Extraction box feeding accurate control processing method and system based on automatic anti-blocking

Abstract

The invention provides an extraction box feeding accurate control processing method and system based on automatic anti-blocking, and relates to the technical field of intelligent control, and the method comprises the following steps: obtaining oil phase distribution information and target concentration, identifying high and low concentration areas in a circulation loop, carrying out differential shearing processing, respectively guiding fine particle oil phase fluid and coarse particle oil phase fluid to filter units with different pore diameters, and filtering the fine particle oil phase fluid and the coarse particle oil phase fluid. And the shearing strength and the backwashing time sequence are adaptively adjusted based on the extraction efficiency and the anti-blocking performance state, so that the purposes of efficient separation in the extraction process, filter blocking prevention and accurate feeding control are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to intelligent control technology, and in particular to an extraction tank feed precise control processing method and system based on automatic anti-blocking. BACKGROUND

[0002] Extraction technology is a widely used separation and purification technology in the fields of chemical industry, metallurgy, environmental protection, etc. It realizes separation by contacting a mixture containing a target component with an extractant and utilizing the solubility difference of components in different phases. The extraction tank, as the core equipment of the extraction process, its feeding control precision directly affects the extraction efficiency and product quality.

[0003] The existing extraction system lacks accurate identification capability for oil phase distribution area, and cannot implement differential treatment for different concentration areas, resulting in low extraction efficiency and high energy consumption. In the high concentration area, due to improper control of droplet size, local aggregation is easily formed, affecting the mass transfer efficiency; while in the low concentration area, excessive shearing will cause energy waste. SUMMARY

[0004] The embodiments of the present application provide an extraction tank feed precise control processing method and system based on automatic anti-blocking, which can solve the problems in the prior art.

[0005] In a first aspect, the embodiments of the present application provide an extraction tank feed precise control processing method based on automatic anti-blocking, comprising:

[0006] Obtaining oil phase distribution information of the to-be-processed feed fluid and target component concentration at the outlet of the extraction tank; collecting time sequence data of the particle size of the oil phase droplets, and identifying the droplet coalescence and acceleration state;

[0007] Identifying the high concentration area and the low concentration area in the circulation loop based on the oil phase distribution information, applying a first shear intensity treatment to the high concentration area to obtain fine particle oil phase fluid, applying a second shear intensity treatment lower than the first shear intensity to the low concentration area to obtain coarse particle oil phase fluid, and enhancing the shear intensity of the high concentration area in advance when the droplet coalescence and acceleration state is identified;

[0008] Directing the fine particle oil phase fluid to a first filtering unit, directing the coarse particle oil phase fluid to a second filtering unit with a pore size larger than the first filtering unit, and obtaining information of each stage of the retentate;

[0009] Determining the extraction efficiency state based on the target component concentration, determining the anti-blocking performance state based on the flux recovery information; and according to the establishment of the extraction efficiency state and the anti-blocking performance state, adjusting the shear intensity, the backwashing time sequence or the combination of the two, to form an adaptive closed loop.

[0010] The step of collecting the particle size time sequence data of the oil phase droplets and identifying the droplet coalescence acceleration state comprises:

[0011] The step of synchronously collecting the particle size distribution data of the oil phase droplets at multiple spatial positions, constructing a particle size-space distribution field, and extracting a spatial propagation speed and a local aggregation degree index representing the droplet coalescence behavior based on the evolution track of the particle size-space distribution field in the time dimension comprises:

[0012] When the product of the spatial propagation speed and the local aggregation degree index exceeds a critical coalescence threshold, it is determined that the droplet coalescence acceleration state is established, and the spatial position information of the coalescence acceleration is coupled with the oil phase distribution information for coupled analysis, so as to locate the boundary of the high-concentration region requiring early enhancement of the shear strength.

[0013] The step of identifying the high-concentration region and the low-concentration region in the circulation loop based on the oil phase distribution information, applying a first shear strength treatment to the high-concentration region to obtain a fine-grained oil phase fluid, applying a second shear strength treatment lower than the first shear strength to the low-concentration region to obtain a coarse-grained oil phase fluid, and early enhancing the shear strength of the high-concentration region when the droplet coalescence acceleration state is identified comprises:

[0014] The step of extracting the oil phase concentration spatial gradient distribution characteristics in the circulation loop based on the oil phase distribution information, identifying the concentration gradient mutation position, taking the concentration value corresponding to the concentration gradient mutation position as a concentration demarcation threshold, and dividing the circulation loop into a high-concentration region and a low-concentration region based on the concentration demarcation threshold comprises:

[0015] The step of calculating the coupling parameters of the oil phase concentration mean value and the particle size distribution variance in the high-concentration region and the low-concentration region respectively, determining the numerical values of the first shear strength and the second shear strength based on the coupling parameters, and making the particle size distribution concentration degree of the fine-grained oil phase fluid obtained after the first shear strength treatment higher than the particle size distribution concentration degree of the coarse-grained oil phase fluid obtained after the second shear strength treatment comprises:

[0016] When the droplet coalescence acceleration state is identified, the coalescence occurrence position in the high-concentration region is located based on the spatial position information corresponding to the droplet coalescence acceleration state, and an enhanced shear strength higher than the first shear strength is applied to the coalescence occurrence position and the adjacent region thereof.

[0017] The step of guiding the fine-grained oil phase fluid to a first filtration unit, guiding the coarse-grained oil phase fluid to a second filtration unit with a larger pore diameter than the first filtration unit, and obtaining the information of each level of the retained material, and performing a backwashing operation on each filtration unit to obtain the flux recovery information comprises:

[0018] The pore size matching range of the first filter unit is determined based on the particle size distribution characteristics of the fine-particle oil phase fluid, and the pore size matching range of the second filter unit is determined based on the particle size distribution characteristics of the coarse-particle oil phase fluid, so that the retention efficiency of the fine-particle oil phase fluid in the first filter unit and the retention efficiency of the coarse-particle oil phase fluid in the second filter unit satisfy a preset efficiency balance relationship.

[0019] During the filtration process, the rate of change of transmembrane pressure difference between the first filtration unit and the second filtration unit is monitored in real time, and the accumulation rate of retentate in each filtration unit is obtained as retentate information at each stage.

[0020] Based on the cumulative rate of entrapment in the information of entrapment at each level, the clogging risk level of each filter unit is predicted. The backwashing triggering time and backwashing intensity parameters of the first filter unit and the second filter unit are set differently according to the clogging risk level. After the backwashing operation is performed, the flux recovery information is obtained by comparing the recovery ratio of transmembrane pressure difference before and after backwashing.

[0021] The steps of predicting the clogging risk level of each filter unit and setting the backwash triggering timing and backwash intensity parameters differently based on the clogging risk level include:

[0022] A time series analysis is performed on the accumulation rate of the retained material, and the first and second derivatives of the accumulation rate of the retained material are calculated. When the first derivative is continuously positive and the second derivative is greater than the acceleration threshold, the filter unit is determined to be in an accelerated clogging state.

[0023] Establish a clogging risk level classification rule, and classify the clogging risk into multiple levels based on the comprehensive score of the numerical range of the accumulating rate of the intercepted material and the current value of the transmembrane pressure difference; establish a differentiated mapping relationship between the clogging risk level and the backwashing parameters, so that the backwashing trigger time interval shortens as the clogging risk level increases, and the backwashing intensity parameter increases as the clogging risk level increases.

[0024] Record the transmembrane pressure differential recovery ratio after each backwashing operation. When the recovery ratio of multiple consecutive backwashes shows a downward trend, increase the clogging risk level of the filter unit and shorten the backwashing trigger time interval accordingly, forming a closed loop of dynamic assessment of clogging risk and adaptive adjustment of backwashing parameters.

[0025] The steps of determining the extraction efficiency state based on the target component concentration and the anti-clogging performance state based on the flux recovery information, and adjusting the shear intensity, backwashing sequence, or a combination of both according to the establishment of the extraction efficiency state and the anti-clogging performance state to form an adaptive closed loop, include:

[0026] Based on the target component concentration, an extraction efficiency evaluation index is constructed that includes the degree of concentration attainment and concentration time-series stability, and the extraction efficiency status is determined according to the extraction efficiency evaluation index; based on the flux recovery information, an anti-clogging performance evaluation index is constructed that includes the degree of recovery sufficiency and the ability to sustain recovery, and the anti-clogging performance status is determined according to the anti-clogging performance evaluation index.

[0027] A four-quadrant decision matrix is ​​constructed to represent the extraction efficiency state and the anti-clogging performance state. When both states are true, the current parameters are maintained. When only the extraction efficiency state is true, the backwashing sequence is adjusted based on the anti-clogging performance evaluation index. When only the anti-clogging performance state is true, the shear intensity is adjusted based on the extraction efficiency evaluation index. When neither state is true, the adjustment weights of the shear intensity and the backwashing sequence are determined based on the degree of deviation between the extraction efficiency evaluation index and the anti-clogging performance evaluation index.

[0028] The adjusted shear strength and backwashing timing are applied to the shearing treatment stage and the backwashing operation stage respectively, forming an adaptive closed loop.

[0029] The steps for constructing a four-quadrant decision matrix for extraction efficiency state and anti-blocking performance state include:

[0030] A two-dimensional decision space is established for the extraction efficiency state and the anti-blocking performance state. The combination of the two states being true and false is divided into four quadrant regions, and each quadrant region corresponds to a set of preset adjustment strategies.

[0031] When both states are true, the system is positioned in the first quadrant and the parameter maintenance strategy is executed.

[0032] When only the extraction efficiency state is established, the second quadrant is located, the recovery sustainability decay coefficient in the anti-clogging performance evaluation index is extracted, and the amount of shortening of the backwashing sequence is calculated based on the decay coefficient.

[0033] When only the anti-blocking performance state is established, it is located in the third quadrant. The fluctuation range of concentration time stability in the extraction efficiency evaluation index is extracted. When the fluctuation range exceeds the stability threshold, the first shear strength of the high concentration region is increased. When the fluctuation range does not exceed the stability threshold but the concentration compliance level is lower than the target value, the second shear strength of the low concentration region is reduced.

[0034] When neither of the two states is true, the system is positioned in the fourth quadrant. The deviation of the extraction efficiency evaluation index and the deviation of the anti-clogging performance evaluation index are calculated. The ratio of the two deviations is used as the adjustment weight allocation coefficient. Based on the adjustment weight allocation coefficient, the amount of shear strength enhancement and the amount of backwashing time shortening are determined simultaneously.

[0035] A second aspect of the present invention provides a precision control and processing system for extraction tank feeding based on automatic anti-clogging, comprising:

[0036] The first unit is used to acquire information on the oil phase distribution of the feed fluid to be treated and the target component concentration at the outlet of the extraction box; to collect time-series data on the particle size of oil phase droplets and to identify the droplet aggregation and acceleration state;

[0037] The second unit is used to identify high-concentration and low-concentration regions in the circulation loop based on the oil phase distribution information, apply a first shear strength treatment to the high-concentration region to obtain fine-particle oil phase fluid, apply a second shear strength treatment lower than the first shear strength to the low-concentration region to obtain coarse-particle oil phase fluid, and enhance the shear strength of the high-concentration region in advance when the droplet coalescence acceleration state is detected.

[0038] The third unit is used to guide the fine-particle oil phase fluid to the first filtration unit and the coarse-particle oil phase fluid to the second filtration unit with a pore size larger than that of the first filtration unit, thereby obtaining information on the intercepted materials at each stage; and to perform backwashing operations on each filtration unit to obtain flux recovery information.

[0039] The fourth unit is used to determine the extraction efficiency status based on the target component concentration and the anti-clogging performance status based on the flux recovery information; according to the establishment of the extraction efficiency status and the anti-clogging performance status, the shear intensity, backwashing sequence or a combination of both are adjusted accordingly to form an adaptive closed loop.

[0040] A third aspect of the present invention,

[0041] An electronic device is provided, comprising:

[0042] processor;

[0043] Memory used to store processor-executable instructions;

[0044] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0045] Fourth aspect of the embodiments of the present invention,

[0046] A computer-readable storage medium is provided, having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0047] This invention achieves directional diversion and precise control of the feed fluid by identifying oil phase distribution information and applying shearing treatments of different intensities to high and low concentration regions, significantly improving the selectivity and efficiency of the extraction process and making the separation effect of the target components more ideal. Attached Figure Description

[0048] Figure 1This is a schematic flowchart of the extraction box feeding precision control processing method based on automatic anti-clogging in an embodiment of the present invention;

[0049] Figure 2 This is a flowchart of a four-quadrant adaptive closed-loop control system based on dual-state evaluation of extraction efficiency and anti-clogging performance. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0052] Figure 1 This is a schematic flowchart of the extraction box feeding precision control method based on automatic anti-clogging in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0053] Acquire the oil phase distribution information of the feed fluid to be treated and the target component concentration at the outlet of the extraction tank; collect the time series data of oil phase droplet size and identify the droplet aggregation acceleration state;

[0054] Based on the oil phase distribution information, high-concentration and low-concentration regions in the circulation loop are identified. A first shear strength is applied to the high-concentration region to obtain fine-particle oil phase fluid, and a second shear strength lower than the first shear strength is applied to the low-concentration region to obtain coarse-particle oil phase fluid. Furthermore, the shear strength of the high-concentration region is enhanced in advance when the droplet coalescence acceleration state is detected.

[0055] The fine-particle oil phase fluid is directed to the first filtration unit, and the coarse-particle oil phase fluid is directed to the second filtration unit with a larger pore size than the first filtration unit to obtain information on the retained substances at each stage; backwashing is performed on each filtration unit to obtain flux recovery information;

[0056] The extraction efficiency state is determined based on the target component concentration, and the anti-clogging performance state is determined based on the flux recovery information. Depending on whether the extraction efficiency state and the anti-clogging performance state are met, the shear intensity, backwashing sequence, or a combination of both are adjusted accordingly to form an adaptive closed loop.

[0057] In one optional implementation, the step of collecting time-series data on the particle size of oil phase droplets and identifying the accelerated state of droplet coalescence includes:

[0058] Oil droplet size distribution data are collected simultaneously at multiple spatial locations to construct a size-spatial distribution field. Based on the evolution trajectory of the size-spatial distribution field in the time dimension, spatial propagation velocity and local aggregation degree indices characterizing droplet aggregation behavior are extracted.

[0059] When the product of the spatial propagation velocity and the local aggregation index exceeds the critical coalescence threshold, the droplet coalescence acceleration state is determined to be established. The spatial location information of the coalescence acceleration is coupled with the oil phase distribution information for analysis to locate the boundary of the high-concentration region that requires pre-enhanced shear strength.

[0060] For example, multiple online oil phase concentration sensors are installed at key locations in the circulation loop. Using capacitive or optical measurement principles, local oil phase volume fraction data is collected every 500 milliseconds and transmitted in real time to the central control unit through a data acquisition system, forming a dataset of the spatial distribution of oil phase concentration covering the entire circulation loop.

[0061] When simultaneously acquiring particle size distribution data of oil droplets at multiple spatial locations, a multi-channel image acquisition device can be configured. This device includes a combination system of a high-speed camera and a microscope, which monitors more than five key spatial locations in the flow field in real time. The image acquisition frequency is set to 200 frames per second, which can effectively capture the dynamic changes of oil droplets at the microscopic scale. The acquired image data is preprocessed using an adaptive edge detection algorithm to extract droplet contour information. A roundness screening mechanism is used to filter out irregular droplets, with a roundness threshold set to 0.85, meaning that droplets with a roundness below 0.85 are judged as non-standard samples and discarded. For droplets that pass the screening, the particle size is calculated using the equivalent circle diameter method, that is, the irregular droplets are equivalent to circles of the same area, and their diameter is used as the characteristic size of the droplet. After the acquisition is completed, a particle size distribution dataset of each spatial location over time is generated.

[0062] The multi-point particle size distribution data obtained above were extended into a continuous distribution field using spatial interpolation techniques. Cubic spline interpolation was used to numerically fill the regions between discrete sampling points, ensuring the smoothness of the distribution field. To improve interpolation accuracy, the weight of each sampling point was adaptively adjusted, assigning different weight coefficients based on the gradient change in droplet concentration. Specifically, regions with a concentration gradient greater than 5% / mm were identified as high-gradient regions, and the weight coefficient for sampling points in these regions was set to 1.5; regions with a concentration gradient not exceeding 5% / mm were identified as low-gradient regions, and the weight coefficient for sampling points in these regions was set to 1.0. The constructed particle size-spatial distribution field contains three dimensions: spatial coordinates, particle size, and corresponding frequency distribution. This distribution field is stored in a gridded format with a grid precision of 0.5 mm. Each grid point contains histogram data for particle size distribution within the range of 5-500 micrometers, and the histogram bar width is set to 5 micrometers.

[0063] To extract the spatial propagation velocity characterizing droplet coalescence behavior, dynamic feature analysis was performed on the time series data of the particle size-spatial distribution field. The displacement vector of the distribution field between adjacent time points was calculated, and cross-correlation analysis was used to identify the overall spatial migration trend of the distribution characteristics. Specifically, for two distribution field images spaced 0.1 seconds apart, the cross-correlation function was calculated using a normalized cross-correlation (NCC) algorithm. Within a search window of ±2 mm, the displacement corresponding to the position of maximum cross-correlation was determined, and the local propagation velocity was obtained by dividing the displacement by the time interval. The implementation process of the normalized cross-correlation algorithm includes: first, zero-mean processing of the two images to eliminate the influence of brightness differences; then, calculating the cross-correlation coefficient for all displacements within the search window, and selecting the displacement corresponding to the maximum cross-correlation coefficient as the actual displacement. In practical applications, when oil droplets flow in a pipeline, the measured spatial propagation velocity typically ranges from 0.05 to 0.5 m / s, with velocity fluctuations not exceeding 10% under stable flow conditions, while velocity fluctuations can reach over 30% under coalescence acceleration conditions.

[0064] The local aggregation index is calculated based on the spatial aggregation characteristics of droplets in the distribution field, using nearest neighbor analysis to quantify the degree of droplet aggregation at the microscale. Within each analysis grid, the ratio of the average nearest neighbor distance to the droplet size is calculated; a smaller ratio indicates a higher degree of droplet aggregation. The specific calculation steps are as follows: first, the center coordinates of all droplets within the grid are identified; then, the Euclidean distance between each droplet and the center of its nearest neighbor droplet is calculated; the average of the nearest neighbor distances of all droplets within the grid is taken; finally, this average is divided by the average diameter of the droplets within the grid to obtain the dimensionless aggregation index. To eliminate random interference, an equal-weighted moving average is applied to data from five consecutive time frames (corresponding to a 0.5-second time window, i.e., 0.025 seconds × 5 consecutive seconds at a sampling frequency of 200 frames / second). In practical applications, the local aggregation index under stable dispersion generally remains in the range of 0.2-0.4, while when droplets begin to show a tendency to coalesce, the index rapidly drops below 0.15.

[0065] The product of spatial propagation velocity and local aggregation index is defined as the coalescence potential energy factor, which comprehensively reflects the dynamic behavior and spatial structure characteristics of droplets. The formula for calculating the coalescence potential energy factor is: Coalescence Potential Energy Factor = Spatial Propagation Velocity (m / s) × Local Aggregation Index (dimensionless). Extensive experimental verification shows that when the coalescence potential energy factor exceeds a critical threshold of 0.03, droplet coalescence exhibits an accelerated trend. This critical threshold was determined based on statistical analysis of experimental data from over 50 sets of oil-water two-phase flow experiments under different operating conditions. When the factor value exceeds 0.03, the probability of observing accelerated coalescence within the following 2-5 seconds reaches over 92%. Taking oil-water two-phase flow as an example, in the flow mode of continuous water phase and dispersed oil phase, when oil droplets with a volume fraction of 30% flow through the sudden expansion region of the pipeline, the coalescence potential energy factor is maintained in the range of 0.015-0.025 in the initial stage of flow, and the system is in a stable state; however, when the factor value exceeds 0.035, obvious droplet coalescence can be observed within 2-5 seconds, and the average particle size growth rate increases sharply from the initial 2 micrometers / second to more than 12 micrometers / second.

[0066] The identified aggregation acceleration points are overlaid onto the oil phase concentration distribution map to form an aggregation risk heatmap. This heatmap uses a pseudo-color representation method, mapping the aggregation potential factor values ​​to a red-yellow-green spectrum. Red areas with a factor value > 0.04 represent high risk, yellow areas with a factor value between 0.03 and 0.04 represent medium risk, and green areas with a factor value < 0.03 represent low risk. High-risk points are grouped using the DBSCAN density clustering algorithm to determine the core range and boundaries of risk areas. The key parameters of the DBSCAN algorithm are set as follows: the neighborhood radius ε is set to 3 mm, and the minimum number of points MinPts is set to 5. That is, when a point has at least 5 high-risk points within its 3 mm neighborhood, that point is assigned to the same cluster. Boundary extraction uses the contour line method. Based on the clustering results, the spatial gradient of the oil phase volume fraction is calculated, and the closed curve with the largest absolute value of the gradient change rate is selected as the boundary of the high-concentration region. In practice, the Sobel operator is used to calculate the gradient field. After Gaussian smoothing of the gradient magnitude, contour lines are extracted. Continuous closed curves with gradient magnitudes greater than a threshold are selected as the final boundary, typically set to 1.5 times the average gradient magnitude. This boundary information is fed back to the shear strength control module to guide the spatial positioning of subsequent shear strength enhancement operations.

[0067] This invention can accurately capture the spatiotemporal evolution characteristics of droplet coalescence behavior, providing early warning of potential clogging risks and enabling precise location of high-concentration region boundaries. This dynamic monitoring mechanism significantly improves the response speed to coalescence phenomena, reduces the probability of filter unit clogging caused by droplet coalescence, and ensures the continuous and stable operation of the extraction process.

[0068] In one optional embodiment, the steps of identifying high-concentration and low-concentration regions in the circulation loop based on the oil phase distribution information, applying a first shear strength treatment to the high-concentration region to obtain a fine-particle oil phase fluid, applying a second shear strength treatment lower than the first shear strength to the low-concentration region to obtain a coarse-particle oil phase fluid, and preemptively increasing the shear strength of the high-concentration region when the droplet coalescence acceleration state is detected include:

[0069] Based on the oil phase distribution information, the spatial gradient distribution characteristics of oil phase concentration in the circulation loop are extracted, the locations of abrupt changes in concentration gradient are identified, the concentration values ​​corresponding to the locations of abrupt changes in concentration gradient are used as concentration boundary thresholds, and the circulation loop is divided into high-concentration regions and low-concentration regions using the concentration boundary thresholds.

[0070] For the high-concentration region and the low-concentration region respectively, the coupling parameters of the mean oil phase concentration and the particle size distribution variance in the region are calculated. Based on the coupling parameters, the values ​​of the first shear strength and the second shear strength are determined, so that the particle size distribution concentration of the fine-particle oil phase fluid obtained after the first shear strength treatment is higher than the particle size distribution concentration of the coarse-particle oil phase fluid obtained after the second shear strength treatment.

[0071] When the accelerated droplet coalescence state is detected, the location of the coalescence occurrence is located in the high-concentration region based on the spatial location information corresponding to the accelerated droplet coalescence state, and an enhanced shear strength higher than the first shear strength is applied to the location of the coalescence occurrence and its adjacent region.

[0072] For example, oil phase distribution information in the circulation loop is acquired by installing multiple online concentration sensors at key locations in the loop. Each sensor records local oil phase concentration data every 500 milliseconds and transmits the data to the central control system. Oil phase concentration data collected over 24 hours is used to construct a spatiotemporal distribution matrix of the oil phase concentration. This matrix contains information on the changing oil phase concentration at various locations in the circulation loop over time, providing a data foundation for subsequent processing.

[0073] The sliding window method was used to extract the spatial gradient distribution characteristics of oil phase concentration. Spatial sampling points were set at 5-cm intervals on the physical model of the loop, and the concentration difference between adjacent sampling points was calculated to obtain discrete gradient values. For a certain loop segment, the measured oil phase concentrations from the start to the end were 3.2%, 3.5%, 3.6%, 5.8%, 6.2%, 6.3%, and 6.1%, respectively. The calculated gradient sequence was 0.3% / 5cm, 0.1% / 5cm, 2.2% / 5cm, 0.4% / 5cm, 0.1% / 5cm, and -0.2% / 5cm. By applying standard deviation analysis to the gradient sequence, the gradient abrupt change threshold was determined to be twice the gradient mean. In this case, the gradient mean was approximately 0.48% / 5cm, and the standard deviation was 0.81% / 5cm; therefore, the gradient abrupt change threshold was set to 0.96% / 5cm. Based on this threshold, a gradient abrupt change was identified between the third and fourth sampling points, corresponding to concentration values ​​of 3.6% and 5.8%. The average of the two values, 4.7%, was taken as the concentration threshold, and the circulation loop was divided into a high-concentration region and a low-concentration region. The high-concentration region was defined as the region with an oil phase concentration ≥ 4.7%, and the low-concentration region was defined as the region with an oil phase concentration < 4.7%.

[0074] For the defined high-concentration and low-concentration regions, the coupling parameter for particle size distribution variance is calculated as the product of the concentration mean and the standard deviation of particle size distribution. In this example, the mean oil phase concentration in the high-concentration region is 6.1%, and particle size distribution data was obtained through three particle size detection points set up in this region. A laser diffractometer was used for particle size detection, with a measurement range of 0.1-1000 micrometers. Measurements were taken continuously for 10 minutes at each detection point, with a sampling interval of 5 seconds, resulting in 120 sets of particle size distribution data. The standard deviation of particle size distribution in the high-concentration region was statistically determined to be 25 micrometers. The mean oil phase concentration in the low-concentration region is 3.4%, and particle size distribution data was obtained using the same measurement method, with a statistically determined standard deviation of 18 micrometers. According to the coupling parameter calculation formula, the coupling parameter for the high-concentration region = 6.1% × 25 micrometers = 152.5%·micrometer, and the coupling parameter for the low-concentration region = 3.4% × 18 micrometers = 61.2%·micrometer.

[0075] Based on the calculated coupling parameters, the first and second shear strengths were determined. The shear strength is directly proportional to the coupling parameters, and the conversion coefficient was determined through preliminary experimental calibration. In the same circulating loop system, oil-water mixtures with coupling parameters of 50-200%·micrometers were treated with different shear strengths ranging from 20-100 N / m². The particle size distribution concentration after treatment was measured, and the optimal conversion coefficient of 0.5 N / m² per (%·micrometer) was obtained through least squares fitting. For different oil-water systems, this coefficient can be adjusted within the range of 0.3-0.7, and the specific value was determined through small-batch experiments. For high-concentration areas, the first shear strength was applied as 152.5 × 0.5 = 76.25 N / m², and for low-concentration areas, the second shear strength was applied as 61.2 × 0.5 = 30.6 N / m².

[0076] Experimental measurements show that the average particle size of the fine-particle oil phase fluid obtained after the first shear strength treatment is 15 micrometers, the standard deviation of the particle size is 8 micrometers, and the particle size distribution concentration is 0.125, where the particle size distribution concentration is expressed as 1 / standard deviation. The average particle size of the coarse-particle oil phase fluid obtained after the second shear strength treatment is 35 micrometers, the standard deviation of the particle size is 15 micrometers, and the particle size distribution concentration is 0.067. The particle size distribution concentration of the fine-particle oil phase fluid is higher than that of the coarse-particle oil phase fluid, which meets the expected design target.

[0077] The identification of the accelerated droplet coalescence state is achieved by continuously monitoring the rate of change in the oil phase particle size distribution. The rate of change in particle size distribution is defined as the change in the standard deviation of particle size distribution within adjacent measurement cycles divided by the time interval, expressed as % / minute. Under normal conditions, the rate of change in particle size distribution remains within ±2% / minute. When the rate of change in particle size distribution in a certain area exceeds 5% / minute for three consecutive measurement cycles, the system determines that the area has entered the accelerated droplet coalescence state, with each measurement cycle being 1 minute. Within each measurement cycle, the particle size detection device collects 120 data points at a sampling interval of 500 milliseconds, calculates the standard deviation of particle size distribution for that cycle, and compares it with the previous cycle to obtain the rate of change.

[0078] In actual operation, after 120 minutes of operation in the circulation loop, the particle size distribution change rate at detection point P37 in the high-concentration area showed abnormal fluctuations. The change rate was measured at 7.2% / min at the 118th minute, 6.8% / min at the 119th minute, and 7.5% / min at the 120th minute, exceeding the judgment threshold of 5% / min for three consecutive measurement cycles. The system confirmed that the P37 location had entered a state of accelerated droplet aggregation.

[0079] Based on the spatial location information corresponding to the droplet coalescence acceleration state, the location of coalescence occurrence is precisely located. After confirming droplet coalescence acceleration at point P37, a high-risk coalescence zone with a radius of 20 cm is automatically delineated in the high-concentration region, centered on P37. This radius is determined based on the average flow velocity of the circulation loop and the coalescence effect diffusion time. The average flow velocity of the circulation loop is 0.2 m / s, and the coalescence effect diffusion time is approximately 100 seconds. The theoretical influence range is calculated as 20 meters according to the formula diffusion distance = flow velocity × time. Considering the local turbulence effect and backflow phenomenon in the circulation loop, the actual influence range is much smaller than the theoretical value. Therefore, 1 / 100 of the theoretical value is taken, and a safety factor of 1.0 is maintained, ultimately determining the radius of the adjacent area to be 20 cm. For this region, the shear strength parameter is adjusted, increasing the original first shear strength of 76.25 N / m² by 50% to 114.38 N / m², which is applied as an enhanced shear strength for this region.

[0080] The enhanced shear strength is achieved by adjusting the rotational speed of the mechanical agitator at the corresponding position in the circulation loop. The relationship between shear strength and rotational speed is expressed by the empirical formula: τ = k·N²·D·C, where τ is the shear strength in Newtons per square meter; N is the rotational speed in revolutions per second; D is the agitator diameter in meters; k is a fluid-related equipment constant, which is 0.85 in this oil-water mixing system; and C is the agitator blade shape correction factor. This system uses a paddle agitator with a diameter of 0.1 meters and a blade shape correction factor C = 2.24.

[0081] The initial rotational speed was 1200 rpm (20 rpm), corresponding to a theoretical shear strength of τ = 0.85 × 20² × 0.1 × 2.24 = 76.16 N / m², which is basically consistent with the set first shear strength of 76.25 N / m². To achieve an enhanced shear strength of 114.38 N / m², the required rotational speed was calculated backward from the above formula: N = 24.5 rpm (1470 rpm). Considering the actual equipment response characteristics and control accuracy, the rotational speed was set to 1800 rpm, corresponding to a theoretical shear strength of τ = 0.85 × 30² × 0.1 × 2.24 = 171.36 N / m². Due to the non-ideal flow of the fluid and energy dissipation in the actual system, the measured local shear strength was approximately 67% of the theoretical value, i.e., 114.81 N / m², which meets the target value of 114.38 N / m².

[0082] The application time for enhancing shear strength was set at 30 minutes. This time was calculated based on the total volume and flow rate of the circulation loop. The total volume of the circulation loop was 1.5 cubic meters, and the flow rate was 3 cubic meters per hour. Therefore, the time required for the fluid to complete one full circulation was 1.5 ÷ 3 = 0.5 hours = 30 minutes. During this period, the changes in the oil phase particle size distribution in this area were continuously monitored.

[0083] After 30 minutes of enhanced shear treatment, the particle size distribution change rate at point P37 and its adjacent area decreased to 3.1% / min, below the 5% / min threshold for accelerated aggregation, indicating that the accelerated aggregation state had been effectively suppressed. At this point, the stirring speed was restored to the initial 1200 rpm, and the shear strength returned to the first shear strength of 76.25 N / m², continuing normal operation.

[0084] This invention can identify and address the problem of uneven oil phase distribution in the circulation loop, apply differentiated shear strength in a targeted manner, and promptly enhance the local shear strength when droplet coalescence acceleration is detected, effectively suppressing the oil droplet coalescence phenomenon.

[0085] In one optional embodiment, the fine-particle oil phase fluid is guided to a first filtration unit, and the coarse-particle oil phase fluid is guided to a second filtration unit with a pore size larger than that of the first filtration unit to obtain information on the filtration stage traps; the step of performing backwashing on each filtration unit to obtain flux recovery information includes:

[0086] The pore size matching range of the first filter unit is determined based on the particle size distribution characteristics of the fine-particle oil phase fluid, and the pore size matching range of the second filter unit is determined based on the particle size distribution characteristics of the coarse-particle oil phase fluid, so that the retention efficiency of the fine-particle oil phase fluid in the first filter unit and the retention efficiency of the coarse-particle oil phase fluid in the second filter unit satisfy a preset efficiency balance relationship.

[0087] During the filtration process, the rate of change of transmembrane pressure difference between the first filtration unit and the second filtration unit is monitored in real time, and the accumulation rate of retentate in each filtration unit is obtained as retentate information at each stage.

[0088] Based on the cumulative rate of entrapment in the information of entrapment at each level, the clogging risk level of each filter unit is predicted. The backwashing triggering time and backwashing intensity parameters of the first filter unit and the second filter unit are set differently according to the clogging risk level. After the backwashing operation is performed, the flux recovery information is obtained by comparing the recovery ratio of transmembrane pressure difference before and after backwashing.

[0089] For example, oil phase fluids typically contain impurity particles of different sizes, and can be classified into fine-particle oil phase fluids and coarse-particle oil phase fluids according to particle size. This embodiment utilizes a multi-stage filtration unit to process oil phase fluids with different characteristics. Fine-particle oil phase fluids are directed to the first filtration unit, and coarse-particle oil phase fluids are directed to the second filtration unit. Information on retained material and flux recovery is obtained by monitoring the operating status of each filtration unit.

[0090] For fine-particle oil-phase fluids, the particle size distribution characteristics were determined using a laser particle size analyzer. The measured particle size range was 0.5-10 μm, with an average particle size of 3.2 μm, of which 90% of the particles were smaller than 7.5 μm. Based on this distribution characteristic, the pore size matching range of the first filtration unit was determined. This range was set according to 1 / 10 to 1 / 3 of the size of 90% of the particles in the fine-particle oil-phase fluid, i.e., 1 / 37.5 to 1 / 3.75 of 7.5 μm, rounded to 0.2-2 μm. Within this range, a polysulfone hollow fiber membrane with a pore size of 1 μm was selected as the filter medium for the first filtration unit. This pore size is approximately 1 / 3 of the average particle size of 3.2 μm, which can balance retention efficiency and filtration flux, ensuring that most impurities in the fine-particle oil-phase fluid can be effectively retained while maintaining a reasonable filtration flux.

[0091] For coarse-particle oil phase fluids, their characteristics were also determined through particle size analysis. The measured particle size range was 10-100 μm, with an average particle size of 35 μm, of which 90% of the particles were larger than 15 μm. Based on this, the pore size matching range of the second filter unit was determined, set according to 1 / 10 to 1 / 3 of the size of 90% of the particles in the coarse-particle oil phase fluid, i.e., 1 / 7.5 to 1 / 1.5 of 15 μm, rounded to 5-20 μm. Within this range, a ceramic tubular membrane with a pore size of 10 μm was selected as the filter medium for the second filter unit, which is approximately 1 / 3.5 of the average particle size of 35 μm. This design allows the second filter unit to effectively trap large particulate impurities in the coarse-particle oil phase fluid, avoiding premature clogging.

[0092] To ensure that the retention efficiencies of the two filtration units meet the preset efficiency balance, a filtration performance evaluation is required. In this embodiment, the preset efficiency balance is defined as the ratio of the retention efficiencies of the two filtration units being within the range of 0.8-1.2, meaning their retention efficiencies are similar. The retention efficiencies are met by adjusting the operating parameters. For the first filtration unit, the feed flow rate is reduced from the initial 50 L / h to 40 L / h, and the operating pressure is reduced from the initial 30 kPa to 25 kPa, increasing the retention efficiency from the initial 78% to 85%. For the second filtration unit, the feed flow rate is reduced from the initial 80 L / h to 70 L / h, while the operating pressure remains constant at 35 kPa, increasing the retention efficiency from the initial 75% to 82%. After adjustment, the ratio of the retention efficiencies is 85% ÷ 82% = 1.04, satisfying the preset efficiency balance.

[0093] During the operation of the filtration system, differential pressure sensors are used to monitor the changes in transmembrane pressure difference in each filtration unit in real time. The initial transmembrane pressure difference of the first filtration unit is 20 kPa, and that of the second filtration unit is 15 kPa. The rate of change of transmembrane pressure difference is calculated by recording the increase in transmembrane pressure difference per unit time. After 2 hours of operation, the transmembrane pressure difference of the first filtration unit increases to 28 kPa, with a rate of change of (28-20)÷2=4 kPa / h; at the same time, the transmembrane pressure difference of the second filtration unit increases to 19 kPa, with a rate of change of (19-15)÷2=2 kPa / h.

[0094] Based on the transmembrane pressure difference change rate data, the retentate accumulation rate of each filtration unit was calculated using a retentate accumulation model. This model adopts a simplified form of the Hermia clogging equation: dm / dt = α·ΔP', where dm / dt is the retentate accumulation rate in g / m²·h; α is the retentate accumulation coefficient, determined through offline calibration experiments; and ΔP' is the transmembrane pressure difference change rate in kPa / h. The calibration experiments used standard suspensions of known concentrations filtered under identical operating conditions. The mass of retentate per unit area per unit time was measured by weighing, and the α value was calculated based on the measured transmembrane pressure difference change rate. For the first filtration unit, α = 0.15 was obtained; for the second filtration unit, α = 0.9 was obtained. Therefore, the retentate accumulation rate of the first filtration unit was calculated to be 0.15 × 4 = 0.6 g / m²·h, and the retentate accumulation rate of the second filtration unit was calculated to be 0.9 × 2 = 1.8 g / m²·h.

[0095] The clogging risk level of each filter unit is assessed based on the cumulative rate of retained material. In this embodiment, a risk level evaluation standard is set: a cumulative rate of retained material less than 0.5 g / m²·h is considered low risk, 0.5-1.0 g / m²·h is considered medium risk, and greater than 1.0 g / m²·h is considered high risk. Accordingly, the cumulative rate of retained material in the first filter unit is 0.6 g / m²·h, and the clogging risk level is medium risk; the cumulative rate of retained material in the second filter unit is 1.8 g / m²·h, and the clogging risk level is high risk.

[0096] Based on different clogging risk levels, differentiated backwashing strategies were adopted for each filtration unit. For the first filtration unit at medium risk, backwashing was triggered when the transmembrane pressure difference increased by 50%. This threshold was determined based on the correlation between pressure difference increase and flux decline. Experiments showed that when the pressure difference increased by 50%, the flux decline was approximately 30%. Backwashing at this point could ensure filtration efficiency while avoiding frequent cleaning. Therefore, backwashing was initiated when the transmembrane pressure difference of the first filtration unit reached 20 × 1.5 = 30 kPa. The backwashing intensity parameters were set as follows: backwashing pressure was 50 kPa, which is twice the normal operating pressure of 25 kPa, sufficient to overcome the adhesion of entrapped material; backwashing time was 30 seconds, calculated based on a membrane area of ​​1.2 m² and a backwashing flow rate of 2.4 L / s, ensuring that the backwashing liquid completely covered the membrane surface; backwashing frequency was once every 4 hours, determined based on the approximately 4 hours required for the pressure difference to increase from 20 kPa to 30 kPa at the medium risk level.

[0097] For the high-risk second filtration unit, the accumulation rate of filtrate is rapid, and backwashing is triggered when the transmembrane pressure difference increases by 30%. Experiments show that when the pressure difference increases by 30%, the flux decreases by approximately 20%, and backwashing at this point can prevent further clogging. Therefore, backwashing is initiated when the transmembrane pressure difference of the second filtration unit reaches 15 × 1.3 = 19.5 kPa. The backwashing intensity parameters are set as follows: backwashing pressure is 70 kPa, which is twice the normal operating pressure of 35 kPa; backwashing time is 45 seconds, calculated based on a membrane area of ​​1.8 m² and a backwashing flow rate of 2.4 L / s; backwashing frequency is once every 2 hours, determined based on the approximately 2 hours required for the pressure difference to increase from 15 kPa to 19.5 kPa under high-risk conditions.

[0098] After the backwashing operation is performed, the transmembrane pressure difference of each filter unit is measured again, and the transmembrane pressure difference recovery ratio is calculated as flux recovery information. The transmembrane pressure difference recovery ratio is defined as the ratio of the pressure difference reduction after backwashing to the pressure difference increase before backwashing. The calculation formula is: Recovery ratio = (ΔP before backwashing - ΔP after backwashing) / (ΔP before backwashing - ΔP initial).

[0099] After backwashing, the transmembrane pressure difference of the first filter unit decreased to 21 kPa, and the recovery rate was (30-21) / (30-20) = 9 / 10 = 90%. After backwashing, the transmembrane pressure difference of the second filter unit decreased to 16 kPa, and the recovery rate was (19.5-16) / (19.5-15) = 3.5 / 4.5 = 78%. Based on this flux recovery information, the backwashing strategy can be further optimized. The recovery rate of the first filter unit reached 90%, indicating a good backwashing effect, and the existing strategy should be maintained. The recovery rate of the second filter unit was only 78%, lower than the target value of 85%, requiring optimization of the treatment measures. To address this, the backwashing intensity of the second filter unit was increased, raising the backwashing pressure from 70 kPa to 85 kPa and extending the backwashing time from 45 seconds to 60 seconds. If the recovery rate was still lower than 80% after three consecutive backwashes, a chemical cleaning step was introduced, using a 0.5% sodium hydroxide solution for 30 minutes to soak and clean the filter unit to improve the flux recovery effect.

[0100] This invention provides precise filtration of oil-phase fluids with different properties, effectively extending the operating cycle of the filtration system, improving overall filtration efficiency, and reducing system maintenance costs. This method is particularly suitable for processing oil-phase fluids containing complex impurities.

[0101] In one optional implementation, the step of predicting the clogging risk level of each filter unit and setting the backwash triggering timing and backwash intensity parameters differently based on the clogging risk level includes:

[0102] A time series analysis is performed on the accumulation rate of the retained material, and the first and second derivatives of the accumulation rate of the retained material are calculated. When the first derivative is continuously positive and the second derivative is greater than the acceleration threshold, the filter unit is determined to be in an accelerated clogging state.

[0103] Establish a clogging risk level classification rule, and classify the clogging risk into multiple levels based on the comprehensive score of the numerical range of the accumulating rate of the intercepted material and the current value of the transmembrane pressure difference; establish a differentiated mapping relationship between the clogging risk level and the backwashing parameters, so that the backwashing trigger time interval shortens as the clogging risk level increases, and the backwashing intensity parameter increases as the clogging risk level increases.

[0104] Record the transmembrane pressure differential recovery ratio after each backwashing operation. When the recovery ratio of multiple consecutive backwashes shows a downward trend, increase the clogging risk level of the filter unit and shorten the backwashing trigger time interval accordingly, forming a closed loop of dynamic assessment of clogging risk and adaptive adjustment of backwashing parameters.

[0105] For example, the rate of change of transmembrane pressure difference for each filtration unit is monitored in real time to characterize the accumulation trend of retained substances. The transmembrane pressure difference value is recorded every 5 minutes, and the increase in pressure difference between two adjacent measurements is divided by the time interval to obtain the current rate of change of transmembrane pressure difference. For example, if the transmembrane pressure difference of a filtration unit is 15 kPa at 12:00 and 15.3 kPa at 12:05, then the rate of change of transmembrane pressure difference during this period is (15.3-15)÷5=0.06 kPa / min, which is 3.6 kPa / h in hourly time. Based on the aforementioned retained substance accumulation model and the retained substance accumulation coefficient α=0.9 for this filtration unit, the retained substance accumulation rate is calculated to be 0.9×3.6=3.24 g / m²·h.

[0106] Time series analysis of the transmembrane pressure difference change rate is performed, and the clogging status of the filter unit is determined by calculating its first and second derivatives. The first derivative represents the trend of the change rate, while the second derivative reflects the degree of acceleration of the change trend. In practice, the transmembrane pressure difference change rate data at the most recent six time points are taken, and the first and second derivatives are calculated using the central difference method. Let the transmembrane pressure difference change rates at the most recent six time points be r1, r2, r3, r4, r5, and r6, with a time interval of Δt = 5 minutes. The first derivative at the current moment is calculated using the central difference formula: f'(t6) = (r6 - r4) / (2Δt); the second derivative is calculated using the three-point central difference formula: f''(t6) = (r6 - 2r5 + r4) / (Δt)². When the first derivative remains positive for three consecutive time points (i.e., the rate of change increases continuously for 15 minutes) and the second derivative exceeds the preset acceleration threshold of 0.001 kPa / min², the filter unit is determined to be in an accelerated clogging state. This acceleration threshold is determined through statistical analysis of historical operating data. When the second derivative exceeds 0.001 kPa / min², the probability of accelerated increase in transmembrane pressure difference within the following 30 minutes reaches over 85%, therefore it is used as the early warning threshold for accelerated clogging.

[0107] A clogging risk level classification rule was established, using the accumulation rate of retained material and the current value of the transmembrane pressure difference as two key indicators for comprehensive scoring. The accumulation rate of retained material was divided into four levels: below 0.5 g / m²·h = 1 point, 0.5-1.0 g / m²·h = 2 points, 1.0-2.0 g / m²·h = 3 points, and above 2.0 g / m²·h = 4 points. The current value of the transmembrane pressure difference was also divided into four levels: below 20 kPa = 1 point, 20-30 kPa = 2 points, 30-40 kPa = 3 points, and above 40 kPa = 4 points. The scores of the two indicators were added together; a total score of 2-3 points was classified as low risk, 4-5 points as medium risk, 6-7 points as high risk, and 8 points as extremely high risk.

[0108] Based on the clogging risk level, a differentiated mapping relationship was established with backwashing parameters. For backwashing trigger conditions, the low-risk unit was set to trigger when the transmembrane pressure differential increased by 40%; the medium-risk unit was set to trigger when the transmembrane pressure differential increased by 50%; the high-risk unit was set to trigger when the transmembrane pressure differential increased by 30%; and the extremely high-risk unit was set to trigger when the transmembrane pressure differential increased by 20%. A maximum time interval was also set as a backup trigger condition: 180 minutes for low-risk units, 120 minutes for medium-risk units, 60 minutes for high-risk units, and 30 minutes for extremely high-risk units. This means that backwashing is forcibly triggered even if the pressure differential increase threshold is not reached when the maximum time interval is reached. Backwashing intensity parameters include backwashing water pressure and backwashing duration: 50 kPa for low-risk units and 30 seconds for medium-risk units; 65 kPa for medium-risk units and 40 seconds for medium-risk units; 80 kPa for high-risk units and 50 seconds for high-risk units; and 95 kPa for extremely high-risk units and 60 seconds for extremely high-risk units.

[0109] To achieve dynamic assessment of clogging risk and adaptive adjustment of backwashing parameters, the effect of each backwashing operation is recorded. The transmembrane pressure differential P1 is measured before backwashing, and P2 is measured immediately after backwashing. The backwash recovery ratio is calculated as (P1-P2) / P1. For example, if the transmembrane pressure differential of a filter unit is 28 kPa before backwashing and 21 kPa after backwashing, the recovery ratio is (28-21) / 28 = 0.25, meaning the pressure differential decreased by 25%. The recovery ratios η1, η2, and η3 for three consecutive backwashes are recorded. If η1 > η2 > η3 and (η1-η3) / η1 > 0.2, meaning the recovery ratio decreases monotonically and the overall decrease exceeds 20%, then the current backwashing parameters are deemed insufficient to effectively control membrane fouling, and the clogging risk level of this unit needs to be increased.

[0110] After the risk level is upgraded, the backwashing parameters are automatically adjusted. For example, if a filter unit that was originally classified as medium risk experiences a decline in backwashing effectiveness after three consecutive backwashes, its risk level is upgraded to high risk. The backwashing trigger condition is adjusted from a 50% increase in differential pressure to a 30% increase in differential pressure, the maximum time interval is shortened from 120 minutes to 60 minutes, the backwashing water pressure is increased from 65 kPa to 80 kPa, and the duration is extended from 40 seconds to 50 seconds. After the risk level is upgraded, a 6-hour observation period begins. During the observation period, if the recovery rate of three consecutive backwashes is higher than 80% of the initial backwash recovery rate and no longer shows a downward trend, the new risk level is maintained; if it still shows a downward trend, the risk level continues to be upgraded until it reaches an extremely high risk level. If, under the extremely high risk level, the strongest backwashing parameters are used and the recovery rate of three consecutive backwashes is still below 30%, an alarm is triggered and offline chemical cleaning is recommended.

[0111] This invention can adopt corresponding backwashing strategies for filter units with different clogging states, avoiding the over-washing or under-washing problems in the traditional uniform backwashing mode, and improving the operating efficiency of membrane filtration and the service life of membrane modules.

[0112] In one optional implementation, the steps of determining the extraction efficiency state based on the target component concentration and the anti-clogging performance state based on the flux recovery information, and adjusting the shear intensity, backwashing sequence, or a combination of both according to the condition of the extraction efficiency state and the anti-clogging performance state to form an adaptive closed loop, include:

[0113] Based on the target component concentration, an extraction efficiency evaluation index is constructed that includes the degree of concentration attainment and concentration time-series stability, and the extraction efficiency status is determined according to the extraction efficiency evaluation index; based on the flux recovery information, an anti-clogging performance evaluation index is constructed that includes the degree of recovery sufficiency and recovery sustainability, and the anti-clogging performance status is determined according to the anti-clogging performance evaluation index.

[0114] A four-quadrant decision matrix is ​​constructed to represent the extraction efficiency state and the anti-clogging performance state. When both states are true, the current parameters are maintained. When only the extraction efficiency state is true, the backwashing sequence is adjusted based on the anti-clogging performance evaluation index. When only the anti-clogging performance state is true, the shear intensity is adjusted based on the extraction efficiency evaluation index. When neither state is true, the adjustment weights of the shear intensity and the backwashing sequence are determined based on the degree of deviation between the extraction efficiency evaluation index and the anti-clogging performance evaluation index.

[0115] The adjusted shear strength and backwashing timing are applied to the shearing treatment stage and the backwashing operation stage respectively, forming an adaptive closed loop.

[0116] Combination Figure 2 The following is an explanation of the four-quadrant adaptive closed-loop control flowchart based on a dual-state evaluation of extraction efficiency and anti-clogging performance. In the extraction system, the concentration data of the target components is first acquired, which can be collected in real time through an online monitoring system. For different target components, corresponding online monitoring methods can be used: proteins are measured at a wavelength of 280 nm using an ultraviolet absorption spectrometer; polyphenols are measured at 765 nm using the Folin-Ciocalteu method; fat-soluble pigments (such as lycopene and β-carotene) are measured using the characteristic wavelength of visible light; and sugars are measured using an online refractometer to determine their refractive index. Taking protein extraction as an example, the sampling frequency is set to once every 10 minutes.

[0117] Record the flux changes of the extraction membrane, including three key flux values: initial flux, flux before backwashing, and flux after backwashing. Initial flux refers to the flux of a new membrane or a fully cleaned membrane under standard operating conditions; flux before backwashing refers to the flux after a period of extraction operation when the membrane surface becomes clogged due to the accumulation of retentate; flux after backwashing refers to the flux restored after a backwashing operation. For example, if the initial system flux is 30 L / m²·h, after 60 minutes of continuous extraction operation, the flux drops to 22 L / m²·h due to retentate accumulation (this is the flux before backwashing). After a backwashing operation (30 seconds), the flux recovers to 26 L / m²·h (this is the flux after backwashing).

[0118] Based on the target component concentration data, an extraction efficiency evaluation index is constructed. This index includes two aspects: concentration attainment level and concentration time-series stability. Concentration attainment level is measured using a relative concentration value, i.e., the ratio of the actual measured concentration to the target concentration. For example, if the target protein concentration is set at 5.0 g / L, and the actual measured concentration is 4.5 g / L, then the concentration attainment level is 4.5 / 5.0 = 0.90. Concentration time-series stability is calculated using the coefficient of variation of multiple consecutive measurements, i.e., the standard deviation divided by the average. For example, if five consecutive measurements are 4.5, 4.6, 4.4, 4.5, and 4.7 g / L, the calculated average is 4.54 g / L, and the standard deviation is 0.11 g / L. Therefore, the concentration time-series stability index (coefficient of variation) is 0.11 / 4.54 ≈ 0.024. The system sets a judgment threshold: when the concentration attainment level ≥ 0.85 and the concentration time-series stability index ≤ 0.05, the extraction efficiency is considered valid.

[0119] Based on flux change information, an evaluation index for anti-clogging performance is constructed, comprising two dimensions: recovery adequacy and recovery sustainability. Recovery adequacy refers to the recovery ratio of flux after backflushing relative to the initial flux, calculated as: Recovery Adequacy = Flux after backflushing / Initial Flux. Continuing the example above, Recovery Adequacy = 26 / 30 ≈ 0.87. This index reflects the immediate effect of backflushing on flux recovery. Furthermore, as a supplementary reference, the flux recovery rate (flux after backflushing / Flux before backflushing) can be calculated; in this example, it is 26 / 22 ≈ 1.18, indicating that backflushing increased the flux by 18%.

[0120] Recovery sustainability is expressed as flux maintenance time, defined as the time required for the flux to decrease from its recovery value to 90% of its initial flux after backflushing, expressed in hours. This indicator reflects the persistence of the backflushing effect. For example, in this case, the initial flux is 30 L / m²·h, and 90% of the initial flux is 27 L / m²·h. After backflushing, the flux is 26 L / m²·h (already below 27 L / m²·h). The system needs to monitor when the flux first drops below 27 L / m²·h. If timing starts immediately after backflushing, and the flux drops to 27 L / m²·h after 3.5 hours, the flux maintenance time is 3.5 hours. If the flux is higher than 27 L / m²·h after backflushing (e.g., recovering to 28 L / m²·h), then timing starts from 28 L / m²·h and continues until it drops to 27 L / m²·h; the time elapsed is the flux maintenance time.

[0121] The threshold for determining the anti-blocking performance status is set as follows: when the recovery adequacy is ≥0.85 and the flux maintenance time is ≥2.0 hours, the anti-blocking performance status is considered valid. In this example, the recovery adequacy of 0.87 ≥ 0.85 meets the requirements, but further observation is needed to see if the flux maintenance time is ≥2.0 hours. Assuming that the monitoring shows the flux maintenance time is 2.5 hours, then the anti-blocking performance status is valid.

[0122] Construct a four-quadrant decision matrix for extraction efficiency state and anti-blocking performance state, and execute corresponding control strategies based on the occurrence of the two states:

[0123] Scenario 1: When both extraction efficiency and anti-clogging performance are satisfactory, maintain the current operating parameters. For example, if the current shear intensity is 1200 rpm and the backwashing sequence is performed every 30 minutes for 30 seconds each time, then continue to maintain these parameter settings. The system enters steady-state operation mode and only performs routine monitoring.

[0124] Scenario 2: When only the extraction efficiency condition is met while the anti-fouling performance condition is not met, it indicates that the extraction effect of the target component is good, but the membrane's anti-fouling performance is insufficient. The system needs to adjust the backwashing sequence based on the anti-fouling performance evaluation index. The adjustment strategy is divided into three sub-cases:

[0125] Sub-case 2a: If the recovery adequacy is <0.85 (insufficient backwash intensity) but the flux maintenance time is ≥2.0 hours (backwash effect is acceptable), then extend the backwash duration to improve the thoroughness of a single backwash. The adjustment amount is calculated as follows: backwash duration increase = base step size × (1 + recovery adequacy deviation), where the base step size is set to 10 seconds, and the recovery adequacy deviation = (0.85 - actual value) / 0.85. For example, if the recovery adequacy is 0.80, the deviation = (0.85 - 0.80) / 0.85 ≈ 0.059, then the duration increase = 10 × (1 + 0.059) ≈ 11 seconds, extending the backwash duration from 30 seconds to 41 seconds.

[0126] Sub-case 2b: If the recovery adequacy is ≥0.85 (single backwash effect is acceptable) but the flux maintenance time is <2.0 hours (backwash effect is insufficient in duration), then increase the backwash frequency to shorten the time interval between two backwashes. The adjustment amount is calculated as follows: backwash interval reduction = base step size × (1 + flux maintenance time deviation), where the base step size is set to 5 minutes, and the flux maintenance time deviation = (2.0 - actual value) / 2.0. For example, if the flux maintenance time is 1.5 hours, the deviation = (2.0 - 1.5) / 2.0 = 0.25, then the interval reduction = 5 × (1 + 0.25) = 6.25 minutes, rounded down to 6 minutes, shortening the backwash interval from 30 minutes to 24 minutes.

[0127] Sub-case 2c: If the recovery adequacy is <0.85 and the flux maintenance time is <2.0 hours (both indicators fail to meet the standard), then the backwash duration and frequency should be adjusted simultaneously. Calculate the adjustment amount according to the two adjustment methods described above and implement them concurrently. For example, if the recovery adequacy is 0.78 and the flux maintenance time is 1.2 hours, then the increase in duration = 10 × (1 + (0.85 - 0.78) / 0.85) ≈ 11 seconds, and the reduction in interval = 5 × (1 + (2.0 - 1.2) / 2.0) = 7 minutes. The final backwash sequence is adjusted to be performed every 23 minutes, with each run lasting 41 seconds.

[0128] Case 3: When only the anti-fouling performance is satisfactory while the extraction efficiency is not, it indicates that the membrane's anti-fouling performance is good, but the extraction effect of the target component is insufficient. The shear strength needs to be adjusted based on the extraction efficiency evaluation index. The adjustment strategy is divided into two sub-cases:

[0129] Sub-case 3a: If the concentration attainment level is <0.85 (extraction concentration is low) but the concentration time-series stability index is ≤0.05 (concentration stability is good), it indicates that the extraction process is stable but the extraction driving force is insufficient, and the shear intensity needs to be increased to enhance mass transfer efficiency. The adjustment amount is calculated as follows: Shear intensity increase = base step size × (1 + concentration attainment level deviation), where the base step size is set to 100 rpm, and the concentration attainment level deviation = (0.85 - actual value) / 0.85. For example, if the concentration attainment level is 0.80, the deviation = (0.85 - 0.80) / 0.85 ≈ 0.059, then the shear intensity increase = 100 × (1 + 0.059) ≈ 106 rpm, which is rounded to 110 rpm, increasing the shear intensity from 1200 rpm to 1310 rpm.

[0130] Sub-case 3b: If the concentration compliance rate is ≥0.85 (average concentration meets the standard) but the concentration time-series stability index is >0.05 (concentration fluctuation is too large), it indicates that the extraction process is unstable and the shear intensity needs to be increased to improve mixing uniformity. The adjustment amount is calculated as follows: Shear intensity increase = base step size × (1 + concentration stability deviation), where concentration stability deviation = (actual value - 0.05) / 0.05. For example, if the concentration time-series stability index is 0.07, the deviation = (0.07 - 0.05) / 0.05 = 0.40, then the shear intensity increase = 100 × (1 + 0.40) = 140 rpm, increasing the shear intensity from 1200 rpm to 1340 rpm.

[0131] Sub-case 3c: If the concentration compliance rate is <0.85 and the concentration time-series stability index is >0.05 (both indices fail to meet the standard), it indicates both insufficient extraction driving force and uneven mixing, requiring a significant increase in shear strength. In this case, the larger of the two deviation values ​​should be used for adjustment. For example, if the concentration compliance rate is 0.78 (deviation 0.082) and the concentration stability index is 0.08 (deviation 0.60), taking the larger deviation of 0.60, the increase in shear strength would be 100 × (1 + 0.60) = 160 rpm, increasing the shear strength from 1200 rpm to 1360 rpm.

[0132] Scenario 4: When both the extraction efficiency and anti-clogging performance statuses are not met, it indicates that both extraction efficiency and anti-clogging performance are problematic, requiring comprehensive adjustment of shear strength and backwashing timing. The adjustment weights are determined based on the combined degree of deviation between the two statuses.

[0133] First, calculate the deviation of each individual indicator: Concentration compliance deviation = (0.85 - actual concentration compliance) / 0.85; Concentration stability deviation = (actual stability index - 0.05) / 0.05 (when actual value > 0.05) or 0 (when actual value ≤ 0.05); Recovery adequacy deviation = (0.85 - actual recovery adequacy) / 0.85; Flux maintenance time deviation = (2.0 - actual flux maintenance time) / 2.0.

[0134] Then calculate the overall deviation: Overall deviation of extraction efficiency = Deviation of concentration compliance + Deviation of concentration stability; Overall deviation of anti-clogging performance = Deviation of recovery sufficiency + Deviation of flux maintenance time; Finally, calculate the adjustment weights: Shear strength adjustment weight = Overall deviation of extraction efficiency / (Overall deviation of extraction efficiency + Overall deviation of anti-clogging performance); Backwashing sequence adjustment weight = Overall deviation of anti-clogging performance / (Overall deviation of extraction efficiency + Overall deviation of anti-clogging performance).

[0135] For example: Assume the monitoring data are as follows: Concentration compliance rate = 0.78, deviation = (0.85 - 0.78) / 0.85 ≈ 0.082; Concentration stability index = 0.08, deviation = (0.08 - 0.05) / 0.05 = 0.60; Recovery adequacy = 0.80, deviation = (0.85 - 0.80) / 0.85 ≈ 0.059; Flux maintenance time = 1.3 hours, deviation = (2.0 - 1.3) / 2.0=0.35; Overall deviation of extraction efficiency status=0.082+0.60=0.682; Overall deviation of anti-clogging performance status=0.059+0.35=0.409; Total deviation=0.682+0.409=1.091; Shear strength adjustment weight=0.682 / 1.091≈0.625; Backwashing sequence adjustment weight=0.409 / 1.091≈0.375.

[0136] The actual adjustment amount is calculated based on weights and deviations: Shear strength adjustment: Since the concentration stability deviation (0.60) is greater than the concentration compliance deviation (0.082), 0.60 is taken as the dominant deviation. The increase in shear strength = 100 × (1 + 0.60) × 0.625 ≈ 100 rpm (rounded after considering weights), and the shear strength is adjusted from 1200 rpm to 1300 rpm.

[0137] Backwashing timing adjustment: Since the deviation in flux maintenance time (0.35) is greater than the deviation in recovery adequacy (0.059), the backwashing frequency is adjusted first. The backwashing interval is shortened by 5 × (1 + 0.35) × 0.375 ≈ 3 minutes (rounded after weighting), reducing the backwashing interval from 30 minutes to 27 minutes. Simultaneously, due to a deviation in recovery adequacy, the backwashing duration is increased by 10 × (1 + 0.059) × 0.375 ≈ 4 seconds, extending the duration from 30 seconds to 34 seconds.

[0138] The final adjustment results are as follows: the shear strength is increased from 1200 rpm to 1300 rpm, and the backwashing sequence is changed from once every 30 minutes for 30 seconds each time to once every 27 minutes for 34 seconds each time.

[0139] To ensure the safe and stable operation of the system, all operating parameters are subject to boundary limits: shear strength adjustment range: 800-2000 rpm. Below 800 rpm, the liquid phase is not sufficiently mixed, failing to effectively disperse the feed and promote mass transfer; above 2000 rpm, excessive shear force causes mechanical damage to the extraction membrane or leads to structural destruction of certain shear-sensitive target components (such as large protein molecules).

[0140] Backwashing frequency adjustment range: once every 60 minutes to once every 10 minutes. When the interval is longer than 60 minutes, the retained products form a dense fouling layer on the membrane surface that is difficult to remove, leading to irreversible blockage; when the interval is shorter than 10 minutes, frequent backwashing will significantly reduce the effective extraction time, reduce system productivity, and at the same time, repeated switching between forward and reverse flow will cause fatigue damage to the membrane structure.

[0141] Backwash duration adjustment range: 10-90 seconds. When the duration is less than 10 seconds, the backwash solution cannot fully penetrate into the membrane pores, resulting in poor cleaning effect; when the duration is more than 90 seconds, excessive backwashing causes the target components extracted to the opposite side of the membrane to be lost in the reverse direction, while increasing the consumption of cleaning solution and the amount of waste liquid generated.

[0142] When the parameters reach the aforementioned boundary limits after automatic adjustment, but the extraction efficiency or anti-clogging performance still fails to meet requirements, a level two alarm is triggered, indicating to the operator that the following abnormalities exist: the extraction membrane has reached the end of its service life and needs replacement; the properties of the feed solution have changed significantly (e.g., abnormally high solid content, abnormally high viscosity, etc.); there is a mechanical fault (e.g., wear of the agitator bearing, leakage of the backwash valve, etc.); the process target setting is unreasonable, such as the target concentration exceeding the thermodynamic equilibrium concentration of the feed solution system. In this case, it is recommended to perform equipment maintenance, re-optimize process parameters, or enhance feed solution pretreatment.

[0143] The adjusted shear strength acts on the shearing process, affecting the contact state between the extract phase and the feed phase, the interface renewal rate, and the mass transfer coefficient. The adjusted backwashing sequence acts on the backwashing operation, affecting the removal efficiency of membrane surface residues and the degree of membrane flux recovery. The two adjustments work together to form an adaptive closed-loop control system.

[0144] The system continuously monitors the target component concentration data and flux changes, with the monitoring cycle consistent with the sampling frequency of the online analyzer (e.g., every 10 minutes). After acquiring each new set of data, the system immediately updates the extraction efficiency status and anti-clogging performance status, and executes the corresponding strategy in the four-quadrant decision matrix based on the latest status. For example, after the adjustment in scenario 4 above (shear intensity increased to 1300 rpm, backwash adjusted to every 27 minutes, 34 seconds each time), after three monitoring cycles (30 minutes), new data is collected: the target component concentration increases from an average of 4.0 g / L to 4.6 g / L, concentration compliance = 4.6 / 5.0 = 0.92 (≥0.85); five consecutive concentration measurements are 4.5, 4.6, 4.7, 4.6, and 4.6 g / L, standard deviation = 0.07 g / L, average value = 4.6 g / L, concentration time-series stability index = 0.07 / 4.6 ≈ 0.015 (≤0.05); the extraction efficiency status is established.

[0145] Meanwhile, after backwashing, the flux recovered to 27.5 L / m²·h (the initial flux was still 30 L / m²·h), and the degree of recovery was approximately 27.5 / 30≈0.92 (≥0.85). The flux maintenance time (the time required to reduce from 27.5 L / m²·h to 27 L / m²·h) was monitored to be 2.8 hours (≥2.0 hours). The anti-clogging performance was established.

[0146] At this point, both the extraction efficiency and anti-clogging performance are satisfactory. Based on the decision in scenario 1, the current operating parameters (shear intensity 1300 rpm, backwash every 27 minutes, 34 seconds each time) are maintained unchanged, completing one full adaptive adjustment cycle. The system enters a new steady-state operation phase, continuing routine monitoring and prepared to respond to any new state changes.

[0147] This adaptive control method, based on dual-state evaluation and four-quadrant decision-making, can dynamically balance the two key objectives of extraction efficiency and anti-clogging performance according to the real-time performance of the extraction process. While ensuring efficient extraction of the target components, it maintains the long-term stable operation of the extraction membrane, significantly improving the intelligence level and economic benefits of the extraction system.

[0148] In one optional implementation, the step of constructing a four-quadrant decision matrix for the extraction efficiency state and the anti-blocking performance state includes:

[0149] A two-dimensional decision space is established for the extraction efficiency state and the anti-blocking performance state. The combination of the two states being true and false is divided into four quadrant regions, and each quadrant region corresponds to a set of preset adjustment strategies.

[0150] When both states are true, the system is positioned in the first quadrant and the parameter maintenance strategy is executed.

[0151] When only the extraction efficiency state is established, the second quadrant is located, the recovery sustainability decay coefficient in the anti-clogging performance evaluation index is extracted, and the amount of shortening of the backwashing sequence is calculated based on the decay coefficient.

[0152] When only the anti-blocking performance state is established, it is located in the third quadrant. The fluctuation range of concentration time stability in the extraction efficiency evaluation index is extracted. When the fluctuation range exceeds the stability threshold, the first shear strength of the high concentration region is increased. When the fluctuation range does not exceed the stability threshold but the concentration compliance level is lower than the target value, the second shear strength of the low concentration region is reduced.

[0153] When neither of the two states is true, the system is positioned in the fourth quadrant. The deviation of the extraction efficiency evaluation index and the deviation of the anti-clogging performance evaluation index are calculated. The ratio of the two deviations is used as the adjustment weight allocation coefficient. Based on the adjustment weight allocation coefficient, the amount of shear strength enhancement and the amount of backwashing time shortening are determined simultaneously.

[0154] For example, based on the aforementioned extraction efficiency state and anti-blocking performance state, a two-dimensional decision space is established and divided into four quadrants. This decision space uses the validity of the extraction efficiency state as the horizontal axis and the validity of the anti-blocking performance state as the vertical axis, forming a 2×2 decision matrix. The first quadrant corresponds to the case where both states are valid; the second quadrant corresponds to the case where only the extraction efficiency state is valid but the anti-blocking performance state is invalid; the third quadrant corresponds to the case where only the anti-blocking performance state is valid but the extraction efficiency state is invalid; and the fourth quadrant corresponds to the case where neither state is valid. Each quadrant has a pre-set set of adjustment strategies, which automatically locate the corresponding quadrant and execute the appropriate strategy based on the real-time monitored state combinations.

[0155] When both extraction efficiency and anti-clogging performance are met, the current operation is determined to be in quadrant 1. At this time, the concentration compliance rate in the extraction efficiency evaluation index is ≥0.85 and the concentration time-series stability index is ≤0.05, while the recovery sufficiency rate in the anti-clogging performance evaluation index is ≥0.85 and the flux maintenance time is ≥2.0 hours. A parameter maintenance strategy is implemented, keeping the current shear intensity setpoint and the current backwash timing setpoint unchanged, performing only routine monitoring without triggering any parameter adjustments. For example, if the current shear intensity in the high-concentration region is 1300 rpm, the shear intensity in the low-concentration region is 900 rpm, the backwash interval is 27 minutes, and the backwash duration is 34 seconds, these parameters remain unchanged in quadrant 1 until a state transition occurs.

[0156] When only the extraction efficiency condition is met while the anti-clogging performance condition is not met, the current operation is determined to be in the second quadrant. In this case, the backwashing sequence needs to be adjusted based on the anti-clogging performance evaluation index to improve anti-clogging performance. The recovery sustainability parameter is extracted from the anti-clogging performance evaluation index. This parameter is characterized by flux maintenance time, defined as the time required for the flux to drop from the recovery value to 90% of the initial flux after backwashing. The recovery sustainability decay coefficient is calculated as follows: (Target flux maintenance time - Actual flux maintenance time) ÷ Target flux maintenance time, where the target flux maintenance time is set to 2.0 hours. For example, if the actual flux maintenance time is 1.5 hours, then the decay coefficient = (2.0 - 1.5) ÷ 2.0 = 0.25. Based on this decay coefficient, the reduction in the backwashing sequence is calculated. Specifically, the calculation method is: base reduction step × (1 + decay coefficient), where the base reduction step is set to 5 minutes. Continuing with the above example, the reduction in the backwashing interval = 5 × (1 + 0.25) = 6.25 minutes, rounded down to 6 minutes. If the current backwash interval is 30 minutes, then the adjusted backwash interval = 30 - 6 = 24 minutes. Simultaneously, if the recovery adequacy is <0.85, the backwash duration needs to be extended accordingly. The extension amount is calculated using the recovery adequacy deviation: Deviation = (0.85 - Actual Recovery Adequacy) ÷ 0.85. The duration increase = base extension step of 10 seconds × (1 + Deviation). For example, if the recovery adequacy is 0.80, the deviation = (0.85 - 0.80) ÷ 0.85 = 0.059, then the duration increase = 10 × (1 + 0.059) ≈ 11 seconds, extending the backwash duration from 30 seconds to 41 seconds.

[0157] When only the anti-clogging performance condition is met while the extraction efficiency condition is not met, the current operation is determined to be in the third quadrant. In this case, the shear intensity needs to be adjusted based on the extraction efficiency evaluation index to improve the extraction effect. First, the fluctuation range of concentration temporal stability in the extraction efficiency evaluation index is extracted. This fluctuation range is represented by the coefficient of variation of multiple consecutive concentration measurements, calculated as standard deviation ÷ average. A stability threshold of 0.05 is set. When the fluctuation range > this threshold, it indicates that the concentration fluctuation is too large, and the first shear intensity in the high-concentration region needs to be increased to enhance mixing uniformity and stabilize the concentration output. The increase in shear intensity = basic enhancement step size 100 rpm × (1 + concentration stability deviation), where deviation = (actual fluctuation range - 0.05) ÷ 0.05. For example, if the fluctuation range is 0.08, the deviation = (0.08 - 0.05) ÷ 0.05 = 0.6, then the increase in shear intensity = 100 × (1 + 0.6) = 160 rpm. If the current first shear intensity is 1200 rpm, then the adjusted shear intensity is 1360 rpm. When the fluctuation range is ≤ the stability threshold but the concentration compliance rate is < the target value of 0.85, it indicates that the concentration is stable but too low. The second shear intensity in the low concentration region needs to be reduced to minimize disturbance to the feed phase and promote the directional migration of the target component to the extraction phase. The reduction in shear intensity = basic adjustment step size 80 rpm × (1 + concentration compliance rate deviation), where deviation = (0.85 - actual concentration compliance rate) ÷ 0.85. For example, if the concentration meets the standard of 0.78, the deviation is (0.85-0.78)÷0.85=0.082, then the reduction in shear strength is 80×(1+0.082)≈87 revolutions per minute. If the current second shear strength is 900 revolutions per minute, then the adjusted value is 813 revolutions per minute.

[0158] When both the extraction efficiency and anti-clogging performance status are not met, the current operation is determined to be in quadrant 4. At this time, it is necessary to simultaneously adjust the shear strength and backwashing sequence, with the adjustment weights based on the deviation of the two types of indicators. The comprehensive deviation of the extraction efficiency evaluation index and the comprehensive deviation of the anti-clogging performance evaluation index are calculated separately. Extraction efficiency comprehensive deviation = concentration compliance deviation + concentration stability deviation, where concentration compliance deviation = (0.85 - actual value) ÷ 0.85, and concentration stability deviation = (actual value - 0.05) ÷ 0.05. Valid only if the actual value > 0.05, otherwise 0. Anti-clogging performance comprehensive deviation = recovery sufficiency deviation + flux maintenance time deviation, where recovery sufficiency deviation = (0.85 - actual value) ÷ 0.85, and flux maintenance time deviation = (2.0 - actual value) ÷ 2.0. Valid only if the actual value < 2.0, otherwise 0. Calculate the adjustment weight allocation coefficients: Shear strength adjustment weight = Overall deviation of extraction efficiency ÷ (Sum of two types of deviations); Backwash timing adjustment weight = Overall deviation of anti-clogging performance ÷ (Sum of two types of deviations). For example, if the concentration compliance level is 0.78 corresponding to a deviation of 0.082, the concentration stability is 0.08 corresponding to a deviation of 0.6, the recovery adequacy is 0.80 corresponding to a deviation of 0.059, and the flux maintenance time is 1.3 hours corresponding to a deviation of 0.35, then the overall deviation of extraction efficiency = 0.082 + 0.6 = 0.682, the overall deviation of anti-clogging performance = 0.059 + 0.35 = 0.409, and the total deviation = 0.682 + 0.409 = 1.091. Shear strength adjustment weight = 0.682 ÷ 1.091 ≈ 0.625, and backwash timing adjustment weight = 0.409 ÷ 1.091 ≈ 0.375. The shear strength enhancement was determined based on weights. The baseline enhancement was calculated as 100 × (1 + 0.6) = 160 rpm, based on a concentration stability dominant deviation of 0.6. The actual enhancement after weighting was 160 × 0.625 = 100 rpm. The backwash time reduction was also determined based on weights. The baseline reduction was calculated as 5 × (1 + 0.35) = 6.75 minutes, based on a flux maintenance time dominant deviation of 0.35. The actual reduction after weighting was 6.75 × 0.375 ≈ 2.5 minutes, rounded to 3 minutes. Simultaneously, the baseline duration extension was calculated as 10 × (1 + 0.059) ≈ 11 seconds, based on a recovery adequacy deviation of 0.059. The actual extension after weighting was 11 × 0.375 ≈ 4 seconds. The final adjustment results in an increase in shear strength in the high-concentration area from 1200 to 1300 rpm, a reduction in the backwash interval from 30 minutes to 27 minutes, and an extension in backwash duration from 30 seconds to 34 seconds.

[0159] Set boundary limits for parameter adjustments: shear strength adjustment range is 800 to 2000 rpm, backflushing interval adjustment range is 10 to 60 minutes, and backflushing duration adjustment range is 10 to 90 seconds. If the adjusted parameters reach the boundary values ​​but the status still does not meet the requirements, an abnormal alarm will be triggered, indicating equipment failure or unreasonable process settings. Equipment maintenance or parameter re-optimization is recommended.

[0160] This invention employs a partitioned adjustment strategy based on a four-quadrant decision matrix. This strategy can accurately pinpoint the root cause of the problem and implement differentiated control based on the coupling state between extraction efficiency and anti-clogging performance. It avoids parameter oscillations caused by blind adjustments, improves the stability and adaptability of extraction, and achieves a dynamic balance between extraction efficiency and membrane flux.

[0161] A second aspect of the present invention provides a precision control and processing system for extraction tank feeding based on automatic anti-clogging, comprising:

[0162] The first unit is used to acquire information on the oil phase distribution of the feed fluid to be treated and the target component concentration at the outlet of the extraction box; to collect time-series data on the particle size of oil phase droplets and to identify the droplet aggregation and acceleration state;

[0163] The second unit is used to identify high-concentration and low-concentration regions in the circulation loop based on the oil phase distribution information, apply a first shear strength treatment to the high-concentration region to obtain fine-particle oil phase fluid, apply a second shear strength treatment lower than the first shear strength to the low-concentration region to obtain coarse-particle oil phase fluid, and enhance the shear strength of the high-concentration region in advance when the droplet coalescence acceleration state is detected.

[0164] The third unit is used to guide the fine-particle oil phase fluid to the first filtration unit and the coarse-particle oil phase fluid to the second filtration unit with a pore size larger than that of the first filtration unit, thereby obtaining information on the intercepted materials at each stage; and to perform backwashing operations on each filtration unit to obtain flux recovery information.

[0165] The fourth unit is used to determine the extraction efficiency status based on the target component concentration and the anti-clogging performance status based on the flux recovery information; according to the establishment of the extraction efficiency status and the anti-clogging performance status, the shear intensity, backwashing sequence or a combination of both are adjusted accordingly to form an adaptive closed loop.

[0166] A third aspect of the present invention provides an electronic device, comprising:

[0167] processor;

[0168] Memory used to store processor-executable instructions;

[0169] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0170] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0171] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for precise control of extraction tank feeding based on automatic anti-clogging, characterized in that, include: Obtain information on the oil phase distribution of the feed fluid to be treated and the target component concentration at the outlet of the extraction tank; Collect time-series data on the particle size of oil phase droplets to identify the accelerated state of droplet coalescence; Based on the oil phase distribution information, high-concentration and low-concentration regions in the circulation loop are identified. A first shear strength is applied to the high-concentration region to obtain fine-particle oil phase fluid, and a second shear strength lower than the first shear strength is applied to the low-concentration region to obtain coarse-particle oil phase fluid. Furthermore, the shear strength of the high-concentration region is enhanced in advance when the droplet coalescence acceleration state is detected. The fine-particle oil phase fluid is directed to the first filtration unit, and the coarse-particle oil phase fluid is directed to the second filtration unit with a pore size larger than that of the first filtration unit, so as to obtain information on the intercepted substances at each stage. Perform backwashing on each filter unit to obtain flux recovery information; The extraction efficiency state is determined based on the target component concentration, and the anti-clogging performance state is determined based on the flux recovery information. Depending on whether the extraction efficiency state and the anti-clogging performance state are met, the shear intensity, backwashing sequence, or a combination of both are adjusted accordingly to form an adaptive closed loop.

2. The method according to claim 1, characterized in that, The steps for collecting time-series data on the particle size of oil-phase droplets and identifying the accelerated state of droplet coalescence include: Oil droplet size distribution data are collected simultaneously at multiple spatial locations to construct a size-spatial distribution field. Based on the evolution trajectory of the size-spatial distribution field in the time dimension, spatial propagation velocity and local aggregation degree indices characterizing droplet aggregation behavior are extracted. When the product of the spatial propagation velocity and the local aggregation index exceeds the critical coalescence threshold, the droplet coalescence acceleration state is determined to be established. The spatial location information of the coalescence acceleration is coupled with the oil phase distribution information for analysis to locate the boundary of the high-concentration region that requires pre-enhanced shear strength.

3. The method according to claim 1, characterized in that, The steps of identifying high-concentration and low-concentration regions in the circulation loop based on the oil phase distribution information, applying a first shear intensity to the high-concentration region to obtain fine-particle oil phase fluid, applying a second shear intensity lower than the first shear intensity to the low-concentration region to obtain coarse-particle oil phase fluid, and preemptively increasing the shear intensity of the high-concentration region when the droplet coalescence acceleration state is detected include: Based on the oil phase distribution information, the spatial gradient distribution characteristics of oil phase concentration in the circulation loop are extracted, the locations of abrupt changes in concentration gradient are identified, the concentration values ​​corresponding to the locations of abrupt changes in concentration gradient are used as concentration boundary thresholds, and the circulation loop is divided into high-concentration regions and low-concentration regions using the concentration boundary thresholds. For the high-concentration region and the low-concentration region respectively, the coupling parameters of the mean oil phase concentration and the particle size distribution variance in the region are calculated. Based on the coupling parameters, the values ​​of the first shear strength and the second shear strength are determined, so that the particle size distribution concentration of the fine-particle oil phase fluid obtained after the first shear strength treatment is higher than the particle size distribution concentration of the coarse-particle oil phase fluid obtained after the second shear strength treatment. When the accelerated droplet coalescence state is detected, the location of the coalescence occurrence is located in the high-concentration region based on the spatial location information corresponding to the accelerated droplet coalescence state, and an enhanced shear strength higher than the first shear strength is applied to the location of the coalescence occurrence and its adjacent region.

4. The method according to claim 1, characterized in that, The fine-particle oil phase fluid is directed to the first filtration unit, and the coarse-particle oil phase fluid is directed to the second filtration unit with a pore size larger than that of the first filtration unit, so as to obtain information on the intercepted substances at each stage. The steps for performing backwashing operations on each filter unit to obtain flux recovery information include: The pore size matching range of the first filter unit is determined based on the particle size distribution characteristics of the fine-particle oil phase fluid, and the pore size matching range of the second filter unit is determined based on the particle size distribution characteristics of the coarse-particle oil phase fluid, so that the retention efficiency of the fine-particle oil phase fluid in the first filter unit and the retention efficiency of the coarse-particle oil phase fluid in the second filter unit satisfy a preset efficiency balance relationship. During the filtration process, the rate of change of transmembrane pressure difference between the first filtration unit and the second filtration unit is monitored in real time, and the accumulation rate of retentate in each filtration unit is obtained as retentate information at each stage. Based on the cumulative rate of entrapment in the information of entrapment at each level, the clogging risk level of each filter unit is predicted. The backwashing triggering time and backwashing intensity parameters of the first filter unit and the second filter unit are set differently according to the clogging risk level. After the backwashing operation is performed, the flux recovery information is obtained by comparing the recovery ratio of transmembrane pressure difference before and after backwashing.

5. The method according to claim 4, characterized in that, The steps of predicting the clogging risk level of each filter unit and setting the backwash triggering timing and backwash intensity parameters differently based on the clogging risk level include: A time series analysis is performed on the accumulation rate of the retained material, and the first and second derivatives of the accumulation rate of the retained material are calculated. When the first derivative is continuously positive and the second derivative is greater than the acceleration threshold, the filter unit is determined to be in an accelerated clogging state. Establish a clogging risk level classification rule, and classify the clogging risk into multiple levels based on the comprehensive score of the numerical range of the accumulating rate of the intercepted material and the current value of the transmembrane pressure difference; establish a differentiated mapping relationship between the clogging risk level and the backwashing parameters, so that the backwashing trigger time interval shortens as the clogging risk level increases, and the backwashing intensity parameter increases as the clogging risk level increases. Record the transmembrane pressure differential recovery ratio after each backwashing operation. When the recovery ratio of multiple consecutive backwashes shows a downward trend, increase the clogging risk level of the filter unit and shorten the backwashing trigger time interval accordingly, forming a closed loop of dynamic assessment of clogging risk and adaptive adjustment of backwashing parameters.

6. The method according to claim 1, characterized in that, The steps of determining the extraction efficiency state based on the target component concentration and the anti-clogging performance state based on the flux recovery information, and adjusting the shear intensity, backwashing sequence, or a combination of both according to the establishment of the extraction efficiency state and the anti-clogging performance state to form an adaptive closed loop, include: Based on the target component concentration, an extraction efficiency evaluation index is constructed that includes the degree of concentration attainment and concentration time-series stability, and the extraction efficiency status is determined according to the extraction efficiency evaluation index; based on the flux recovery information, an anti-clogging performance evaluation index is constructed that includes the degree of recovery sufficiency and the ability to sustain recovery, and the anti-clogging performance status is determined according to the anti-clogging performance evaluation index. A four-quadrant decision matrix is ​​constructed to represent the extraction efficiency state and the anti-clogging performance state. When both states are true, the current parameters are maintained. When only the extraction efficiency state is true, the backwashing sequence is adjusted based on the anti-clogging performance evaluation index. When only the anti-clogging performance state is true, the shear intensity is adjusted based on the extraction efficiency evaluation index. When neither state is true, the adjustment weights of the shear intensity and the backwashing sequence are determined based on the degree of deviation between the extraction efficiency evaluation index and the anti-clogging performance evaluation index. The adjusted shear strength and backwashing timing are applied to the shearing treatment stage and the backwashing operation stage respectively, forming an adaptive closed loop.

7. The method according to claim 6, characterized in that, The steps for constructing a four-quadrant decision matrix for extraction efficiency state and anti-blocking performance state include: A two-dimensional decision space is established for the extraction efficiency state and the anti-blocking performance state. The combination of the two states being true and false is divided into four quadrant regions, and each quadrant region corresponds to a set of preset adjustment strategies. When both states are true, the system is positioned in the first quadrant and the parameter maintenance strategy is executed. When only the extraction efficiency state is established, the second quadrant is located, the recovery sustainability decay coefficient in the anti-clogging performance evaluation index is extracted, and the amount of shortening of the backwashing sequence is calculated based on the decay coefficient. When only the anti-blocking performance state is established, it is located in the third quadrant. The fluctuation range of concentration time stability in the extraction efficiency evaluation index is extracted. When the fluctuation range exceeds the stability threshold, the first shear strength of the high concentration region is increased. When the fluctuation range does not exceed the stability threshold but the concentration compliance level is lower than the target value, the second shear strength of the low concentration region is reduced. When neither of the two states is true, the system is positioned in the fourth quadrant. The deviation of the extraction efficiency evaluation index and the deviation of the anti-clogging performance evaluation index are calculated. The ratio of the two deviations is used as the adjustment weight allocation coefficient. Based on the adjustment weight allocation coefficient, the amount of shear strength enhancement and the amount of backwashing time shortening are determined simultaneously.

8. A precision control and processing system for the extraction tank feed based on automatic anti-clogging, used to implement the method of any one of claims 1-7, characterized in that, include: The first unit is used to obtain information on the oil phase distribution of the feed fluid to be treated and the target component concentration at the outlet of the extraction box; Collect time-series data on the particle size of oil phase droplets to identify the accelerated state of droplet coalescence; The second unit is used to identify high-concentration and low-concentration regions in the circulation loop based on the oil phase distribution information, apply a first shear strength treatment to the high-concentration region to obtain fine-particle oil phase fluid, apply a second shear strength treatment lower than the first shear strength to the low-concentration region to obtain coarse-particle oil phase fluid, and enhance the shear strength of the high-concentration region in advance when the droplet coalescence acceleration state is detected. The third unit is used to guide the fine-particle oil phase fluid to the first filtration unit and the coarse-particle oil phase fluid to the second filtration unit with a pore size larger than that of the first filtration unit, so as to obtain information on the intercepted substances at each stage. Perform backwashing on each filter unit to obtain flux recovery information; The fourth unit is used to determine the extraction efficiency status based on the target component concentration and the anti-clogging performance status based on the flux recovery information; according to the establishment of the extraction efficiency status and the anti-clogging performance status, the shear intensity, backwashing sequence or a combination of both are adjusted accordingly to form an adaptive closed loop.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images