes the correlation between parameters such as magnetic particle concentration, magnetization, carrier medium viscosity, and action time window with the degree of particle agglomeration, which is used for the formulation design, process parameter selection, agglomeration risk prediction, and manufacturing process quality assessment of magnetically mixed liquids. Background Technology

[0002] Functional liquids containing magnetic particles are widely used in the preparation of magnetically responsive materials, the molding of magnetic composite materials, additive manufacturing, coating, curing, encapsulation, and the fabrication of micro- and nano-functional devices. In such systems, magnetic particles such as neodymium iron boron, ferrite, iron tetroxide, and nickel powder are typically dispersed in resins, hydrogel precursors, elastomer precursors, solvents, or other liquid carrier media to impart magnetic response, programmable magnetization, magnetic actuation, or magnetic sensing functions to the materials. When the mixed magnetic liquid is in an uncured or unformed state, the magnetic particles still possess a certain degree of freedom of movement and are prone to chaining, agglomeration, or network aggregation under the combined influence of factors such as external magnetic fields, pre-magnetized remanent magnetization, interparticle magnetic dipole interactions, carrier viscous resistance, and limited process time windows. In additive manufacturing and photopolymerization printing processes, the degree of agglomeration of magnetic particles directly affects the subsequent performance of the mixed magnetic liquid. Specifically, particle agglomeration may lead to uneven distribution of local magnetic fillers, changes in light transmittance, curing boundary disturbances, material interface widening, and decreased magnetization coding accuracy. During coating, potting, or composite molding processes, particle agglomeration can cause localized performance fluctuations, uneven deposition, structural defects, or decreased batch consistency. Therefore, predicting the degree of magnetic particle agglomeration in advance, before the magnetically mixed liquid enters printing, coating, curing, or other manufacturing steps, is of great significance for formulation selection, process window determination, and quality control.

[0003] Currently, the prediction of the degree of magnetic particle agglomeration in mixed magnetic liquids mainly relies on empirical judgment, limited experimental screening, or post-event microscopic image analysis. Existing technologies have significant shortcomings: First, there is a lack of predictive models that can unify and correlate material formulation parameters, magnetization parameters, and fluid properties. The actual agglomeration process is influenced by multiple factors, including magnetic particle concentration, particle size, remanence or magnetization intensity, carrier medium viscosity, interaction time window, and external disturbance conditions. Judging solely based on a single parameter often fails to reflect the true agglomeration trend. For example, increasing the magnetic field strength usually enhances the magnetic dipole attraction between particles, but in high-viscosity media or within a short time window, particles may not be able to complete effective migration and capture; increasing particle concentration usually increases the probability of particle encounters, but under strong magnetization conditions, agglomeration growth may be limited by local saturation, cluster demagnetization effects, or spatial capture range. Second, existing methods are mostly at the qualitative or semi-quantitative empirical level, typically comparing agglomeration images under different concentrations, magnetic field strengths, or matrix materials through a small number of preliminary experiments, and then manually selecting relatively suitable process parameters. This method is not only costly in terms of trial and error and has a long experimental cycle, but it is also difficult to extend to new types of magnetic particles, different particle size systems, or different liquid carrier media. This leads to a reliance on accumulated experience in establishing process windows, lacking calculable, transferable, and reusable parameter selection criteria. Furthermore, existing methods typically cannot directly output predicted values ​​for agglomeration risk or degree based on input parameters before manufacturing. For additive manufacturing, coating, curing, and encapsulation processes, agglomeration status further affects material distribution uniformity, curing boundary stability, local magnetic response consistency, and the quality of the final part. Post-processing detection using microscopic images is insufficient to guide timely formulation design and process control. Therefore, there is an urgent need in this field to develop a method and system for predicting the degree of magnetic particle agglomeration in mixed magnetic liquids. This system should comprehensively consider factors such as magnetic particle concentration, magnetization, carrier medium viscosity, and finite interaction time to establish a quantitative predictive relationship between process input parameters and agglomeration degree, providing technical support for formulation screening, process window determination, and manufacturing process quality assessment of mixed magnetic liquids. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies that mainly rely on experience-based judgment, repeated experimental screening, or post-event image analysis to evaluate the agglomeration state of magnetic particles in mixed magnetic liquids. This invention provides a method and system for predicting the degree of agglomeration of magnetic particles in mixed magnetic liquids. Based on input parameters such as the magnetic particle concentration, magnetization parameters, carrier medium viscosity, particle size, and interaction time window of the mixed magnetic liquid, the method predicts the degree of agglomeration of magnetic particles in the liquid carrier medium, providing a quantitative basis for formula screening, process parameter setting, and manufacturing process quality assessment of mixed magnetic liquids.

[0005] To achieve the above objectives, the present invention provides the following solution: This invention provides a method for predicting the degree of magnetic particle aggregation in a magnetically mixed liquid. The magnetically mixed liquid is a functional liquid system formed by dispersing magnetic particles in a liquid carrier medium, including but not limited to magnetically mixed inks, magnetically mixed resins, photocurable magnetically mixed precursor liquids, magnetically mixed hydrogel precursor liquids, magnetically mixed elastomer precursor liquids, magnetically mixed coating liquids, or other liquid dispersion systems containing magnetic fillers. The method includes the following steps: Step 1: Obtain the input parameters of the mixed magnetic liquid to be predicted. The input parameters include at least the magnetic particle concentration parameter, the carrier medium viscosity parameter, and the magnetization input parameter. The magnetic particle concentration parameter is used to characterize the content or number density of magnetic particles in the mixed magnetic liquid. The carrier medium viscosity parameter is used to characterize the viscous resistance encountered by the magnetic particles when they move in the liquid carrier medium. The magnetization input parameter is used to characterize at least one of the applied magnetic field, pre-magnetization intensity, magnetization current, magnetization time, or equivalent magnetization intensity.

[0006] Step 2: Determine the equivalent magnetization or equivalent magnetic dipole moment of the magnetic particles based on the magnetization input parameters. Specifically, the magnetizing current, applied magnetic field strength, or pre-magnetization conditions can be converted into the equivalent magnetization or equivalent magnetic dipole moment of a single particle based on a preset magnetization calibration relationship; the magnetization calibration relationship can be a linear calibration relationship, a saturation-type calibration relationship, a lookup table relationship, an interpolation relationship, or a functional relationship obtained from experimental calibration.

[0007] Step 3: Based on the magnetic dipole interaction between magnetic particles and the viscous resistance of the carrier medium, determine the capture range or capture capability parameter of the magnetic particles within a preset action time window. The preset action time window is the time allowed for relative migration and aggregation of magnetic particles during the printing, coating, curing, transfer, settling, stirring, or other manufacturing steps of the magnetically mixed liquid. The capture range increases with increasing equivalent magnetization or equivalent magnetic dipole moment, increases with increasing action time window, and decreases with increasing carrier medium viscosity.

[0008] Step 4: Based on the magnetic particle concentration parameter, equivalent magnetization or equivalent magnetic dipole moment, carrier medium viscosity parameter, and capture range, calculate a potential clustering descriptor to characterize the aggregation trend of magnetic particles. The potential clustering descriptor characterizes the degree to which magnetic particles mutually capture, chain, aggregate, or network within the stated action time window.

[0009] In one embodiment, the potential agglomeration descriptor is positively correlated with the magnetic particle concentration parameter, positively correlated with the equivalent magnetization or equivalent magnetic dipole moment, and negatively correlated with the carrier medium viscosity parameter. Preferably, the potential agglomeration descriptor can be represented as: N = C·ρ a ·M b ·η c Where N is the potential agglomeration descriptor, ρ is the magnetic particle concentration parameter, M is the equivalent magnetization or magnetization parameter related to the equivalent magnetic dipole moment, η is the carrier medium viscosity parameter, C is the proportionality coefficient, and a, b, and c are exponential parameters; where a and b are positive numbers and c is a negative number.

[0010] Step 5: Perform saturation mapping on the potential cluster descriptor to obtain the predicted value of magnetic particle clustering degree. The saturation mapping is used to characterize the nonlinear change characteristic of the predicted output tending to a plateau value after the clustering degree increases to a certain range.

[0011] In one embodiment, the predicted value of the magnetic particle agglomeration degree is the predicted agglomeration index AI. pred And satisfy: AI pred = A·N γ / (N γ + N s γ ) Where A is the upper limit parameter of the clustering index, γ is the parameter that adjusts the steepness of the mapping curve, and N... s is a semi-saturated cluster descriptor, and N is a potential cluster descriptor.

[0012] Step 6: Output the predicted value of the magnetic particle agglomeration degree, and generate agglomeration risk assessment result or process parameter recommendation result based on the predicted value. The agglomeration risk assessment result includes at least one of low agglomeration risk, medium agglomeration risk, and high agglomeration risk; the process parameter recommendation result includes a recommended magnetic particle concentration range, a recommended magnetization intensity range, a recommended carrier medium viscosity range, a recommended action time window, or a combination of the above parameters.

[0013] In a preferred embodiment, to characterize the effect of magnetic particle concentration on agglomeration degree under different magnetization intensities, the method further includes constructing an effective concentration index related to the magnetization input parameter. The effective concentration index varies with the magnetization input parameter and is used to correct for the influence of the magnetic particle concentration parameter on the potential agglomeration descriptor.

[0014] Preferably, the effective concentration index satisfies: a eff (I) = a0 / [1+(I / I ρ ) p ] Where I is the magnetizing current or equivalent magnetization input parameter, a0 is the concentration exponent under low magnetization conditions, and I ρ is the characteristic magnetization input parameter that affects the decay due to concentration, and p is the decay shape parameter.

[0015] Accordingly, the potential cluster descriptor can be represented as: M(I) = M max ·[1-exp(-I / I0)] Among them, M max I0 is the limiting magnetization parameter, and I0 is the magnetization saturation characteristic parameter.

[0016] In a preferred embodiment, the exponent c of the viscosity parameter of the carrier medium is taken as -2 / 3, or determined based on experimental calibration, historical data fitting, or material system calibration.

[0017] In a preferred embodiment, the method further includes a parameter calibration step: obtaining multiple sets of mixed magnetic liquid samples with known magnetic particle concentration parameters, carrier medium viscosity parameters, and magnetization input parameters; obtaining the measured agglomeration index or agglomeration grade corresponding to each sample; and calibrating C, a, b, c, A, γ, Ns, a0, and I based on the measured agglomeration index or agglomeration grade. ρ p, M max The parameters are fitted or calibrated with at least one parameter in I0 to obtain predictive model parameters applicable to specific magnetic particles, liquid carrier media, or manufacturing process conditions.

[0018] In a preferred embodiment, the measured clustering index is obtained by at least one of the following methods: microscopic image analysis, online imaging detection, offline image detection, simulation results, historical database, or manually calibrated samples.

[0019] In a preferred embodiment, the method further includes a model update step: after obtaining new mixed magnetic liquid test samples and corresponding measured agglomeration degree, the new sample data is added to the calibration dataset, and the prediction model parameters are updated to improve the prediction consistency under different batches of materials, different equipment, or different process conditions.

[0020] This invention also provides a system for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid, comprising: The parameter acquisition module is used to acquire at least one of the following parameters: magnetic particle concentration parameter, carrier medium viscosity parameter, magnetization input parameter, particle size parameter, and action time window parameter of the magnetically mixed liquid to be predicted; A magnetization conversion module is used to determine the equivalent magnetization intensity or equivalent magnetic dipole moment of the magnetic particles based on the magnetization input parameters. The capture capability calculation module is used to calculate the capture range or capture capability parameters of magnetic particles based on the magnetic dipole interaction between magnetic particles, the viscous resistance of the carrier medium, and the action time window. The cluster descriptor calculation module is used to calculate potential cluster descriptors based on magnetic particle concentration parameters, equivalent magnetization or equivalent magnetic dipole moment, carrier medium viscosity parameters, and capture range. The clustering index prediction module is used to perform saturation mapping on the potential clustering descriptor to obtain a predicted value of the degree of magnetic particle clustering. The output module is used to output the predicted value of the agglomeration degree of the magnetic particles and generate agglomeration risk assessment results or process parameter recommendation results.

[0021] In a preferred embodiment, the system further includes a parameter calibration module, which is used to fit, update, or calibrate the prediction model parameters based on known input parameters and measured clustering indices.

[0022] In a preferred embodiment, the system further includes a database module for storing model parameters, measured agglomeration index, and agglomeration risk threshold under different magnetic particle types, particle sizes, carrier media, viscosity ranges, magnetization conditions, and manufacturing process conditions.

[0023] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores computer program instructions, and the processor is configured to implement, when executing the computer program instructions, the method for predicting the degree of agglomeration of magnetic particles in a mixed magnetic liquid as described above.

[0024] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for predicting the degree of agglomeration of magnetic particles in a mixed magnetic liquid as described in any of the preceding claims.

[0025] The present invention has the following beneficial effects: This invention integrates influencing factors such as magnetic particle concentration, magnetization, carrier medium viscosity, and action time window into a unified prediction model. This allows for quantitative prediction of magnetic particle agglomeration before the actual printing, coating, curing, or forming of the mixed magnetic liquid, reducing the uncertainty of relying solely on repeated trial and error or post-processing image analysis. The agglomeration prediction process is constructed based on the competitive relationship between magnetic dipole attraction and viscous resistance, enabling the prediction results to reflect physical trends such as enhanced magnetization promoting agglomeration, increased viscosity inhibiting particle migration, and a limited time window restricting the particle capture range. This results in clear parameter meanings and interpretability. By introducing a saturation mapping relationship, the potential agglomeration descriptor is converted into agglomeration index prediction value, which can adapt to the nonlinear and saturation changes in magnetic particle agglomeration under strong magnetization, high concentration, or limited space conditions, thereby improving the applicability of the prediction results under different operating conditions. The model parameters can be calibrated and updated using measured agglomeration indices, microscopic image analysis results, online detection results, or historical databases, making the method applicable to different magnetic particle types, particle sizes, carrier media, and manufacturing processes. The predicted agglomeration degree, agglomeration risk level, or recommended process parameters output by this invention can be used for formula screening, magnetization parameter selection, viscosity control, manufacturing process quality assessment, and determination of subsequent printing, coating, or curing process windows for mixed magnetic liquids. Attached Figure Description

[0026] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings: Figure 1 A flowchart of a method for predicting the degree of agglomeration of magnetic particles in a mixed magnetic liquid provided by an embodiment of the present invention; Figure 2 This is a schematic diagram of the agglomeration mechanism of magnetic particles in a mixed magnetic liquid provided in an embodiment of the present invention; Figure 3 A schematic diagram illustrating the process of converting magnetization input parameters into equivalent magnetization intensity or equivalent magnetic dipole moment, as provided in embodiments of the present invention. Figure 4 A schematic diagram of a model for calculating potential agglomeration descriptors based on particle concentration, magnetization, carrier medium viscosity, and interaction time window, provided in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the saturation mapping relationship between potential cluster descriptors and clustering degree prediction values ​​provided in an embodiment of the present invention. Figure 6 This is a flowchart of model parameter calibration and update provided in an embodiment of the present invention; Figure 7 This is a structural block diagram of a system for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid, provided by an embodiment of the present invention. Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of protection of this application. Any equivalent substitutions or conventional modifications made to parameters, formulas, module divisions, or application scenarios based on the inventive concept of this application shall fall within the scope of protection of this application.

[0028] This application provides a method and system for predicting the degree of magnetic particle agglomeration in a magnetically mixed liquid. The magnetically mixed liquid refers to a functional liquid system formed by dispersing magnetic particles in a liquid carrier medium, including but not limited to magnetically mixed inks, magnetically mixed resins, photocurable magnetically mixed precursor liquids, magnetically mixed hydrogel precursor liquids, magnetically mixed elastomer precursor liquids, magnetically mixed coating liquids, and other liquid dispersion systems containing magnetic fillers. The magnetic particles can be neodymium iron boron particles, samarium cobalt particles, ferrite particles, iron tetroxide particles, nickel particles, or other particles with magnetic response characteristics. The liquid carrier medium can be a resin, hydrogel precursor liquid, elastomer precursor liquid, solvent system, or other liquid medium capable of dispersing magnetic particles. See also... Figure 1 The method for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid provided in this embodiment includes the following steps: S101, Obtain the input parameters of the mixed magnetic fluid to be predicted.

[0029] Specifically, for image acquisition, inverted microscopes and industrial cameras can be used to obtain observation images under constant illumination conditions. First, image denoising and artifact removal are performed on the original image, followed by binarization segmentation, marking magnetic particle regions as 1 and non-magnetic particle regions as 0. After small connected component removal, a quality mask representing the real particle regions can be generated. For image denoising, spatial filtering, transform domain filtering, or deep learning-based methods can be used. For artifact removal, JPEG block artifact removal algorithms, interpolation or deep learning restoration methods to repair damaged pixels, algorithms to correct lens distortion, and artifact removal through multi-frame fusion or deep learning can be employed. The image optimization processes—image denoising, artifact removal, and small connected component removal—can be selectively used as needed.

[0030] S102, determine the equivalent magnetization or equivalent magnetic dipole moment of the magnetic particles based on the magnetization input parameters.

[0031] Specifically, see Figure 2In one implementation, S201 first acquires the magnetizing current I or the applied magnetic field strength B, and then converts it into the equivalent magnetization M of the magnetic particles using the magnetization calibration relationship in S202. The magnetization calibration relationship can be obtained through experimental calibration, table lookup, interpolation, hysteresis curve fitting, or a preset function. For magnetic particles with saturation magnetization characteristics, the equivalent magnetization M can be represented by the following saturation-type function: M(I) = M max ·[1-exp(-I / I0)] Among them, M max I is the limiting magnetization parameter, I0 is the magnetization saturation characteristic parameter, and I is the magnetizing current or equivalent magnetization input parameter.

[0032] In another implementation, S204 can further calculate the equivalent magnetic dipole moment m of a single particle based on the particle volume V: m=M·V Where V represents the volume of the magnetic particle. For approximately spherical particles, V can be taken as V = πd³ / 6, where d is the particle diameter. For non-spherical particles, an equivalent volume or a volume determined by particle morphology parameters can be used.

[0033] S103, based on the magnetic dipole interaction between magnetic particles, the viscous resistance of the carrier medium, and the action time window, determines the capture range or capture capability parameters of the particles.

[0034] Specifically, see Figure 3 When the magnetically mixed liquid is in an unsolidified or unformed state, the magnetic particles possess a certain degree of freedom of motion. After magnetization, the magnetic particles can be equivalent to particle units with magnetic dipole moments. The magnetic dipole interaction between adjacent particles promotes particle approach, chaining, aggregation, or the formation of network structures; simultaneously, the viscous resistance of the carrier medium inhibits particle migration. Whether particles can be captured within a finite time depends on the relationship between magnetic dipole attraction, viscous resistance, and the time window of action.

[0035] In one embodiment, the particle capture process can be represented as an overdamped motion process. The equivalent attraction between particles increases with the increase of the equivalent magnetic dipole moment m, decreases with the increase of the viscosity η of the carrier medium, and increases with the increase of the action time window T. From this, the parameter R characterizing the particle capture range can be obtained. abs The R abs The following trend relationship is satisfied: R abs ∝ F(m,T,η) Among them, Rabs increases with increasing m, increases with increasing T, and decreases with increasing η.

[0036] Preferably, it can be expressed in the following form: R abs =[Rc 5 +κ·m²·T / η]¹ / 5 Where Rc is the particle contact distance or equivalent minimum spacing, and κ is a coefficient related to particle shape, interaction orientation, and unit conversion. It should be noted that the above expression is only one implementation method; in practical applications, lookup table relationships, empirical functions, experimental calibration functions, or capture range functions obtained through numerical simulation can also be used.

[0037] S104. Calculate the potential agglomeration descriptor based on the magnetic particle concentration parameter, equivalent magnetization or equivalent magnetic dipole moment, carrier medium viscosity parameter, and action time window.

[0038] Specifically, see Figure 4 The potential agglomeration descriptor N characterizes the degree to which magnetic particles are likely to be captured, chained, aggregated, or networked within a finite time window under a given mixed magnetic liquid formulation and magnetization conditions. This potential agglomeration descriptor N is positively correlated with the magnetic particle concentration parameter ρ, the equivalent magnetization intensity M or the equivalent magnetic dipole moment m, and the action time window T, and negatively correlated with the carrier medium viscosity η.

[0039] In one implementation, the potential cluster descriptor N can be represented in the following form: N=C·ρᵃ·Mᵇ·ηᶜ Where C is a proportionality coefficient, ρ is the magnetic particle concentration parameter, M is the equivalent magnetization intensity, η is the carrier medium viscosity parameter, and a, b, and c are exponential parameters. Preferably, a and b are positive numbers, and c is a negative number.

[0040] In another implementation, to further characterize the effect of particle concentration on the degree of agglomeration under different magnetization intensities, an effective concentration index a related to the magnetization input can be introduced. eff (I) The potential cluster descriptor N is represented as: N=C·ρ^a eff (I)·[M(I)]ᵇ·ηᶜ Where M(I) is the equivalent magnetization determined by the magnetization input parameter I, a eff (I) is the effective concentration index related to the magnetization input. Preferably, a eff (I) Satisfies: a eff (I)=a0 / [1+(I / I ρ )^p] Where a0 is the concentration index under low magnetization conditions, I ρThe characteristic magnetization input parameter is p, where p is the decay shape parameter. This setting describes how, under higher magnetization conditions, the degree of agglomeration is gradually affected by factors such as local saturation, limited capture range, or redistribution of the magnetic field within the cluster, thus causing changes in concentration sensitivity.

[0041] In another embodiment, the viscosity index c of the carrier medium can be fixed to a theoretically or empirically determined value, or it can be obtained through experimental calibration. Preferably, c can be -2 / 3, or it can be calibrated according to the specific type of magnetic particles, the liquid carrier medium, and the manufacturing process conditions.

[0042] S105, perform saturation mapping on the potential agglomeration descriptor to obtain the predicted value of the agglomeration degree of magnetic particles.

[0043] Specifically, see Figure 5 Since the degree of particle agglomeration in actual mixed magnetic fluids typically does not increase linearly indefinitely, but rather exhibits a gradual saturation trend under conditions of strong magnetization, high concentration, or confined space, this embodiment further maps the potential agglomeration descriptor N to a predicted agglomeration degree value. This predicted agglomeration degree value can be the predicted agglomeration index AI. pred It can also be an output quantity that is monotonically related to the clustering risk level, clustering degree level, or other clustering degree.

[0044] In one implementation, the clustering index prediction value AI pred satisfy: AI pred =A·N γ / (N γ +Ns γ ) Where A is the upper limit parameter for the predicted clustering index, γ is the parameter controlling the steepness of the mapping curve, Ns is the semi-saturated clustering descriptor, and N is the potential clustering descriptor. From this formula, it can be seen that when N is small, AI... pred AI increases significantly with increasing N; when N is large, AI... pred It gradually approaches the plateau value A. This saturation mapping relationship can reflect the nonlinear variation trend of the predicted output under strong clustering conditions.

[0045] In another implementation, a logistic function, piecewise function, lookup table function, normalization function, or other monotonically saturated function obtained by fitting experimental data can be used to map the potential cluster descriptor N to a clustering degree prediction value.

[0046] S106, outputs cluster risk assessment results or process parameter recommendation results based on the predicted clustering degree.

[0047] Specifically, AI can be predThe risk of clustering is determined by comparing the data with a preset threshold, classifying it as low, medium, or high. For example, when AI... pred When the risk is less than the first threshold, output low clustering risk; when AI pred When the value is between the first and second thresholds, the output shows the clustering risk; when AI... pred If the value exceeds the second threshold, a high agglomeration risk is output. The first and second thresholds can be determined through historical experimental data, material system calibration, quality requirements, or manufacturing process requirements.

[0048] In one implementation, if the output indicates a high risk of agglomeration, the system can further provide suggestions for adjusting process parameters, including at least one of the following: reducing the magnetic particle concentration, reducing the magnetization input parameters, increasing the carrier medium viscosity, shortening the particle migration time window, changing the mixing or curing sequence, and adjusting the sample preparation waiting time. If the output indicates a low risk of agglomeration, the system can indicate that the current mixed magnetic liquid formulation and magnetization parameters meet the preset agglomeration risk requirements. If the output indicates a medium risk of agglomeration, the system can indicate that further experimental verification or local parameter optimization is required.

[0049] See Figure 6 This embodiment also provides a method for model parameter calibration and updating. The method includes: S601, acquire input parameters for multiple sets of mixed magnetic liquid samples. The input parameters include at least one of the following: magnetic particle concentration parameter ρ, carrier medium viscosity parameter η, magnetization input parameter I, particle size parameter d, and action time window T.

[0050] S602, Obtain the measured clustering degree corresponding to each group of samples. The measured clustering degree can be the clustering index obtained through microscopic image analysis, or it can be the clustering level from online detection results, offline detection results, simulation results, manually calibrated levels, or historical databases.

[0051] S603, establish the error function between the predicted and measured values, and fit the model parameters. The model parameters include C, a, b, c, A, γ, and N. s 、a0、I ρ p, M max At least one of I0 and I0. Preferably, the model parameters can be determined using least squares method, weighted least squares method, nonlinear optimization, grid search, Bayesian optimization or other parameter fitting methods.

[0052] In one implementation, if the measured clustering index is AI meas The predicted clustering index is AI. pred The model parameters can then be determined by minimizing the following error function: min Σ w i ·[AI meas,i-AI pred,i ]² Where i is the sample number, w i Let be the weight corresponding to the i-th sample. The weight can be determined based on repeated measures variance, image quality, sample confidence, or process importance.

[0053] S604 stores the fitted model parameters in the database for subsequent prediction of the aggregation degree of similar mixed magnetic liquid samples.

[0054] S605: When new sample data is obtained, the new input parameters and the measured degree of clustering are added to the calibration dataset, and the model parameters are updated. This model update step can improve the consistency of prediction results under different batches of materials, different equipment, and different process conditions.

[0055] In one specific application embodiment, the mixed magnetic liquid to be predicted is a photocurable resin system containing NdFeB magnetic particles. First, the mass fraction ρ of the magnetic particles, the viscosity η of the photocurable resin, the magnetizing current I, and the particle size d are obtained. Then, based on the magnetizing current I and a pre-calibrated magnetization function, the equivalent magnetization M(I) is calculated, and the equivalent magnetic dipole moment m is obtained by combining the particle volume. Subsequently, the potential agglomeration descriptor N is calculated based on the magnetic particle concentration parameter, the equivalent magnetization, the viscosity of the carrier medium, and the action time window. Finally, N is substituted into the saturation mapping function to obtain the predicted agglomeration index AI. pred If AI pred If the preset threshold is exceeded, the formulation is deemed to have a high risk of agglomeration under the current magnetization conditions, and it is recommended to reduce the particle concentration, reduce the magnetization intensity, or increase the viscosity of the carrier medium; if AI pred If the value is below the preset threshold, the formula is determined to meet the preset agglomeration risk requirements and can proceed to the subsequent printing, coating, or curing steps.

[0056] See Figure 7 This application also provides a device 700 for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid. The system includes a parameter acquisition module 701, a magnetization conversion module 702, a capture capability calculation module 703, an agglomeration descriptor calculation module 704, an agglomeration index prediction module 705, a parameter calibration module 706, an output module 707, and a database module 708.

[0057] The parameter acquisition module 701 is used to acquire at least one of the following parameters: magnetic particle concentration parameter, carrier medium viscosity parameter, magnetization input parameter, particle size parameter, and action time window parameter of the mixed magnetic liquid to be predicted.

[0058] The magnetization conversion module 702 is used to determine the equivalent magnetization intensity or equivalent magnetic dipole moment of magnetic particles based on the magnetization input parameters. The magnetization conversion module can access magnetization calibration curves, magnetization functions, lookup tables, or experimental calibration databases.

[0059] The capture capability calculation module 703 is used to calculate the capture range or capture capability parameters based on the magnetic dipole interaction between magnetic particles, the viscous resistance of the carrier medium, and the action time window.

[0060] The cluster descriptor calculation module 704 is used to calculate potential cluster descriptors based on magnetic particle concentration parameters, equivalent magnetization or equivalent magnetic dipole moment, carrier medium viscosity parameters, and action time window.

[0061] The clustering index prediction module 705 is used to perform saturation mapping on potential clustering descriptors to obtain clustering index prediction values ​​or clustering degree prediction values.

[0062] The parameter calibration module 706 is used to fit, update, or calibrate the model parameters based on the known input parameters and the measured degree of clustering.

[0063] The output module 707 is used to output predicted values ​​of agglomeration degree, agglomeration risk level, or recommended process parameters. The output module can output prediction results through a display interface, data file, communication interface, control system interface, or host computer software.

[0064] Database module 708 is the preferred module, used to store model parameters, measured agglomeration index, agglomeration risk threshold and historical prediction results under different magnetic particle types, particle sizes, carrier media, viscosity ranges, magnetization conditions and manufacturing process conditions.

[0065] See Figure 8 This application also provides a computer device 800. The computer device includes a processor 801, a memory 802, a communication bus 804, an input device 805, and an output device 806. The memory stores a computer program 803, and when the processor executes the computer program, it implements the method for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid as described in any of the above embodiments.

[0066] The input interface can be used to receive user-inputted parameters such as magnetic particle concentration, carrier medium viscosity, magnetization parameters, particle size, and action time window. It can also receive data transmitted from external detection equipment, databases, or a host computer. The output interface can output predicted agglomeration index values, agglomeration risk levels, recommended process parameters, or model calibration results. The computer equipment can be a desktop computer, industrial computer, server, embedded computing device, edge computing device, or other electronic equipment with data processing capabilities.

[0067] This application also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the method for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid as described in any of the above embodiments. The computer-readable storage medium can be a read-only memory, random access memory, flash memory, hard disk, solid-state drive, optical disk, portable storage medium, or other medium capable of storing a computer program.

[0068] It should be noted that the specific function forms, parameter names, threshold settings, and module divisions listed in the above embodiments are merely preferred embodiments for illustrating the technical solution of this application and do not constitute a limitation on the scope of protection of this application. In practical applications, those skilled in the art can adjust or replace the magnetization conversion relationship, potential agglomeration descriptor function, saturation mapping function, and risk threshold according to different magnetic particle types, particle size distributions, carrier media, manufacturing processes, and detection conditions. As long as the prediction of the degree of magnetic particle agglomeration is based on the input parameters of the mixed magnetic liquid, it falls within the inventive scope of this application.

Claims

1. A method for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid, characterized in that, Includes the following steps: S1: Obtain the input parameters of the mixed magnetic liquid to be predicted. The input parameters include magnetic particle concentration parameters, carrier medium viscosity parameters, magnetization input parameters, and action time window parameters. S2: Determine the magnetization characterization quantity of the magnetic particles based on the magnetization input parameters. The magnetization characterization quantity includes the equivalent magnetization intensity or the equivalent magnetic dipole moment. S3: Based on the magnetic dipole interaction between magnetic particles, the viscous resistance of the carrier medium, and the action time window parameters, calculate the capture range or capture capability parameters of magnetic particles. S4: Based on the magnetic particle concentration parameter, the magnetization characterization quantity, the carrier medium viscosity parameter, and the capture range or capture capacity parameter, calculate a potential clustering descriptor to characterize the clustering trend of magnetic particles. S5: Perform saturation mapping on the potential agglomeration descriptor to obtain the predicted value of the agglomeration degree of magnetic particles; S6: Output the clustering risk assessment result or process parameter recommendation result based on the predicted value of the magnetic particle agglomeration degree.

2. The method according to claim 1, characterized in that, The magnetically mixed liquid is a functional liquid system formed by dispersing magnetic particles in a liquid carrier medium. The magnetically mixed liquid includes at least one of the following: magnetically mixed ink, magnetically mixed resin, photocurable magnetically mixed precursor liquid, magnetically mixed hydrogel precursor liquid, magnetically mixed elastomer precursor liquid, magnetically mixed coating liquid, or liquid dispersion system containing magnetic fillers.

3. The method according to claim 1, characterized in that, The magnetic particle concentration parameter is at least one of the following: mass fraction, volume fraction, area fraction, particle number density, or proxy variable monotonically related to particle content; the magnetization input parameter is at least one of the following: magnetizing current, applied magnetic field strength, magnetic flux density, pre-magnetization time, equivalent magnetization intensity, or equivalent magnetic dipole moment.

4. The method according to claim 1, wherein step S2 comprises: The magnetization input parameters are converted into the magnetization characterization quantity based on the magnetization calibration model; wherein the magnetization calibration model includes at least one of the following: linear calibration model, nonlinear calibration model, saturation calibration model, hysteresis curve fitting model, demagnetization correction model, lookup table model, interpolation model, regression model, or function model obtained by experimental calibration.

5. The method according to claim 1, characterized in that, In step S3, the capture range or capture capability parameter is jointly determined by the magnetic dipole attraction term, the viscous resistance term, and the action time window term, and increases with the increase of the magnetization characterization quantity, increases with the increase of the action time window parameter, and decreases with the increase of the carrier medium viscosity parameter.

6. The method according to claim 1, characterized in that, The potential cluster descriptor N is calculated using a product model, a power-law model, a lookup table model, a regression model, or an experimentally calibrated prediction model; wherein, in a product model or a power-law model, the potential cluster descriptor N satisfies: N = C·ρ a ·M b ·η c Where ρ is the magnetic particle concentration parameter, X is the magnetization characterization quantity, η is the carrier medium viscosity parameter, C(T,d) is the coefficient related to the action time window parameter T and the particle size parameter d, and a, b, c are exponential parameters, with a and b being positive numbers and c being a negative number.

7. The method according to claim 6, characterized in that, The magnetic particle concentration parameter index in the potential agglomeration descriptor N is an effective concentration index related to the magnetization input parameter or magnetization characterization quantity, and the effective concentration index changes with the magnetization input parameter or magnetization characterization quantity.

8. The method according to claim 1, characterized in that, In step S5, the saturation mapping includes at least one of Hill-type saturation mapping, Logistic-type saturation mapping, piecewise saturation mapping, lookup table mapping, or monotonic saturation mapping obtained by experimental calibration.

9. The method according to claim 8, characterized in that, The predicted value of the agglomeration degree of the magnetic particles is the predicted agglomeration index AI. pred In the Hill-type saturation mapping, the predicted clustering index AI pred Satisfy: AI pred =A·N γ / (N γ + N s γ ), where A is the upper limit parameter for the predicted value of the clustering index, N is the potential cluster descriptor, γ is the parameter for the steepness of the mapping curve, and N s This is a descriptor for semi-saturated aggregates.

10. The method according to claim 1, characterized in that, It also includes model parameter calibration and update steps: Obtain the input parameters and corresponding measured agglomeration degree of multiple sets of mixed magnetic liquid samples; Calculate the predicted value of the degree of magnetic particle agglomeration based on the input parameters; Based on the difference between the measured degree of agglomeration and the predicted value of the degree of agglomeration of magnetic particles, the model parameters are fitted or calibrated. After obtaining a new mixed magnetic liquid sample and its corresponding measured agglomeration degree, the input parameters and measured agglomeration degree of the new mixed magnetic liquid sample are added to the calibration dataset, and the model parameters are updated. The measured degree of agglomeration is obtained by at least one of the following methods: microscopic image analysis, online imaging detection, offline image detection, simulation results, historical database, or manually calibrated samples.

11. A system for predicting the degree of magnetic particle agglomeration in a mixed magnetic liquid, characterized in that, include: The parameter acquisition module is used to acquire the magnetic particle concentration parameters, carrier medium viscosity parameters, magnetization input parameters, and action time window parameters of the mixed magnetic liquid to be predicted. A magnetization conversion module is used to determine the equivalent magnetization intensity or equivalent magnetic dipole moment of the magnetic particles based on the magnetization input parameters. The capture capability calculation module is used to calculate the capture range or capture capability parameters of magnetic particles based on the magnetic dipole interaction between magnetic particles, the viscous resistance of the carrier medium, and the action time window parameters. The cluster descriptor calculation module is used to calculate potential cluster descriptors based on the magnetic particle concentration parameters, equivalent magnetization or equivalent magnetic dipole moment, carrier medium viscosity parameters, and capture range or capture capability parameters. The clustering index prediction module is used to perform saturation mapping on the potential clustering descriptor to obtain a predicted value of the degree of magnetic particle clustering. The parameter calibration module is used to fit, update, or calibrate the model parameters based on the input parameters of multiple sets of mixed magnetic liquid samples and the corresponding measured degree of aggregation. The output module is used to output agglomeration risk assessment results or process parameter recommendation results based on the predicted value of the agglomeration degree of the magnetic particles.

12. A computer device comprising a memory and a processor, the memory storing computer program instructions, the processor being configured to, when executing the computer program instructions, implement the method steps of any one of claims 1 to 11.

13. A computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method steps of any one of claims 1 to 10.

Images