Ultrafiltration permeate regulation method and device based on gradient sponge structure dialysis membrane

Abstract

The application discloses an ultrafiltration permeation regulation method and device based on a gradient sponge structure dialysis membrane and relates to the technical field of hemodialysis. An intelligent response membrane assembly containing a nonlinear aperture gradient is constructed; a signal is collected and a concentration polarization critical index is calculated; when the threshold value is exceeded, a multi-frequency coupling pulse control instruction containing a mechanical resonance excitation waveform and a phase reverse matching osmotic pressure oscillation waveform is generated; mechanical resonance excitation is applied to the blood side to induce membrane layer micro-breathing deformation, and an osmotic pressure oscillation waveform is injected into the dialysate side to form a reverse solute diffusion field; microgel deformation feedback signals are monitored in real time, waveform parameters are dynamically adjusted, and the concentration polarization critical index is returned to a safe interval until the ultrafiltration flux is adaptively regulated. The device comprises an intelligent gradient membrane assembly, a fluid power regulation module, an osmotic pressure wave generation module, a sensing feedback array and a central processing unit, can effectively inhibit membrane pollution and concentration polarization, and improves the system operation stability.

Description

Technical Field

[0001] This invention relates to the field of hemodialysis technology, specifically to an ultrafiltration osmosis regulation method and apparatus based on a gradient sponge structure dialysis membrane. Background Technology

[0002] Hemodialysis is a crucial treatment for end-stage renal disease patients, and the performance of its core component, the dialyzer, directly determines the efficiency of toxin removal and patient prognosis. Ultrafiltration, the core step in dialysis for removing excess water, is primarily driven by transmembrane pressure across the semipermeable membrane. However, in clinical practice, dialysis membranes are highly susceptible to concentration polarization and membrane fouling. When blood flows across the membrane surface, large molecules such as plasma proteins and cell debris adsorb and deposit on the membrane surface and at the pore inlets, forming a dense gel layer. This reduces the effective filtration pore size and dramatically increases hydraulic resistance. To maintain the target ultrafiltration rate, clinicians are often forced to increase transmembrane pressure or blood flow velocity, which not only increases the risk of hemolysis and clotting but also accelerates irreversible pore blockage, leading to premature dialyzer failure. Existing dialysis membranes are mostly symmetrical or asymmetrical structures with finger-like macropores. These structures are prone to pore collapse or insufficient mechanical strength under high pressure and lack effective mechanisms to cope with dynamic fouling. Therefore, how to fundamentally improve the antifouling ability of membranes and maintain high-flux stability by starting with the design of membrane microstructure has been a long-standing technical challenge in the field of blood purification.

[0003] Existing technical solutions mainly focus on two directions: surface modification of membrane materials and optimization of pore structure. On the one hand, researchers widely employ blending modification or surface grafting techniques to introduce hydrophilic polymers such as polyvinylpyrrolidone (PVP) and polyethylene glycol (PEG) in an attempt to form a hydration layer on the membrane surface to hinder protein adsorption. This strategy has been reported in detail in several authoritative publications, including the *Journal of Membrane Science*. On the other hand, improving the non-solvent-induced phase separation (NIPS) process and controlling the composition of the coagulation bath to prepare sponge-like asymmetric membranes with gradient pore distribution has become the mainstream approach in the industry. Compared with existing finger-like pore structures, sponge-like pore structures have more uniform stress distribution and higher mechanical strength, effectively mitigating pore deformation under high pressure. However, existing gradient sponge structure membranes are essentially still passive static structures, and their pore size distribution remains fixed once prepared. Faced with the dynamic changes in blood rheology and the continuous accumulation of contaminants during dialysis, this static structure cannot adaptively adjust. Once the gel layer breaks through the surface protection and enters the deep pores, the existing membrane structure is helpless and can only rely on stopping the machine for cleaning or replacing the dialyzer, which seriously affects the continuity and efficiency of treatment.

[0004] To further address the limitations of static membranes, some existing technologies attempt to introduce external dynamic operating modes or smart responsive materials. For example, periodic backflushing technology is widely used in industrial filtration and some dialysis strategies, briefly reversing the pressure flow to flush the membrane surface. However, as related studies have pointed out, this macroscopic pressure reversal easily causes hemodynamic fluctuations in patients and is difficult to remove stubborn contaminants in deep pores. Furthermore, research on smart membrane materials has made some progress, such as using temperature-sensitive polymers (e.g., PNIPAM) to modify the membrane surface, attempting to adjust pore size through temperature changes. However, human body temperature is relatively constant, lacking sufficient driving temperature difference to achieve effective dynamic pore size adjustment, and the thermal response speed is slow, failing to meet the second-level real-time control requirements during dialysis. Other studies have explored using ultrasound or magnetic fields to assist in antifouling, but introducing exogenous physical fields often leads to increased equipment complexity and poses potential biosafety risks, making large-scale clinical application difficult. Summary of the Invention

[0005] To address the aforementioned technical problems, this application discloses an ultrafiltration osmosis adjustment method and apparatus based on a gradient sponge structure dialysis membrane. The ultrafiltration osmosis adjustment method based on the gradient sponge structure dialysis membrane specifically includes: A smart response membrane module with a nonlinear pore size gradient is constructed. The smart response membrane module includes a gradient sponge-like porous support layer with an exponentially decreasing pore size from the blood side to the dialysate side, and the pore walls of the gradient sponge-like porous support layer are embedded with microgel actuation units that are sensitive to both shear force and osmotic pressure. Acquire real-time transmembrane flux signal and solute concentration difference signal across the membrane, and calculate concentration polarization critical exponent based on real-time transmembrane flux signal and solute concentration difference signal across the membrane; When the concentration polarization critical index exceeds a preset threshold, a multi-frequency coupled pulse control command is generated. The multi-frequency coupled pulse control command includes a mechanical resonance excitation waveform whose frequency varies with the depth of the gradient sponge layer and an osmotic pressure oscillation waveform whose phase is inversely matched. According to the multi-frequency coupled pulse control command, the mechanical resonance excitation waveform is applied to the blood side to induce the gradient sponge-like porous support layer to produce a micro-breathing deformation, and the osmotic pressure oscillation waveform is injected into the dialysate side to form a reverse solute diffusion field. The deformation feedback signal of the microgel actuation unit is monitored in real time, and the spectral distribution of the mechanical resonance excitation waveform and the amplitude of the osmotic pressure oscillation waveform are dynamically adjusted until the concentration polarization critical index falls back to the safe range, thereby achieving adaptive adjustment of the ultrafiltration flux.

[0006] Preferably, the specific steps for constructing the smart response membrane module with a nonlinear pore size gradient include: Hollow fiber membranes were prepared using a solvent-free phase separation method combined with microfluidic spinning technology, by dynamically adjusting the volume ratio of non-solvent to solvent in the coagulation bath. Controlling the phase separation rate to form pores Along the film thickness direction A gradient sponge-like porous support layer exhibiting a nonlinear exponential decay distribution; Wherein, aperture distribution function formula: In the formula, The initial pore size on the blood side surface. The minimum pore size on the dialysate side. The gradient decay coefficient is... It is a non-linear shape factor. This is a dynamic function of the viscosity of the coagulation bath as a function of time. Poly(N-isopropylacrylamide)-acrylic acid copolymer microgels are grafted onto the pore wall surface of the gradient sponge-like porous support layer to form shear force-resistant microgels. With local osmotic pressure A dual-sensitivity microgel actuation unit, wherein the swelling ratio of the microgel actuation unit is... Defined as a dynamic adjustment variable for local porosity.

[0007] Preferably, the step of acquiring the real-time transmembrane flux signal and the solute concentration difference signal across the membrane, and calculating the concentration polarization critical exponent, specifically involves: Obtain blood inlet pressure Export pressure and dialysate side pressure Calculate instantaneous transmembrane pressure Simultaneously, the blood concentration was measured using a conductivity meter. Concentration on the dialysate side Constructing the concentration polarization critical index The calculation model is as follows: In the formula, For instantaneous transmembrane flux, To account for the equivalent boundary layer thickness after microgel deformation, is the effective diffusion coefficient of the solute in the gradient sponge structure. The solute concentration at the membrane surface. This is the shear force fluctuation sensitivity coefficient. This refers to the wall shear force. when When the system is determined to have entered a high-risk region of concentration polarization, the multi-frequency coupled pulse control command generation mechanism is triggered.

[0008] Preferably, the specific methods for generating multi-frequency coupled pulse control commands include: Based on the depth-layered characteristics of the gradient sponge-like porous support layer, the membrane thickness is divided into... The discrete resonant unit, for the th discrete resonant unit Calculate the natural hydroelastic frequency of each resonant element. ; Constructing mechanical resonance excitation waveforms Its spectral distribution for: In the formula, For the first The target amplitude of the layer For bandwidth parameters, This is the phase offset, and With depth It increases linearly to create a traveling wave effect; Constructing phase-inversely matched osmotic pressure oscillation waveforms So that its main frequency component is The fundamental frequency component remains The phase difference is expressed by the formula: ,in For the corresponding frequency The phase angle of the mechanical wave. For the first The amplitude of the oscillation intensity of each frequency component.

[0009] Preferably, the application of excitation via multi-frequency coupled pulse control commands specifically includes: Drive the blood-side proportional valve at frequency Rapidly switch the opening degree to generate the mechanical resonance excitation waveform. Superimposed on the basic transmembrane pressure, this forces the pore walls of the gradient sponge-like porous support layer to undergo radial displacement. ; The radial displacement With local shear stress field The coupling relationship satisfies the viscoelastic dynamics equation: In the formula, The equivalent density of the membrane skeleton, The position-dependent damping coefficient, For microgel swelling rate The dynamic stiffness coefficient, For fluid stress tensor, This serves as the driving force for the microgel under osmotic pressure oscillation. Different concentrations of replacement fluid are injected by a high-precision metering pump on the dialysate side to generate the osmotic pressure oscillation waveform, forming a periodic reverse solvent drag force inside the membrane pores.

[0010] Preferably, the method for real-time monitoring of the deformation feedback signal of the microgel actuation unit includes: Using a fiber Bragg grating embedded within the membrane module housing, the axial strain of the membrane module caused by microgel swelling / shrinkage was obtained. Establish axial strain With average swelling rate of microgel Mapping model: In the formula, This represents the inverse Fourier transform. For the frequency domain representation of the strain signal, For sensor transfer function, For the length of the membrane Equivalent modulus of composite materials in the direction of orientation; Based on calculations Invert the actual tortuosity factor of the current membrane pores. This serves as the basis for dynamically adjusting the spectral distribution of the mechanical resonance excitation waveform.

[0011] Preferably, the dynamic adjustment of the spectral distribution of the mechanical resonance excitation waveform and the amplitude of the osmotic pressure oscillation waveform specifically involves: Construct the system energy efficiency optimization objective function To minimize concentration polarization resistance and maximize effective flux stability, the formula is: In the formula, These are the weighting coefficients, This represents the total input power of the system. Using the MPC algorithm, and As a state variable, with and As a control variable, in each control cycle Solve the above objective function internally and output the optimal spectral parameter set for the next period. and optimal amplitude set If the swelling rate of microgels at a certain depth is detected If the temperature falls below the critical dehydration threshold, the frequency of that layer will automatically increase. amplitude This is to enhance the amplitude of local breathing deformation.

[0012] Preferably, the process of adaptively adjusting the ultrafiltration flux until the concentration polarization critical index falls back to a safe range, thereby achieving adaptive maximization of the ultrafiltration flux, specifically includes: Continuous monitoring of transmembrane flux volatility variance Rate of change of protein concentration in reflux solution When the following conditions are met, self-cleaning is deemed complete and the system switches to steady-state operation mode: In the formula, For flux stability threshold, For frequency locking tolerance, This is the current driving frequency; If the calculated membrane hydraulic resistance is obtained without changing the external pressure... A continuous downward trend has emerged, that is... If the duration exceeds the set window, it confirms that the gel layer inside the membrane pores has been synergistically stripped away by mechanical resonance and osmotic fluctuations.

[0013] The ultrafiltration osmosis conditioning device based on a gradient sponge structure dialysis membrane includes: A smart gradient membrane assembly, which is filled with a hollow fiber membrane having a nonlinear pore size gradient and microgel actuation units embedded in the pore walls. A fluid dynamics control module, connected to the blood-side inlet of the intelligent gradient membrane assembly, is used to generate multi-frequency coupled pulsed blood flow; An osmotic pressure wave generating module is connected to the dialysate side of the intelligent gradient membrane assembly and is used to inject periodically changing dialysate to form a reverse solute concentration wave. A sensor feedback array is distributed at the inlet and outlet of the intelligent gradient membrane module and on the surface of the housing to collect pressure, flow rate, conductivity and membrane strain signals in real time. The central processing unit is electrically connected to the fluid dynamics control module, the osmotic pressure wave generation module, and the sensor feedback array, respectively, and is used to execute the above-mentioned ultrafiltration osmosis regulation method.

[0014] Preferably, the core innovation of the intelligent gradient membrane module lies in the microstructure design of its internal hollow fiber membrane: The hollow fiber membrane has a continuous gradient sponge-like pore distribution in its cross-section, and the pore walls are composed of a matrix polymer material and nanoscale dual-response microgel particles dispersed therein. The particle size of the dual-response microgel particles Matrix pore size at the location To meet specific size matching requirements: Furthermore, the dual-response microgel particles are grafted onto the pore wall surface via chemical bonds, and their grafting density is... Along the film thickness direction It exhibits a Gaussian distribution: In the formula, This is the center location of the gradient transition region. For the distribution width parameter, The structure configuration allows the microgel particles to act as independent nanoactuators under the action of external multi-frequency pulses, generating local eddy current disturbances inside the pores, rather than just causing macroscopic deformation of the overall pores, thereby disrupting the adsorption equilibrium of the solute at the pore throat at the molecular scale.

[0015] Compared with the prior art, the technical solution of this application has the following technical effects: This invention achieves precise optimization of the spatial distribution of membrane pore structure by constructing a smart responsive membrane module with a nonlinear pore size gradient. The gradient sponge-like porous support layer, combined with the microgel actuation unit, can dynamically adjust the pore morphology according to the local fluid environment, effectively reducing the deposition resistance of solute on the membrane surface.

[0016] This invention utilizes multi-frequency coupled pulse control commands to excite the membrane skeleton to generate mechanical resonance and osmotic pressure oscillation at a specific frequency. The dual excitation mechanism forms micro-eddies and reverse solvent drag forces inside the membrane pores, actively destroying the basis for the formation of concentration polarization layers. Through the waveform design with phase reversal matching, the system can effectively remove deep contaminants without causing macroscopic blood flow fluctuations, thereby significantly extending the continuous operation cycle of the membrane module and reducing the need for cleaning and maintenance.

[0017] This invention employs a real-time feedback-based dynamic adjustment strategy, enabling the ultrafiltration permeation regulation process to possess a high degree of intelligence and adaptability. By monitoring the concentration polarization critical index and microgel deformation signal, it automatically optimizes the spectral distribution and amplitude parameters of the excitation waveform, avoiding ineffective energy consumption and preventing membrane structure damage caused by over-excitation, thus achieving a balance between therapeutic safety and clearance efficiency.

[0018] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings.

[0019] The above and other objects, advantages and features of this application will become more apparent to those skilled in the art from the following detailed description of specific embodiments in conjunction with the accompanying drawings. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0021] Based on the description of the figures and their corresponding technical content in the document, the titles of the figures are as follows: Figure 1A schematic diagram of the overall process of ultrafiltration osmosis regulation method based on gradient sponge structure dialysis membrane; Figure 2 Schematic diagram of the preparation of the gradient sponge-like porous support layer and microgel grafting of the smart response membrane module; Figure 3 A schematic diagram illustrating the process of generating multi-frequency coupled pulse control commands and constructing mechanical resonance and osmotic pressure oscillation waveforms; Figure 4 Schematic diagram of the principle of multi-frequency coupled pulse excitation application and reverse solute diffusion field formation; Figure 5 A schematic diagram of the overall structure and module connection of an ultrafiltration osmosis regulation device based on a gradient sponge structure dialysis membrane; Figure 6 A visualization of the dynamic evolution of the flow field structure and microgel distribution inside the membrane pores under different microgel grafting densities; Figure 7 Schematic diagram of instantaneous velocity vector field distribution and micro-vortex chain formation inside membrane pores under four spectral bandwidth modes; Figure 8 A schematic diagram illustrating the dynamic response of the membrane pores, flow velocity power spectral density, and coherence analysis under different phase differences; Figure 9 A comparison of ultrafiltration flux decline and transmembrane pressure growth trends between adaptive control systems and static gradient systems; Figure 10 Scanning electron microscope images of the surface morphology and pore structure of the membrane modules in the experimental and control groups after continuous operation. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. In the following description, specific details such as specific configurations and components are provided merely to help fully understand the embodiments of this application. Therefore, those skilled in the art should understand that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. In addition, for clarity and brevity, descriptions of known functions and structures are omitted in the embodiments.

[0023] It should be understood that the phrase "an embodiment" or "this embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "an embodiment" or "this embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0024] Furthermore, reference numerals and / or letters may be repeated in different examples within this application. Such repetition is for the purpose of simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or settings discussed.

[0025] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" in this article describes another type of relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the related objects before and after it are in an "or" relationship.

[0026] In this article, the term "at least one" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, "at least one of A and B" can mean: A exists alone, A and B exist simultaneously, or B exists alone.

[0027] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion.

[0028] Example 1 mainly describes an ultrafiltration osmosis regulation method based on a gradient sponge structure dialysis membrane, such as... Figure 1 As shown, it specifically includes: A smart response membrane module with a nonlinear pore size gradient is constructed. The smart response membrane module includes a gradient sponge-like porous support layer with an exponentially decreasing pore size from the blood side to the dialysate side, and the pore walls of the gradient sponge-like porous support layer are embedded with microgel actuation units that are sensitive to both shear force and osmotic pressure. Acquire real-time transmembrane flux signal and solute concentration difference signal across the membrane, and calculate concentration polarization critical exponent based on real-time transmembrane flux signal and solute concentration difference signal across the membrane; When the concentration polarization critical index exceeds a preset threshold, a multi-frequency coupled pulse control command is generated. The multi-frequency coupled pulse control command includes a mechanical resonance excitation waveform whose frequency varies with the depth of the gradient sponge layer and an osmotic pressure oscillation waveform whose phase is inversely matched. According to the multi-frequency coupled pulse control command, the mechanical resonance excitation waveform is applied to the blood side to induce the gradient sponge-like porous support layer to produce a micro-breathing deformation, and the osmotic pressure oscillation waveform is injected into the dialysate side to form a reverse solute diffusion field. The deformation feedback signal of the microgel actuation unit is monitored in real time, and the spectral distribution of the mechanical resonance excitation waveform and the amplitude of the osmotic pressure oscillation waveform are dynamically adjusted until the concentration polarization critical index falls back to the safe range, thereby achieving adaptive adjustment of the ultrafiltration flux.

[0029] Furthermore, such as Figure 2 As shown, a smart response membrane module with a nonlinear pore size gradient was constructed, and a smart response membrane module with both nonlinear pore size gradient distribution characteristics and dual-sensitive microgel actuation units was prepared. The core structure is a gradient sponge-like porous support layer. This support layer exhibits a continuous exponential decay characteristic of pore size from the blood side to the dialysate side, and the pore walls are embedded with microgel actuation units that have dual sensitive response characteristics to shear force and osmotic pressure. The entire preparation process integrates a composite process system of non-solvent-induced phase separation method and microfluidic spinning technology, and at the same time, it uses polymer material surface grafting modification technology to complete the embedding of microgel actuation units in the pore walls. Hollow fiber membrane substrates were prepared using a non-solvent-induced phase separation method combined with microfluidic spinning technology. During the preparation process, the volume ratio of non-solvent to solvent in the coagulation bath system was dynamically adjusted. This enables real-time and precise control of the phase separation rate. Compared with volume ratio Satisfying linear relationship ,in This is the phase separation rate control coefficient. Both the phase separation rate and the pore size parameter are determined by the composition of the coagulation bath system and the process temperature. Along the film thickness direction A strictly nonlinear exponential decay distribution is formed, in which the film thickness direction The measurement origin is set at the blood-side surface of the gradient sponge-like porous support layer, and the measurement endpoint is the dialysate-side surface. , The total thickness of the membrane, The value increases monotonically from the blood side to the dialysate side, and the volume ratio of the coagulation bath... As time The dynamic function is dynamically adjusted in real time as the microfluidic spinning process progresses, and the adjustment rule follows... ,in This is the upper limit of the volume ratio. The rate constant is adjusted to the volume ratio to ensure the matching between the phase separation rate and the pore size gradient distribution, resulting in a pore size distribution that follows a mathematical formula. This formula is a nonlinear exponential decay model, where The integral term represents the change in viscosity of the coagulation bath over time. The integration interval is the entire time dimension of the contact between the hollow fiber membrane and the coagulation bath during microfluidic spinning, from 0 to the real-time process time. Dynamic function of coagulation bath viscosity satisfy ,in The initial viscosity of the coagulation bath. The viscosity variation coefficient is... The viscosity change index is in the formula. The initial pore size of the blood-side surface is the characteristic pore diameter of the initial formation of the blood-side surface of the gradient sponge-like porous support layer. The minimum pore size on the dialysate side is the characteristic pore diameter of the dialysate-side surface layer of the gradient sponge-like porous support layer. It represents the lower limit critical parameter of the entire pore size gradient distribution and satisfies a strict numerical relationship. This parameter directly determines the lower limit of the pore size on the dialysate side; The gradient attenuation coefficient is a core characteristic parameter that controls the overall attenuation rate of the pore size from the blood side to the dialysate side. It is a positive constant, and its value is positively correlated with the attenuation rate of the pore size. The larger the value, the faster the attenuation rate of the pore size from the blood side to the dialysate side. The nonlinear shape factor is a key parameter for controlling the degree of nonlinearity in aperture attenuation. It is a positive constant, and its magnitude determines the nonlinear curvature of the aperture attenuation curve. The values ​​correspond to different nonlinear decay trends and can be precisely set according to actual ultrafiltration permeation needs.

[0030] After the gradient sponge-like porous support layer was prepared, a polymer surface grafting modification technique was used to graft poly(N-isopropylacrylamide) and acrylic acid copolymer microgels onto its pore wall surface. The grafting reaction kinetics of the microgel grafting process followed a first-order reaction kinetic model. ,in The grafting reaction rate constant is... For real-time grafting density, To achieve the maximum grafting density, microgel particles are embedded into the pore wall surface through chemical bonding to form microgel actuation units. These microgel actuation units are capable of responding to shear forces. With local osmotic pressure Dual sensitive response characteristics, local osmotic pressure Calculated from the solute concentration difference across the membrane, satisfying ,in The gas constant is... The system's thermodynamic temperature is... The swelling ratio is the molar concentration difference of the solute across the membrane. And swelling rate Local porosity is explicitly defined as a dynamically adjusting variable of local porosity. With swelling rate satisfy ,in Based on the basic porosity, the microgel actuation units undergo swelling or contraction deformation reactions according to changes in shear force and local osmotic pressure in their environment; the swelling rate... With shear force Local osmotic pressure The coupling relationship satisfies ,in The basic swelling ratio of the microgel. Shear force sensitivity coefficient As the osmotic pressure sensitivity coefficient, this deformation directly alters the local porosity and swelling ratio at corresponding locations in the gradient sponge-like porous support layer. The numerical change of the swelling ratio exhibits a strict dynamic response relationship with local shear force and local osmotic pressure. The real-time changes in shear force and osmotic pressure directly drive the swelling ratio. The dynamic adjustment enables precise control of local porosity.

[0031] Furthermore, the real-time transmembrane flux signal and the solute concentration difference signal across the membrane are acquired, and the concentration polarization critical exponent is calculated, specifically as follows: The system collects core physical signals such as pressure and concentration during the dialysis process in real time through multi-dimensional sensing devices. Based on the collected real-time signals, it calculates the instantaneous transmembrane flux and the solute concentration difference across the membrane. Then, it uses a dedicated concentration polarization critical index calculation model to accurately solve the index. Multi-dimensional real-time signal acquisition is performed. High-precision pressure sensing modules are installed at the blood-side inlet, blood-side outlet, and dialysate side of the intelligent response membrane module to monitor the blood-side inlet pressure. Blood outlet pressure and dialysate side pressure Continuous real-time data acquisition, all pressure parameters are time-varying. The dynamic function can accurately reflect the real-time changes in pressure at different locations during dialysis. Based on the three sets of collected pressure signals, the instantaneous transmembrane pressure is calculated using fluid dynamics pressure calculation formulas. ,satisfy Instantaneous transmembrane pressure is a core parameter reflecting the pressure difference across the membrane and is also an important basis for calculating instantaneous transmembrane flux. Meanwhile, high-precision conductivity meters are installed at both the blood-side inlet and outlet and the dialysate-side inlet and outlet of the smart response membrane module, utilizing the linear correlation between solute concentration and conductivity. ,in This is the concentration-conductivity conversion coefficient. Based on the correction value, To measure conductivity, the blood solute concentration was obtained in real time using a conductivity meter. Compared with the solute concentration on the dialysate side Solute concentration difference across the membrane Both types of concentration parameters are time-dependent. The dynamic function accurately reflects the real-time changes in solute concentration across the membrane, thereby obtaining the real-time value of the solute concentration difference across the membrane. Based on the real-time transmembrane pressure collected and calculated, combined with membrane hydraulic resistance Instantaneous transmembrane flux signal was calculated using Darcy's law. ,satisfy Membrane hydraulic resistance Related to membrane pore structure, satisfying ,in The membrane pore tortuosity factor. Using the average pore size of the membrane and combining it with the solute concentration difference signal across the membrane, a specific concentration polarization critical index is constructed. The computational model is used to solve for the concentration polarization critical exponent in real time. The mathematical formula for the computational model is as follows: ,in Instantaneous transmembrane flux refers to the volume of solution passing through a unit membrane area per unit time, and its value change directly reflects the ultrafiltration efficiency of the membrane. To account for the equivalent boundary layer thickness after microgel deformation, is the actual effective thickness of the solute boundary layer on the membrane surface after the microgel actuation unit undergoes swelling / contraction deformation. is the corrected boundary layer characteristic parameter, related to the microgel swelling ratio. Dynamic correlation ,in Based on the thickness of the boundary layer, This is the swelling ratio correction factor. The greater the degree of microgel deformation, the greater the correction range for the equivalent boundary layer thickness. The effective diffusion coefficient of the solute in the gradient sponge structure is given by the blood-side solute concentration. With system temperature A bivariate function that satisfies ,in Based on the basic diffusion coefficient, The concentration influence coefficient is... This is the temperature influence coefficient. The effective diffusion coefficient is the core parameter that reflects the diffusion ability of solutes within the gradient sponge-like porous support layer, with the reference temperature as the reference temperature. The solute concentration at the membrane surface specifically refers to the solute concentration on the blood-side surface of the gradient sponge-like porous support layer. It is a direct characteristic parameter of concentration polarization and is expressed as a time-dependent parameter. The dynamic function of concentration polarization indicates that the higher the value of the solute on the membrane surface, the more severe the concentration polarization, and that the solute concentration on the membrane surface and the bulk concentration satisfy the mass transfer equilibrium relationship. ; The shear force fluctuation sensitivity coefficient is used to quantify the influence of wall shear force fluctuation on the concentration polarization critical index. Its value is precisely set according to the structural characteristics of the gradient sponge-like porous support layer and the response characteristics of the microgel actuation unit. Wall shear force refers to the shear force exerted by blood on the blood sidewalls of the gradient sponge-like porous support layer during blood flow on the membrane surface. Its value is closely related to blood flow velocity and membrane surface structure, and must satisfy certain conditions. ,in For blood dynamic viscosity, For blood flow rate, The hydraulic diameter of the flow channel; It is the first partial derivative of the wall shear force with respect to time, which reflects the real-time rate of change of the wall shear force, and its value can reflect the degree of fluctuation of the shear force. The concentration polarization logarithm is calculated by taking the natural logarithm of the ratio of the solute concentration at the membrane surface to the difference in solute concentration across the membrane. This logarithm represents the degree of concentration polarization. The larger the value of this logarithm, the higher the solute enrichment at the membrane surface and the more significant the concentration polarization phenomenon. After completing the real-time calculation of the concentration polarization critical index, a preset threshold for the concentration polarization critical index is set. This threshold is precisely set based on the ultrafiltration performance of the dialysis membrane, the composition characteristics of the dialysis fluid, and the tolerance to concentration polarization. It is a positive constant. When the calculated real-time concentration polarization critical exponent meets the requirements... Upon entering a high-risk concentration polarization zone, the system immediately determines that it has entered the zone and simultaneously triggers the generation mechanism for subsequent multi-frequency coupled pulse control commands. This enables real-time determination of concentration polarization risk and automatic triggering of subsequent control measures, adjusting the concentration polarization risk level accordingly. With the critical exponent It is used to quantify the degree of risk of concentration polarization.

[0032] Furthermore, such as Figure 3 As shown, a multi-frequency coupled pulse control command is generated based on the concentration polarization critical exponent exceeding the threshold, specifically as follows: Based on the deep layered structure characteristics of the gradient sponge-like porous support layer, the membrane is divided into discrete resonant units. The mechanical resonance excitation waveform with frequency varying with the depth of the gradient sponge layer is constructed by combining the inherent hydroelastic frequency of each unit. At the same time, an osmotic pressure oscillation waveform that is phase-matched with this waveform is constructed. The two types of waveforms together form a multi-frequency coupled pulse control command. Based on the pore size gradient distribution and structural characteristics of the gradient sponge-like porous support layer along the film thickness direction, discrete resonant units are divided along the film thickness direction. The overall thickness of the gradient sponge-like porous support layer is uniformly divided into... A discrete resonant unit, unit thickness , The value is a positive integer, set according to the overall membrane thickness, pore size gradient variation characteristics, and control precision requirements. The discrete resonant unit is divided based on a high degree of matching with the pore size gradient distribution characteristics of the gradient sponge layer, ensuring that each discrete resonant unit is an independent resonant response unit with its own unique intrinsic hydroelastic frequency. All units have a uniform thickness and are numbered sequentially along the membrane thickness direction from the blood side to the dialysate side. , No. The thickness range of each unit is ; For the divided first The inherent hydroelastic frequency of a discrete resonant element is calculated using formulas from fluid elasticity theory. ,satisfy ,in For the first The dynamic stiffness coefficient of the element, For the first The equivalent density of the membrane skeleton of the unit cell, For the first The effective volume of the unit, and the frequency is an inherent physical characteristic parameter of the discrete resonant unit. Its value is determined by the unit's pore size, porosity, skeleton density, and fluid medium properties. Discrete resonant units at different depths possess different inherent hydroelastic frequencies due to differences in pore size and structural characteristics, achieving a differentiated frequency distribution with varying depths of the gradient sponge layer. Average aperture of the unit , is the core characterization parameter of the unit aperture; Based on the inherent hydroelastic frequencies of each discrete resonant unit, a mechanical resonance excitation waveform is constructed. The time-domain expression of this waveform is: Its spectral distribution Follow mathematical formula ,in Angular frequency is the angular velocity representation of frequency, and is related to frequency. Satisfying fixed mathematical relations ; For the first The target amplitude of each discrete resonant element is the core parameter for controlling the resonance amplitude of that element. It is a positive constant, and its magnitude is positively correlated with the resonance deformation amplitude of the element. The amplitude distribution follows... ,in This represents the total fundamental amplitude. For the first The bandwidth parameter of each discrete resonant element is a positive constant used to adjust the bandwidth range of the resonant frequency of that element, satisfying the following conditions: ,in The larger the value, the wider the bandwidth and the stronger the frequency adaptability of the resonance. For the first The phase offset of each discrete resonant element, and The value varies with the depth numbering of the discrete resonant unit. It exhibits a strictly linear increase, satisfying ,in For the initial phase, The phase increment is used to create a traveling wave effect that propagates along the film thickness direction within the gradient sponge-like porous support layer, thereby enhancing the effect of the resonant excitation on the film. This is a Gaussian attenuation term. Its core function is to regulate the attenuation characteristics of the spectrum distribution near the inherent hydroelastic frequency, so that the energy of the resonant excitation is mainly concentrated near the inherent hydroelastic frequency of each discrete resonant unit, ensuring the resonant excitation is targeted and effective, and avoiding energy loss at invalid frequencies. For the first The resonant cosine oscillation term of a discrete resonant unit reflects the time-dependent changes in that unit. The resonant oscillation law is the core oscillation unit that constitutes the mechanical resonant excitation waveform; After constructing the mechanical resonance excitation waveform, an osmotic pressure oscillation waveform with a phase mismatch to this waveform is constructed. The core technical requirement is to make the dominant frequency component of the osmotic pressure oscillation waveform resonate with the mechanical resonance excitation waveform. The fundamental frequency component remains strictly Phase difference, fundamental frequency of mechanical resonance excitation waveform This achieves phase coordination and inverse matching between the two types of waveforms, enhancing the synergistic effect of subsequent regulation. The mathematical formula for the osmotic pressure oscillation waveform is: ,in The number of effective frequency components satisfies , Baseline osmotic pressure is the reference osmotic pressure value on the dialysate side. It serves as the fundamental reference parameter for osmotic pressure oscillations and is determined by the concentration of the basic components of the dialysate, satisfying the following conditions: ,in This represents the basic solute concentration of the dialysate. The frequency component ordinal number of the osmotic pressure oscillation waveform is consistent with the frequency component ordinal number of the mechanical resonance excitation waveform, thus achieving correspondence between the frequency components of the two types of waveforms. For the first The amplitude of the oscillation intensity of each frequency component is the core parameter for controlling the intensity of osmotic pressure oscillation at that frequency component. It is a positive constant, and its magnitude is positively correlated with the amplitude of osmotic pressure oscillation at that frequency component, satisfying the following condition: ,in This is the baseline amplitude of the osmotic pressure oscillation. For the first The oscillation frequency of each frequency component is exactly the same as the frequency of the corresponding frequency component in the mechanical resonance excitation waveform, that is... This enables precise frequency matching between the two types of waveforms. For the corresponding frequency The phase angle of the mechanical wave is directly taken from the phase parameter of that frequency component in the mechanical resonance excitation waveform, i.e. This ensures the correlation of phases; For the first The osmotic pressure sinusoidal oscillation term with frequency components, among which... The phase offset is the core setting for achieving inverse phase matching with the fundamental frequency component of the mechanical resonance excitation waveform. This phase offset establishes a coordinated inverse phase relationship between the osmotic pressure oscillation waveform and the mechanical resonance excitation waveform, providing a phase basis for the subsequent formation of the reverse solute diffusion field and the coordinated regulation of membrane deformation. Furthermore, the sinusoidal oscillation term can be converted through trigonometric identities. This achieves a unified correlation with the oscillation form of the mechanical resonance excitation waveform.

[0033] Furthermore, such as Figure 4 As shown, excitation is applied according to the multi-frequency coupled pulse control command to form a reverse solute diffusion field, specifically as follows: By using fluid dynamics control equipment and osmotic pressure control equipment, mechanical resonance excitation waveforms and osmotic pressure oscillation waveforms matching the control commands are applied to the blood side and dialysate side, respectively, to achieve the synchronous formation of the micro-amplitude breathing deformation of the gradient sponge-like porous support layer on the blood side and the reverse solute diffusion field on the dialysate side. At the same time, the synergy between membrane deformation and reverse solute diffusion is ensured, and the effective suppression of concentration polarization is achieved through the coupling effect of the two. On the blood side, a proportional valve connected to the blood-side inlet of the smart response membrane assembly is driven by a dedicated hydrodynamic control module, and the opening degree of the proportional valve... The mechanical resonance excitation waveform satisfies ,in Based on the opening degree, To maximize the amplitude of the mechanical resonance excitation waveform, the proportional valve operates at the inherent hydroelastic frequency of each discrete resonant unit. Rapid valve opening switching is performed, with the switching frequency precisely matched to the inherent hydroelastic frequency of each discrete resonant unit, thereby constructing the mechanical resonance excitation waveform. Precisely superimposed on the basic transmembrane pressure, forming a composite transmembrane pressure excitation that dynamically changes over time. ,in Based on the transmembrane pressure, this composite transmembrane pressure excitation is transmitted along the membrane thickness direction to each discrete resonant unit of the gradient sponge-like porous support layer, forcing the pore walls of the gradient sponge-like porous support layer to produce a small radial displacement. ,in For the positional parameters in the film thickness direction, For time parameters, radial displacement With composite transmembrane pressure excitation satisfied ,in As a displacement conversion coefficient, it can accurately reflect the degree of radial micro-deformation of the pore wall. The pore wall undergoes periodic radial expansion and contraction with changes in the composite transmembrane pressure excitation, which constitutes the micro-breathing deformation of the gradient sponge-like porous support layer. This deformation will change the effective pore size and porosity of the membrane pores in real time, and the dynamic effective pore size... This disrupts the solute enrichment layer on the membrane surface; Radial displacement of the pore walls of the gradient sponge-like porous support layer With local shear stress field There exists a strict viscoelastic dynamic coupling relationship, and the local shear stress field and fluid velocity satisfy... ,in The radial velocity of the fluid within the membrane pores follows the viscoelastic dynamics equation. ,in The equivalent density of the membrane skeleton is the equivalent density value of the gradient sponge-like porous support layer skeleton structure. It is a core parameter characterizing the physical properties of the membrane skeleton and satisfies... ,in For the basic density of the membrane skeleton, This refers to local porosity, measured in kilograms per cubic meter (kg / m³). ); It is the second partial derivative of the radial displacement with respect to time, reflecting the acceleration of the radial deformation of the hole wall; The position-dependent damping coefficient is located in the film thickness direction. A single-valued function that satisfies ,in As the basic damping coefficient, the gradient sponge layer at different locations has different damping characteristics due to differences in pore size, porosity and skeleton structure. Its value is determined by the membrane structure characteristics at the corresponding location. It is the first-order partial derivative of radial displacement with respect to time, reflecting the rate of radial deformation of the hole wall; For microgel swelling rate The dynamic stiffness coefficient is in the direction of film thickness. With microgel swelling rate A bivariate function that satisfies ,in Based on the basic stiffness coefficient, The swelling ratio stiffness correction factor is the microgel swelling ratio. The change in swelling ratio will directly alter the stiffness characteristics of the gradient sponge layer skeleton. The greater the swelling ratio, the smaller the stiffness of the membrane skeleton, and vice versa. The divergence of the fluid stress tensor. The Hamiltonian operator represents the spatial partial derivative operation, satisfying... , The fluid stress tensor generated during blood flow inside the membrane pores is a core tensor parameter that reflects the spatial distribution and changes of fluid forces. Its divergence can quantify the spatial variation trend of fluid forces inside the membrane pores. The driving force for the microgel under osmotic pressure oscillation is the osmotic pressure oscillation waveform. A single-valued function that satisfies ,in As the actuation force conversion coefficient, the dynamic change of osmotic pressure will drive the microgel actuation unit to swell or shrink and deform, thereby generating an actuation force acting on the pore wall. This actuation force is an important driving force for the deformation of the pore wall. While applying a mechanical resonance excitation waveform to the blood side, a high-precision metering pump is driven by a dedicated osmotic pressure wave generation module on the dialysate side, controlling the injection flow rate of the metering pump. The oscillation waveform satisfies ,in Inject traffic based on the foundation, To maximize the amplitude of the osmotic pressure oscillation waveform, replacement fluids of varying concentrations are precisely injected into the dialysate-side channel of the smart response membrane module. The concentration of the replacement fluid... satisfy The injection concentration and injection rate of the replacement fluid are based on the osmotic pressure oscillation waveform. The parameters are dynamically adjusted in real time, and by utilizing the dynamic changes in the concentration of the replacement fluid, an osmotic pressure oscillation waveform that perfectly matches the multi-frequency coupled pulse control command is generated on the dialysate side. This osmotic pressure oscillation waveform is transmitted along the membrane thickness from the dialysate side to the blood side, at a rate of [missing information]. satisfy A periodic reverse solvent drag force is formed inside the membrane pores of the gradient sponge-like porous support layer. ,satisfy ,in This is the drag force coefficient. The direction of this reverse solvent drag force is completely opposite to the direction of the positive solvent drag force generated by transmembrane pressure on the blood side. (Positive solvent drag force...) Through the periodic action of the reverse solvent drag force, a reverse solute diffusion field is constructed inside the membrane pores, and the diffusion flux of the solute in the reverse diffusion field... This diffusion field drives the enriched solute on the membrane surface and inside the membrane pores to diffuse in the reverse direction towards the dialysate side, effectively breaking the enrichment equilibrium of the solute and suppressing concentration polarization.

[0034] Furthermore, adaptive adjustment of ultrafiltration flux is achieved by monitoring deformation feedback signals and dynamically adjusting parameters, specifically as follows: By using high-precision sensing devices to monitor the deformation feedback signal of the microgel actuator in real time, the core parameters of the membrane structure are retrieved based on the feedback signal. Then, by constructing an energy efficiency optimization objective function and combining it with a model predictive control algorithm, the spectral distribution of the mechanical resonance excitation waveform and the amplitude of the osmotic pressure oscillation waveform are dynamically and precisely adjusted until the concentration polarization critical index falls back to the safe range, thereby achieving adaptive adjustment of the ultrafiltration flux. By utilizing a fiber Bragg grating sensor array embedded within the housing of the smart response membrane module, real-time acquisition of deformation feedback signals from the microgel actuation unit is achieved, including the center wavelength offset of the fiber Bragg grating. With strain satisfaction ,in The initial center wavelength, To achieve an effective elastic-optical coefficient, the fiber Bragg grating sensing array is uniformly distributed along the axial and radial directions of the membrane module, enabling precise capture of the axial strain of the membrane module caused by the swelling / contraction deformation of the microgel actuation unit. axial strain For time The dynamic function is a direct characterization signal of the deformation of the microgel actuating unit. Its magnitude is strictly linearly correlated with the degree of deformation of the microgel. The unit of measurement is dimensionless strain value, and the radial strain and axial strain satisfy the Poisson's ratio relationship. ,in Given the Poisson's ratio of the membrane material, the radial deformation characteristics can be derived from the axial strain. Based on the acquired axial strain First, the time-domain signal is converted into a frequency-domain signal using a forward Fourier transform. ,satisfy Then, axial strain and average swelling rate of microgels were reconstructed. A proprietary mapping model is used to accurately calculate the average swelling ratio of microgels. The mathematical formula for the mapping model is as follows: ,in The inverse Fourier transform is the core mathematical operation for converting frequency domain signals into time domain signals, satisfying... This transformation achieves signal domain conversion, ensuring the matching of parameter calculations; It is a frequency domain representation of the axial strain signal, which can reflect the distribution characteristics of the axial strain signal at different frequencies; The transfer function of the fiber Bragg grating sensor is the angular frequency. A single-valued function that satisfies ,in For the sensor's base gain, The signal delay time is determined by the sensor's own physical characteristics. It can accurately reflect the conversion relationship between the sensor's input signal and output signal, ensuring the accuracy of the sensing signal. The effective length of the membrane module is denoted as , which is the effective working length along the axial direction of the membrane module. The upper limit of the integration operation interval is , and the lower limit of integration is 0. For the length of the membrane The equivalent modulus of the composite material in the direction of film thickness is... With average swelling rate of microgel A bivariate function that satisfies ,in The modulus of the matrix polymer material. The microgel modulus reflects the dynamic change of the modulus characteristics of the membrane composite material with the membrane thickness, position, and swelling rate of the microgel. Its value is determined by the modulus of the matrix polymer material and the modulus of the microgel. This is the integral term of the equivalent modulus of the composite material along the effective length of the membrane. This integral is used to quantify the overall equivalent modulus characteristics of the membrane module, providing a basis for calculating the average swelling rate of the microgel. Based on the average swelling rate of the microgel calculated using a mapping model By utilizing the correlation between membrane pore structure parameters and microgel swelling rate, the actual tortuosity factor of the current gradient sponge-like porous support layer membrane pores can be obtained through inversion. ,satisfy ,in As the basic tortuosity factor, This is the correction coefficient for the swelling ratio tortuosity factor, and the actual tortuosity factor of the membrane pores. This is a core structural parameter reflecting the tortuosity of the channels within the membrane pores. Its value is dynamically correlated with the microgel swelling rate; a higher microgel swelling rate results in greater tortuosity of the membrane pore channels and a larger tortuosity factor, and vice versa. The inversion result of this parameter serves as the direct core basis for subsequent dynamic adjustment of the mechanical resonance excitation waveform spectrum distribution. Simultaneously, the real-time value of membrane hydraulic resistance can be inverted through the tortuosity factor. This enables full-parameter characterization of membrane structure properties; After completing the inversion of the core parameters of the membrane structure, a system energy efficiency optimization objective function is constructed. Using this as the optimization criterion for parameter adjustment, the core optimization direction of this objective function is to minimize concentration polarization resistance and maximize effective flux stability, while also taking into account the rationalization of system energy consumption. The mathematical formula is: ,in These are the weighting coefficients for the concentration polarization critical exponent, the rate of change of transmembrane flux, and the total input power of the system, respectively. They are all positive constants and satisfy strict normalization relations. The values ​​of each weighting coefficient are precisely set according to the actual control requirements of ultrafiltration permeation, quantifying the weight ratio of each optimization objective in the total optimization objective; The safety threshold for the concentration polarization critical index, i.e., the target control threshold of the system, is compared with the preset threshold for the concentration polarization critical index mentioned earlier. For the same parameter, it is a positive constant. The absolute value of the first derivative of the instantaneous transmembrane flux with respect to time reflects the real-time fluctuation of the transmembrane flux. The smaller the value, the more stable the transmembrane flux, satisfying the condition... , The total input power of the system is the time. The dynamic function satisfies ,in The mechanical resonance excitation input power, The input power for osmotic pressure oscillation reflects the real-time energy consumption of the system during the control process, and the unit of measurement is watt (W). The integral term of the total system input power over time, with the integral interval from 0 to the real-time control time. It can quantify the cumulative energy consumption during the control process.

[0035] The model predictive control (MPC) algorithm is used to solve the energy efficiency optimization objective function. The state-space equation of the model predictive control is as follows: , where state variables Control variables Output variables , The state-space matrix represents the average swelling ratio of the microgels. and concentration polarization critical index As state variables of the system, these variables can reflect the system's operating state and structural characteristics in real time, and can represent the target amplitude in the mechanical resonance excitation waveform. and the oscillation intensity amplitude in the oscillation waveform As control variables of the system, these variables are core parameters that can be adjusted in real time, while a fixed control period is set. Predicting the time domain With control time domain satisfy In each control cycle Within the current state variable values, the model predictive control algorithm is used to accurately solve the energy efficiency optimization objective function, obtaining the optimal solution that minimizes the objective function value. The optimal spectral parameter set of the mechanical resonance excitation waveform for the next control cycle is then output. and the optimal amplitude set of osmotic pressure oscillation waveforms The optimal solution satisfies and This allows for precise dynamic adjustment of the excitation waveform parameters. Based on conventional parameter adjustments, specific parameter adjustments are made for the special conditions of microgel dehydration. If the swelling rate of the microgel at a certain depth is detected through deformation feedback signals and swelling rate calculations... Below the preset critical dehydration threshold This threshold is the minimum swelling ratio at which the microgel maintains normal responsiveness; it is a positive constant and satisfies... This will automatically trigger a special adjustment mechanism, increasing the inherent hydroelastic frequency corresponding to that depth layer. Target amplitude, adjusted amplitude This enhances the micro-breathing deformation amplitude of the depth-gradient sponge-like porous support layer, thereby promoting fluid flow within the membrane pores and increasing the fluid velocity within the pores. ,in It serves as the flow rate increment coefficient, replenishes moisture to the microgel, and restores the normal shear force and osmotic pressure dual-sensitive response characteristics of the microgel actuation unit.

[0036] Based on the dynamically adjusted parameters, optimized mechanical resonance excitation waveforms and osmotic pressure oscillation waveforms are continuously applied to both the blood and dialysate sides, while the concentration polarization critical index is monitored in real time. The numerical change until the concentration polarization critical index Falling back to the safe threshold Within the designated safe zone, i.e. This enables adaptive adjustment of ultrafiltration flux, with adaptive maximization of ultrafiltration flux achieved through dual criteria: steady-state operation mode switching and continuous monitoring of transmembrane flux. volatility variance Rate of change of protein concentration in reflux solution The difference between the inherent hydroelastic frequency of each discrete resonant unit and the current driving frequency, and the variance of the transmembrane flux fluctuation satisfy the following: ,in The number of samples, For single sampling throughput, For average flux, when the condition set is met When the self-cleaning process of the membrane module is completed, the system automatically switches to steady-state operation mode. The flux stability threshold is the upper limit of the variance of transmembrane flux fluctuation. It is a positive constant. A value below this threshold indicates that the transmembrane flux is in a stable ultrafiltration state. The protein concentration in the reflux solution reflects protein adsorption and gel formation on the membrane surface. The unit of measurement is moles per liter. The protein concentration and gel layer thickness satisfy the following relationship: ,in The thickness of the gel layer. This is the concentration-thickness conversion factor; For the first The current driving frequency of a discrete resonant unit is the actual resonant frequency applied to that unit; The frequency lock tolerance is the upper limit of the frequency difference, a positive constant. A value below this indicates a precise match between the driving frequency and the inherent hydroelastic frequency, satisfying the requirements. .

[0037] The determination of gel layer peeling inside the membrane pores is based on the premise that all external pressure conditions, such as transmembrane pressure, remain unchanged. The pressure and flux signals are kept constant, and the membrane hydraulic resistance is calculated by real-time acquisition. Membrane hydraulic resistance is a core parameter reflecting the patency of membrane pores. A higher value indicates a more severe degree of pore blockage. If the calculated membrane hydraulic resistance shows a continuous decreasing trend, it satisfies the following condition. And the duration of this downward trend Exceeding the preset time window This time window is set based on the control precision and membrane structure characteristics to meet the requirements. This confirms that the solute gel layer inside the membrane pores of the gradient sponge-like porous support layer has been completely stripped away by the synergistic effect of mechanical resonance and osmotic fluctuations, and the membrane hydraulic resistance after gel layer stripping is... The membrane's ultrafiltration performance recovers to normal levels, achieving adaptive maximization and stabilization of ultrafiltration flux, and steady-state ultrafiltration flux. This is the optimal ultrafiltration flux value.

[0038] This embodiment details how an intelligent responsive membrane module containing a nonlinear pore size gradient and dual-sensitive microgel actuation units was constructed. By combining concentration polarization critical index monitoring and multi-frequency coupled pulse modulation, adaptive regulation of ultrafiltration flux was achieved. The synergistic effect of membrane breathing deformation induced by mechanical resonance excitation and the reverse solute diffusion field, along with dynamically adjusted waveform parameters adapted to the real-time state of the membrane, avoids ineffective energy consumption and structural damage. While ensuring treatment safety, this significantly improves flux stability and anti-fouling ability during dialysis.

[0039] Example 2 describes in detail an ultrafiltration osmosis regulation device based on a gradient sponge structure dialysis membrane, used to implement the above-mentioned ultrafiltration osmosis regulation method based on a gradient sponge structure dialysis membrane, such as... Figure 5 As shown, it specifically includes: A smart gradient membrane assembly, which is filled with a hollow fiber membrane having a nonlinear pore size gradient and microgel actuation units embedded in the pore walls. The fluid dynamics control module, connected to the blood-side inlet of the smart gradient membrane assembly, is used to generate multi-frequency coupled pulsed blood flow; The osmotic pressure wave generation module is connected to the dialysate side of the smart gradient membrane module and is used to inject periodically changing dialysate to form a reverse solute concentration wave. The sensor feedback array is distributed at the inlet and outlet of the smart gradient membrane module and on the surface of the housing to collect pressure, flow rate, conductivity and membrane strain signals in real time. The central processing unit is electrically connected to the fluid dynamics control module, the osmotic pressure wave generation module, and the sensor feedback array, respectively, and is used to execute the ultrafiltration osmosis regulation method described above.

[0040] Furthermore, the hollow fiber membrane filled inside the intelligent gradient membrane module features a microstructure and material composite design. This hollow fiber membrane is the core carrier for achieving coordinated regulation of mechanical resonance excitation and osmotic pressure oscillation. Its cross-section exhibits a continuous gradient sponge-like pore distribution along the membrane thickness direction. The pores show a nonlinear exponential decay characteristic from the blood side to the dialysate side, forming a gradient sponge-like porous support layer that matches the ultrafiltration osmotic regulation method. This gradient sponge-like pore distribution is not a simple gradual change in pore size, but a continuous sponge-like structure prepared by a non-solvent-induced phase separation method combined with microfluidic spinning technology. The pores are interconnected, and the pore size distribution strictly follows the nonlinear exponential decay law, providing a suitable structural basis for the embedding and function of the microgel actuation unit.

[0041] The pore walls of the hollow fiber membrane are composed of a matrix polymer material and dispersed nanoscale dual-response microgel particles. The matrix polymer material provides a structural framework for the gradient sponge-like porous support layer, ensuring the mechanical strength and pore structure stability of the membrane. The nanoscale dual-response microgel particles, as the core functional particles, are fixed to the pore wall surface through chemical bonding, forming microgel actuation units that are sensitive to both shear force and osmotic pressure. A strict size matching relationship is maintained between the dual-response microgel particles and the matrix pore size. ,in The particle size of the dual-response microgel particles, The size of the matrix pores at the location of the particles is crucial for the microgel particles to function as independent nanoactuators. This ensures that the microgel particles will not be lost with the fluid due to their small size, nor will they clog the membrane pores due to their large size. At the same time, it provides sufficient space for the swelling and shrinkage deformation of the microgel particles, ensuring that the deformation can effectively change the local porosity and membrane flow characteristics.

[0042] The grafting density of the dual-response microgel particles on the pore wall surface is not uniformly distributed; its grafting density... Along the film thickness direction It exhibits a Gaussian distribution and follows the formula ,in To achieve the maximum grafting density, This is the center location of the gradient transition region. As the distribution width parameter, this Gaussian grafting method enables the microgel particles to achieve a high-density distribution in the gradient transition region along the membrane thickness direction, while exhibiting a gradient decay distribution on the blood side and dialysate side. This matches the pore size gradient distribution of the membrane, allowing the effect strength of the microgel actuation unit to match the ultrafiltration characteristics of the membrane pores, thus forming a stronger deformation regulation capability in the gradient transition region where concentration polarization is prone to occur.

[0043] The microstructure design of the intelligent gradient membrane module enables the dual-response microgel particles to become independent nanoactuators under the action of external multi-frequency pulses, rather than merely causing overall macroscopic deformation of the membrane pores. This allows for the application of mechanical resonance excitation waveforms to the blood side by the hydrodynamic control module and osmotic pressure oscillation waveforms to the dialysate side by the osmotic pressure wave generation module. During this process, the shear force and osmotic pressure inside the membrane pores dynamically change with the excitation waveform and oscillation waveform. The dual-response microgel particles undergo precise swelling or contraction deformation according to the changes in local shear force and osmotic pressure. This deformation can generate local eddy current disturbances inside the pores, directly acting on the adsorption sites of solute at the pore throat, disrupting the adsorption balance of solute at the pore throat at the molecular scale, effectively preventing solute enrichment and gel layer formation. At the same time, combined with the micro-breathing deformation of the membrane pores caused by mechanical resonance excitation and the reverse solute diffusion field formed by osmotic pressure oscillation, it achieves efficient suppression of concentration polarization. The realization of this core function is based on the unique microstructure design of the intelligent gradient membrane module, which is also the core technical support for the adaptive adjustment of ultrafiltration flux in this device.

[0044] Furthermore, the fluid dynamics control module, as the execution module that works in conjunction with the intelligent gradient membrane module, is sealed to the blood-side inlet of the intelligent gradient membrane module through a pressure-resistant pipeline. Internally, it integrates core components such as a high-precision proportional valve, a pulse generation unit, and a fluid dynamics pump. Based on control commands issued by the central processing unit, it can precisely adjust the valve opening and fluid output pressure. It rapidly switches the proportional valve opening using the inherent fluid elastic frequency of each discrete resonant unit, accurately superimposing the mechanical resonance excitation waveform onto the baseline transmembrane pressure. This generates a multi-frequency coupled pulsed blood flow that matches the control commands, providing precise mechanical resonance excitation to the gradient sponge-like porous support layer of the intelligent gradient membrane module, inducing it to produce micro-amplitude respiratory deformation. The module's power output accuracy and frequency response characteristics are compatible with the structural characteristics of the intelligent gradient membrane module, ensuring that the mechanical resonance excitation can be effectively transmitted to the discrete resonant units at various depths of the membrane, achieving precise dynamic control of the membrane pore structure.

[0045] Furthermore, the osmotic pressure wave generation module is connected to the dialysate side of the intelligent gradient membrane module. It forms a loop connection with the dialysate inlet and outlet via a high-precision metering pipeline. Internally, it integrates core components such as a high-precision metering pump, a multi-concentration replacement fluid storage unit, and a concentration mixing valve. Based on control commands from the central processing unit, it can precisely control the injection flow rate and mixing ratio of replacement fluids of different concentrations. By injecting periodically varying dialysate, it generates an osmotic pressure oscillation waveform on the dialysate side that is inversely phase-matched to the mechanical resonance excitation waveform. ,in This is the baseline osmotic pressure on the dialysate side. For the first The amplitude of the oscillation intensity of each frequency component, The oscillation frequency is... To correspond to the phase angle of the mechanical wave, this module can accurately match the frequency and phase characteristics of the mechanical resonance excitation waveform, ensuring that the dominant frequency component of the osmotic pressure oscillation waveform and the fundamental frequency component of the mechanical resonance excitation waveform remain consistent. The phase difference causes the osmotic pressure oscillation waveform to be transmitted along the membrane thickness direction, forming a periodic reverse solvent drag force inside the membrane pores, constructing a reverse solute diffusion field, which works synergistically with the microgel actuation unit deformation and membrane pore breathing deformation of the smart gradient membrane module to achieve efficient suppression of solute enrichment.

[0046] Furthermore, the sensor feedback array is distributed across the blood-side inlet, blood-side outlet, dialysate-side inlet, dialysate-side outlet, and the surface of the membrane module housing of the intelligent gradient membrane module. The array consists of various sensing units, including high-precision pressure sensors, flow sensors, conductivity meters, and fiber Bragg grating strain sensors. The arrangement and detection accuracy of each sensing unit are adapted to the structural characteristics of the intelligent gradient membrane module. The pressure and flow sensors can collect the blood-side inlet pressure in real time. Export pressure Dialysis fluid side pressure In addition to transmembrane flow signals, the conductivity meter can detect the solute concentration on the blood and dialysate sides in real time. The fiber Bragg grating strain sensor is embedded inside the membrane module housing, closely fitted to the hollow fiber membrane of the smart gradient membrane module, and can collect the axial strain of the membrane module caused by the swelling / contraction of the microgel actuation unit in real time. All sensing units transmit the collected signals such as pressure, flow rate, conductivity, and membrane strain to the central processing unit in real time, providing accurate real-time data support for concentration polarization critical index calculation, excitation waveform parameter adjustment, and ultrafiltration flux status determination, ensuring the closed-loop operation of the entire control system.

[0047] Furthermore, the central processing unit is bidirectionally electrically connected to the fluid dynamics control module, osmotic pressure wave generation module, and sensor feedback array via an industrial control bus. Internally, it integrates core components such as a high-speed computing chip, algorithm storage unit, instruction output unit, and data acquisition unit. It pre-stores all the algorithms and control logic of the aforementioned ultrafiltration osmosis regulation method and can receive various acquisition signals transmitted from the sensor feedback array in real time. Through built-in algorithms, it accurately calculates core parameters such as instantaneous transmembrane pressure, instantaneous transmembrane flux, concentration polarization critical index, and average swelling rate of microgels. When the concentration polarization critical index exceeds a preset threshold, it can quickly generate waveforms including mechanical resonance excitation waveforms and osmotic pressure oscillation waveforms. The system generates multi-frequency coupled pulse control commands and issues precise execution commands to the fluid dynamics control module and the osmotic pressure wave generation module. Simultaneously, it can invert the deformation state of the microgel actuation unit in real time based on the membrane strain signal collected by the sensor feedback array. Through the MPC algorithm, it dynamically adjusts the spectral distribution of the mechanical resonance excitation waveform and the amplitude parameters of the osmotic pressure oscillation waveform until the concentration polarization critical index falls back to the safe range, thereby achieving adaptive adjustment of the ultrafiltration flux. The unit's computing speed and control precision can meet the requirements of real-time regulation, ensuring the coordinated operation of each module and the precise execution of the control commands. It is the core control component for realizing the automation and intelligent ultrafiltration osmotic regulation of the entire device.

[0048] This embodiment describes in detail that the device as a whole is highly automated and intelligent, and can complete pollution monitoring, command generation and parameter adjustment without human intervention, which greatly extends the continuous operation cycle of the membrane module, reduces the need for cleaning and maintenance, and is suitable for the actual application scenarios of clinical dialysis.

[0049] Based on Example 1 or 2, to verify the effectiveness and stability of the ultrafiltration osmosis regulation method based on the gradient sponge structure dialysis membrane described in this application, the optimal grafting density of the microgel actuation unit on the pore wall was determined. A hollow fiber membrane module was used in the experiment, maintaining a constant blood flow rate of 200 mL / min and a constant dialysate flow rate of 500 mL / min. The baseline transmembrane pressure was set at 150 mmHg, and a solute model of vitamin B12 with a molecular weight of approximately 12000 Da was selected to simulate a medium-molecular-weight toxin. Under these fixed conditions, five groups with the same nonlinear pore size gradient (blood-side pore size 120 nm, dialysate-side pore size 10 nm, gradient attenuation coefficient 1.2 μm) were prepared. -1 However, for membrane modules with different microgel grafting densities, the grafting density was set to 5 grafts / μm. 2 20 cells / μm 2 45 cells / μm 2 80 cells / μm 2 and 110 / μm 2 During the experiment, the system monitored the concentration polarization critical index in real time. When the index exceeded the threshold of 1.5, it automatically triggered a multi-frequency coupled pulse control command. The mechanical resonance excitation frequency range covered 50Hz to 2000Hz, and the osmotic pressure oscillation amplitude was controlled within ±8kPa. After 4 hours of continuous operation, the steady-state ultrafiltration flux maintenance rate and the thickness of the fouling layer on the membrane surface were recorded for each group. Data showed that the grafting density was 5 grafts / μm. 2 Due to the sparse distribution of actuating units, the group could not form a continuous deformation field, resulting in a flux maintenance rate of only 62% and a contamination layer thickness of 18.5 μm; as the density increased to 20 units / μm... 2 and 45 / μm 2 The flux maintenance rate significantly increased to 78% and 91%, respectively, while the contamination layer thickness decreased to 12.3 μm and 6.8 μm, respectively; when the density further increased to 80 particles / μm... 2 At this point, the flux maintenance rate reached a peak of 94.5%, and the thinnest fouling layer thickness was 4.2 μm. At this point, the respiratory deformation amplitude of the membrane pores and the fluid shear force achieved optimal matching. However, when the density continued to increase to 110 pores / μm... 2 At that time, due to the excessive occupation of pore space by the microgel, the basic porosity decreased. Although the contamination layer thickness remained at 4.5 μm, the initial flux decreased significantly, and the steady-state flux maintenance rate dropped to 83%. This process clearly reveals the nonlinear influence of grafting density on the regulation effect. Through a combination of sophisticated computational fluid dynamics (CFD) simulation and magnetic resonance imaging (MRI), such as... Figure 6As shown, the dynamic evolution of the flow field structure and microgel distribution inside the membrane pores under different grafting densities is presented with high-resolution four-dimensional spatiotemporal visualization. The dynamic demonstration shows the superposition of color-coded flow velocity vector field and semi-transparent gel volume fraction cloud map, which shows the whole process from the local low-velocity region under sparse distribution to the uniform high-velocity streamline under optimal density and then to the increase of flow resistance under over-dense distribution.

[0050] Based on the determination of the optimal grafting density, the spectral distribution parameters of the mechanical resonance excitation waveform in the multi-frequency coupled pulse control command were optimized, and the microgel grafting density was fixed at 80 grafts / μm. 2 Other basic parameters remain the same as above; the experimental variable is set as the spectral bandwidth coefficient of the mechanical resonance excitation waveform. Examine narrowband focusing ( =10Hz), mid-band equalization ( =50Hz), broadband coverage ( =150Hz) and ultra-wideband diffusion ( The effects of four modes (300Hz, 360Hz, 10 ... The changes and overall energy consumption indicators were analyzed. The results show that the narrowband focusing mode only produces strong resonance at a specific depth, resulting in a mass transfer coefficient as high as 2.5 × 10⁻⁶ at the inlet. -5 The speed is m / s, but the output speed is only 0.8×10 m / s. -5 The mass transfer rate was m / s, resulting in extremely uneven cleaning performance; in the mid-band equalization mode, the difference in mass transfer coefficient across different axial segments decreased, and the average value increased to 1.9 × 10⁻⁶ m / s. -5 m / s; the broadband coverage mode achieves effective excitation across the entire depth range, with the mass transfer coefficients at the inlet, middle, and outlet ends remaining stable at 2.1 × 10 m / s. -5 m / s, 2.0×10 -5 m / s and 1.95×10 -5 The speed is m / s, the overall uniformity is optimal, and the energy consumption per unit flux is controlled at 0.45 kWh / m. 3 The low level; although the ultrawideband diffusion mode has a wide coverage, due to energy dispersion, the resonance intensity of each layer is insufficient, and the average mass transfer coefficient drops back to 1.6×10. -5 The speed is m / s, but the energy consumption rises to 0.62 kWh / m 3 Table 1 details a comparison of key performance indicators across the four spectral modes, demonstrating the necessity of a gradient matching strategy that adapts frequency to depth for overcoming the inherent non-uniformity of gradient membrane structures. Figure 7 As shown, Figure 7 Composed of a heatmap and four sub-maps, it displays the instantaneous distribution cloud maps of the velocity vector field inside the membrane pores under four spectral modes. Figure 7 (a) shows that in the narrowband mode, the vortex is concentrated only in a local area. Figure 7 (b) and Figure 7 (d) in the figure shows that the eddy current intensity is insufficient or disordered in the mid-band and ultra-wideband modes, respectively. Figure 7 (c) clearly shows the continuous, layered, and appropriately strong micro-vortex chains formed along the film thickness direction under broadband coverage mode. This flow field structure effectively disrupts the concentration polarization boundary layer at all depths, which corroborates the data in Table 1. Figure 7 (e) in the figure is a thermogram of the flow velocity distribution along the film thickness under different spectral bandwidth modes. It shows that the broadband coverage mode maintains a uniform and stable high flow velocity distribution throughout the entire film thickness range, further verifying that this mode can achieve the equalization of fluid disturbance in the deep and surface pores of the gradient sponge layer, and comprehensively improve the anti-fouling and mass transfer uniformity.

[0051] Table 1 Comparison of membrane module performance indicators under different spectral bandwidth coefficients

[0052] Note: The subscripts in / mid / out represent the inlet, mid, and outlet positions of the membrane module, respectively. This is the axial uniformity index; the closer the value is to 1, the more uniform the mass transfer along the path. Specific energy consumption per unit of water production; This represents the hourly growth rate of pollution resistance.

[0053] Based on the above-determined optimal parameter combination (microgel grafting density 80 grafts / μm) 2 (with a spectral bandwidth coefficient of 150Hz), this study verifies the fine-tuning capability of the phase-reversed osmotic pressure oscillation waveform for adaptive adjustment of ultrafiltration flux. While keeping the mechanical excitation parameters constant, the phase difference between the osmotic pressure oscillation waveform and the mechanical resonance waveform is changed. Four operating conditions were set up with phase differences of 0° (in-phase), 90° (orthogonal), 180° (reverse matching), and 270° (anti-orthogonal). Simultaneously, different levels of simulated protein contamination load (BSA concentrations of 10 g / L, 30 g / L, and 50 g / L) were introduced to test the system's robustness. Experimental monitoring indicators included transient flux fluctuation amplitude, steady-state flux recovery time, and solute retention rate stability. Data recording showed that under the in-phase (0°) condition, the superposition of mechanical deformation and osmotic pressure fluctuations caused instantaneous pore closure or excessive expansion, resulting in severe flux oscillations with a standard deviation as high as 15.2 L / (m²). 2•h), and is prone to microgel fatigue damage; under orthogonal (90°) and anti-orthogonal (270°) conditions, flux fluctuations are somewhat reduced, but under high contamination loads (50g / LBSA), the steady-state flux recovery time is as long as 45 minutes; only under the phase-reverse matching (180°) condition, the mechanical expansion stage corresponds exactly to the osmotic pressure reverse drag stage, and the synergistic effect of the two is maximized, reducing the standard deviation of flux fluctuation to 2.1L / (m 2 The system achieves high performance (·h), and under a high pollution load of 50 g / L, it only takes 12 minutes to bring the concentration polarization critical index back to the safe range, maintaining a steady-state flux of over 92% of the initial value, with solute rejection fluctuations of less than 1.5%. Table 2 summarizes the comprehensive performance data under different phase differences and pollution loads. (Recovery time) (Standard deviation of flux) (Stability of retention rate) and (Dynamic adjustment index), among which the 180° phase difference group showed the best performance at all pollution concentrations. The value (>0.95) highlights the core value of precise phase matching and the synchronous analysis results of membrane pore dynamic response and multi-frequency coupled excitation, such as... Figure 8 As shown in Figure 8, (a) displays the millisecond-level dynamic oscillation curves of membrane pore radius over time under four phase differences: 0°, 90°, 180°, and 270°. This intuitively demonstrates the difference in the regulation of the amplitude and rhythm of membrane pore breathing micro-deformation by phase matching. (b) is the power spectral density analysis diagram of the flow velocity in the pore under the corresponding operating conditions. It clearly identifies the 100 Hz mechanical excitation peak, the 90 Hz osmotic pressure oscillation peak, and the difference frequency and sum frequency harmonics generated by their coupling, verifying the effective transmission of multi-frequency coupled pulses and the enhancement of flow field disturbance. (c) is the coherence analysis diagram of membrane pore deformation and flow velocity signal. The high coherence peaks at 90 Hz and 100 Hz prove the strong phase locking and efficient energy transfer between the two. Overall, it reveals that optimizing phase matching can enhance membrane pore disturbance, disrupt solute adsorption balance, and ultimately achieve the technical effect of suppressing concentration polarization and membrane fouling, and stabilizing ultrafiltration flux.

[0054] Table 2 Comparison of changes in ultrafiltration flux and membrane resistance before and after multi-frequency coupled pulse modulation under different BSA pollution loads.

[0055] Note: Phase difference ( The phase difference between the mechanical excitation and the osmotic pressure oscillation is represented by 0.5; the contaminant concentration is the contaminant concentration simulated by bovine serum albumin. The time required for the concentration polarization index to recover to the safe region; The standard deviation of ultrafiltration flux reflects the degree of fluctuation. The stability score is calculated based on the solute rejection rate. The index is dynamically adjusted to comprehensively measure response speed and accuracy; Max The maximum increment of transmembrane pressure during operation; the micro-damage rate is the micro-damage rate of the microgel actuation unit.

[0056] To comprehensively evaluate the overall performance and structural stability of the technical solution in this application during long-term operation, a fourth implementation method was verified. This method integrates all the aforementioned optimal parameters (grafting density 80 grafts / μm). 2 The system (with a spectral bandwidth of 150Hz and a phase difference of 180°) underwent continuous operation for 48 hours under complex and variable conditions simulating clinical dialysis. During the experiment, the blood flow rate fluctuated randomly between 150-250 mL / min, the transmembrane pressure varied stepwise between 100-200 mmHg, and the solute concentration was periodically adjusted to simulate in vivo metabolic fluctuations. The system operated automatically throughout the entire process without manual intervention, recording in real time the total ultrafiltration volume, transmembrane pressure trends, membrane module integrity indicators, and flux recovery rate after cleaning, such as... Figure 9 As shown, Figure 9 (a) Figure 9 (b) shows the ultrafiltration flux decay trend and transmembrane pressure growth trend, respectively. During the 48-hour operation period, the ultrafiltration flux of the membrane module controlled by the method of this application remained within a narrow range of 45-48 L / (m²·h), and the transmembrane pressure growth rate was extremely low, increasing by only 8 kPa after 48 hours. In contrast, the control group (relying solely on the static gradient structure) without this adjustment strategy experienced a surge of 50 kPa in transmembrane pressure after 12 hours, forcing a shutdown. Figure 10 As shown, Figure 10 (a) Figure 10 (b) shows the verification results for the experimental and control groups, respectively. After the operation, the membrane modules were scanned for microstructure. It was found that the experimental group membrane had almost no gel layer deposition, the pore structure was intact, and there was no obvious detachment or breakage of the microgel actuation units. The flux recovery rate was as high as 98.5%. In stark contrast, the microstructure of the control group membrane module after forced shutdown was clearly visible in the scanning electron microscope (SEM) image. The image clearly shows that the membrane surface was completely covered by a dense and thick protein gel layer. The thickness of this contaminant layer was uneven, and in some areas it even formed blocky accumulations, completely blocking the membrane pore entrances. In the high-magnification view, the originally clear gradient sponge pore structure was no longer visible. Instead, there was a hardened network of contaminants, and some pore walls showed obvious collapse and distortion due to long-term exposure to abnormal high pressure.

[0057] Through multi-set parameter optimization and robustness testing, the optimal microgel grafting density and spectral bandwidth parameters were determined, achieving uniform and efficient fouling removal across the entire depth range of the membrane layer. The 180° phase-reverse matching design maximizes the synergistic effect of mechanical excitation and osmotic pressure oscillation, significantly reducing flux fluctuations and recovery time. Under different BSA fouling loads, the system maintains excellent dynamic adjustment capabilities, stable solute rejection rate, and low microgel damage rate. Long-term variable operating condition testing further demonstrates that this technology can effectively suppress abnormal increases in transmembrane pressure and flux decay, while maintaining good membrane structural integrity.

[0058] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any changes, modifications, substitutions, integrations, and parameter changes made to these embodiments within the spirit and principles of the present invention, without departing from the principles and spirit of the present invention, through conventional substitutions or to achieve the same function, fall within the scope of protection of the present invention.

Claims

1. A method for ultrafiltration osmosis adjustment based on a gradient sponge structure dialysis membrane, characterized in that, include: A smart response membrane module with a nonlinear pore size gradient is constructed. The smart response membrane module includes a gradient sponge-like porous support layer with an exponentially decreasing pore size from the blood side to the dialysate side, and the pore walls of the gradient sponge-like porous support layer are embedded with microgel actuation units that are sensitive to both shear force and osmotic pressure. Acquire real-time transmembrane flux signal and solute concentration difference signal across the membrane, and calculate concentration polarization critical exponent based on real-time transmembrane flux signal and solute concentration difference signal across the membrane; When the concentration polarization critical index exceeds a preset threshold, a multi-frequency coupled pulse control command is generated. The multi-frequency coupled pulse control command includes a mechanical resonance excitation waveform whose frequency varies with the depth of the gradient sponge layer and an osmotic pressure oscillation waveform whose phase is inversely matched. According to the multi-frequency coupled pulse control command, the mechanical resonance excitation waveform is applied to the blood side to induce the gradient sponge-like porous support layer to produce a micro-breathing deformation, and the osmotic pressure oscillation waveform is injected into the dialysate side to form a reverse solute diffusion field. The deformation feedback signal of the microgel actuation unit is monitored in real time, and the spectral distribution of the mechanical resonance excitation waveform and the amplitude of the osmotic pressure oscillation waveform are dynamically adjusted until the concentration polarization critical index falls back to the safe range, thereby achieving adaptive adjustment of the ultrafiltration flux.

2. The ultrafiltration osmosis adjustment method based on a gradient sponge structure dialysis membrane according to claim 1, characterized in that, The specific steps for constructing a smart response membrane module with a nonlinear pore size gradient include: Hollow fiber membranes were prepared using a solvent-free phase separation method combined with microfluidic spinning technology, by dynamically adjusting the volume ratio of non-solvent to solvent in the coagulation bath. Controlling the phase separation rate to form pores Along the film thickness direction A gradient sponge-like porous support layer exhibiting a nonlinear exponential decay distribution; Wherein, aperture distribution function The formula is: In the formula, The initial pore size on the blood side surface. The minimum pore size on the dialysate side. The gradient decay coefficient is... It is a non-linear shape factor. This is a dynamic function of the viscosity of the coagulation bath as a function of time. Poly(N-isopropylacrylamide)-acrylic acid copolymer microgels are grafted onto the pore wall surface of the gradient sponge-like porous support layer to form shear force-resistant microgels. With local osmotic pressure A dual-sensitivity microgel actuation unit, wherein the swelling ratio of the microgel actuation unit is... Defined as a dynamic adjustment variable for local porosity.

3. The ultrafiltration osmosis adjustment method based on a gradient sponge structure dialysis membrane according to claim 2, characterized in that, The process of acquiring real-time transmembrane flux signals and solute concentration difference signals across the membrane, and calculating the concentration polarization critical exponent, specifically involves: Obtain blood inlet pressure Export pressure and dialysate side pressure Calculate instantaneous transmembrane pressure Simultaneously, the blood concentration was measured using a conductivity meter. Concentration on the dialysate side Constructing the concentration polarization critical index The calculation model is as follows: In the formula, For instantaneous transmembrane flux, To account for the equivalent boundary layer thickness after microgel deformation, is the effective diffusion coefficient of the solute in the gradient sponge structure. The solute concentration at the membrane surface. This is the shear force fluctuation sensitivity coefficient. This refers to the wall shear force. when At that time, the system is determined to have entered a high-risk region for concentration polarization. A multi-frequency coupled pulse control command generation mechanism is triggered based on a preset threshold.

4. The ultrafiltration osmosis adjustment method based on a gradient sponge structure dialysis membrane according to claim 1, characterized in that, The specific components of generating the multi-frequency coupled pulse control command include: Based on the depth-layered characteristics of the gradient sponge-like porous support layer, the membrane thickness is divided into... The discrete resonant unit, for the th discrete resonant unit Calculate the natural hydroelastic frequency of each resonant element. ; Constructing mechanical resonance excitation waveforms Its spectral distribution for: In the formula, For the first The target amplitude of the layer For bandwidth parameters, This is the phase offset, and With depth It increases linearly to create a traveling wave effect; Constructing phase-inversely matched osmotic pressure oscillation waveforms So that its main frequency component is The fundamental frequency component remains The phase difference is expressed by the formula: ,in For the corresponding frequency The phase angle of the mechanical wave. For the first The amplitude of the oscillation intensity of each frequency component.

5. The ultrafiltration osmosis adjustment method based on a gradient sponge structure dialysis membrane according to claim 4, characterized in that, Excitation is applied via multi-frequency coupled pulse control commands, specifically including: Drive the blood-side proportional valve at frequency Rapidly switch the opening degree to generate the mechanical resonance excitation waveform. Superimposed on the basic transmembrane pressure, this forces the pore walls of the gradient sponge-like porous support layer to undergo radial displacement. ; The radial displacement With local shear stress field The coupling relationship satisfies the viscoelastic dynamics equation: In the formula, The equivalent density of the membrane skeleton, The position-dependent damping coefficient, For microgel swelling rate The dynamic stiffness coefficient, For fluid stress tensor, This serves as the driving force for the microgel under osmotic pressure oscillation. Different concentrations of replacement fluid are injected by a high-precision metering pump on the dialysate side to generate the osmotic pressure oscillation waveform, forming a periodic reverse solvent drag force inside the membrane pores.

6. The ultrafiltration osmosis adjustment method based on a gradient sponge structure dialysis membrane according to claim 2, characterized in that, The method for real-time monitoring of the deformation feedback signal of the microgel actuator includes: Using a fiber Bragg grating embedded within the membrane module housing, the axial strain of the membrane module caused by microgel swelling / shrinkage was obtained. Establish axial strain With average swelling rate of microgel Mapping model: In the formula, This represents the inverse Fourier transform. For the frequency domain representation of the strain signal, For sensor transfer function, For the length of the membrane Equivalent modulus of composite materials in the direction of orientation; Based on calculations Invert the actual tortuosity factor of the current membrane pores. This serves as the basis for dynamically adjusting the spectral distribution of the mechanical resonance excitation waveform.

7. The ultrafiltration osmosis adjustment method based on a gradient sponge structure dialysis membrane according to claim 6, characterized in that, The dynamic adjustment of the spectral distribution of the mechanical resonance excitation waveform and the amplitude of the osmotic pressure oscillation waveform specifically includes: Construct the system energy efficiency optimization objective function To minimize concentration polarization resistance and maximize effective flux stability, the formula is: In the formula, These are the weighting coefficients, This represents the total input power of the system. Using the MPC algorithm, and As a state variable, with and As a control variable, in each control cycle Solve the above objective function internally and output the optimal spectral parameter set for the next period. and optimal amplitude set If the swelling rate of microgels at a certain depth is detected If the temperature falls below the critical dehydration threshold, the frequency of that layer will automatically increase. amplitude This is to enhance the amplitude of local breathing deformation.

8. The ultrafiltration osmosis adjustment method based on a gradient sponge structure dialysis membrane according to claim 1, characterized in that, The process continues until the concentration polarization critical index falls back to a safe range, achieving adaptive adjustment of the ultrafiltration flux. This is achieved by adaptively maximizing the ultrafiltration flux, specifically including: Continuous monitoring of transmembrane flux volatility variance Rate of change of protein concentration in reflux solution When the following conditions are met, self-cleaning is deemed complete and the system switches to steady-state operation mode: In the formula, For flux stability threshold, For frequency locking tolerance, This is the current driving frequency; If the calculated membrane hydraulic resistance is obtained without changing the external pressure... A continuous downward trend has emerged, that is... If the duration exceeds the set window, it confirms that the gel layer inside the membrane pores has been synergistically stripped away by mechanical resonance and osmotic fluctuations.

9. An ultrafiltration osmosis conditioning device based on a gradient sponge structure dialysis membrane, comprising: A smart gradient membrane assembly, which is filled with a hollow fiber membrane having a nonlinear pore size gradient and microgel actuation units embedded in the pore walls. A fluid dynamics control module, connected to the blood-side inlet of the intelligent gradient membrane assembly, is used to generate multi-frequency coupled pulsed blood flow; An osmotic pressure wave generating module is connected to the dialysate side of the intelligent gradient membrane assembly and is used to inject periodically changing dialysate to form a reverse solute concentration wave. A sensor feedback array is distributed at the inlet and outlet of the intelligent gradient membrane module and on the surface of the housing to collect pressure, flow rate, conductivity and membrane strain signals in real time. The central processing unit is electrically connected to the fluid dynamics control module, the osmotic pressure wave generation module and the sensor feedback array, respectively, and is used to execute the ultrafiltration osmosis regulation method according to any one of claims 1 to 8.

10. The ultrafiltration osmosis conditioning device based on a gradient sponge structure dialysis membrane according to claim 9, characterized in that, The core innovation of the intelligent gradient membrane module lies in the microstructure design of its internal hollow fiber membrane: The hollow fiber membrane has a continuous gradient sponge-like pore distribution in its cross-section, and the pore walls are composed of a matrix polymer material and nanoscale dual-response microgel particles dispersed therein. The particle size of the dual-response microgel particles Matrix pore size at the location To meet specific size matching requirements: Furthermore, the dual-response microgel particles are grafted onto the pore wall surface via chemical bonds, and their grafting density is... Along the film thickness direction It exhibits a Gaussian distribution: In the formula, This is the center location of the gradient transition region. For the distribution width parameter, The structure configuration allows the microgel particles to act as independent nanoactuators under the action of external multi-frequency pulses, generating local eddy current disturbances inside the pores, rather than just causing macroscopic deformation of the overall pores, thereby disrupting the adsorption equilibrium of the solute at the pore throat at the molecular scale.

Images