Six-channel micrometer high-precision filtration separation method

Abstract

The present application relates to the technical field of fluid precision separation, and discloses a six-channel micrometer high-precision filtration separation method, which comprises the following steps: controlling fluid cooling to a supersaturation precipitation temperature to generate colloidal flocculates; sequentially conveying the fluid through a six-stage physical separation sequence composed of gravity sedimentation, centrifugal cyclone, magnetic adsorption and multi-stage filtration medium; collecting terminal filtration differential pressure in real time and calculating the resistance growth rate; when the resistance growth rate exceeds a critical compaction threshold, feedback adjusting the front-end cooling rate or conveying temperature to optimize the flocculate structure rigidity; by establishing a reverse feedback regulation mechanism between the terminal rheological resistance characteristics and the front-end thermodynamic precipitation parameters, the technical problem that soft colloidal impurities penetrate the filtration medium due to shear deformation is solved, and the separation precision and process operation stability of complex heterogeneous fluid systems are improved.

Description

Technical Field

[0001] This invention relates to the field of precision fluid separation technology, and more specifically, to a six-channel micron-level high-precision filtration and separation method. Background Technology

[0002] Currently, in the process of industrial fluid purification, for heterogeneous systems containing trace amounts of suspended solids or colloidal impurities, the industry generally adopts a physical field synergistic graded separation process. This process combines mechanical coarse filtration, centrifugal sedimentation, gravity stratification, and terminal pressure filtration. By utilizing the selective effect of different physical fields on the particle size or density of impurities, liquid-solid separation can be achieved without the introduction of chemical reagents.

[0003] However, when dealing with soft flocculant systems rich in heat-sensitive colloids, waxes, or thixotropic properties, existing fluid transport modes based on constant flow rate or constant pressure gradients exhibit inherent limitations. These soft impurities possess non-Newtonian fluid characteristics and low yield stress thresholds, and the continuous shear stress field applied by conventional transport equipment often ignores this rheological property. For example, the utility model patent CN213192743U discloses a high-efficiency multi-stage filtration device for colloidal impurities. This scheme uses multi-stage filter grids with different pore sizes to remove colloidal and fibrous impurities through physical interception and grading. While hierarchical progressive screening logic is effective in processing rigid particles or long fibers, it is difficult to apply to thixotropic soft colloids due to the lack of process awareness and dynamic control in the open-loop processing mode. The device lacks the ability to intervene in the fluid thermodynamic state in real time and has no feedback mechanism based on the characteristics of terminal resistance. When soft colloids accumulate on the grid surface and are subjected to accumulated hydraulic pressure, they are prone to rheological extrusion deformation. Once the colloids undergo liquefaction-like deformation, their size becomes smaller than the grid aperture and penetrates the medium, leading to secondary contamination in downstream processes. This static passive mode of filtering without regard to the environment cannot adapt to fluctuations in raw material composition or changes in the compaction state of the filter cake, making it difficult to break through the bottleneck in separation accuracy.

[0004] Therefore, how to construct a dynamic transport and separation mechanism that can actively adapt to the rheological properties of soft impurities, and prevent colloids from deforming, penetrating or becoming dense and blocked while ensuring throughput, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] This invention provides a six-channel micron-level high-precision filtration and separation method, which includes the following steps:

[0006] S1. Temperature-controlled precipitation step: The solid-liquid mixture is pumped into a temperature-controlled regulating tank, and the fluid temperature is controlled to be reduced to the supersaturated precipitation temperature at a preset cooling rate, so that the dissolved impurities in the fluid are transformed into suspended colloidal flocs.

[0007] S2, Multi-stage physical separation step: The solid-liquid mixture after step S1 is sequentially transported through a six-stage physical separation sequence consisting of a gravity settling zone, a centrifugal hydrocyclone, a metal interception net, a magnetic adsorption bed, a deep fiber filter element and a terminal microporous membrane, so as to progressively intercept impurities of different particle sizes.

[0008] S3, Resistance feature extraction step: When the fluid passes through the terminal microporous membrane, real-time pressure difference data ΔP(t) on both sides of the terminal microporous membrane inlet and outlet are collected at a preset sampling frequency, and the resistance growth rate k, which characterizes the degree of compaction of the filter cake layer, is calculated based on the real-time pressure difference data.

[0009] S4. Source feedback control step, which compares the resistance growth rate k with the preset critical compaction threshold. Comparison; when the resistance growth rate k is greater than the critical compaction threshold If the filter cake layer is determined to be in an over-compacted state, a feedback adjustment action is executed: the cooling rate in step S1 is reduced, or the delivery temperature of the fluid to the terminal microporous membrane in step S2 is increased, in order to increase the average particle size of the colloidal flocs or reduce their compressibility, until the resistance growth rate k collected subsequently returns to less than or equal to the critical compaction threshold. Within the range.

[0010] Preferably, the specific execution actions of each physical separation sequence in step S2 are as follows: Gravity sedimentation separation, controlling the fluid to flow through the sedimentation tank in a laminar flow state, so that particles with a Stokes diameter greater than 100 μm settle to the bottom of the tank by utilizing density difference; Centrifugal cyclone separation, guiding the fluid tangentially into the hydrocyclone, using the centrifugal force field to separate particles with a density greater than the fluid matrix to the hydrocyclone wall; Metal mesh interception, allowing the fluid to pass through a rigid metal mesh with a pore size of 50 μm to 80 μm to intercept hard particles; Magnetic adsorption separation, guiding the fluid to flow through a flow channel equipped with a permanent magnet array, using the magnetic field to adsorb ferromagnetic particles in the fluid; Deep fiber filtration, guiding the fluid through a non-woven fiber medium with a pore size that decreases gradually along the flow direction, retaining soft colloids of 10 μm to 50 μm; Terminal microporous membrane separation, guiding the fluid through a microporous membrane with a defined retention pore size to remove submicron particles, which serve as the collection point for the real-time differential pressure data ΔP(t) in step S3.

[0011] Preferably, in step S1, the cooling rate is controlled between 0.5°C / min and 2.0°C / min; and during the cooling process, mechanical stirring is applied to the fluid, with the shear rate of the stirring controlled at 10. Up to 50 between.

[0012] Preferably, in step S3, the calculation process of the resistance growth rate k includes: obtaining the pressure difference-time series within the current filtering cycle; removing pressure pulsation noise with a frequency higher than 50Hz from the series; performing a first-order differential operation on the denoised series to obtain the slope of the pressure difference change with time, and using this slope as the resistance growth rate k.

[0013] Preferably, deep fiber filtration further includes a media regeneration step based on flow rate decay: real-time monitoring of the flow rate of fluid through the nonwoven fiber media; when the flow rate decays to 60% of the initial flow rate, it is determined that the pores of the nonwoven fiber media are blocked; using a cleaning medium of the same type as the filtered fluid, the nonwoven fiber media is impacted in the form of reverse pulses to cause the trapped colloids to fall off.

[0014] Preferably, step S4 further includes a flow-limiting logic based on shear stress: calculating the actual shear stress of the fluid on the surface of the terminal microporous membrane based on the current fluid delivery flow rate and the flow channel geometry parameters of the terminal microporous membrane. ; the actual shear stress With the preset colloidal yield stress Compare; when When, maintain the current delivery flow rate; if This reduces the transport flow rate.

[0015] Preferably, the terminal microporous membrane separation further includes a vibration cleaning step: during the filtration process, an ultrasonic generator is used to apply mechanical vibration at a frequency of 20kHz to 40kHz to the terminal microporous membrane; the mechanical vibration is used to break the adhesion of colloidal particles on the surface of the terminal microporous membrane.

[0016] Preferably, the method also utilizes a blocking feature factor. To evaluate the clogging status of the separated sequences; where, clogging characteristic factors Calculate using the following formula: Where ΔP(t) is the real-time differential pressure data measured at the current time t. The initial pressure difference of the terminal microporous membrane in a clean state is given by μ, the dynamic viscosity of the fluid is given by v, the apparent flow rate of the fluid through the terminal microporous membrane is given by v, and the cumulative operating time is given by t since the start of the current filtration cycle. The source feedback control step further includes: when the calculated clogging characteristic factor When the signal exhibits an exponential growth trend over time t, it is determined that a nonlinear deep blockage has occurred, triggering a feedback adjustment action.

[0017] Preferably, between gravity sedimentation separation and centrifugal cyclone separation, a water coalescence removal step is further included: guiding the fluid flow through a coalescence assembly made of hydrophilic fibers; causing tiny dispersed water droplets in the fluid to adsorb and coalesce into large droplets with a diameter greater than 500 μm on the surface of the hydrophilic fibers; and using the density difference between oil and water to separate and discharge the large droplets from the fluid.

[0018] Preferably, the feedback adjustment action follows the following tiered execution strategy: when the resistance growth rate k exceeds the critical compaction threshold... When the amplitude is less than 10%, reduce the drive frequency of the fluid transfer pump in the physical separation sequence; when the resistance growth rate k exceeds the critical compaction threshold... When the magnitude is between 10% and 30%, the cooling rate in step S1 is reduced while the driving frequency is decreased; when the resistance growth rate k exceeds the critical compaction threshold... When the amplitude is greater than 30%, the filtration and delivery are suspended, the heating program is started to raise the fluid temperature in the temperature control zone to redissolve the colloidal particles, and step S1 is restarted at a cooling rate lower than the original set value.

[0019] The embodiments of the present invention have at least the following beneficial effects:

[0020] 1. In six-stage high-precision micron-level filtration separation, a pressure filtration mechanism based on rheological characteristic matching eliminates the risk of shear deformation and penetration of soft impurities. Addressing the thixotropic and shear-sensitive physical properties of colloidal flocs precipitated at low temperatures, this invention abandons the conventional constant pressure or constant speed transport mode in the pressure filtration stage. Instead, it establishes a dynamic pressure response mechanism based on the transmembrane pressure difference growth rate, limiting the driving pressure gradient during the pressure filtration process between the yield stress threshold and the critical densification pressure of the colloidal filter cake layer. When the filter cake layer is detected to have entered the critical state of compression densification, periodic pressure relaxation actions are performed. Utilizing the viscoelastic rebound characteristics of the colloidal filter cake itself, the compressed micro-capillary channels are induced to undergo volume recovery and pore regeneration. This precise control of the fluid transport shear field avoids liquefaction-like deformation or extrusion effects caused by excessive compaction of soft impurities, ensuring that colloidal particles are always retained by the medium in a solid phase with a three-dimensional skeleton structure. This achieves efficient removal of soft matter in the deep filtration stage and slows down the clogging process of the filter medium.

[0021] 2. Constructing a reverse thermodynamic optimization closed loop across physical dimensions and establishing an adaptive separation steady state: This invention utilizes the terminal fine filtration unit as a diagnostic probe for the separation efficiency of the entire system, establishing a reverse feedback link between fluid dynamic resistance characteristics and upstream thermodynamic phase change parameters. By analyzing the pressure difference data of the fine filtration unit, the initial pressure difference component characterizing the basic viscosity of the fluid and the dynamic pressure difference component characterizing the media blockage rate are decoupled and analyzed. Based on the evolution trend of these two components, the system dynamically corrects the temperature control setpoint of the upstream low-temperature induced precipitation stage, enabling the separation system to automatically converge to the optimal balance point between viscosity resistance and impurity precipitation under conditions of batch fluctuations or component differences in feed oil. This prevents energy consumption redundancy and pumping resistance overload caused by excessively low temperatures, and also eliminates the hidden danger of dissolved colloids escaping to the terminal and adsorbing into a film in the micropores due to incomplete phase change, ensuring the system stability of the separation process under unsteady-state input.

[0022] 3. Achieving step-by-step unloading and phase regulation of the physical field gradient to ensure long-term operation with micron-level precision: This invention constructs a load-level stripping system for complex heterogeneous systems by sequentially coupling mechanical sieving, centrifugal force field, gravity sedimentation, thermodynamic phase change, and shear-controlled pressure filtration. The preceding physical field is not simply pretreatment, but rather gradually improves the physical properties of the fluid by specifically removing impurities of different particle sizes, densities, and phases: high-density particles are removed using centrifugal and sedimentation fields, dissolved solutes are converted into suspended solids using low-temperature fields, and most of the colloidal load is retained using the pressure filtration field. This step-by-step purification process changes the boundary conditions of the terminal fine filtration medium, requiring it to treat only trace amounts of rigid particles, thereby eliminating the tendency of colloids to form films and concentration polarization on the surface of micron-level filter pores. This allows the terminal filtration unit to stably maintain an absolute filtration precision of 4 to 6 microns under low-resistance, high-flux laminar flow conditions, achieving a dual improvement in filter consumable life and separation efficiency. Attached Figure Description

[0023] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings, in which several embodiments of the invention are illustrated by way of example and not limitation, wherein:

[0024] Figure 1 This is a process flow diagram of the six-channel micron-level high-precision filtration and separation method of the present invention;

[0025] Figure 2 This is a functional module and interaction logic diagram of the six-channel micron high-precision filtration and separation system of the present invention. Detailed Implementation

[0026] The principles and spirit of the present invention will now be described with reference to several exemplary embodiments in conjunction with the accompanying drawings. It should be understood that these embodiments are provided merely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0027] A six-channel micron-level high-precision filtration and separation method, comprising the following steps:

[0028] S1. Temperature-controlled precipitation step: The solid-liquid mixture is pumped into a temperature-controlled regulating tank, and the fluid temperature is controlled to be reduced to the supersaturated precipitation temperature at a preset cooling rate, so that the dissolved impurities in the fluid are transformed into suspended colloidal flocs.

[0029] S2, Multi-stage physical separation step: The solid-liquid mixture after step S1 is sequentially transported through a six-stage physical separation sequence consisting of a gravity settling zone, a centrifugal hydrocyclone, a metal interception net, a magnetic adsorption bed, a deep fiber filter element and a terminal microporous membrane, so as to progressively intercept impurities of different particle sizes.

[0030] S3, Resistance feature extraction step: When the fluid passes through the terminal microporous membrane, real-time pressure difference data ΔP(t) on both sides of the terminal microporous membrane inlet and outlet are collected at a preset sampling frequency, and the resistance growth rate k, which characterizes the degree of compaction of the filter cake layer, is calculated based on the real-time pressure difference data.

[0031] S4, Source Feedback Control Steps, will reduce the resistance growth rate. With the preset critical compaction threshold Comparison; when the resistance growth rate k is greater than the critical compaction threshold If the filter cake layer is determined to be in an over-compacted state, a feedback adjustment action is executed: the cooling rate in step S1 is reduced, or the delivery temperature of the fluid to the terminal microporous membrane in step S2 is increased, in order to increase the average particle size of the colloidal flocs or reduce their compressibility, until the resistance growth rate k collected subsequently returns to less than or equal to the critical compaction threshold. Within the range.

[0032] Preferably, the specific execution actions of each physical separation sequence in step S2 are as follows: Gravity sedimentation separation, controlling the fluid to flow through the sedimentation tank in a laminar flow state, so that particles with a Stokes diameter greater than 100 μm settle to the bottom of the tank by utilizing density difference; Centrifugal cyclone separation, guiding the fluid tangentially into the hydrocyclone, using the centrifugal force field to separate particles with a density greater than the fluid matrix to the hydrocyclone wall; Metal mesh interception, allowing the fluid to pass through a rigid metal mesh with a pore size of 50 μm to 80 μm to intercept hard particles; Magnetic adsorption separation, guiding the fluid to flow through a flow channel equipped with a permanent magnet array, using the magnetic field to adsorb ferromagnetic particles in the fluid; Deep fiber filtration, guiding the fluid through a non-woven fiber medium with a pore size that decreases gradually along the flow direction, retaining soft colloids of 10 μm to 50 μm; Terminal microporous membrane separation, guiding the fluid through a microporous membrane with a defined retention pore size to remove submicron particles, which serve as the collection point for the real-time differential pressure data ΔP(t) in step S3.

[0033] Preferably, in step S1, the cooling rate is controlled between 0.5°C / min and 2.0°C / min; and during the cooling process, mechanical stirring is applied to the fluid, with the shear rate of the stirring controlled at 10. Up to 50 between.

[0034] Preferably, in step S3, the calculation process of the resistance growth rate k includes: obtaining the pressure difference-time series within the current filtering cycle; removing pressure pulsation noise with a frequency higher than 50Hz from the series; performing a first-order differential operation on the denoised series to obtain the slope of the pressure difference change with time, and using this slope as the resistance growth rate k.

[0035] Preferably, deep fiber filtration further includes a media regeneration step based on flow rate decay: real-time monitoring of the flow rate of fluid through the nonwoven fiber media; when the flow rate decays to 60% of the initial flow rate, it is determined that the pores of the nonwoven fiber media are blocked; using a cleaning medium of the same type as the filtered fluid, the nonwoven fiber media is impacted in the form of reverse pulses to cause the trapped colloids to fall off.

[0036] Preferably, step S4 further includes a flow-limiting logic based on shear stress: calculating the actual shear stress of the fluid on the surface of the terminal microporous membrane based on the current fluid delivery flow rate and the flow channel geometry parameters of the terminal microporous membrane. ; the actual shear stress With the preset colloidal yield stress Compare; when When, maintain the current delivery flow rate; if This reduces the transport flow rate.

[0037] Preferably, the terminal microporous membrane separation further includes a vibration cleaning step: during the filtration process, an ultrasonic generator is used to apply mechanical vibration at a frequency of 20kHz to 40kHz to the terminal microporous membrane; the mechanical vibration is used to break the adhesion of colloidal particles on the surface of the terminal microporous membrane.

[0038] Preferably, the method also utilizes a blocking feature factor. To evaluate the clogging status of the separated sequences; where, clogging characteristic factors Calculate using the following formula: Where ΔP(t) is the real-time differential pressure data measured at the current time t. The initial pressure difference of the terminal microporous membrane in a clean state is given by μ, the dynamic viscosity of the fluid is given by v, the apparent flow rate of the fluid through the terminal microporous membrane is given by v, and the cumulative operating time is given by t since the start of the current filtration cycle. The source feedback control step further includes: when the calculated clogging characteristic factor When the signal exhibits an exponential growth trend over time t, it is determined that a nonlinear deep blockage has occurred, triggering a feedback adjustment action.

[0039] Preferably, between gravity sedimentation separation and centrifugal cyclone separation, a water coalescence removal step is further included: guiding the fluid flow through a coalescence assembly made of hydrophilic fibers; causing tiny dispersed water droplets in the fluid to adsorb and coalesce into large droplets with a diameter greater than 500 μm on the surface of the hydrophilic fibers; and using the density difference between oil and water to separate and discharge the large droplets from the fluid.

[0040] Preferably, the feedback adjustment action follows the following tiered execution strategy: when the resistance growth rate k exceeds the critical compaction threshold... When the amplitude is less than 10%, reduce the drive frequency of the fluid transfer pump in the physical separation sequence; when the resistance growth rate k exceeds the critical compaction threshold... When the magnitude is between 10% and 30%, the cooling rate in step S1 is reduced while the driving frequency is decreased; when the resistance growth rate k exceeds the critical compaction threshold... When the amplitude is greater than 30%, the filtration and delivery are suspended, the heating program is started to raise the fluid temperature in the temperature control zone to redissolve the colloidal particles, and step S1 is restarted at a cooling rate lower than the original set value.

[0041] Example 1: This example is applied to the treatment of a high-viscosity non-Newtonian fluid system containing heat-sensitive colloids and trace amounts of waxes. The impurity components in this system exhibit shear-thinning characteristics and are prone to deformation and penetration under conventional pressure fields. Separation is achieved through a multi-level physical field synergy and source feedback mechanism. The specific process is as follows: Step S1, temperature-controlled precipitation, is performed by pumping the solid-liquid mixture into a temperature-controlled regulating tank. The fluid temperature is controlled to decrease to the supersaturated precipitation temperature at a cooling rate of 0.5℃ / min to 2.0℃ / min. During this cooling process, a shear rate of 10 is applied to the fluid. Up to 50 Mechanical stirring, utilizing thermodynamic solubility differences, induces dissolved colloidal components in the fluid to transform into suspended colloidal flocs with specific physical forms, executing the S2 multi-stage physical separation step. The fluid sequentially passes through a six-stage physical separation sequence to progressively retain impurities of different particle sizes. The fluid flows through a settling tank in laminar flow, allowing particles with a Stokes diameter greater than 100 μm to settle due to density differences. It then enters a centrifugal hydrocyclone tangentially, where centrifugal force separates particles with a density greater than the fluid matrix. Next, it flows through a channel equipped with a permanent magnet array for magnetic adsorption. Afterward, it passes through a rigid metal mesh with pore sizes of 50 μm to 80 μm to intercept hard particles. Subsequently, it undergoes deep fiber filtration through a nonwoven fiber medium with pore sizes gradually decreasing along the flow direction, retaining soft colloids of 10 μm to 50 μm. Finally, the fluid passes through a terminal microporous membrane with a defined retention pore size, which removes submicron particles. The system uses microparticles as differential pressure acquisition nodes. When fluid passes through the terminal microporous membrane, the S3 resistance feature extraction step is executed. To eliminate the interference of temperature regulation in step S4 on the fluid's physical properties, a temperature-viscosity lookup table for this specific fluid is pre-written into non-volatile memory. This lookup table covers a range from 10℃ to 60℃ with a step interval of 0.5℃. In each calculation cycle, the central processing unit reads the real-time fluid temperature at the terminal microporous membrane and obtains the current fluid dynamic viscosity value using linear interpolation via a lookup table. The system acquires real-time differential pressure data ΔP(t) at a preset sampling frequency of 200Hz across the inlet and outlet of the terminal microporous membrane. By removing pressure pulsation noise with a frequency higher than 50Hz and performing first-order differential operations on the denoised sequence, the resistance growth rate k, characterizing the compaction degree of the filter cake layer, is calculated. Simultaneously, the system calculates the resistance growth rate k based on the real-time differential pressure data ΔP(t) and the initial differential pressure. The fluid dynamic viscosity μ and apparent velocity v are determined according to the formula. Calculate the blocking feature factor , where t is the cumulative running time since the start of the current filtering cycle.

[0042] Finally, the S4 source feedback control step is executed, and the calculated resistance growth rate k is compared with the preset critical compaction threshold. In comparison, when the resistance growth rate k is greater than the critical compaction threshold... or blocking feature factors When the filter cake layer exhibits an exponential growth trend over time t, it is determined that the filter cake layer is in a state of over-compaction or nonlinear deep blockage. The system then executes feedback adjustment actions: the adjustment strategy follows the blockage type priority logic: if the alarm is triggered by an excessive growth acceleration ratio (i.e., exponential growth), it indicates that fine particles are undergoing deep penetration, and the system prioritizes reducing the cooling rate in step S1, with a step size of 0.2℃ per minute; if the alarm is only triggered by the resistance growth rate exceeding the linear threshold, and the above ratio does not exceed the limit, it indicates that the filter cake layer has undergone mechanical compaction, and the system prioritizes reducing the cooling rate in step S1, or increasing the delivery temperature of the fluid to the terminal microporous membrane in step S2. This adjustment action changes the precipitation morphology of colloidal flocs, increasing their average particle size or reducing their compressibility, until the subsequently collected resistance growth rate k returns to less than or equal to the critical compaction threshold. Within the range, a reverse feedback closed loop was established between the terminal rheological resistance characteristics and the front-end thermodynamic precipitation parameters to ensure that soft impurities maintain suitable rigidity or size characteristics for physical retention when they reach the terminal filter medium.

[0043] Example 2: In a pilot-scale verification scenario for the fine separation of high-viscosity biomass fluids, the test subject was a standardized test liquid simulating industrial field conditions, with a matrix viscosity of [missing value] at 40°C. The sample contained 2.5 wt% thermosensitive phospholipid colloid and 0.5 wt% microcrystalline wax. This study verified the actual separation efficiency and long-term operational stability of the six-channel micron-level high-precision filtration separation method under conditions of fluctuating fluid rheological properties and unstable component phases. The experimental platform was equipped with a co-rotating twin-screw pump with variable frequency drive to provide a pulsating flow field. Background pressure noise with a frequency of 50 Hz and an amplitude of 0.02 MPa was introduced into the pipeline to simulate pumping disturbances in a real industrial environment. The data acquisition system included a high-frequency pressure sensor with a sampling frequency of 200 Hz and an online laser particle size analyzer to capture the pressure difference across the terminal microporous membrane and the particle distribution of the filtered fluid in real time. An experimental sample group using the technical solution of this invention was constructed. During the operation of the experimental sample group, the cooling rate of the temperature control tank was set to 1.2 °C / min, and the shear rate of the mechanical agitator was maintained at [missing information]. After the system starts up, the differential pressure data ΔP(t) of the terminal microporous membrane is monitored in real time. The data shows that during the initial operation phase, the resistance growth rate k remains in the low range of 0.002 kPa / s to 0.005 kPa / s, which is lower than the preset critical compaction threshold. (Set at 0.015 kPa / s); simultaneously, the blocking characteristic factor It exhibits a linear growth trend with time t, and the slope is approximately Under this steady state, the turbidity of the filtered fluid measured by the online laser particle size analyzer remained below 5 NTU, and no escaped particles with a diameter greater than 5 μm were detected. This indicates that the colloidal flocs precipitated at the front end have sufficient structural rigidity and can be effectively intercepted by the deep fiber filter without penetration. This is to verify the shear rate range ( to To demonstrate the engineering necessity of this approach, a first control sample group was established, which increased the shear rate of mechanical stirring to [a certain value]. With other conditions remaining constant, the experimental data exhibited obvious nonlinear degradation characteristics: within 15 minutes of the start of operation, the pressure difference ΔP(t) of the terminal microporous membrane did not rise slowly as expected, but instead showed irregular fluctuations; the resistance growth rate k repeatedly exceeded 0.020 kPa / s in a short period of time. Although it had not yet reached complete blockage, the laser particle size analyzer captured a surge in the number of particles with a diameter of 10 μm to 20 μm in the filtered fluid. Physical analysis showed that the excessively high shear rate caused the nascent colloidal flocs to break into fine fragments. These fragments exhibited stronger rheological properties under pressure, thus penetrating the deep fiber filter element and accumulating on the surface of the terminal microporous membrane, leading to the failure of separation accuracy. This result confirms that the upper limit of the shear rate in this invention is a key boundary for preventing the floc structure from breaking down and maintaining its physical interception size.

[0044] Next, to verify the thermodynamic rationality of the cooling rate range (0.5℃ / min to 2.0℃ / min), a second control group was established. This control group had a cooling rate increased to 3.5℃ / min, aiming to test the separation efficiency under rapid cooling. The experimental results showed that the blocking characteristic factor... It exhibited exponential growth characteristics in its early stages of operation, and the correlation coefficient of its fitted curve was... Reaching 0.98, after only 40 minutes of operation, the terminal differential pressure exceeded the alarm limit of 0.3 MPa. Upon dissection, it was discovered that the filter cake layer exhibited a dense gel-like structure instead of the expected loose and porous structure. This indicated that the excessively rapid cooling rate caused the colloids to precipitate through a burst of nucleation, forming a large number of fine, soft particles with a huge specific surface area. These particles rapidly filled the pores of the filter medium. Conversely, when the test sample group triggered the source feedback control mechanism, finely adjusting the cooling rate from 1.2℃ / min to 0.8℃ / min, subsequent data collection showed... The curve slope quickly returned to the linear range, and the permeability of the filter cake layer was restored. This comparative data proves that the cooling rate range defined in this invention is a thermodynamically necessary condition for inducing colloid growth into large-size, low-compressibility flocs. Finally, to verify the synergistic effect of the source feedback control mechanism, a third control group (partially missing control group) was established. This group cut off the feedback loop from S4 to S1 and S2, and only performed open-loop control. When the raw material batch of the test fluid fluctuated, causing a temporary increase of 10% in the colloidal content, the terminal pressure difference of this group rapidly increased, and the resistance growth rate k continuously exceeded the critical compaction threshold. The process lasted for 20 minutes, eventually causing the system to shut down due to high-pressure interlocking. The turbidity of the filtered fluid soared to 45 NTU. In contrast, the test sample group with a feedback mechanism automatically increased the fluid delivery temperature by 2°C and reduced the cooling rate within 3 minutes of detecting the abnormal increase in the k value, thus suppressing the deterioration trend of the k value and ensuring the continuous and stable operation of the unit under fluctuating raw material conditions.

[0045] Example 3: In an engineering scenario involving the initial process deployment of a new type of bio-fermentation broth, to ensure that the system's control parameters are adapted to the rheological characteristics of this specific fluid, a standardized critical compaction threshold calibration procedure is performed before starting continuous separation operations. The system introduces the fluid to be processed into the test loop, maintains constant front-end temperature control precipitation conditions, starts the delivery pump, and executes a step-by-step flow ramp-up program. At each flow step, the system maintains operation for 300 seconds and records the pressure difference data across the terminal microporous membrane. As the flow rate increases, when the ratio of the pressure drop across the filter cake layer to the flow rate no longer remains constant but exhibits a nonlinear positive abrupt change in the second derivative, it is determined that the colloidal particles inside the filter cake layer begin to undergo irreversible rheological compression deformation. The system captures the resistance growth rate corresponding to this abrupt change point as the ultimate collapse value. And the critical compaction threshold for normal system operation. Set as This calibration process transforms the abstract threshold setting into a measured process based on the physical limit of the filter cake layer yield stress, ensuring... Applicability to systems with different viscosity; limit collapse value in calibration procedure execution logic. The locking mechanism follows a real-time signal analysis and judgment algorithm. It performs a 50-sampling-point moving average filter on the acquired pressure-time raw sequence to remove high-frequency fluid turbulence noise. Second-order difference operations are then performed on the denoised data to extract the acceleration characteristics of the drag growth rate as a function of pressure difference. When the second-order difference value exceeds three times the reference noise variance for five consecutive cycles, it is determined that irreversible collapse deformation has occurred in the filter cake layer pore structure, triggering yield point signal latching. The average drag growth rate one second prior to this moment is automatically captured as the ultimate collapse value. According to the formula The calculated value is written to non-volatile memory as the critical compaction threshold for the current batch separation operation, in order to eliminate the possibility of control baseline drift caused by differences in operator subjective judgment.

[0046] After determining the baseline parameters, the system enters an adaptive operation mode. Its core lies in executing a set of feedback decision-making logic based on two-factor feature decoupling to address the problem that a single indicator cannot accurately guide multi-variable adjustment. The controller operates in each sampling period... Within 200ms, simultaneously calculate the resistance growth rate k and the blocking characteristic factor. When the system detects an abnormal alarm, it calculates... rate of change over time ,like Exceeding the preset depth clogging limit indicates that the current main failure mode is that fine particles penetrate the filter cake layer and are adsorbed inside the membrane pores, meaning the colloidal precipitate particle size is too small. In this case, the system prioritizes S1 path adjustment, reducing the cooling rate in steps of 0.1℃ / min to extend the crystal nucleation time and increase the flocculant particle size. Conversely, if... It is within the normal linear range, but the resistance growth rate k alone exceeds... This indicates that the particle size is acceptable but the structural rigidity is insufficient, resulting in flattening and creep under pressure. At this point, the system locks the S1 parameter and executes the S2 path adjustment, raising the fluid transport temperature by 1.0℃. It then uses the temperature-sensitive characteristic of colloidal solubility to fine-tune the solid-liquid interfacial tension, enhancing the shear modulus of the flocculants. To ensure the signal-to-noise ratio of the input signal of the above control logic in an industrial electromagnetic interference environment, the system forcibly executes a signal preprocessing procedure based on a sliding time window before calculating the resistance growth rate k. The original differential pressure signal sequence is passed through a moving average filter of length N=50 to filter out high-frequency pumping pulsations and fluid turbulence noise. The first derivative is calculated using a five-point quadratic Savitzky-Golay smoothing differential algorithm to obtain a smooth and trend-sensitive k value. This algorithm ensures that the feedback loop will not malfunction due to instantaneous pressure fluctuations, enabling the entire separation system to achieve dynamic optimization and precise locking of thermodynamic phase change parameters and fluid dynamic transport parameters solely based on physical sensor data in an unattended state.

[0047] Example 4: In the pre-application stage of industrial applications requiring rapid adaptation to different batches of raw materials, to address the risk of parameter drift caused by differences in the fundamental physical properties of fluids, this embodiment of the invention constructs and executes a standardized on-site deployment pre-calibration procedure. Through a series of controlled micro-trial and error experiments, the characteristic fingerprint of the current batch of fluid is established, and a baseline parameter set for system operation is generated accordingly. Before formal feeding, the system extracts 10L of standard sample fluid into the test loop and performs a linear cooling scan at an extremely low rate of 0.1℃ / min under a constant stirring rate. During this process, a high-sensitivity turbidimeter monitors the abrupt changes in the fluid's optical properties in real time, and the corresponding temperatures are recorded as the initial nucleation temperatures. The system is based on Automatically calculates and sets the target supersaturation temperature for the S1 temperature-controlled precipitation step. The calculation formula is set as follows ,in The preset subcooling optimization value is derived from the system's internal database of historical best operating conditions. For the initial cold start operation, the default value of this parameter in the database is initialized to 2.0℃. The system has a self-learning update mechanism: after each complete separation operation, if the cleaning cycle of the terminal microporous membrane exceeds 4 hours and the turbidity after filtration is less than 1 NTU, the system will record the subcooling value set for that operation and replace the original value in the database with this value as the starting benchmark for the next batch of the same raw materials. This calibration step ensures that no matter how the raw material batch fluctuates, the system always starts near the thermodynamic optimum, avoiding incomplete precipitation due to insufficient subcooling or explosive nucleation due to excessive subcooling.

[0048] After establishing the thermodynamic baseline, the next step is to perform on-site optimization of the kinetic parameters, maintaining the fluid temperature at [temperature value missing]. The system starts the variable frequency delivery pump to circulate the fluid in the test loop. During the circulation, the controller performs a sinusoidal flow disturbance test with a disturbance frequency set from 0.1Hz to 1.0Hz. By synchronously acquiring the dynamic pressure difference response across the microporous membrane at the terminal, the system uses Fast Fourier Transform (FFT) to analyze the dynamic impedance spectrum of the filter cake layer. Based on the amplitude characteristics of the low-frequency band of the impedance spectrum, a low-frequency amplitude-modulus mapping table is used for lookup. This mapping table was established in the laboratory stage through the following calibration process: a rheometer is used to perform dynamic mechanical tests on standard filter cakes with different compaction degrees to determine their storage modulus at a frequency of 0.5Hz; at the same time, the pressure response amplitude in the flow disturbance test under this state is recorded. Multiple sets of pressure response amplitudes and storage modulus data are linearly fitted to generate a slope coefficient, which is stored in the controller. During field operation, the system extracts the pressure amplitude at 0.5Hz, multiplies it by the slope coefficient, and the system reversely calculates the equivalent compressive modulus of the current filter cake layer. Based on this measured modulus value, the system automatically corrects the critical compaction threshold. The setting and correction logic follow an inverse relationship, that is, when the measured modulus When the temperature is low (indicating the filter cake is soft), the temperature will automatically decrease. To implement stricter pressure growth limits, this parameter adaptive initialization process based on field-measured rheological characteristics eliminates the risk of membrane blockage during trial operation caused by improper setting of empirical parameters, ensuring that the system can directly enter a highly efficient and stable separation condition in a cold start state.

[0049] Example 5: In the system integration, debugging, and procedure solidification phase for large-scale continuous production, to ensure the reproducibility of the above-mentioned multi-level physical field coordination mechanism across different batches of raw materials and varying operating conditions, this example developed and implemented a standardized offline calibration procedure for the critical compaction threshold and an adaptive control parameter initialization process. Through a set of controlled stepped pressure scan experiments, the core control parameter, the critical compaction threshold, was transformed from an abstract concept into a specific physical quantity value for a particular fluid to be processed, and the initial operating baseline of the system was set accordingly. The critical compaction threshold calibration was then performed. For offline calibration, a 50L sample of the fluid to be processed is injected into a separate test loop. This loop is equipped with a deep fiber filter and a terminal microporous membrane of the same specifications as the main system. The calibration is performed at a constant temperature (e.g., 40℃) and a constant stirring shear rate (e.g., 30). Under these conditions, the feed pump is started, and the fluid is controlled to pass through the filter medium at a preset initial flow rate (e.g., 0.5 m / s). A step-by-step pressurization program is executed, increasing the transmembrane pressure difference by 0.01 MPa every 300 seconds. Simultaneously, real-time pressure difference data and flow rate data are collected at a frequency of 100 Hz. The system calculates and records the resistance growth rate k at each pressure step in real time. When a non-linear positive abrupt change in the second derivative of k with the increase of pressure difference is detected (i.e., ... When the pressure difference reaches a certain point, it is determined that the colloidal particles inside the filter cake layer begin to undergo irreversible rheological compression deformation. The corresponding pressure difference is the rheological yield point of the fluid system, and the system records the drag growth rate corresponding to this yield point as the ultimate collapse value. And the critical compaction threshold in formal production. Set as This serves as a safety boundary to prevent filter cake densification.

[0050] In determining Next, the initial settings for the temperature-controlled precipitation parameters are executed. Based on the viscosity-temperature characteristic curve of the fluid to be processed, which is obtained through prior rheometer testing and stored in the control system database, the system determines the theoretical optimal shear rate range of the fluid at different temperatures, combined with the previously calibrated values. The controller automatically generates an initial parameter matrix including the cooling rate, target supersaturation temperature, and stirring shear rate. For example, for high-viscosity fluids, the system will automatically match a lower cooling rate, such as 0.8℃ / min, to extend the crystal nucleation time, and set a higher shear rate, such as 40℃ / min. To prevent local agglomeration; conversely, for fluids with lower viscosity, a higher cooling rate (e.g., 1.5℃ / min) is matched to improve production efficiency. In the initial stage of system startup, the controller will perform open-loop control based on this parameter matrix, and seamlessly switch to closed-loop adaptive control mode with the feedback of real-time resistance characteristic data in step S3. This ensures that the system can quickly converge to the optimal separation condition in the cold start stage, avoiding initial operational instability or premature blockage of the medium due to improper parameter settings.

[0051] The above description is only a few preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, technical solutions formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A six-channel micron-level high-precision filtration and separation method, characterized in that, The method includes the following steps: S1. Temperature-controlled precipitation step: The solid-liquid mixture is pumped into a temperature-controlled regulating tank, and the fluid temperature is controlled to be reduced to the supersaturated precipitation temperature at a preset cooling rate, so that the dissolved impurities in the fluid are transformed into suspended colloidal flocs. S2, Multi-stage physical separation step: The solid-liquid mixture after step S1 is sequentially transported through a six-stage physical separation sequence consisting of a gravity settling zone, a centrifugal hydrocyclone, a metal interception net, a magnetic adsorption bed, a deep fiber filter element and a terminal microporous membrane, so as to progressively intercept impurities of different particle sizes. S3, Resistance feature extraction step: When the fluid passes through the terminal microporous membrane, real-time pressure difference data ΔP(t) on both sides of the terminal microporous membrane inlet and outlet are collected at a preset sampling frequency, and the resistance growth rate k, which characterizes the degree of compaction of the filter cake layer, is calculated based on the real-time pressure difference data. S4. Source feedback control step, which compares the resistance growth rate k with the preset critical compaction threshold. Comparison; when the resistance growth rate k is greater than the critical compaction threshold If the filter cake layer is determined to be in an over-compacted state, a feedback adjustment action is executed: the cooling rate in step S1 is reduced, or the delivery temperature of the fluid to the terminal microporous membrane in step S2 is increased, in order to increase the average particle size of the colloidal flocs or reduce their compressibility, until the resistance growth rate k collected subsequently returns to less than or equal to the critical compaction threshold. Within the range.

2. The six-channel micron-level high-precision filtration and separation method according to claim 1, characterized in that, The specific actions of each physical separation sequence in step S2 are as follows: Gravity sedimentation separation, controlling the fluid to flow through the sedimentation tank in a laminar flow state, so that particles with a Stokes diameter greater than 100μm settle to the bottom of the tank by utilizing density difference; Centrifugal cyclone separation, guiding the fluid tangentially into the hydrocyclone, using the centrifugal force field to separate particles with a density greater than that of the fluid matrix to the hydrocyclone wall; Metal mesh interception, allowing the fluid to pass through a rigid metal mesh with a pore size of 50μm to 80μm to intercept hard particles; Magnetic adsorption separation guides the fluid through a flow channel equipped with a permanent magnet array, using the magnetic field to adsorb ferromagnetic particles in the fluid; deep fiber filtration guides the fluid through a nonwoven fiber medium with pore sizes that decrease in gradient along the flow direction, retaining soft colloids of 10μm to 50μm. Terminal microporous membrane separation guides fluid through a microporous membrane with a defined pore size to remove submicron particles and serves as the acquisition point for real-time differential pressure data ΔP(t) in step S3.

3. The six-channel micron-level high-precision filtration and separation method according to claim 1, characterized in that, In step S1, the cooling rate is controlled between 0.5℃ / min and 2.0℃ / min; and during the cooling process, mechanical stirring is applied to the fluid, with the shear rate of the stirring controlled at 10. Up to 50 between.

4. The six-channel micron-level high-precision filtration and separation method according to claim 1, characterized in that, In step S3, the calculation process of the resistance growth rate k includes: obtaining the pressure difference-time series within the current filtering cycle; removing pressure pulsation noise with a frequency higher than 50Hz from the series; performing a first-order differential operation on the denoised series to obtain the slope of the pressure difference change with time, and using this slope as the resistance growth rate k.

5. The six-channel micron-level high-precision filtration and separation method according to claim 2, characterized in that, Deep fiber filtration also includes a media regeneration step based on flow rate decay: real-time monitoring of the fluid velocity through the nonwoven fiber media; when the flow rate decays to the initial flow rate... When the nonwoven fiber medium is found to be blocked, a cleaning medium of the same type as the filtered fluid is used to impact the nonwoven fiber medium in the form of a reverse pulse, causing the trapped colloids to fall off.

6. The six-channel micron-level high-precision filtration and separation method according to claim 1, characterized in that, Step S4 also includes a flow-limiting logic based on shear stress: based on the current fluid delivery flow rate and the flow channel geometry parameters of the terminal microporous membrane, the actual shear stress of the fluid on the surface of the terminal microporous membrane is calculated. ; the actual shear stress With the preset colloidal yield stress Compare; when When, maintain the current delivery flow rate; if This reduces the transport flow rate.

7. The six-channel micron-level high-precision filtration and separation method according to claim 1, characterized in that, Terminal microporous membrane separation also includes a vibration cleaning step: during the filtration process, an ultrasonic generator is used to apply mechanical vibration at a frequency of 20kHz to 40kHz to the terminal microporous membrane; the mechanical vibration is used to break the adhesion of colloidal particles on the surface of the terminal microporous membrane.

8. The six-channel micron-level high-precision filtration and separation method according to claim 1, characterized in that, This method also utilizes a blocking feature factor To evaluate the clogging status of the separated sequences; where, clogging characteristic factors Calculate using the following formula: Where ΔP(t) is the real-time differential pressure data measured at the current time t. The initial pressure difference of the terminal microporous membrane in a clean state is given by μ, the dynamic viscosity of the fluid is given by v, the apparent flow rate of the fluid through the terminal microporous membrane is given by v, and the cumulative operating time is given by t since the start of the current filtration cycle. The source feedback control step further includes: when the calculated clogging characteristic factor When the signal exhibits an exponential growth trend over time t, it is determined that a nonlinear deep blockage has occurred, triggering a feedback adjustment action.

9. The six-channel micron-level high-precision filtration and separation method according to claim 2, characterized in that, Between gravity sedimentation separation and centrifugal cyclone separation, there is also a water coalescence removal step: guiding the fluid through a coalescence assembly made of hydrophilic fibers; causing tiny dispersed water droplets in the fluid to be adsorbed on the surface of the hydrophilic fibers and coalesce into large droplets with a diameter greater than 500 μm; and using the density difference between oil and water to separate and discharge the large droplets from the fluid.

10. The six-channel micron-level high-precision filtration and separation method according to claim 1, characterized in that, Feedback adjustment actions follow the following tiered execution strategy: when the resistance growth rate k exceeds the critical compaction threshold... When the amplitude is less than 10%, reduce the drive frequency of the fluid transfer pump in the physical separation sequence; when the resistance growth rate k exceeds the critical compaction threshold... When the magnitude is between 10% and 30%, the cooling rate in step S1 is reduced while the driving frequency is decreased; when the resistance growth rate k exceeds the critical compaction threshold... When the amplitude is greater than 30%, the filtration and delivery are suspended, the heating program is started to raise the fluid temperature in the temperature control zone to redissolve the colloidal particles, and step S1 is restarted at a cooling rate lower than the original set value.

Images