A membrane concentration process for plant extracts

Abstract

The application discloses a membrane concentration process for plant extract. The process sends pretreated plant water extract or alcohol-water mixed extract into at least four sections of serial reverse osmosis membrane assemblies in turn, wherein different sections adopt a combination of low-pressure low-permeation type membrane, medium-pressure medium-permeation type membrane and medium-pressure high-permeation type membrane. By controlling the total dissolved solids, conductivity and polyphenol, polysaccharide and protein content of the plant extract raw material, and setting the system operation pH, transmembrane flux, flow channel flow rate and inter-stage osmotic pressure difference, the multi-section concentration process is kept stable. In the process, the second section and the third section are subjected to flow rate control, pH micro-amplitude fluctuation, feed TDS periodic disturbance and temperature regulation, and the local flow rate is increased as necessary to weaken the influence of the deposition layer. The process can maintain a high transmembrane flux for a long operation period, concentrate the plant extract to a high degree, reduce the development speed of membrane pollution, and improve the continuous operation stability and concentration efficiency.

Description

Technical Field

[0001] This invention relates to the technical field of membrane concentration, and in particular to a membrane concentration process for plant extracts. Background Technology

[0002] Plant extracts are complex systems obtained by extracting natural plant materials through water or a mixture of water and alcohol. These systems typically contain a variety of components, including polyphenols, soluble polysaccharides, proteins, small peptides, and inorganic ions. They are characterized by a wide variety of components, complex solute interactions, and a rapid increase in osmotic pressure with increasing concentration. In the production of food, health products, and traditional Chinese medicine, it is often necessary to concentrate plant extracts to increase the content of active ingredients and reduce the energy consumption of subsequent evaporation.

[0003] In existing technologies, the concentration of plant extracts mainly relies on membrane separation and evaporation. For example, after coarse filtration to remove suspended impurities, reverse osmosis membranes are used for pre-concentration, followed by evaporation to achieve the target concentration. Because plant extracts contain a large number of organic macromolecules and surfactants, irreversible deposits easily form on the membrane surface, leading to problems such as rapid decline in transmembrane flux, increased operating pressure differential, and unstable inter-stage load distribution. For plant extracts with high TDS, membrane fouling often develops much faster than in inorganic salt systems, making it even more difficult to maintain stable concentration.

[0004] As the concentration ratio increases further, components such as polyphenols, polysaccharides, and proteins in the plant extracts are more likely to aggregate, associate, and compact on the membrane surface and in the adjacent area, causing the deposited layer to gradually transform from an initial loose state to a dense and compact morphology. Once the deposited layer forms a dense structure, it significantly hinders solute diffusion and water molecule migration, causing a continuous increase in transmembrane mass transfer resistance and leading to a faster rate of flux decline. To maintain the operating performance of the membrane module, existing systems typically require frequent use of methods such as pressurization, forced hydraulic flushing, or chemical cleaning to remove the deposited layer. However, these measures often only provide temporary flux recovery, and long-term frequent operation can also cause membrane material fatigue or performance degradation, affecting the lifespan of the membrane module.

[0005] On the other hand, the composition and physicochemical properties of plant extracts fluctuate significantly. The total dissolved solids content, conductivity, polyphenol and polysaccharide ratios, and even pH can vary between different batches of raw materials, causing the same membrane system to exhibit significant differences when dealing with different raw materials. For example, some raw materials can induce significant interfacial deposition at low concentrations, while others will suddenly experience a sharp decline in flux at high TDS. Due to the lack of effective control over interfacial deposition behavior, existing processes are less adaptable to processing extracts from different plant sources, and are prone to problems such as unstable operation, difficulty in increasing final concentration, or increased energy consumption.

[0006] It can be seen that plant extracts tend to form dense deposits that are difficult to remove during membrane concentration. Furthermore, the deposition behavior is influenced by the properties of the raw materials, operating conditions, and solute coupling, making it difficult to maintain stable transmembrane flux over the long term. Therefore, the key technical challenge currently faced is how to effectively inhibit the development of the deposited layer into a dense structure during the concentration of plant extracts, enabling the membrane system to maintain stable transmembrane flux under continuous operating conditions. Summary of the Invention

[0007] The purpose of this invention is to overcome the above-mentioned problems in the prior art and provide a membrane concentration process for plant extracts. This process can effectively inhibit the densification of the deposited layer and maintain the stability of the transmembrane flux of the membrane system during long-term operation.

[0008] To achieve the above objectives, the present invention provides a membrane concentration process for plant extracts, comprising:

[0009] Plant extracts are sequentially fed into a multi-stage membrane concentration system consisting of at least four reverse osmosis membrane modules connected in series. The first stage uses a low-pressure, low-permeability membrane, the second and third stages use medium-pressure, medium-permeability membranes, and the final stage uses a medium-pressure, high-permeability membrane.

[0010] The plant extract raw material is a plant aqueous extract or an alcohol-water mixed extract that has undergone coarse filtration and impurity removal pretreatment. The properties of the raw material include: total dissolved solids (TDS) of 10-20 g / L, total polyphenol content of 50-1000 mg / L (calculated as gallic acid), soluble polysaccharide content of 0.1-5 g / L, soluble protein and / or small molecule peptide content of 0.05-2 g / L, feed conductivity of 1-10 mS / cm, and pH of the liquid of 4.5-6.8.

[0011] The multi-stage membrane concentration system operates at a pH range of 5.5–7.0, and the overall transmembrane flux is controlled at 10–25 L / (m²). 2 ·h);

[0012] The flow velocity in the second and third sections is 1.5 to 2.0 m / s.

[0013] The first stage uses a low-pressure, low-permeability membrane; the second and third stages use medium-pressure, medium-permeability membranes; and the final stage uses a medium-pressure, high-permeability membrane. This combination of membrane types allows the system to establish distinct solute migration rates and interfacial deposition characteristics between stages from the initial stage of operation. Large molecules such as polyphenols, soluble polysaccharides, and proteins in plant extracts have high surface activity. Their adsorption or desorption behavior on the membrane surface is not only affected by the membrane material but also closely related to the transmembrane osmotic pressure difference, feed solution ionic strength, feed solution pH, and local fluid shear conditions. This invention, through the combined control of these factors, enables the formation of interfacial behavior patterns with significant spatial partitioning characteristics within the multi-stage system.

[0014] The first stage has the highest transmembrane flux and migration driving force, causing the solute to migrate significantly faster in the boundary layer than it resides on the membrane surface. In this region dominated by driving force, polyphenols and polysaccharides are unlikely to form a stable adsorption layer on the membrane surface, and most migrate with the flow, thus significantly inhibiting fouling accumulation in the initial stage. The second and third stages, under the combined influence of membrane type, operating pressure, and flow rate, form an interface region with moderate migration speed and moderate residence time. In this region, polyphenols and polysaccharides are more likely to undergo weak adsorption on the membrane surface, providing the necessary prerequisite for the formation of the interface layer. Since the flow rate in this region is limited to 1.5 to 2.0 m / s, the shear force is neither too low to cause rapid compaction of the deposited layer nor too high to prevent the interface layer from forming. Therefore, this stage forms a dynamic interface state with weak adsorption and weak exfoliation. The final stage has the lowest migration driving force and uses a high-permeability membrane, which alleviates the osmotic pressure limitation under high-salt conditions. Stable flux is maintained in the final stage of concentration, thereby increasing the overall concentration factor and reducing the subsequent evaporation load.

[0015] The TDS, polyphenols, polysaccharides, proteins, conductivity, and pH of the plant extract raw materials are limited to specific ranges, ensuring that the charge state of the solute near the membrane surface, ion inclusion, hydrogen bond network strength, and intermolecular interactions are all within a range that can be effectively controlled by operating conditions. Within this range, the interaction forces between various organic compounds are at a moderate strength, preventing the formation of a strong adsorption layer due to excessive charge or the complete absence of interfacial deposition due to insufficient charge, thus providing a basis for the formation of a reversible interfacial layer.

[0016] Within the pH range of 5.5 to 7.0, the phenolic hydroxyl groups of polyphenols partially dissociate, and the amino and carboxyl groups in proteins are in a weakly charged or near-neutral state. Therefore, their interaction with membrane materials is mainly achieved through weak electrostatic interactions, π–π interactions between aromatic rings, and a hydrogen bond network with water molecules as bridges. The binding energies of these interactions are generally within a range that can be broken by factors such as local fluid shear forces, osmotic pressure differences, and micro-perturbations in ionic strength. In other words, the interactions are sufficient to form an adsorption layer, but can also loosen or desorb when interfacial conditions change. This reversible characteristic provides the necessary physical basis for subsequent interfacial cycling induced by solute migration, shear forces, and local concentration fields.

[0017] The overall transmembrane flux range was set to 10 to 25 L / (m²). 2The flow rate (h) is designed to achieve a suitable balance between the migration rate of the solute from the bulk liquid phase to the membrane surface and the residence time of the solute on the membrane surface. Within this flux range, the rate at which the solute reaches the membrane surface is not too high, thus preventing the continuous dispersion of deposits and the formation of an interfacial buffer layer. Simultaneously, the migration rate is not too low, preventing the solute from continuously remaining on the membrane surface, stacking together, and rapidly forming a dense and difficult-to-remove fouling layer. This maintains a relatively loose and non-dense structure in the membrane deposit layer, making it easier to loosen and redistribute under disturbance conditions in the middle region, thereby maintaining reversible dynamic adsorption behavior. Under the combined influence of membrane type, migration driving force, feedstock properties, operating pH, and flow rate, the multi-stage membrane system forms a spatially partitioned interfacial system composed of a high-migration zone, a medium-adsorption zone, and a low-driving zone. This system causes polyphenols and polysaccharides to exhibit different interfacial behaviors in different regions, from rapid migration in the initial stage to weak adsorption in the middle stage, and then to low-driving flux maintenance in the final stage, forming a cyclical process of migration, adsorption, loosening, and redistribution. This cyclic process is essentially an interface update mechanism induced by operating conditions, which prevents the contaminant layer from accumulating unidirectionally and keeps it in a dynamic and reversible state. This significantly reduces the long-term flux decline trend of multi-stage systems, improves the stability of continuous concentration, and meets the engineering requirements for high-concentration of plant extracts.

[0018] As a further improvement of the present invention, by adjusting the outlet pressure of the inter-section booster pump, the steady-state transmembrane osmotic pressure difference along each section of the flow direction satisfies the following relationship:

[0019] The transmembrane osmotic pressure difference in the first segment is the maximum value. The transmembrane osmotic pressure differences in the second and third segments are controlled at 80% to 95% of the transmembrane osmotic pressure difference in the first segment, and are both higher than the transmembrane osmotic pressure difference in the last segment. This forms a transmembrane osmotic pressure difference distribution with the first segment as the peak, the second and third segments forming a relatively stable plateau, and a significant decrease in the last segment.

[0020] By adjusting the outlet pressure of the inter-stage booster pumps, a transmembrane osmotic pressure gradient distribution is achieved under steady-state operating conditions, with the first stage having the highest peak pressure, the second and third stages forming a buffer plateau, and the final stage showing a significant decrease. The first stage has the highest transmembrane osmotic pressure gradient, which provides the greatest migration driving force for polyphenols, polysaccharides, and proteins in the feed solution upon entering the system. The rate of solute transfer to the membrane surface is significantly faster than its retention and adsorption rate on the membrane surface. Therefore, it is difficult for a persistent structured deposition layer to form in the first stage, and the tendency for fouling accumulation is significantly weakened.

[0021] The transmembrane osmotic pressure difference in the second and third stages is limited to 80% to 95% of that in the first stage, reducing the driving force for solute migration in this stage to a moderate level. Under this driving force, the rate at which the solute reaches the membrane surface is close to its residence time on the membrane surface, making it easier for polyphenols and polysaccharides to form a loose, non-dense adsorption layer on the membrane surface.

[0022] The transmembrane osmotic pressure differential at the end of the concentration stage is set to a minimum, eliminating the strong driving force on the high-concentration feed solution at the concentration end. Under these low-driving conditions, the solute is less likely to form new deposits, and existing loose deposits are less likely to be compacted, thus avoiding rapid fouling caused by high osmotic pressure at the end. The low pressure differential design at the end also reduces the mechanical load on the membrane element, which is beneficial for maintaining flux stability and membrane life under high TDS conditions.

[0023] As a further improvement of the present invention, the pH of the multi-stage membrane concentration system is controlled at 5.8 to 6.5, and the pH of the second and third stages fluctuates within this range of ±0.1 to ±0.3.

[0024] The operating pH of the multi-stage membrane concentration system is limited to a narrow range of 5.8 to 6.5, keeping polyphenols, proteins, small peptides, and other functional groups containing carboxyl, phenolic hydroxyl, and amino groups in the plant extract in a weakly or partially ionized state. Within this pH range, the phenolic hydroxyl groups of polyphenols are not completely deprotonated, and protein molecules have both positively and negatively charged groups on their surfaces, exhibiting an overall weak charge or near-isoelectric state. This charge state results in the interaction between the solute and the membrane surface being dominated by weak electrostatic attraction, hydrogen bonding, and hydrophobic interactions. The interaction energy is in a moderate range, capable of forming an adsorption layer but easily broken by changes in external conditions, thus facilitating the formation of a reversible interfacial deposition layer.

[0025] Secondly, within the aforementioned operating pH range, the pH of the second and third sections is allowed to fluctuate slightly between ±0.1 and ±0.3, causing the local chemical environment of these sections to slowly oscillate between weakly acidic and near-neutral states. With each slight change in pH, the degree of ionization of polyphenol and protein molecules changes slightly, and the number and distribution of charged groups are subtly adjusted. This results in a continuous, slow rearrangement of electrostatic interactions, hydrogen bond numbers, and strengths with the film surface and within the deposition layer. Each small increase or decrease in pH causes partial dissociation and recombination of the existing hydrogen bond network and electrostatic pairing relationships within the deposition layer, making it difficult for the deposition layer to maintain a highly ordered and dense structure over a long period, instead tending towards a more loose and defect-prone state.

[0026] By controlling the pH fluctuations, the transformation of the deposited layer into a dense, highly cross-linked fouling layer can be effectively suppressed. This reduces the compaction and mechanical strength of the deposited layer, making it more susceptible to localized delamination or migration under subsequent flow channel shear forces, osmotic pressure fluctuations, and concentration disturbances. Through this micro-disturbance control at the chemical environment level, this invention delays the irreversible fouling process of the mid-section membrane module without introducing additional cleaning or severe impact conditions, thus improving flux retention during long-term operation.

[0027] As a further improvement of the present invention, the flow velocity of the second and third sections is higher than that of the preceding and following sections. The flow velocity of the first section is 2.0 to 3.0 m / s, and the flow velocity of the last section is maintained at 1.0 to 1.5 m / s.

[0028] The flow velocities in the second and third stages are limited to be higher than those in their adjacent stages, while the first stage maintains a relatively high velocity of 2.0 to 3.0 m / s, and the final stage maintains a relatively low velocity of 1.0 to 1.5 m / s. This creates a hydrodynamic gradient structure along the flow direction in the entire multi-stage membrane system, transitioning from a high-shear zone to a moderate-shear zone and then to a low-shear zone. This structure not only affects the transport mode of the solute in the flow channels but also significantly alters the velocity of the solute reaching the membrane surface, the mechanical state of the deposited layer, and the stability of the deposited layer between stages.

[0029] In the first stage, the high flow rate causes the feed to immediately experience a strong shear flow field upon entering the membrane module. This high shear keeps the boundary layer thickness low, significantly shortens the residence time of solutes near the membrane surface, and makes it difficult for polyphenols, polysaccharides, and protein molecules to form stable initial deposition nuclei on the membrane surface. Consequently, the growth rate of the fouling layer is significantly reduced in this stage. Therefore, the high-shear region in the first stage not only mitigates the initial deposition risk but also provides a more uniform feed condition for subsequent stages, making the mid-stage deposition process more controllable.

[0030] In the second and third stages, the flow rate is moderate, and the fluid shear force and solute adsorption force reach an equilibrium state that allows for the formation of a thin layer of deposition without causing compaction. The deposition layer structure formed by molecules such as polyphenols and polysaccharides on the film surface tends to be loose and contains a large number of micropores. The shear force can continuously disturb the microscopic accumulation inside the deposition layer, keeping the deposition layer in a loose state and preventing it from transforming into a dense structure. This shear force level is precisely within the range that can maintain the interfacial circulation state, enabling a stable dynamic process of weak adsorption and weak exfoliation to form in the middle stage.

[0031] In the final stage, the reduced flow rate allows for a longer residence time in the high TDS concentrate. Under low shear conditions, the tendency for the deposited layer to detach decreases, and the interface becomes more stable, which is beneficial for maintaining permeability under high concentration conditions. The low shear range in the final stage, combined with the high-permeability membrane, ensures that the risk of fouling at the end of the concentration is not exacerbated by insufficient driving force or excessive flow rate, thus guaranteeing steady-state operation under high TDS conditions.

[0032] As a further improvement of the present invention, the change range of instantaneous transmembrane osmotic pressure difference between the first and second segments reaches more than 10% of the steady-state transmembrane osmotic pressure difference of the first segment; the difference of steady-state transmembrane osmotic pressure difference between the second and third segments does not exceed ±5%; and the steady-state transmembrane osmotic pressure difference between the third segment and the last segment is reduced to 10-30% of the steady-state osmotic pressure difference of the third segment.

[0033] The instantaneous transmembrane osmotic pressure difference between the first and second stages exceeds 10% of the steady-state transmembrane osmotic pressure difference in the first stage, causing a temporary increase in the driving force for solute migration during the transition between stages. Under the influence of this instantaneous pressure fluctuation, easily deposited components such as polyphenols, polysaccharides, and proteins in the plant extract experience a stronger migration drive in the interfacial thin layer that was not fully deposited in the first stage. Their residence behavior near the membrane surface is disrupted, and the original weakly adsorbed layer loosens or partially migrates. This instantaneous driving force disturbance effectively weakens the local deposition nuclei formed in the early stages at the interface, making it difficult for the deposition structure to be compacted at the transition point between stages, thereby slowing down the development of the fouling layer.

[0034] The steady-state transmembrane osmotic pressure difference between the second and third sections does not exceed 5%, ensuring a continuous and stable driving force level for both sections. This uniform osmotic pressure plateau maintains a balance between solute interfacial migration, adsorption, and desorption in the middle section. The polyphenol and polysaccharide deposits on the membrane surface maintain a loose structure within this driving force range, preventing rapid compaction due to sudden increases in pressure difference or complete failure to form deposits due to sudden decreases in pressure difference, thus contributing to long-term flux stability.

[0035] The steady-state osmotic pressure difference between the third and final stages decreases by 10% to 30% of the steady-state osmotic pressure difference in the third stage, significantly reducing the migration driving force of the high-concentration feed solution entering the final stage. Under these low driving force conditions, the solute is unlikely to form a new thick deposition layer in the final stage, and the existing deposition layer lacks the conditions for further compaction, thus avoiding the interface hardening phenomenon caused by high osmotic pressure that is commonly seen at the end of the concentration process. The low driving force environment in the final stage gives the membrane element higher operational tolerance under high TDS conditions, reduces the mechanical stress of the membrane material, and improves the reliability of the high-concentration process.

[0036] As a further improvement of the present invention, the permeate return flow rate is adjusted within the range of ±10% to ±20% of the average value in a cycle of 20 to 60 minutes, and the feed pressure and / or feed flow rate of the second and third stages are adjusted accordingly within the cycle, so that the fluctuation range of the feed TDS of the second and third stages relative to the cycle average value within the cycle is 1.0 to 3.0 g / L, wherein the fluctuation range of the feed TDS of the first and last stages within the cycle is less than 1.0 g / L.

[0037] The permeate reflux flow rate is adjusted according to the above technical scheme to cause regular fluctuations in the ionic strength, solute concentration, and solvent structure within the system during this cycle. Since the second and third stages are located in the regions of moderate driving force and strongest reversibility of interfacial deposition in the system, the periodic reflux disturbance will cause the feed TDS in these two stages to fluctuate relative to the periodic average value, resulting in continuous slow disturbances in the local concentration polarization thickness, interfacial deposition layer density, and solute residence time on the membrane surface.

[0038] When the reflux rate increases, the ionic strength of the feed solution entering the second and third stages increases accordingly. The hydrogen bond network near the interface becomes tighter, the hydration structure inside the deposition layer tends to shrink, and the binding force of polyphenols and polysaccharides in the deposition layer is slightly enhanced, temporarily maintaining the relative stability of the interfacial deposition layer. When the reflux rate decreases, the ionic strength in the feed decreases, the local hydration structure becomes looser, the number of hydrogen bonds decreases, and the binding relationship between the originally weakly interacting deposition particles is partially disrupted. The deposition layer changes from a relatively compact state to a loose state, especially under the shear force in the middle stage, making it more prone to local peeling, interfacial rearrangement, and partial migration.

[0039] This periodic disturbance enables the membrane interface to form a stable, controllable, and sustainable cyclic state, thereby enhancing the long-term operational stability of the multi-stage membrane concentration system under high-load conditions and reducing overall energy consumption.

[0040] As a further improvement of the present invention, the third section gradually forms a temperature rise of 2 to 5°C along the flow direction, and the operating temperature of the third section is maintained at 28 to 33°C, which is higher than the operating temperatures of the first and second sections.

[0041] The third section generates a temperature rise of 2–5°C along the flow direction, causing the local temperature of the liquid to smoothly transition from the relatively low temperature of the previous section to a medium temperature range of 28–33°C. This temperature range happens to be within the sensitive area where polyphenols in the plant extract undergo mild auto-oxidation, weak intermolecular polymerization, and loosening of the sediment layer structure, making the third section a key region where significant changes occur in the solute interfacial behavior.

[0042] Within this temperature range, polyphenol molecules undergo limited auto-oxidation due to the slightly enhanced oxidizing power of their phenolic hydroxyl groups. Because the oxidation process is incomplete, the interactions between polymer segments are limited, resulting in a flexible structure in the deposited layer that does not form a dense or difficult-to-peel cross-linked layer. As the temperature gradually increases, the stability of the hydrogen bond network within the deposited layer decreases, the solvation structure of water molecules around the deposited particles loosens, and the bonding strength of the interfacial layer decreases. This temperature-induced structural loosening allows the weakly adsorbed structure formed in the second stage of the deposited layer to become further loosened in the third stage, making it easier for localized detachment or dispersion to occur under the moderate shear forces within the third stage.

[0043] Furthermore, under the condition of a slight temperature rise in the third stage, the viscosity of the feed liquid decreases slightly, which increases the solute migration rate and reduces the local concentration polarization intensity. The reduction in concentration polarization decreases the back-diffusion resistance on the surface of the sediment layer, making it easier for loose parts in the sediment layer to be carried away by the migrating flow.

[0044] As a further improvement of the present invention, the transmembrane flux and the 280nm UV absorbance on the concentrate side are monitored online for the second and third reverse osmosis membrane modules. When the transmembrane flux decreases by 5-15% compared to the initial operating flux and the UV absorbance increases by 0.05-0.20 compared to the initial value, the flow rate of the second and third sections is increased by 5-20% relative to their set flow rate and maintained for 10-60 minutes.

[0045] Online monitoring of transmembrane flux and 280 nm UV absorbance on the concentrate side in the second and third stages allows for real-time reflection of the interfacial deposition state and fouling development trend of the mid-section membrane module. Polyphenols in plant extracts exhibit a distinct characteristic absorption peak at 280 nm, and an increase in absorbance typically stems from localized enrichment of polyphenol structures on the membrane surface or within the deposition layer. Therefore, an increase of 0.05–0.20 nm in UV absorbance compared to the initial operating value indicates a detectable accumulation of aromatic organic matter in the membrane deposition layer, while a 5–15% decrease in transmembrane flux further reflects an increase in interfacial migration resistance. The simultaneous changes in these two indicators suggest that the deposition layer has evolved from an initially loose structure to a more compact one.

[0046] After detecting the aforementioned interface deposition signal, the flow rates in the second and third sections are increased by 5–20% relative to their set values ​​and maintained for 10–60 minutes, causing the middle section film surface to undergo a disturbance with significantly enhanced shear force in a short period of time. Under the action of enhanced shear force, the deposited layer is more prone to local fracture, fragment detachment, or structural loosening, causing loose aggregates and weakly adsorbed particles in the deposited layer to be redispersed into the bulk fluid, thereby weakening the densification tendency of the deposited layer.

[0047] This short-term enhanced shear response will not cause irreversible damage to the membrane module because the flow rate increase is significantly lower than the limit of operation and the duration is limited. Its mechanism of action is mainly to break down the localized dense regions that gradually form inside the deposition layer due to long-term steady-state operation, restoring the structure of the interface layer to a loose and easily detachable state.

[0048] As a further improvement of the present invention, the second and third segments adopt the same membrane type, and the effective membrane area of ​​their membrane modules is smaller than that of the first segment but larger than that of the last segment.

[0049] The second and third stages use the same membrane type, and the effective membrane area of ​​these two stages is limited to be smaller than that of the first stage but larger than that of the last stage. This creates membrane area zones with significant load distribution differences in the multi-stage membrane concentration system. Membrane area not only determines the permeate flux density per unit membrane area but also profoundly affects the solute migration rate, deposition rate, and effective range of interfacial shear forces. This invention, through its area distribution structure, makes the second and third stages key regions in the system where deposition is easily formed and most suitable for maintaining reversible interfacial cycling.

[0050] The effective membrane area in the middle section is set to be smaller than that in the first section, resulting in a slightly higher actual permeation flux per unit area. At this flux density, the solute migration rate is moderate, allowing the interfacial deposition layer to form without rapid compaction. Furthermore, the effective membrane area in the middle section is larger than that in the final section, resulting in a lower flux load per unit area and weaker interfacial shear forces in the final section, thereby reducing the deposition risk at the high TDS concentration end.

[0051] As a further improvement of the present invention, when running continuously for 24 hours, the decrease in transmembrane flux in the second and third stages does not exceed 15% of the initial flux.

[0052] The present invention, by adopting the above technical solution, has the following beneficial effects:

[0053] This invention achieves a multi-stage membrane system with distinct spatial zoning by synergistically controlling multiple parameters, including membrane type combination, inter-stage osmotic pressure difference, flow rate gradient, operating pH, periodic reflux disturbance, third-stage temperature rise, and membrane area distribution. Within this environment, the first stage possesses the highest solute migration driving force and maximum shear force, making it difficult for polyphenols and polysaccharides in the feed to form dense deposits on the membrane surface. The osmotic pressure difference, flow rate, and membrane area of ​​the second and third stages are simultaneously within a range suitable for the coexistence of weak adsorption and loosening, allowing the deposit layer to undergo a reversible process of formation, loosening, and redistribution in the middle stage. The 2-5°C temperature rise along the flow direction in the third stage further induces mild auto-oxidation and weak polymerization of polyphenols, maintaining a flexible and non-dense deposit structure. Periodic permeate reflux disturbance causes regular changes in ionic strength and interfacial bonding force in the middle stage, keeping the deposit layer in a dynamic equilibrium of structural loosening. The final stage operates in a low-driving-force and low-membrane-load region, maintaining stable permeability at the high TDS concentration end and avoiding secondary compaction of the deposit layer.

[0054] Through the interaction of the aforementioned multiple factors, the middle section becomes the core region of the reversible deposition structure. After formation, the deposition layer continuously undergoes processes such as hydrogen bond loosening, shear disturbance rupture, and migration flow carry-away, preventing it from developing into an irreversible hard fouling layer. This significantly reduces the rate of transmembrane flux decline over time, enabling the system to maintain good flux stability even under high concentration ratios and high recovery rates. This achieves high concentration of high-TDS plant extracts under low-pressure conditions, significantly reduces subsequent evaporation load, extends membrane module lifespan, improves system energy efficiency, and enhances continuous operation stability. Detailed Implementation

[0055] The following provides a detailed description of specific embodiments of the present invention. It should be understood that the specific embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit the scope of the invention.

[0056] Unless otherwise defined, all scientific and technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art.

[0057] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0058] Unless otherwise specified, all materials and reagents used in the following examples are commercially available.

[0059] The present invention will now be described in detail with reference to specific embodiments, which are intended to understand rather than limit the invention.

[0060] Example 1

[0061] This embodiment provides a membrane concentration process for plant extracts.

[0062] A compound plant extract (obtained by mixing and extracting green tea, rosemary, and mulberry leaves in a mass ratio of 3:1:1) produced by a plant extraction factory was selected as raw material. After coarse filtration and impurity removal pretreatment, a plant aqueous extract was obtained. The physicochemical properties of this plant aqueous extract are as follows: total dissolved solids (TDS) of the feed is 15 g / L, total polyphenol content is 300 mg / L (calculated as gallic acid), soluble polysaccharide content is 1.2 g / L, total soluble protein and small molecule peptide content is 0.5 g / L, feed conductivity is 4.5 mS / cm, and pH of the solution is 6.2.

[0063] The above-mentioned plant aqueous extract was placed in a raw material tank and then passed through a pretreatment system, sequentially through a 50μm security filter and a 5μm precision filter to remove residual suspended impurities, before being fed into a four-stage reverse osmosis multi-stage membrane concentration system connected in series. The configuration and operating conditions of the four membrane modules are as follows.

[0064] (1) The first section uses low-pressure, low-permeability spiral wound reverse osmosis membrane element A1, with each membrane element having an effective membrane area of ​​approximately 37m². 2 The design working pressure is 1.6–2.0 MPa, the operating temperature is 25–28℃, and the feed flow rate for this section is 12 m³ / s. 3 / h, concentrate outlet flow rate is 9.0m³ / h. 3 / h, with a production water flow rate of 3.0m³ / h. 3 / h. By adjusting the feed pressure, the steady-state transmembrane osmotic pressure difference in the first stage is maintained at 1.5 MPa, the linear velocity in the first channel is controlled at 2.5 m / s, and the overall transmembrane flux is approximately 22 L / (m²). 2 ·h).

[0065] (2) The second and third sections use the same type of medium-pressure, medium-permeability membrane element B1, with each membrane element having an effective membrane area of ​​approximately 34 m². 2 The second stage feed is supplied by the concentrate from the first stage. The operating pressure of the second stage is adjusted to 1.3 MPa by an inter-stage booster pump, so that the steady-state transmembrane osmotic pressure difference of the second stage is 1.25 MPa. The third stage feed is supplied by the concentrate from the second stage. The operating pressure of the third stage is adjusted to 1.3 MPa by an inter-stage booster pump, so that the steady-state transmembrane osmotic pressure difference of the third stage is 1.2 MPa.

[0066] The transmembrane osmotic pressure differences in the second and third sections are approximately 83% and 80% of those in the first section, respectively, and are both higher than those in the final section. The flow velocities in the second and third sections are controlled at 1.8 m / s and 1.8 m / s, respectively, both higher than the velocities in the preceding and following sections. The overall transmembrane flux remains stable at 18–20 L / (m²). 2 ·h) interval.

[0067] The third section is equipped with a plate heat exchanger along the flow direction, so that the feed temperature is 28°C at the front end of the third section and 31°C at the end of the third section, which means that a temperature rise of about 3°C ​​is formed along the flow direction.

[0068] (3) The final section uses a medium-pressure, high-permeability membrane element C1 with an effective membrane area of ​​approximately 30 m². 2 The feed is provided by the third-stage concentrate, and the operating pressure is controlled at 1.0 MPa, resulting in a steady-state transmembrane osmotic pressure differential of approximately 0.9 MPa at the end stage, which is about 25% of the steady-state transmembrane osmotic pressure differential of the third stage. The flow velocity in the end stage is controlled at 1.2 m / s, and the overall flux is approximately 15 L / (m²). 2 ·h).

[0069] pH disturbance control is as follows:

[0070] During system operation, diluted acid or diluted alkali is continuously added to the raw material tank and circulation loop through the acid-base dosing system to stabilize the overall operating pH of the system at 5.8–6.3. The diluted acid is a 0.1 wt% dilute sulfuric acid solution, prepared by mixing industrial-grade sulfuric acid and deionized water at a mass ratio of 1:1000. The diluted alkali is a 0.05 wt% sodium hydroxide solution, prepared by mixing industrial-grade sodium hydroxide and deionized water at a mass ratio of 1:2000.

[0071] Specifically, independent pH adjustment tanks are set up at the front ends of the second and third sections, respectively. An online pH meter is installed at the inlet of each tank. The diluted acid or diluted alkali solution is intermittently added to each pH adjustment tank via a metering pump, according to a fixed 60-minute cycle. The specific addition method is as follows: addition is performed every 5-10 minutes, with each addition lasting 20-40 seconds, so that the pH value of the second and third sections produces regular slight fluctuations within the cycle.

[0072] The online monitoring data is recorded as follows:

[0073] (1) In the second segment, within a 60-minute period, the pH value rose from 5.9 to 6.1 and then fell back to 5.9, with a fluctuation of approximately ±0.1.

[0074] (2) During the third segment, the pH value slowly fluctuated between 6.0 and 6.2 within a 60-minute period, with a fluctuation of approximately ±0.1.

[0075] (3) The fluctuation values ​​of both segments did not exceed the set range of ±0.2, and remained within the system operating pH range of 5.8 to 6.3.

[0076] Instantaneous transmembrane osmotic pressure differential is controlled as follows:

[0077] Between the first and second stages, by controlling the outlet pressure of the inter-stage booster pump, the inlet pressure of the second stage fluctuates to a certain extent relative to the outlet pressure of the first stage within a 30-minute cycle, so that the instantaneous transmembrane osmotic pressure difference between the first and second stages reaches about 12% of the steady-state transmembrane osmotic pressure difference of the first stage.

[0078] Online records show that after the periodic pressure fine-tuning began, the transmembrane osmotic pressure difference in the first stage increased by about 0.2 MPa for a short period before returning to a steady-state value, creating periodic instantaneous driving force disturbances between the outlet of the first stage and the inlet of the second stage. The steady-state transmembrane osmotic pressure difference between the second and third stages was consistently controlled within ±3%, while the steady-state transmembrane osmotic pressure difference between the third stage and the final stage decreased from 1.2 MPa to 0.9 MPa, a reduction of approximately 25%.

[0079] Permeate reflux control:

[0080] The system has a product water return pipeline at the outlet of the final product water stage. The final product water is divided into return flow and discharge flow in a fixed ratio through a three-way valve. The opening of the return branch of the three-way valve is set to 70%, that is, 70% of the final product water flows back to the raw material tank, and the remaining 30% is discharged as system product water.

[0081] To create a defined periodic backflow disturbance, a backflow regulating valve is installed on the backflow pipeline, and the valve opening is controlled by a PLC to perform fixed-cycle reciprocating adjustments. The adjustment method is as follows: with a complete cycle of 30 minutes, the opening of the backflow regulating valve is increased to 115% of the cycle average opening for the first 15 minutes of the cycle, and then decreased to 85% of the cycle average opening for the last 15 minutes. Under this fixed program, the backflow rate exhibits a regular rise and fall within the cycle.

[0082] Simultaneously, the feed control valves and reflux regulating valves of the second and third stages are linked by a PLC to achieve synchronized opening, causing the feed flow rate of both stages to undergo compensatory fine-tuning in sync with the rise and fall of the reflux flow rate. Through this defined control logic, the feed TDS of the second and third stages generates clear periodic fluctuations around their respective average values.

[0083] The specific monitoring data for Example 1 are as follows:

[0084] (1) Actual fluctuation range of TDS in the second stage feed

[0085] When the reflux regulating valve is at an average opening of 115%, the TDS of the second-stage feed rises to 23.8 g / L;

[0086] When the valve opening drops to 85% of the average opening, the TDS of the second-stage feed decreases to 22.1 g / L;

[0087] The average TDS of the second-stage feed was 22.9 g / L, with a periodic fluctuation of ±0.85 g / L (corresponding to a total fluctuation of 1.7 g / L).

[0088] (2) Actual fluctuation range of TDS in the third stage feed

[0089] During the high reflux stage (115% opening), the TDS of the third-stage feed increased to 37.8 g / L;

[0090] During the low reflux phase (85% opening), the concentration decreased to 35.2 g / L;

[0091] The average value of the third period was 36.5 g / L, and the period fluctuation range was ±1.3 g / L (corresponding to a total fluctuation of 2.6 g / L).

[0092] (3) The TDS fluctuation of the first and last feeds is minimal.

[0093] The TDS of the first feed varied within the range of 14.7 to 15.3 g / L, with a total fluctuation of 0.6 g / L;

[0094] The TDS of the final feed varied within the range of 78.9 to 79.5 g / L, with a total fluctuation of 0.6 g / L.

[0095] Ultraviolet absorbance photometric testing and control:

[0096] Online flow meters and online transmembrane flux calculation modules were installed on the concentrate outlet pipelines of the second and third sections, respectively. An online ultraviolet absorption detector was installed at the concentrate outlet of the third section to monitor the absorbance at a wavelength of 280 nm.

[0097] Under the conditions of this embodiment, after the system starts up and operates stably, the initial average transmembrane flux of the second and third stages is approximately 19 L / (m²). 2 (h) The initial value of the 280nm UV absorbance on the concentrate side of the third stage was 0.35. After continuous operation for about 6 hours, when the average transmembrane flux of the second and third stages decreased by about 10% relative to the initial operating flux, and the 280nm UV absorbance on the concentrate side of the third stage increased by 0.12, the flow velocity in the second and third stages was increased by about 10% from the original set value, i.e., from 1.8 m / s to about 2.0 m / s, and maintained for 20 minutes, by adjusting the frequency converter of the circulating pump. After the flow velocity was increased, the transmembrane flux gradually recovered to more than 95% of the initial value, and the UV absorbance dropped by about 0.07, indicating that partial peeling and redispersion of the deposited layer on the middle section membrane surface occurred.

[0098] Under the above conditions, the system operated continuously for 24 hours, and the overall transmembrane flux remained consistently between 15 and 22 L / (m²). 2 Within the range of ·h), the average transmembrane flux decrease in the second and third stages does not exceed 12% of the initial flux. At the end of 24 hours of operation, the overall system recovery rate is approximately 75%, the concentrate TDS increases from 15 g / L of the feed to approximately 80 g / L, and the product water TDS is below 500 mg / L. No significant irreversible fouling has occurred in the system, and it can continue to operate without chemical cleaning.

[0099] This embodiment demonstrates that when the TDS, polyphenol, polysaccharide, and protein content, conductivity, and pH of the plant extract raw material are all within a defined range, a combination of a first-stage low-pressure, low-permeability membrane, a second and third-stage medium-pressure, medium-permeability membrane, and a final-stage medium-pressure, high-permeability membrane, along with specific transmembrane osmotic pressure differential distribution, flow rate gradient, operating pH control, temperature gradient, and periodic permeate reflux disturbance, can achieve high concentration of high-TDS plant extracts under low operating pressure conditions. Simultaneously, it significantly slows down the fouling development of the mid-stage membrane module, keeping flux decline under 24-hour continuous operation conditions below 15%.

[0100] Example 2

[0101] This embodiment provides a membrane concentration process for plant extracts, which has the same system structure as that in Embodiment 1, except that the properties of the plant extract raw materials used are different.

[0102] A compound extract of medicinal and edible plants was selected as the raw material. The raw material formula included a mixed aqueous extract of Citrus aurantium, Glycyrrhiza uralensis, and Lycium barbarum in a mass ratio of 2:1:1. After coarse filtration, impurity removal, and pretreatment with a 1μm filter, the plant aqueous extract was obtained. The physicochemical properties of the raw material are as follows: total dissolved solids (TDS) of the feed is 12 g / L, total polyphenol content is 120 mg / L (calculated as gallic acid), soluble polysaccharide content is 0.3 g / L, total soluble protein and small molecule peptide content is 0.08 g / L, feed conductivity is 2.5 mS / cm, and pH of the solution is 5.4.

[0103] The above-mentioned plant aqueous extract was fed into a four-stage reverse osmosis membrane concentration system connected in series. The first stage used a low-pressure, low-permeability membrane element A1 with an effective membrane area of ​​37 m². 2 The second and third sections use the same medium-pressure, medium-permeability membrane element B1, with each element having an effective membrane area of ​​34 m². 2 The final stage uses a medium-pressure, high-permeability membrane element C1 with an effective membrane area of ​​30m². 2 .

[0104] pH adjustment tanks were installed at the front ends of the second and third stages, respectively. By intermittently adding 0.1wt% dilute sulfuric acid or 0.05wt% sodium hydroxide solution, the online pH values ​​of the second and third stages fluctuated within ±0.1 to ±0.2. Online records showed that the pH in the second stage changed slowly and periodically within the range of 5.8 to 6.0, and the pH in the third stage changed slowly and periodically within the range of 5.9 to 6.1.

[0105] In the first stage, the steady-state transmembrane osmotic pressure differential is maintained at 1.4 MPa by adjusting the feed pressure, the flow velocity in the channel is 2.2 m / s, and the overall transmembrane flux is approximately 20 L / (m²). 2 (h). The steady-state transmembrane osmotic pressure difference in the second section is set to 1.2 MPa, and the steady-state transmembrane osmotic pressure difference in the third section is 1.18 MPa, which are 86% and 84% of the first section, respectively, both greater than the steady-state transmembrane osmotic pressure difference of 0.92 MPa in the final section. The flow velocity in both the second and third sections is set to 1.7 m / s. In the third section, a temperature rise of approximately 2.5°C is generated along the flow direction through the plate heat exchanger, raising the operating temperature from 27°C at the beginning of the section to 29.5°C at the end.

[0106] The permeate return system returns 65% of the final permeate to the feed tank via a return pipeline, while the remaining 35% is discharged as purified water for the system. The return regulating valve on the return pipeline is controlled by a PLC program, with an opening adjustment cycle of 30 minutes: in the first 15 minutes of the cycle, the valve opening is increased to 115% of the average opening; in the last 15 minutes of the cycle, the valve opening is decreased to 85% of the average opening. Under this fixed program, the return flow exhibits a definite rise and fall within the cycle.

[0107] Because the feed control valves and reflux regulating valves of the second and third stages are controlled by a linkage logic, the feed flow rates of the second and third stages change slightly synchronously with the reflux flow rate, thus creating periodic TDS disturbances. Specific monitoring results are as follows:

[0108] (1) Actual fluctuation value of TDS of the second stage feed

[0109] During the stable operation phase, the conductivity data calculated over a 30-minute period are as follows:

[0110] When the reflux valve opening is at an average opening of 115%, the TDS of the second-stage feed rises to 23.8 g / L;

[0111] When the opening of the reflux valve drops to 85% of the average opening, the TDS of the second-stage feed drops to 22.0 g / L;

[0112] The average value over the period was 23.0 g / L, with a fluctuation range of ±0.9 g / L (corresponding to a total fluctuation of 1.8 g / L).

[0113] (2) Actual fluctuation value of TDS of the third stage feed

[0114] The online conductivity data for the third stage feed is as follows:

[0115] During the high reflux stage (valve opening 115%), TDS rose to 36.5 g / L;

[0116] During the low reflux phase (valve opening 85%), TDS decreased to 34.2 g / L;

[0117] The average value over the period was 35.4 g / L, with a fluctuation range of ±1.15 g / L (corresponding to a total fluctuation of approximately 2.3 g / L).

[0118] (3) The actual fluctuation values ​​of TDS of the feed in the first and last stages are extremely small.

[0119] The monitoring records are as follows:

[0120] The TDS of the first feed varies between 11.8 and 12.3 g / L, with a total range not exceeding 0.5 g / L;

[0121] The TDS of the final feed varies between 69.5 and 70.1 g / L, with a total range not exceeding 0.6 g / L.

[0122] The system reached stable operating conditions 5 hours after startup. The initial transmembrane flux of both the second and third stages remained at approximately 18 L / (m²). 2 (h). After approximately 8 hours of operation, when online monitoring showed a decrease of about 8% in transmembrane flux compared to the initial value, and an increase in the 280 nm UV absorbance on the concentrate side of the third stage from the initial 0.28 to 0.35, the flow rates in the second and third stages were increased by 10% and maintained for 15 minutes. After a short period of enhanced shearing, the transmembrane flux recovered to 94%–96% of the initial value, and the UV absorbance decreased to approximately 0.30.

[0123] Under the above operating mode, the system operated continuously for 24 hours. The decrease in transmembrane flux in the second and third stages did not exceed 13% of the initial flux, and no rapid membrane fouling or irreversible flux decay occurred. The TDS of the permeate was below 600 mg / L, while the TDS of the concentrate increased to approximately 72 g / L, and the overall system recovery rate remained at approximately 70%.

[0124] Comparative Example 1

[0125] This comparative example uses the exact same membrane concentration system, operating procedure, membrane configuration, transmembrane osmotic pressure differential platform setting, pH micro-fluctuation strategy, flow rate control strategy, temperature rise setting, and permeate reflux disturbance method as Example 1. The difference lies in the properties of the plant extract raw material used in this comparative example. The properties of the raw material are as follows:

[0126] Total dissolved solids (TDS): 28 g / L;

[0127] Total polyphenol content: 1500 mg / L;

[0128] Soluble polysaccharides: 6.5 g / L;

[0129] Soluble proteins and small peptides: 3.1 g / L;

[0130] Electrical conductivity: 12.2 mS / cm;

[0131] pH: 5.9.

[0132] The results show:

[0133] From the 4th hour onwards, the transmembrane flux in the second and third stages decreased significantly, with an average decrease of 35% over 24 hours;

[0134] From the 6th hour onwards, the 280nm UV absorbance on the concentrate side of the third section increased from 0.35 to 0.58, indicating significant densification of the sediment layer;

[0135] The terminal membrane element exhibits a rapid increase in differential pressure under high TDS conditions, requiring chemical cleaning after 24 hours.

[0136] Comparative Example 2

[0137] Except for the operating pH, the system structure and operating conditions of this comparative example are completely the same as those of Example 1.

[0138] The overall system pH is controlled between 4.2 and 4.4.

[0139] The online pH fluctuations in the second and third stages were limited to the range of 4.2 to 4.5, with a fluctuation range of ±0.1.

[0140] All other operating conditions (including transmembrane pressure differential plateau, flow rate gradient, temperature rise, reflux cycle disturbance, etc.) are the same as in Example 1.

[0141] Results after 24 hours of operation showed that transmembrane flux decreased rapidly by 20% in the first 6 hours, with a total decrease of 42% over 24 hours.

[0142] The absorbance at 280 nm on the concentrate side of the third section continued to increase to 0.63, indicating that the deposition layer was in a hardened state.

[0143] The low pH level causes polyphenol molecules to hardly dissociate, significantly enhancing the stability of the deposition layer. Even with perturbations by increasing flow rate, it is difficult to restore the flux.

[0144] Comparative Example 3

[0145] This comparative example uses the same raw materials and the same membrane system structure as Example 1, and all operating strategies are kept consistent, except that the transmembrane flux is set as follows:

[0146] First segment: 32L / (m 2 ·h)

[0147] Second and third sections: 28-30 L / (m 2 ·h)

[0148] Final stage: 26L / (m) 2 ·h)

[0149] To achieve the aforementioned high throughput, the overall feed pressure was increased by 20%.

[0150] The test results after 24 hours showed that: starting from the second hour, the second and third deposition layers thickened rapidly; the transmembrane flux decreased by 50% within 24 hours; the online UV absorbance increased sharply from the initial 0.35 to 0.70; and an irreversible increase in differential pressure occurred at the 10th hour.

[0151] Comparative Example 4

[0152] The raw materials, membrane system structure, pH micro-perturbation, transmembrane pressure plateau, temperature rise, and reflux cycle perturbation in this comparative example are the same as in Example 1. The difference lies in the flow channel velocity settings as follows:

[0153] First section flow velocity: maintain 2.5 m / s;

[0154] Second section flow velocity: 1.0 m / s;

[0155] The flow velocity in the third section is 1.1 m / s.

[0156] Final flow velocity: 1.2 m / s;

[0157] The results showed that the second and third sections had significant compaction of the deposited layer after 4 hours due to insufficient shearing; the transmembrane flux decreased by 38% within 24 hours; the UV absorbance increased to 0.55, and the deposited layer structure could not be restored by perturbation operations that increased the flow rate; the third section of the deposited layer showed obvious hardening, and the concentration polarization was significantly aggravated.

[0158] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A membrane concentration process for plant extracts, characterized in that, The plant extract is sequentially sent to a multi-stage membrane concentration system composed of at least four reverse osmosis membrane assemblies arranged in series, wherein the first stage adopts a low-pressure and low-permeability membrane, the second and third stages adopt a medium-pressure and medium-permeability membrane, and the last stage adopts a medium-pressure and high-permeability membrane. The plant extract is sequentially sent to a multi-stage membrane concentration system composed of at least four reverse osmosis membrane assemblies arranged in series, wherein the first stage adopts a low-pressure and low-permeability membrane, the second and third stages adopt a medium-pressure and medium-permeability membrane, and the last stage adopts a medium-pressure and high-permeability membrane. The flow channel flow rate of the second and third stages is 1.5-2.0 m / s. The pH control of the multi-stage membrane concentration system is 5.5-7.0, and the transmembrane flux of the whole system is controlled at 10-25 L / (m 2 ·h); By adjusting the outlet pressure of the booster pump between stages, the steady-state transmembrane osmotic pressure difference along the flow direction of each stage satisfies the following relationship:

2. A process for membrane concentration of plant extract according to claim 1, wherein, The transmembrane osmotic pressure difference of the first stage is the maximum, the transmembrane osmotic pressure difference of the second and third stages is controlled at 80-95% of the transmembrane osmotic pressure difference of the first stage, and is higher than the transmembrane osmotic pressure difference of the last stage, thereby forming a transmembrane osmotic pressure difference distribution with the first stage as the peak, the second and third stages forming a relatively stable platform, and the last stage significantly decreasing. The pH control of the multi-stage membrane concentration system is 5.8-6.5, and the pH of the second and third stages within this range produces a fluctuation of ±0.1-±0.

3.

3. A process for concentrating a plant extract according to claim 1, wherein, The change amplitude of the instantaneous transmembrane osmotic pressure difference of the first stage to the second stage is more than 10% of the steady-state transmembrane osmotic pressure difference of the first stage; the steady-state transmembrane osmotic pressure difference difference between the second stage and the third stage is not more than ±5%; and the steady-state transmembrane osmotic pressure difference between the third stage and the last stage is reduced to 10-30% of the steady-state transmembrane osmotic pressure difference of the third stage.

4. The process for membrane concentration of plant extract according to claim 1, wherein, The water production flow rate is adjusted within ±10%-±20% of the average value for a period of 20-60 min, and the feed pressure and / or feed flow rate of the second and third stages are adjusted correspondingly within the period, so that the feed TDS of the second and third stages fluctuates within 1.0-3.0 g / L relative to the average value of the period, and the feed TDS of the first and last stages fluctuates within the period by less than 1.0 g / L.

5. The process for membrane concentration of plant extract according to claim 1, wherein, The third stage gradually forms a temperature rise of 2-5°C along the flow direction, and the operating temperature of the third stage is maintained at 28-33°C, which is higher than the operating temperature of the first and second stages.

6. The process for membrane concentration of plant extract according to claim 1, wherein, The second and third stages of reverse osmosis membrane assemblies are detected online for transmembrane flux and 280 nm ultraviolet absorbance on the concentrated water side, when the transmembrane flux decreases by 5-15% compared to the initial operating flux and the ultraviolet absorbance increases by 0.05-0.20 compared to the initial value, the flow channel flow rate of the second and third stages is increased by 5-20% relative to the set flow rate, and maintained for 10-60 min.

7. The process for membrane concentration of plant extract according to claim 1, wherein, The second and third stages adopt the same membrane type, and the effective membrane area of the membrane assemblies thereof is smaller than the effective membrane area of the first stage and larger than the effective membrane area of the last stage.

8. The process for membrane concentration of plant extract according to claim 1, wherein, The transmembrane flux of the second and third stages decreases by no more than 15% of the initial flux within 24 h of continuous operation.

9. The process for membrane concentration of plant extract according to claim 1, wherein, ​