Controllable filtration method based on surface property enhancement of porous materials

Abstract

The application discloses a controllable filtering method based on surface performance enhancement of porous materials and relates to the technical field of controllable filtering of porous materials, and comprises the following steps: on the basis of identifying that there is a surface active component in a liquid, generating an asymmetric liquid bridge evolution mapping based on a tension behavior identification matrix, determining whether an asymmetric contact liquid bridge is formed and local wetting deviation is caused by a three-dimensional contact front trajectory, a left-right contact area ratio and tension vector distribution; based on the asymmetric liquid bridge evolution mapping, a response deviation indication map is constructed, and the non-synergy between the wetting response change rate of the membrane hole edge, the liquid bridge contact center deviation angle and the import flow rate is calculated to identify whether the membrane hole is out of control. The application solves the problem of membrane hole opening and closing out of control caused by asymmetric contact liquid bridges, realizes local wetting deviation identification and hierarchical dynamic regulation, and improves filtering accuracy and system stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of controllable filtration of porous materials, and particularly relates to a controllable filtration method based on surface performance enhancement of porous materials. BACKGROUND

[0002] Controllable filtration based on surface performance enhancement of porous materials refers to modifying the surface of porous materials in a physical or chemical manner to improve its functionality in the filtration process, thereby achieving precise separation and control of target substances in terms of time, space or concentration. Existing related technologies usually improve the wettability, adsorbability, anti-pollution or selective permeability of the surface of porous materials by introducing hydrophilic / hydrophobic coatings, static charge modification, surface roughness control, and responsive polymer coatings (such as temperature-sensitive, pH-sensitive, and light-responsive materials). These modification methods can dynamically adjust the permeability of the filtration pores according to external environmental conditions (such as temperature, pH, electric field or light), thereby achieving "controllable" filtration behavior. The entire process usually includes the following key links: first, the selection and preparation of porous base materials, such as polymer membranes, ceramic foams or metal meshes; second, the implementation of surface modification treatment processes, including chemical grafting, plasma treatment, and nano-coating deposition technologies; third, the integrated design with the filtration system, so that the modified material can work cooperatively with the control module (such as sensors, power supplies, and micro-processing units); and finally, in the actual filtration process, the filtration behavior is adjusted and feedback optimized in real time by controlling external stimulation parameters.

[0003] The existing technology has the following disadvantages:

[0004] In the existing controllable filtration method based on surface performance enhancement of porous materials, a hydrophobic nanoparticle modification layer is usually introduced on the surface of the porous membrane to achieve the regulation of the liquid wetting behavior, thereby achieving controllable filtration of the opening and closing state of the pores. However, in the case of a small amount of surface active ingredients in the liquid, when the liquid droplet contacts the membrane surface, due to the uneven distribution of surface tension, an asymmetric contact liquid bridge is formed, which makes the liquid only establish a wetting connection on one side of the membrane pores, while the other side is still in a gas phase isolation state. This asymmetric wetting state will cause the local membrane pore edge to be wet-induced in advance, resulting in a mismatch of the opening and closing response direction of the membrane pores, and thus leading to the failure of pore control. The existing technology relies on the overall surface wetting state to determine the pore control behavior, and cannot accurately identify whether the membrane pores have lost control in terms of opening and closing according to the local wetting deviation phenomenon caused by the asymmetric contact liquid bridge on the membrane surface in the presence of surface active ingredients in the liquid. This may lead to the system judging that the membrane pores are still in a closed state in the control logic, while in fact a local permeation path has been formed, allowing non-target substances to penetrate, damaging the filtration selectivity, and increasing the risk of membrane contamination, thereby reducing the system stability and filtration reliability.

[0005] The above information disclosed in the Background section is only for enhancing the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art that is already known to those of ordinary skill in the art. SUMMARY

[0006] The purpose of the present application is to provide a controllable filtration method based on the performance enhancement of the surface of porous materials to solve the problems in the background.

[0007] In order to achieve the above-mentioned purpose, the present application provides the following technical solution: a controllable filtration method based on the performance enhancement of the surface of porous materials, specifically comprising the following steps:

[0008] S1, constructing a droplet interfacial tension behavior recognition matrix, collecting the diffusion speed of the droplet contact front tension gradient, the droplet profile change rate and the contact angle convergence path, for identifying whether there is a surface active component in the liquid;

[0009] S2, on the basis of identifying that there is a surface active component in the liquid, generating an asymmetric liquid bridge evolution mapping based on the tension behavior recognition matrix, determining whether an asymmetric contact liquid bridge is formed and causes local wetting deviation through three-dimensional contact front trajectory, left and right contact area ratio and tension vector distribution;

[0010] S3, based on the asymmetric liquid bridge evolution mapping, constructing a response deviation indication map, calculating the non-synergy between the membrane hole edge wetting response change rate, the liquid bridge contact center deviation angle and the imported flow rate, for identifying whether the membrane hole is out of control;

[0011] S4, based on the membrane hole state identified in the response deviation indication map, constructing an adaptive channel correction control mapping, outputting correction indicators corresponding to the membrane hole position, the out-of-control amplitude and the behavior level;

[0012] S5, based on the correction indicators of the adaptive channel correction control mapping, implementing graded dynamic regulation for the membrane hole area, including adjusting the surface energy distribution, the contact angle driving parameter and the liquid injection parameter, realizing real-time correction of the membrane hole opening and closing state.

[0013] Preferably, S1 specifically comprises the following steps:

[0014] Collecting a continuous image sequence of the droplet contacting the surface of the porous membrane, obtaining the diffusion speed change of the droplet contact front tension gradient between each time frame, forming a droplet contact front tension gradient diffusion speed data column;

[0015] Extracting the spatial change rate of the droplet edge profile boundary in the time dimension in the image sequence, and calculating the evolution trajectory of the convergence direction of the droplet bottom contact angle, forming a joint feature matrix of the droplet profile change rate and the contact angle convergence path;

[0016] The liquid drop contact front tension gradient diffusion velocity data column is uniformly normalized with the joint feature matrix, a liquid drop interface tension behavior identification matrix is constructed, and mode feature extraction is performed on the liquid drop interface tension behavior identification matrix. If it is identified that the tension diffusion velocity abnormally intensifies and there is a nonlinear coupling jump between the contact angle convergence path, it is determined that there is a surface active ingredient in the liquid.

[0017] Preferably, S2 specifically includes the following steps:

[0018] S201, on the basis of identifying the presence of a surface active ingredient in the liquid, reading the tension gradient values, edge spatial position values and contact angle change direction values of the liquid drop contact edge at different time points in the tension behavior identification matrix as the X-axis, Y-axis and Z-axis components of the three-dimensional space respectively, generating a three-dimensional point cloud set of the liquid drop contact behavior, and constructing a liquid bridge evolution mapping diagram;

[0019] S202, using an image processing algorithm to calculate the contact area pixel areas of the left and right sides of the liquid drop in each time frame respectively, obtaining the ratio between the left and right contact areas, and labeling the area ratio to the point at the corresponding time point in the three-dimensional liquid bridge evolution mapping diagram as a parameter of the contact imbalance degree;

[0020] S203, according to the direction change trend and amplitude difference of the tension vector at each time point in the tension behavior identification matrix, determining whether the three-dimensional point cloud is continuously offset in a single spatial direction, judging whether the left and right contact area ratio is continuously less than or greater than 1, and identifying whether the tension vector is concentrated in one direction. If the three-dimensional point cloud has a spatial offset trend, the left and right contact area ratio is not equal to 1 continuously, and the tension vector distribution has a single direction concentration, it is determined that the liquid drop forms an asymmetric contact liquid bridge and causes local wetting deviation.

[0021] Preferably, S202 specifically includes:

[0022] An image processing algorithm is applied to the image sequence of the liquid drop contacting the surface of the porous membrane for gray scale preprocessing and edge enhancement, the contact boundary profile of the liquid drop and the porous membrane in each frame of image is extracted, and the left and right contact areas are divided by the liquid drop symmetry axis;

[0023] Based on the extracted left and right contact area boundaries, the number of contact pixels in each area is counted respectively to obtain the left and right contact area pixel areas in each frame of image;

[0024] The pixel areas of the left and right contact areas in each frame of image are calculated by ratio to form an area ratio sequence, and each area ratio is labeled as a parameter to the point at the corresponding time point in the three-dimensional liquid bridge evolution mapping diagram, which is used to quantify the area asymmetry degree of the left and right contact areas of the liquid drop.

[0025] Preferably, S3 specifically comprises the following steps:

[0026] S301, extract the spatial coordinates of the wetting front of the membrane pore edge at each time point in the asymmetric liquid bridge evolution mapping, calculate the Euclidean distance change value between adjacent time points, form the membrane pore edge wetting response change rate sequence, which is used as the first coordinate dimension of the response shift indicator map;

[0027] S302, according to the liquid bridge contact center coordinates corresponding to each time point in the membrane pore edge wetting response change rate sequence, combined with the membrane pore center reference line, calculate the shift angle thereof, and construct the angle difference between consecutive time points as the liquid bridge contact center shift angle sequence, which is used as the second coordinate dimension of the response shift indicator map;

[0028] S303, collect the introduction flow rate value corresponding to each time point, construct the introduction flow rate sequence with the same time dimension as the membrane pore edge wetting response change rate sequence and the liquid bridge contact center shift angle sequence, and synchronously compare the first derivative change trend of the three sequences on the time axis. If the membrane pore edge wetting response change rate increases in a positive growth form in a continuous time interval, the angle between the liquid bridge contact center and the membrane pore center is continuously greater than the initial angle threshold in the continuous time interval, and the shift angle change rate is positive, while the first derivative value of the introduction flow rate is zero in the continuous time interval, it is determined that the three sequences have a non-synergistic evolution trend, thereby identifying that the membrane pore is in an open-close out-of-control state.

[0029] Preferably, S301 specifically comprises:

[0030] Extract the time stamp of each frame of image in the asymmetric liquid bridge evolution mapping and the three-dimensional spatial coordinates of the wetting front of the membrane pore edge at the corresponding time point, including the lateral position, the height position and the contact interface distance value, and sort them in time sequence;

[0031] Based on the three-dimensional spatial coordinates between adjacent time points, the Euclidean distance between each pair of consecutive points is calculated as the displacement of the wetting front in unit time, and the wetting response speed value corresponding to each time interval is obtained;

[0032] The wetting response speed values in all consecutive time intervals are constructed into a time sequence to form the membrane pore edge wetting response change rate sequence, and the sequence is set as the first coordinate dimension of the response shift indicator map for subsequent trend analysis and shift identification.

[0033] Preferably, S302 specifically comprises:

[0034] Based on each time point in the membrane pore edge wetting response change rate sequence, extract the spatial coordinates of the liquid bridge contact center in the corresponding frame of image, and extract the membrane pore center reference point coordinates, construct the spatial angle model of the liquid bridge contact center connecting line and the membrane pore center reference line;

[0035] The angle of deviation at each time point is calculated by using the three-dimensional vector angle formula, and the time-angle relationship mapping table is constructed by numerical normalization processing;

[0036] The change value of the angle of deviation between consecutive time points is extracted to form a sequence of the angle of deviation of the liquid bridge contact center, and the sequence is used as the second coordinate dimension of the response deviation indicator map for joint analysis of the membrane hole deviation trend with the first coordinate dimension.

[0037] Preferably, S4 specifically is:

[0038] Based on the sequence of the change rate of the membrane hole edge wetting response, the sequence of the angle of deviation of the liquid bridge contact center, and the sequence of the introduced flow rate in the response deviation indicator map, the spatial position coordinates, the wetting response speed range, the total amount of change of the angle of deviation, and the first derivative features of the corresponding flow rate of each membrane hole in the consecutive time interval are extracted, which are respectively set as the membrane hole position parameter, the out-of-control amplitude parameter, and the behavior level parameter;

[0039] The three-dimensional mapping tensor is constructed by taking the membrane hole position parameter, the out-of-control amplitude parameter, and the behavior level parameter as input variables, each mapping unit in the three-dimensional mapping tensor corresponds to a unique membrane hole state combination, and the adaptive correction response type corresponding to each state combination is defined in the three-dimensional mapping tensor;

[0040] Based on the membrane hole state combination located in the three-dimensional mapping tensor, the correction index corresponding to the combination is output, which includes three types of parameters: liquid injection rate adjustment factor, contact angle driving compensation coefficient, and surface energy local regulation threshold, which are used for subsequent hierarchical dynamic regulation of the membrane hole region.

[0041] Preferably, S5 specifically is:

[0042] According to the liquid injection rate adjustment factor output in the three-dimensional mapping tensor, the numerical adjustment of the driving control parameter of the liquid injection unit is performed, and the instantaneous flow rate of the liquid into the membrane hole region is finely adjusted by using the step motor controlled micropump, so as to realize the precise control of the liquid supply amount;

[0043] According to the contact angle driving compensation coefficient, the electric field intensity applied to the surface of the membrane hole region is adjusted in real time by the electrowetting control platform, the interfacial tension between the droplet and the membrane hole surface is dynamically changed, and then the contact angle change path is regulated, the liquid bridge shape is stabilized, and the wetting uniformity is improved;

[0044] According to the local regulation threshold of surface energy, the corresponding local thermal radiation power of the light-thermal response material excitation system output is controlled, the surface energy distribution state of the material in the target membrane hole peripheral region is changed, the spatial grading regulation of the local surface wetting ability is realized on the basis of maintaining the overall structure stability, the grading dynamic regulation in different states of the membrane hole region is realized, and the responsiveness and precision of the membrane hole opening and closing state correction are improved.

[0045] In the above technical solutions, the technical effects and advantages provided by the application are as follows:

[0046] 1、The application accurately identifies the existing state of the surface active ingredient in the liquid by constructing a liquid droplet interfacial tension behavior recognition matrix, and breaks through the limitation of traditional filtering methods that only rely on overall wetting behavior for judgment by combining an asymmetric liquid bridge evolution mapping diagram and a three-dimensional point cloud behavior modeling, thereby realizing structured modeling and dynamic identification of the local wetting deviation phenomenon caused by asymmetric contact liquid bridges for the first time.

[0047] 2、The application introduces a three-dimensional mapping tensor construction method based on membrane hole state parameters, realizes coupled modeling of the membrane hole position, wetting deviation amplitude and behavior level, and combines three types of correction indicators output by the tensor, i.e., a liquid injection rate adjustment factor, a contact angle driving compensation coefficient and a surface energy local regulation threshold, to establish an adaptive channel correction control mechanism with spatial grading, behavior typing and response fine-tuning capabilities. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art according to these drawings.

[0049] Figure 1 The flowchart of the controllable filtering method based on the surface performance enhancement of the porous material according to the present application is shown. Detailed Implementation

[0050] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0051] This invention provides, for example Figure 1 The controllable filtration method based on enhanced surface properties of porous materials, as shown, specifically includes the following steps:

[0052] S1. Construct a droplet interfacial tension behavior recognition matrix, collect the droplet contact leading edge tension gradient diffusion rate, droplet profile change rate and contact angle convergence path, and use it to identify whether there are surface-active components in the liquid.

[0053] In this embodiment, S1 specifically includes the following steps:

[0054] A continuous image sequence of droplet contact with the porous membrane surface is acquired to obtain the diffusion rate change of the tension gradient at the droplet contact front edge between time frames, forming a data series of the diffusion rate of the tension gradient at the droplet contact front edge.

[0055] To capture the diffusion rate changes of the droplet contact edge tension gradient across time frames during the droplet's contact with the porous membrane surface, a high-speed variable frame rate image acquisition device is first used to continuously capture the droplet's descent to contact process, ensuring sufficient temporal resolution to capture the dynamic evolution of the leading edge tension. By extracting the grayscale boundary of the droplet-membrane contact area in each frame, the real-time position of the droplet's leading edge can be identified. Combining the grayscale gradient distribution of the contact area with the changes in the leading edge boundary contour, a tension gradient distribution map of the droplet contact edge can be constructed using image differencing and gradient vector mapping. Based on this, the diffusion rate of the tension gradient per unit time is calculated using the inter-frame position changes and gradient amplitude changes, generating tension diffusion rate parameters frame by frame. Arranging the corresponding tension diffusion rate values ​​in each frame in chronological order, while maintaining consistency with spatial coordinates, forms a data series of droplet contact edge tension gradient diffusion rates. Each data point in this series is calculated from the rate of change of the local tension gradient at the leading edge in the image frame, reflecting the dynamic evolution of tension disturbance in the initial stage of droplet contact, and is a key foundation for subsequently building the recognition model.

[0056] Extract the spatial rate of change of the droplet edge contour boundary in the time dimension from the image sequence, and calculate the evolution trajectory of the contact angle convergence direction at the bottom of the droplet to form a joint feature matrix of the droplet contour change rate and the contact angle convergence path;

[0057] In analyzing the process of droplet contact with porous membrane, the droplet edge profile boundary in the image sequence can be extracted by an edge detection-based contour tracking algorithm, such as applying Canny edge operator combined with active contour model to locate the pixel-level droplet outer contour in each frame of image, extract the spatial position change of the contour boundary points in consecutive frames, and calculate the position increment in time dimension, so as to obtain the spatial change rate of the droplet edge profile boundary in time dimension. In order to calculate the evolution trajectory of the convergence direction of the droplet bottom contact angle, the contact point between the droplet and the membrane surface can be identified in each frame of image, the contact angles on both sides are measured, and the change direction and amplitude of the contact angle over time are recorded. Taking a liquid containing a surface active component as an example, in the initial contact stage, the left contact angle may gradually decrease while the right contact angle is relatively stable, and then the convergence direction evolution trajectory shows a continuous trajectory line with the angle shifting to the left. By synchronously combining the droplet profile change rate data extracted in each frame of image with the time sequence coordinates of the contact angle convergence direction, a joint feature matrix of the droplet profile change rate and the contact angle convergence path can be constructed, wherein each row corresponds to a time node, and each column represents the quantization parameters of the boundary change rate and the angle convergence direction, respectively. The matrix can comprehensively describe the coupled dynamic process of the droplet shape and the contact behavior, and provide accurate physical feature basis for subsequent identification model.

[0058] The droplet contact front tension gradient diffusion velocity data column and the joint feature matrix are uniformly normalized to construct a droplet interface tension behavior identification matrix, and the pattern feature of the droplet interface tension behavior identification matrix is extracted. If it is identified that the tension diffusion velocity abnormally intensifies and there is a nonlinear coupling jump between the contact angle convergence path, it is determined that there is a surface active component in the liquid.

[0059] In order to realize the identification of the surface active component in the liquid, the droplet contact front tension gradient diffusion velocity data column and the joint feature matrix composed of the droplet profile change rate and the contact angle convergence path need to be uniformly analyzed. Since there are significant differences in the physical quantity level, value range and unit between the two types of data, the data need to be normalized first. Common methods include Z-score standardization or minimum-maximum linear mapping, so as to ensure the consistency of feature weights in numerical calculation between different dimensions. By aligning the time axis of the normalized tension diffusion velocity data column and the joint feature matrix, and fusing according to the time node, a multi-channel droplet interface tension behavior identification matrix is obtained, wherein each row represents the comprehensive behavior state at a time, and each column represents the normalized value of a certain physical feature. The matrix is used as a high-dimensional state feature input for subsequent identification of whether there is an interface disturbance caused by a surface active component.

[0060] The recognition process adopts an anomaly detection method based on pattern recognition, such as convolution feature extraction combined with principal component mapping, to extract the interactive coupling features between the tension diffusion and the contact angle convergence path by recognizing the cooperative change trend between different channels in the matrix. When the tension diffusion speed continuously rises and the contact angle convergence path presents a nonlinear jump trajectory in a short time interval, that is, the angle direction suddenly changes by more than a set threshold in consecutive frames, and both of them show a coupling mutation behavior in the feature space, it can be judged that the droplet behavior is affected by the interfacial active substance. The key of this recognition method is to include the time continuity and the interactive nature of physical quantities into the judgment basis by constructing the behavior mapping relationship, thereby avoiding the misjudgment caused by single numerical anomaly and realizing high-precision judgment of the existence state of the surface active component in the liquid.

[0061] S2, on the basis of recognizing the existence of the surface active component in the liquid, generating an asymmetric liquid bridge evolution mapping based on the tension behavior recognition matrix, determining whether an asymmetric contact liquid bridge is formed and causes local wetting deviation by the three-dimensional contact front trajectory, the left and right contact area ratio, and the tension vector distribution;

[0062] In this embodiment, S2 specifically includes the following steps:

[0063] S201, on the basis of recognizing the existence of the surface active component in the liquid, reading the tension gradient value, the edge space position value and the contact angle change direction value of the droplet contact edge in the tension behavior recognition matrix at different time points as the X-axis, Y-axis and Z-axis components of the three-dimensional space respectively, generating a three-dimensional point cloud set of the droplet contact behavior, and constructing a liquid bridge evolution mapping;

[0064] In order to realize the visualization and dynamic recognition of the formation process of the asymmetric liquid bridge, the key physical behaviors of the droplet in the contact evolution process on the surface of the porous membrane need to be modeled and expressed in three-dimensional space. By calling the continuous time frame data in the tension behavior recognition matrix, the tension gradient intensity of the droplet contact edge at different times, the planar projection position of the edge in the spatial coordinate system, and the vector angle of the contact angle change direction are extracted, and are respectively assigned as the X-axis, Y-axis and Z-axis coordinate components of the three-dimensional coordinate axis. By sequentially organizing these time series data into a three-dimensional point cloud set, the dynamic evolution trend of the contact boundary between the droplet and the membrane surface can be intuitively displayed. The point cloud set not only constructs the spatial deviation path of the liquid bridge caused by tension disturbance in the contact process, but also reveals the structural evolution trajectory of the asymmetric wetting gradually formed, thereby providing a morphological basis for subsequent uneven feature labeling and discrimination.

[0065] The tension gradient value is obtained by calculating the rate of change of the pixel intensity boundary at the front of the droplet over time, and is obtained by analyzing the difference between the gray values of adjacent boundary pixels in the high-frequency frame image and mapping it using a tension conversion function. The edge spatial position value is based on the two-dimensional coordinate information of the droplet contact boundary in the image plane, which is projected to the reference coordinate system of the membrane surface after sub-pixel level edge fitting, to obtain the edge point displacement information in the X and Y axis directions. The contact angle change direction value is obtained by tracking the change trend of the left and right sides of the contact angle at the bottom of the droplet, combining the slope and evolution direction of the contact angle convergence path, fitting the angle change trajectory into a vector, and mapping it into the Z-axis coordinate value of the three-dimensional point cloud. These three types of values are derived from high-resolution image processing and continuous time series analysis, and have objective measurability and parameter consistency, which are the core data basis for constructing the liquid bridge evolution mapping graph.

[0066] S202, using image processing algorithms to calculate the contact area pixel area of the left and right sides of the droplet in each time frame, respectively, to obtain the ratio between the left and right contact areas, and to mark the area ratio to the corresponding time point on the three-dimensional liquid bridge evolution mapping graph as a parameter of the contact imbalance degree;

[0067] S203, according to the direction change trend and amplitude difference of the tension vector at each time point in the tension behavior recognition matrix, determine whether the three-dimensional point cloud is continuously offset in a single spatial direction, whether the left and right contact area ratio is continuously less than or greater than 1, and whether the tension vector is concentrated in one direction, if the three-dimensional point cloud has a spatial offset trend, the left and right contact area ratio is not equal to 1 continuously, and the tension vector distribution is concentrated in one direction, it is determined that the droplet forms an asymmetric contact liquid bridge and causes local wetting deviation.

[0068] In the process of judging whether the droplet forms an asymmetric contact liquid bridge and causes local wetting deviation, the multi-dimensional data in the tension behavior recognition matrix need to be analyzed. The direction and amplitude of the tension vector are extracted from the data at consecutive time points, and an evolution trend curve is constructed using the vector change path to determine whether the trend is continuously offset in a single spatial direction. If the offset direction is stable, it means that the droplet contact behavior has a directional deviation in space. At the same time, combined with the area ratio sequence, if the ratio continuously deviates from 1 in multiple consecutive frames, it reflects that the left and right contact areas are in an unbalanced state for a long time. Further, through density analysis of the spatial distribution of the tension vector, if most of the vectors are concentrated in one direction, it means that the local tension driving has a single direction. When the three types of indicators meet the requirements in the time dimension, it indicates that the droplet has formed a biased wetting trend in the dynamic contact process, and the formation of the asymmetric contact liquid bridge can be determined accordingly, and it is considered that it has caused local wetting deviation, leading to inconsistent response of the membrane hole edge.

[0069] The tension behavior recognition matrix contains the liquid drop tension gradient vector information corresponding to different time points, the direction change trend is realized by tracking the rotation track of the vector in the three-dimensional coordinate system, and the amplitude difference is obtained by standard deviation statistics on the vector length. The three-dimensional point cloud is the mapping coordinate set of the tension behavior at different time nodes. If the set shows a continuous deviation trend from the original central track in space, it is considered to have deviated. The continuous inequality of the area ratio to 1 indicates that the wetting center of gravity deviates, and the aggregation direction of the tension vector points to a stable deviation behavior. The aggregation of the tension vector is judged by the local density function, and the vector clustering algorithm based on kernel density estimation is usually used for processing. The cross verification of the three judgments enhances the robustness and accuracy of the judgment, and provides accurate support for identifying the asymmetric wetting state.

[0070] In this embodiment, S202 is specifically:

[0071] An image processing algorithm is applied to the image sequence of the liquid drop contacting the surface of the porous membrane to perform gray scale preprocessing and edge enhancement, the contact boundary profile of the liquid drop and the porous membrane in each frame of image is extracted, and the left and right contact regions are divided based on the liquid drop symmetry axis;

[0072] In order to identify the unbalanced behavior between the liquid drop and the porous membrane during the contact process, image processing is required for the whole process of the liquid drop contacting the membrane surface. The acquired liquid drop contact image sequence is processed frame by frame by using an image processing algorithm to remove light interference and enhance edge features. The contact area boundary curve of the liquid drop profile and the porous membrane boundary is extracted by an edge enhancement operator, and based on the symmetry of the liquid drop shape structure in the image, the liquid drop symmetry axis is determined to divide the liquid drop contact area into left and right. This processing method can realize the spatial behavior division of the liquid drop contact area in the time sequence image, provide structured input for subsequent contact area ratio calculation and asymmetric evolution analysis, and is a necessary data preparation link for building a spatial evolution model.

[0073] The image processing algorithm can be constructed by integrating the Canny edge detection operator, the Sobel filter and the edge enhancement model based on the Gaussian kernel function, and is usually realized by customizing the image preprocessing process based on OpenCV. The liquid drop symmetry axis can be determined by the projection center point of the liquid drop top end maximum bending point in the image vertical axis direction, and the long axis direction of the profile circumscribed ellipse is combined for secondary correction to form a vertical line consistent with the long axis direction of the liquid drop profile as the division line. This symmetry axis division method can maximize the restoration of the left-right structure symmetry in the natural shape of the liquid drop, ensure the physical meaning uniformity of the left and right contact regions, and ensure the direction consistency and data stability of the area ratio calculation.

[0074] Based on the extracted left and right contact area boundaries, the number of contact pixels in each area is counted respectively to obtain the pixel area of the left contact area and the pixel area of the right contact area in each frame of image;

[0075] In order to accurately quantify the difference in wetting behavior between the left and right sides of the droplet during contact with the porous membrane, the number of effective contact pixels in each area needs to be counted based on the left and right contact area boundaries obtained through image processing. The region filling algorithm (such as Flood Fill or mask filling based on contour closure) can be used to mark the pixel area within the left and right contact boundaries as highlighted areas in the binary image. By traversing the image frame by frame, the total number of white pixel points in the left and right mask areas in each frame is counted, and the pixel area of the left contact area and the pixel area of the right contact area are calculated. For example, if in a certain frame of image, the left contact area is marked as 2450 pixel points, and the right contact area is marked as 1670 pixel points, it means that the left contact area of the droplet with the porous membrane is larger than the right contact area, and there is a wetting asymmetry trend. This pixel area counting method can truly reflect the spatial distribution characteristics of local wetting expansion during contact, and is the basic data source for constructing the contact imbalance parameter and identifying the asymmetric liquid bridge evolution state. The accuracy of the above statistical behavior depends on the accuracy of boundary extraction and region division, so the integrity of image preprocessing has a direct impact on this process.

[0076] The pixel areas of the left and right contact areas in each frame of image are calculated by ratio, and the area ratio sequence is constructed. Each area ratio is marked as a parameter on the corresponding time point in the three-dimensional liquid bridge evolution mapping, which is used to quantify the area asymmetry degree of the left and right contact areas of the droplet.

[0077] In order to continuously track the dynamic evolution of the left and right wetting behaviors during the contact between the droplet and the porous membrane, the pixel areas of the left and right contact regions in each frame of image are compared to form an area ratio sequence that can reflect the symmetry of the contact. The ratio can be obtained by dividing the left contact pixel area by the right contact pixel area in each frame of image, and a value equal to 1 indicates symmetric contact, and a value less than 1 or greater than 1 indicates asymmetric contact. For example, if the left area is 2200 pixels and the right area is 1100 pixels in a frame of image, the area ratio of the frame is 2, indicating that the droplet mainly expands to the left. The ratio sequence is arranged in the order of time frames and mapped to the points corresponding to the time points in the three-dimensional liquid bridge evolution diagram, and used as an additional scalar parameter of the points for accurately quantifying the imbalance degree of the contact behavior of the droplet in the three-dimensional space. This process not only provides the contact structure information of the time sequence evolution, but also can be used to capture the formation trend of the asymmetric wetting state, which has a basic supporting significance for subsequent judgment of whether there is local wetting deviation. The stability of the area ratio is closely related to the accuracy of image processing, so the consistency and accuracy of the region division and pixel recognition need to be ensured.

[0078] S3, based on the asymmetric liquid bridge evolution mapping diagram, constructing a response deviation indication map, by calculating the non-synergy between the wetting response change rate of the membrane hole edge, the liquid bridge contact center deviation angle and the introduction flow rate, for identifying whether the membrane hole is out of control of opening and closing;

[0079] In this embodiment, S3 specifically includes the following steps:

[0080] S301, extracting the spatial coordinates of the membrane hole edge wetting front edge of each time point in the asymmetric liquid bridge evolution mapping diagram, calculating the Euclidean distance change value between adjacent time points to form a membrane hole edge wetting response change rate sequence, which is used as the first coordinate dimension of the response deviation indication map;

[0081] S302, according to the liquid bridge contact center coordinates corresponding to each time point in the membrane hole edge wetting response change rate sequence, combining the membrane hole center reference line to calculate the deviation angle thereof, and constructing the angle difference between the continuous time points as a liquid bridge contact center deviation angle sequence, which is used as the second coordinate dimension of the response deviation indication map;

[0082] S303, collect the import flow rate value corresponding to each time point, construct the import flow rate sequence with the same time dimension as the membrane hole edge wetting response change rate sequence and the liquid bridge contact center offset angle sequence, and compare the first derivative change trend of the three groups of sequences on the time axis. If the membrane hole edge wetting response change rate increases in a positive growth form in a continuous time interval, the angle between the liquid bridge contact center and the membrane hole center is continuously greater than the initial angle threshold in the continuous time period and the offset angle change rate is positive, and the first derivative value of the import flow rate is zero in the continuous time interval, it is determined that the three groups of sequences have a non-synergistic evolution trend, thereby identifying that the membrane hole is in an open-close out-of-control state.

[0083] In the membrane hole state identification process, the import flow rate value corresponding to the time stamp of each frame of image is collected, a time sequence is constructed, the time axis of which is consistent with the membrane hole edge wetting response change rate sequence and the liquid bridge contact center offset angle sequence, thereby forming three time-synchronized data sets. The first derivative of the three groups of sequences on the time axis is calculated respectively to obtain the rate trend of each with time. The first derivative of the import flow rate indicates whether the liquid input condition changes, the first derivative of the membrane hole edge wetting response change rate reflects the dynamic acceleration degree of the wetting front, and the first derivative of the liquid bridge contact center offset angle is used to monitor the development speed of the spatial offset. By synchronously analyzing the three derivative sequences, if the membrane hole edge wetting response change rate shows a continuous positive growth trend in a continuous time interval, the angle between the liquid bridge contact center and the membrane hole center is always greater than the initial angle threshold and the offset angle change rate is positive, that is, the spatial offset continues to increase, and the first derivative of the import flow rate is zero, that is, the liquid injection does not change, it can be determined that this response behavior is not caused by external input, but an abnormal out-of-control performance of the internal response logic of the system. This non-synergistic evolution trend means that the membrane hole has failed to maintain normal opening and closing response based on the input logic, and there is a potential risk of opening and closing out-of-control, which needs to trigger the regulation mechanism for intervention and correction. This method breaks through the limitation of single parameter in accurately identifying complex membrane surface out-of-control behavior by establishing a dynamic synergistic analysis mechanism among multiple behavior variables.

[0084] In this embodiment, S301 is specifically:

[0085] The time stamp of each frame of image in the asymmetric liquid bridge evolution mapping and the three-dimensional spatial coordinates of the membrane hole edge wetting front at the corresponding time point are extracted, including the lateral position, the height position and the contact interface distance value, and are sorted in time sequence;

[0086] The timestamp of each frame of image in the asymmetric liquid bridge evolution mapping and the three-dimensional spatial coordinates of the wetting front of the membrane hole edge at the corresponding time point can be obtained by collecting high-frame-rate image sequences during droplet contact and embedding accurate timestamps for each frame of image to establish a one-to-one correspondence between the image and the time point. Each frame of image is identified by the wetting front area of the membrane hole edge through image segmentation algorithm, and the three-dimensional coordinate information of the area is extracted by three-dimensional reconstruction method, which is represented as lateral position, vertical height and relative distance to the droplet contact interface. The lateral position can be calibrated by the number of pixel rows in the image plane, the vertical height can be restored by combining the optical focal plane displacement or stereo vision measurement, and the contact interface distance value is obtained by scaling the pixel spacing between the droplet boundary and the membrane surface. After combining the three-dimensional coordinate data of all frames with the corresponding timestamps and sorting them in chronological order, a time-continuous wetting front three-dimensional coordinate sequence is formed, which is used for subsequent calculation of the wetting response rate of the membrane hole edge and supports the dynamic identification of local wetting behavior.

[0087] Based on the three-dimensional spatial coordinates between adjacent time points, the Euclidean distance between each pair of consecutive points is calculated as the displacement of the wetting front per unit time, and the wetting response speed value corresponding to each time period is obtained;

[0088] The Euclidean distance between each pair of consecutive points based on the three-dimensional spatial coordinates between adjacent time points can be realized by performing spatial difference operation on the three-dimensional coordinate data extracted from the wetting front of the membrane hole edge at consecutive time points. Let the coordinates of the two consecutive time points be point and point , the three-dimensional Euclidean distance formula is used to calculate the wetting front displacement distance in this time period, and then divided by the time interval Δt between the two time points to obtain the weting response speed value in this time period. The whole process can be automatically completed by programming to call high-precision floating-point operation interface and build speed sequence. Taking the coordinates of the wetting front of the droplet at time as (0.2, 0.5, 0.1), and the coordinates at time as (0.3, 0.55, 0.15), and the time interval as 0.01 seconds as an example, the three-dimensional displacement is , and the corresponding speed is 0.122 ÷ 0.01 = 12.2 units / sec. This calculation method ensures that the wetting response behavior has a quantitative expression basis in space and time, and provides accurate data support for building the wetting response rate sequence of the membrane hole edge.

[0089] The wetting response speed values in all consecutive time periods are combined to form a time sequence, which forms the wetting response rate sequence of the membrane hole edge, and the sequence is set as the first coordinate dimension of the response offset indication map for subsequent trend analysis and offset identification.

[0090] The wetting response speed values calculated in all continuous time periods are arranged in chronological order to form a one-dimensional numerical sequence with a clear time index, that is, the wetting response rate sequence of the membrane hole edge. The sequence can be realized by programming language to build a time series object, and each element is composed of a corresponding timestamp and a wetting speed value. This sequence not only records the dynamic evolution law of the membrane hole edge wetting front with time during the whole droplet contact process, but also provides an analysis basis for identifying trend fluctuations, mutation characteristics and nonlinear behavior in wetting behavior. When constructing the response offset indicator map, the sequence can be used as the first coordinate dimension, which can be mapped with other behavior metrics (such as offset angle, flow rate change) to perform multi-dimensional visual analysis of local behavior. As the horizontal or vertical axis input, the wetting response rate can not only show the acceleration and deceleration trend of the wetting response, but also identify whether the membrane hole is in an out-of-control state through the synergy or deviation relationship with other physical quantities, so it has key data organization and analysis value.

[0091] In this embodiment, S302 is specifically:

[0092] Based on each time point in the wetting response rate sequence of the membrane hole edge, the spatial coordinates of the liquid bridge contact center in the corresponding frame image are extracted, and the membrane hole center reference point coordinates are extracted to construct a spatial angle model of the liquid bridge contact center connecting line and the membrane hole center reference line.

[0093] At each time point, the image of the liquid bridge contact area is subjected to feature point extraction to obtain the spatial coordinates of the liquid bridge contact center, which can be obtained by identifying the barycentric position of the contact contour boundary of the droplet bottom and the membrane hole. At the same time, the spatial coordinates of the geometric center of the membrane hole extracted from the initial membrane structure design are taken as the membrane hole center reference point. Based on the spatial position relationship between the liquid bridge contact center and the membrane hole center reference point, a spatial vector is constructed as the liquid bridge contact center connecting line, and a reference direction vector is constructed as the membrane hole center reference line in the direction of the membrane hole normal or the membrane thickness. The angle between the two vectors is calculated by the three-dimensional vector angle formula to establish a spatial angle model between the liquid bridge contact center connecting line and the membrane hole center reference line. This model can quantitatively represent whether the liquid bridge wetting path deviates from the membrane hole axis direction with an angle value, thereby revealing whether the membrane hole wetting evolves asymmetrically, judging the matching degree of the droplet contact offset and the opening and closing response, and further laying a spatial analysis foundation for subsequent construction of the liquid bridge contact center offset angle sequence. The significance of constructing this angle model is to convert the complex wetting behavior into a spatial angle change problem, improving the accuracy and traceability of local offset identification.

[0094] The angle between the liquid bridge contact center and the membrane hole center is traversed in time sequence, the offset angle at each time point is calculated by using the three-dimensional vector angle formula, and numerical normalization processing is performed to construct a time-angle relationship mapping table;

[0095] First, according to the time sequence of the image acquisition sequence, the spatial coordinates of the liquid bridge contact center and the spatial coordinates of the membrane hole center reference point are extracted frame by frame, and the liquid bridge direction vector corresponding to each frame of image in the time sequence is constructed. At the same time, the membrane hole normal direction vector in each frame of image is extracted as a unified reference direction. The included angle between the two vectors is calculated by using the three-dimensional vector included angle formula θ = arccos[(A·B) / (|A||B|)], wherein A is the liquid bridge contact direction vector, B is the membrane hole reference direction vector, · represents the dot product of the vectors, |A| and |B| are the lengths of the vectors, respectively. The included angle between the two vectors is the liquid bridge deflection angle, which reflects the degree of deflection of the liquid bridge from the center of the membrane hole. In order to make the deflection angles at different time points comparable, the calculated angle values need to be linearly normalized to fall within a unified numerical interval. Finally, each time point and the corresponding normalized angle value are constructed as a time-angle mapping pair to form a time-angle relationship mapping table, which is the basis for judging the deflection dynamic evolution trend of the liquid bridge. On this basis, it can be identified whether the deflection angle has a sustained growth or a periodic oscillation behavior, so as to assist in judging whether the membrane hole has entered a response out-of-control state.

[0096] The deflection angle change value between consecutive time points is extracted to form a liquid bridge contact center deflection angle sequence, and the sequence is used as the second coordinate dimension of the response deflection indication map for joint analysis of the membrane hole deflection trend with the first coordinate dimension.

[0097] After constructing the time-angle relationship mapping table, the deflection angle of the liquid bridge contact center corresponding to each pair of adjacent time points is read in turn, and the difference value is calculated to obtain the change amplitude of the deflection angle in that time period. The deflection angle change values between all consecutive time points are arranged in time sequence to form a liquid bridge contact center deflection angle sequence. The sequence reflects the dynamic change trend of the liquid bridge deflection angle with time, and can reveal whether the liquid bridge has a sustained deflection, a sudden deflection or an oscillation deflection behavior pattern. The deflection angle sequence is used as the second coordinate dimension of the response deflection indication map, and the membrane hole edge wetting response change rate sequence forms a two-dimensional joint analysis coordinate system, so that the identification of the membrane hole opening and closing state not only depends on a single angle change or displacement speed, but is based on a comprehensive judgment of the coupling trend of multiple features, thereby improving the identification accuracy and the judgment sensitivity of the opening and closing out-of-control. This method can early warn potential response deflection abnormalities when the membrane hole has not yet opened completely, effectively enhancing the feedforward ability of the control system.

[0098] S4, based on the membrane hole state identified in the response deflection indication map, constructing an adaptive channel correction control mapping, and outputting correction indicators corresponding to the membrane hole position, out-of-control amplitude and behavior level;

[0099] In this embodiment, S4 is specifically:

[0100] Based on the sequence of the change rate of the wetting response of the membrane hole edge, the sequence of the contact center offset angle of the liquid bridge, and the sequence of the introduction flow rate identified in the response offset indication map, the spatial position coordinates of each membrane hole in the continuous time interval, the wetting response speed range, the total amount of angle change, and the first derivative of the corresponding flow rate are extracted, which are respectively set as the membrane hole position parameter, the out-of-control amplitude parameter, and the behavior level parameter.

[0101] The sequence of the change rate of the wetting response of the membrane hole edge, the sequence of the contact center offset angle of the liquid bridge, and the sequence of the introduction flow rate are recorded in a unified time dimension, which can be used for joint analysis of the dynamic behavior characteristics of the membrane hole in the local area. By extracting the three-dimensional space trajectory of each membrane hole in the continuous time interval, its corresponding spatial position coordinates can be obtained for subsequent positioning identification. Based on the sequence of the change rate of the wetting response, the difference between the maximum response speed and the minimum response speed of the membrane hole in the interval is extracted as the wetting response speed range to represent the fluctuation amplitude of the out-of-control dynamics. The total amount of angle change in the sequence of the contact center offset angle of the liquid bridge at each time point can be used to evaluate the offset degree, thereby forming the total amount of angle change parameter. At the same time, the first derivative value of each time point in the sequence of the introduction flow rate can be used to analyze the flow rate stability, extract the derivative range, change mode, and other characteristics, and combine the offset trend to determine the behavior level. Finally, the spatial position coordinates, the wetting response speed range, the total amount of angle change, and the first derivative of the introduction flow rate are defined as the membrane hole position parameter, the out-of-control amplitude parameter, and the behavior level parameter, respectively, which are used as input to the control model.

[0102] In specific implementation, first, the wetting response speed value, the contact center offset angle, and the introduction flow rate value of each membrane hole at each time in the response offset indication map are obtained, and a corresponding time axis of the three numerical sequences is established using a programming tool (such as Python combined with OpenCV and NumPy). By traversing the values of each membrane hole in a time window of 10 images, the average value of the center coordinates is calculated as the spatial position coordinates of the membrane hole, the difference between the maximum value and the minimum value in the wetting response speed sequence is calculated as the wetting response speed range, the total amount of angle change is accumulated to obtain the total amount of angle change, and the first derivative of the introduction flow rate sequence is calculated to extract the maximum, minimum, and fluctuation mode. After all the extracted parameters are standardized, they are input into the adaptive channel correction control mapping model for identifying and quantifying the current membrane hole state.

[0103] The membrane hole position parameter, the out-of-control amplitude parameter, and the behavior level parameter are used as input variables to construct a three-dimensional mapping tensor, and each mapping unit in the three-dimensional mapping tensor corresponds to a unique membrane hole state combination. In the three-dimensional mapping tensor, the adaptive correction response type corresponding to each state combination is defined.

[0104] The membrane pore position parameter, the out-of-control amplitude parameter, and the behavior level parameter respectively represent the working state of the membrane pore from the three dimensions of space, dynamic amplitude, and behavior trend. These three parameters form a three-dimensional variable space, which can be used for classification and control response mapping of the membrane pore state. On this basis, a three-dimensional mapping tensor is constructed, and each dimension corresponds to a discrete interval of parameter value. For example, the membrane pore position is divided into multiple spatial grids, the out-of-control amplitude is divided into low, medium, and high levels, and the behavior level is divided into stable, slight deviation, and serious deviation. Each mapping unit of the three-dimensional mapping tensor represents a specific parameter combination, which is a state node. The node stores a predefined adaptive correction response type. The adaptive correction response type includes pressure adjustment, flow rate increase or decrease, local closing or opening control, etc., which can be highly matched with the actual state of the membrane pore to achieve targeted regulation. Through this structured mapping method, the corresponding correction strategy can be quickly retrieved from the complex parameter combination, improving the accuracy and real-time performance of the correction response.

[0105] In specific implementation, the three groups of parameters can be binned by setting parameter interval division rules, and three-dimensional array or tensor structure is used for storage and management of mapping relationship. For example, a three-dimensional matrix is constructed using the NumPy library in Python, and each dimension corresponds to the discrete value range of the position parameter, the out-of-control amplitude parameter, and the behavior level parameter. For each specific parameter combination, the corresponding correction response is set in the tensor in advance. For example, the unit with spatial position in the upper left area, high out-of-control amplitude, and serious deviation behavior level is set as “local pressure closing + feedback calibration”, while the middle area with slight deviation state corresponds to “maintain current flow rate + dynamic observation”. After identifying the membrane pore state parameter combination, the system can locate the corresponding unit in the tensor and call the response strategy stored in that position, thereby realizing a control output mechanism with clear structure, complete logic, and fast response.

[0106] Based on the membrane pore state combination located in the three-dimensional mapping tensor, the correction index corresponding to the combination is output, including three types of parameters: liquid injection rate adjustment factor, contact angle driving compensation coefficient, and surface energy local adjustment threshold, which are used for subsequent hierarchical dynamic regulation of the membrane pore region.

[0107] Each membrane pore state combination unit in the three-dimensional mapping tensor stores a set of correction indicators matching the current membrane pore state. After identifying the specific state combination of the membrane pore, the system corresponds to a unique position in the three-dimensional mapping tensor, and extracts the pre-set correction indicators in this position by table lookup. The correction indicators include three types of parameters: liquid injection rate adjustment factor, contact angle driving compensation coefficient, and surface energy local adjustment threshold. The liquid injection rate adjustment factor is used to accurately control the liquid inflow speed at the target membrane pore to alleviate the dynamic instability caused by sudden changes in flow rate; the contact angle driving compensation coefficient is used to apply a small amount of adjustment to the wetting tension of the droplet on the membrane pore surface, affecting the droplet deformation and spreading path; the surface energy local adjustment threshold is used to control the surface energy distribution of the material in the peripheral area of the membrane pore, thereby regulating the local wetting behavior. These three types of parameters together constitute the core regulation mechanism of membrane pore control, realizing adaptive intervention on the local state. The output correction indicators not only have pertinence and quantifiability, but also have a highly matched response mechanism with the membrane pore state combination, improving the self-repairing and dynamic regulation ability of the system.

[0108] In specific implementation, by binding the three-dimensional mapping tensor with the correction indicator table in an index mapping manner, after identifying the membrane pore state parameter combination, the three-dimensional index of the state combination is used as a query key to directly call the corresponding correction parameters. For example, the liquid injection rate adjustment factor can control the driving frequency of the micropump through a numerical coefficient, realizing nanoliter-level flow rate adjustment; the contact angle driving compensation coefficient can dynamically change the surface tension driving level by controlling the voltage amplitude of the electrowetting platform; the surface energy local adjustment threshold can correspondingly regulate the heating power of the photo-thermal responsive material in a specific area, thereby inducing local surface energy changes. The output process of the entire correction indicator can be automatically completed by the decision engine embedded in the control system, completing identification, mapping and execution preparation within milliseconds, ensuring that the dynamic regulation of the membrane pore area is both timely and accurate.

[0109] S5, based on the adaptive channel correction control mapping, implements hierarchical dynamic regulation of the membrane pore area, including adjusting the surface energy distribution, the contact angle driving parameter and the liquid injection parameter, to realize real-time correction of the membrane pore opening and closing state.

[0110] In this embodiment, S5 is specifically:

[0111] According to the liquid injection rate adjustment factor output in the three-dimensional mapping tensor, the numerical adjustment of the driving control parameter of the liquid injection unit is performed, and the instantaneous flow rate of the liquid into the membrane pore area is finely adjusted by using the step motor controlled micropump, to realize precise control of the liquid supply amount;

[0112] In the liquid transport system, the liquid injection rate adjustment factor is one of the correction parameters output as a three-dimensional mapping tensor, reflecting the adjustment amplitude of the required liquid supply under the current membrane hole state. In order to realize the response control of this parameter, it is necessary to map this factor as a control input to the drive control module of the liquid injection unit. The drive control module usually adopts a step-by-step electric control method, which accurately adjusts the working rhythm of the micropump by controlling the pulse frequency and duty cycle, thereby realizing the fine-grained adjustment of the liquid injection rate. The step-by-step electric control micropump has high responsiveness and high repeatability, and can dynamically control the liquid flow rate with microliter accuracy, ensuring real-time controllability of liquid supply in the membrane hole area. With the assistance of the feedback mechanism of the real-time fluid behavior monitoring system, this adjustment method can accurately respond to the rapid changes of the membrane hole state, preventing insufficient liquid supply or local liquid accumulation from interfering with the wetting process.

[0113] In a specific implementation, the control system first indicates that the current liquid flow rate should be reduced by 15% according to the liquid injection rate adjustment factor obtained from the three-dimensional mapping tensor, for example 0.85. The control instruction will be transmitted to the drive unit of the step-by-step electric control micropump, reducing the original set pulse frequency from 1000Hz to 850Hz, and adjusting the pulse duty cycle in real time to ensure that the pump cavity opening time and closing interval meet the new target rate. The entire control logic is closed-loop processed in the local control chip and directly controls the micropump execution module through the high-speed digital-to-analog conversion interface. The system will perform feedback correction according to the flow value collected by the actual flow rate sensor, ensuring that the adjusted flow rate is accurately and stably maintained within the target rate range, thereby maintaining the control accuracy and response timeliness of the membrane hole wetting state.

[0114] According to the contact angle driving compensation coefficient, the electric field intensity applied to the surface of the membrane hole area is adjusted in real time through the electrowetting control platform, dynamically changing the interfacial tension between the droplet and the membrane hole surface, thereby regulating the contact angle change path, stabilizing the liquid bridge shape and improving the wetting uniformity;

[0115] In the wetting regulation system based on the electrowetting effect, the contact angle driving compensation coefficient is a key correction index output in the three-dimensional mapping tensor, used to indicate the demand intensity of the current membrane hole area contact angle regulation. The compensation coefficient corresponds to the adjustment factor of the electric field intensity in the electrowetting control platform through the mapping relationship, and then drives the electric field output module to apply different intensity local electric fields to the membrane hole area. The electrowetting effect can significantly change the interfacial tension between the droplet and the solid surface, thereby causing reversible changes in the contact angle of the droplet. The greater the electric field intensity, the lower the interfacial tension of the droplet, the smaller the contact angle, and the more easily the droplet spreads, and vice versa. By dynamically adjusting the amplitude and duration of the applied electric field, the system can control the change path of the droplet contact boundary in real time, thereby maintaining the symmetry and stability of the liquid bridge structure, improving the wetting uniformity of the droplet on the surface of the porous material, and reducing the risk of asymmetric wetting behavior caused by surface activity disturbance.

[0116] In actual implementation, the system first analyzes the contact angle driving compensation coefficient, for example, the coefficient is +0.12, indicating that the electrowetting effect needs to be enhanced to reduce the current contact angle. The control platform then applies an enhanced electric signal to the electrode array at the bottom of the membrane hole area through a high-frequency pulse power supply, and adjusts the local electric field strength from the initial 1.5 V / μm to 1.68 V / μm (the enhancement amplitude corresponds to the compensation coefficient). At the same time, the system monitors the droplet edge morphology in real time, analyzes the contact angle evolution path through image recognition algorithm, and compares the monitoring results with the target contact angle trajectory. If the contact angle change trend is consistent with the target compensation direction, the electric field strength remains unchanged; if the response lags, the electric field amplitude is further fine-tuned until the contact angle change rate tends to be stable. The whole regulation process relies on the closed-loop feedback mechanism of software and hardware cooperation to ensure that the contact angle adjustment response is consistent with the target compensation direction, thereby maintaining the geometric symmetry and wetting stability of the liquid bridge structure.

[0117] According to the local surface energy regulation threshold, the control light-thermal response material excitation system outputs the corresponding local thermal radiation power to change the surface energy distribution state of the target membrane hole peripheral area material, realizing the spatial hierarchical regulation of the local surface wetting ability on the basis of maintaining the overall structure stability, thereby realizing the hierarchical dynamic regulation of the membrane hole area in different states and improving the responsiveness and precision of the membrane hole opening and closing state correction.

[0118] In the dynamic control system based on the local surface energy regulation mechanism, the surface energy local regulation threshold is used to indicate the required local wetting ability adjustment degree of the target membrane hole area. After receiving the regulation threshold, the light-thermal response material excitation system drives the infrared microwave source or near-infrared laser to output local thermal radiation with a specified power density in the selected area around the membrane hole. Through thermal radiation, the light-thermal response material produces thermal-induced phase change, thermal expansion or thermal-induced molecular configuration rearrangement, thereby changing the exposure state of the material surface polar group or the order of molecular arrangement, and then realizing the local enhancement or inhibition of surface energy. In this process, the surface energy of different positions around the membrane hole is adjusted as needed, realizing the spatial hierarchical wetting regulation ability. This hierarchical regulation mechanism can dynamically change the spreading range and wetting contact behavior of the liquid droplet around the membrane hole according to the real-time state change of the membrane hole, thereby improving the response consistency of the liquid droplet on the porous interface and the stability and precision of the membrane hole opening and closing state adjustment.

[0119] In specific implementation, the control system first analyzes the surface energy local regulation threshold output by the three-dimensional mapping tensor, for example, the output threshold is +0.25 J / m 2 , indicating that the wetting ability needs to be enhanced on the left side of the membrane hole. The system then selects the light-thermal response polymer film layer coated on the area as the regulation medium, and drives the near-infrared laser array to output 0.6 W / cm 2The power density of the laser locally heats the target area for 3 seconds. During the heating process, the conformation of the polymer groups inside the material changes, exposing more hydrophilic end groups, and the surface energy of this area increases from 1.2 J / m 2 to 1.45 J / m 2 . The spreading rate of the droplet on this area increases significantly, the contact angle decreases significantly, and the wetting area expands, thereby compensating for the lack of local wetting ability. The entire regulation process operates in coordination with the electrowetting compensation mechanism, and through the hierarchical regulation of the surface energy field, it achieves fine dynamic control of the different wetting states of the membrane hole area. This method can quickly respond to changes in the membrane hole state without changing the overall performance of the material, and improve the dynamic adaptability and regulation accuracy of the membrane hole correction.

[0120] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, server, or data center to another via wired or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0121] It should be understood that the size of the sequence number of each process in the various embodiments of the present application does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0122] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0123] In several embodiments provided in the present application, it should be understood that the disclosed system and method can be implemented in other manners. For example, the embodiments described above are merely schematic. For example, the division of units is only a logical function division. There can be another division manner for the actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0124] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e. can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to the actual needs to achieve the purposes of the embodiments of the present application.

[0125] In addition, each functional unit in the embodiments of the present application can be integrated in a processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in a unit.

[0126] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of controlled filtration based on the enhancement of the properties of the surface of a porous material, characterized in that, Specifically comprising the following steps: S1, construct a droplet interfacial tension behavior recognition matrix, collect the droplet contact front tension gradient diffusion velocity, droplet profile change rate and contact angle convergence path, for identifying whether there is a surface active ingredient in the liquid; S2, on the basis of identifying that there is a surface active ingredient in the liquid, generate an asymmetric liquid bridge evolution mapping based on the tension behavior recognition matrix, determine whether an asymmetric contact liquid bridge is formed and causes local wetting deviation by three-dimensional contact front trajectory, left and right contact area ratio and tension vector distribution; S3, based on the asymmetric liquid bridge evolution mapping, construct a response deviation indication map, calculate the non-synergy between the membrane hole edge wetting response change rate, the liquid bridge contact center deviation angle and the import flow rate, for identifying whether the membrane hole is out of control; S4, based on the membrane hole state identified in the response deviation indication map, construct an adaptive channel correction control mapping, output correction indicators corresponding to the membrane hole position, loss of control amplitude and behavior level; S5, based on the correction indicators of the adaptive channel correction control mapping, implement hierarchical dynamic regulation for the membrane hole area, including adjusting the surface energy distribution, the contact angle driving parameter and the liquid injection parameter, to realize real-time correction of the membrane hole opening and closing state.

2. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 1, characterized in that, S1 specifically comprises the following steps: Collect a continuous image sequence of the droplet contacting the surface of the porous membrane, obtain the diffusion velocity change of the droplet contact front tension gradient between each time frame, and form a droplet contact front tension gradient diffusion velocity data column; Extract the spatial change rate of the droplet edge profile boundary in the time dimension in the image sequence, and calculate the evolution trajectory of the convergence direction of the droplet bottom contact angle, to form a joint feature matrix of the droplet profile change rate and the contact angle convergence path; Uniformly normalize the droplet contact front tension gradient diffusion velocity data column and the joint feature matrix, construct a droplet interfacial tension behavior recognition matrix, and extract the pattern features of the droplet interfacial tension behavior recognition matrix, if the tension diffusion velocity is abnormally intensified and there is a nonlinear coupling jump between the contact angle convergence path, it is determined that there is a surface active ingredient in the liquid.

3. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 1, characterized in that, S2 specifically comprises the following steps: S201, on the basis of identifying that there is a surface active ingredient in the liquid, read the tension gradient values, edge spatial position values and contact angle change direction values of the droplet contact edge at different time points in the tension behavior recognition matrix, as the X-axis, Y-axis and Z-axis components of the three-dimensional space respectively, generate a three-dimensional point cloud set of droplet contact behavior, and construct a liquid bridge evolution mapping; S202, calculate the contact area pixel area of the droplet on the left and right sides in each time frame respectively by using image processing algorithm, obtain the ratio between the left and right contact areas, and mark the area ratio to the point at the corresponding time point in the three-dimensional liquid bridge evolution mapping as a parameter of contact imbalance degree; S203, according to the tension behavior recognition matrix direction change trend and amplitude difference of each time point tension vector, determine whether the three-dimensional point cloud is continuously offset to a single spatial direction, judge whether the left and right contact area ratio is continuously less than or greater than 1, identify whether the tension vector is concentrated in one direction, if the three-dimensional point cloud has a spatial offset trend, the left and right contact area ratio is not equal to 1 continuously, and the tension vector distribution exists in a single direction, it is determined that the droplet forms an asymmetric contact liquid bridge and causes local wetting offset.

4. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 3, characterized in that, S202 specifically: The image processing algorithm is applied to the image sequence of the droplet contacting the surface of the porous membrane for gray scale preprocessing and edge enhancement, the contact boundary profile of the droplet and the porous membrane in each frame of image is extracted, and the left and right contact regions are divided by the droplet symmetry axis as the boundary; Based on the extracted left and right contact region boundaries, the number of contact pixels in each region is counted respectively, and the left and right contact region pixel areas in each frame of image are obtained; The pixel areas of the left and right contact regions in each frame of image are calculated by ratio, an area ratio sequence is formed, and each area ratio is marked as a parameter on the point of the corresponding time point in the three-dimensional liquid bridge evolution mapping, which is used to quantify the area asymmetry degree of the left and right contact regions of the droplet.

5. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 1, wherein, S3 specifically includes the following steps: S301, extract the membrane pore edge wetting front space coordinates of each time point in the asymmetric liquid bridge evolution mapping, calculate the Euclidean distance change value between adjacent time points, form the membrane pore edge wetting response change rate sequence, which is the first coordinate dimension of the response offset indication map; S302, according to the liquid bridge contact center coordinates corresponding to each time point in the membrane pore edge wetting response change rate sequence, combined with the membrane pore center reference line, the angle of deviation is calculated, and the angle difference between continuous time points is constructed as the liquid bridge contact center deviation angle sequence, which is the second coordinate dimension of the response offset indication map; S303, collect the import flow rate value corresponding to each time point, construct the import flow rate sequence with the same time dimension as the membrane pore edge wetting response change rate sequence and the liquid bridge contact center deviation angle sequence, and compare the first derivative change trend of the three sequences on the time axis synchronously, if the membrane pore edge wetting response change rate increases in a positive growth form in a continuous time interval, the angle between the liquid bridge contact center and the membrane pore center is continuously greater than the initial angle threshold in a continuous time period and the deviation angle change rate is positive, and the first derivative value of the import flow rate is zero in the continuous time interval, it is determined that the three sequences have a non-synergistic evolution trend, and it is identified that the membrane is in an open-close out-of-control state.

6. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 5, wherein, S301 specifically: The time stamp of each frame of image in the asymmetric liquid bridge evolution mapping and the three-dimensional space coordinates of the membrane pore edge wetting front corresponding to the time point are extracted, including the horizontal position, the height position and the contact interface distance value, and are sorted in time sequence; Based on the three-dimensional space coordinates between adjacent time points, the Euclidean distance between each pair of continuous points is calculated as the displacement of the wetting front in unit time, and the wetting response speed value corresponding to each time period is obtained; The wetting response speed values in all continuous time periods are formed into a time sequence to form a membrane hole edge wetting response change rate sequence, and the sequence is set as a first coordinate dimension of a response offset indication map for subsequent trend analysis and offset identification.

7. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 6, characterized in that, S302 specifically is: Based on each time point in the membrane hole edge wetting response change rate sequence, the spatial coordinates of the liquid bridge contact center in the corresponding frame image are extracted, and the membrane hole center reference point coordinates are extracted to construct a spatial angle model of the liquid bridge contact center connecting line and the membrane hole center reference line; In time sequence, the angle formed between the liquid bridge contact center and the membrane hole center is traversed, the offset angle at each time point is calculated using the three-dimensional vector angle formula, and numerical normalization processing is performed to construct a time-angle relationship mapping table; The offset angle change value between consecutive time points is extracted to form a liquid bridge contact center offset angle sequence, and the sequence is taken as a second coordinate dimension of the response offset indication map for joint analysis of the membrane hole offset trend with the first coordinate dimension.

8. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 1, wherein, S4 specifically is: Based on the membrane hole edge wetting response change rate sequence, the liquid bridge contact center offset angle sequence and the imported flow rate sequence identified in the response offset indication map, the spatial position coordinates, the wetting response speed range, the offset angle change total amount and the corresponding flow rate first derivative features of each membrane hole in the continuous time interval are extracted, which are respectively set as the membrane hole position parameter, the out-of-control amplitude parameter and the behavior level parameter; Taking the membrane hole position parameter, the out-of-control amplitude parameter and the behavior level parameter as input variables, a three-dimensional mapping tensor is constructed, each mapping unit in the three-dimensional mapping tensor corresponds to a unique membrane hole state combination, and an adaptive correction response type corresponding to each state combination is defined in the three-dimensional mapping tensor; Based on the membrane hole state combination located in the three-dimensional mapping tensor, the correction index corresponding to the combination is output, including liquid injection rate adjustment factor, contact angle driving compensation coefficient and surface energy local regulation threshold, which are used for subsequent execution of hierarchical dynamic regulation of the membrane hole region.

9. The controllable filtration method based on the surface performance enhancement of porous materials according to claim 1, wherein, S5 specifically is: According to the liquid injection rate adjustment factor output in the three-dimensional mapping tensor, the numerical adjustment of the driving control parameter of the liquid injection unit is performed, the instantaneous flow rate of the liquid into the membrane hole region is finely adjusted by using the step electric control micropump, and the precise control of the liquid supply amount is realized; According to the contact angle driving compensation coefficient, the electric field intensity applied to the surface of the membrane hole region is adjusted in real time through the electrowetting control platform, the interfacial tension between the droplet and the membrane hole surface is dynamically changed, the contact angle change path is regulated, the liquid bridge shape is stabilized, and the wetting balance is improved; According to the surface energy local regulation threshold, the corresponding local heat radiation power output by the photo-thermal response material excitation system is controlled to change the surface energy distribution state of the material in the target membrane hole peripheral region, realizing spatial hierarchical regulation of local surface wetting ability on the basis of maintaining the stability of the overall structure, thereby realizing hierarchical dynamic regulation of the membrane hole region in different states, and improving the responsiveness and precision of the membrane hole opening and closing state correction.

Images