Data fusion processing method and system applied to volcanic rock water purification system

Abstract

The application provides a data fusion processing method and system applied to a volcanic rock water purification system, first acquires volcanic rock filter material pore evolution data, pollutant adsorption and migration data and filter pool operation feedback data, then performs time sequence correlation capture processing on the three types of data, identifies a shape transformation node and a transformation amplitude, constructs a dynamic correlation network based on the transformation node and the amplitude, serially connects interaction relationships of filter material pores, pollutant migration and operation parameter adjustment, generates filter material state self-adaptive adjustment rules according to the dynamic correlation network, describes optimal matching modes of filter material pores and operation parameters under different pollutant migration states, and generates operation parameter dynamic correction instructions according to the filter material state self-adaptive adjustment rules, including filter speed and backwashing cycle adjustment parameters, so that the purification efficiency and stability of the volcanic rock water purification system can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a data fusion processing method and system applied to a volcanic rock water purification system. BACKGROUND

[0002] In the field of operation and management of the volcanic rock water purification system, the traditional method mainly focuses on the monitoring and analysis of single data. For example, for the volcanic rock filter material, only the initial pore structure parameters are usually detected periodically, and the dynamic change of the pore morphology of the filter material with time during the water purification process is ignored. In terms of pollutant treatment, only the initial concentration and the final removal rate of the pollutants in the water body are often concerned, and the specific position change and morphology transformation process of the pollutants in the filter material layer are lack of in-depth research and monitoring. For the operation of the filter tank, the preset fixed filtration rate and backwashing period are usually followed, and the actual influence of the filtration rate adjustment and backwashing operation on the filter material state is rarely considered.

[0003] Due to the lack of comprehensive correlation analysis of the volcanic rock filter material pore evolution, pollutant adsorption and migration, and filter tank operation feedback, the existing method is difficult to comprehensively and accurately grasp the actual operation state of the volcanic rock water purification system. In the actual operation process, the operation parameters cannot be adjusted in time according to the dynamic change of the system, and problems such as filter material blockage and unstable pollutant removal effect are prone to occur, which affects the overall purification efficiency and stability of the volcanic rock water purification system, and cannot meet the increasingly strict sewage treatment requirements and water resource reuse standards. SUMMARY

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present application, the present application embodiment provides a data fusion processing method applied to a volcanic rock water purification system, which comprises:

[0005] obtaining volcanic rock filter material pore evolution data, pollutant adsorption and migration data, and filter tank operation feedback data, the volcanic rock filter material pore evolution data recording the time sequence change of the pore morphology between filter material particles with the water purification process, the pollutant adsorption and migration data recording the position change and morphology transformation of the pollutants in the filter material layer in the water body, and the filter tank operation feedback data recording the filter material state change caused by the filtration rate adjustment and backwashing operation;

[0006] performing time sequence correlation capture processing on the volcanic rock filter material pore evolution data, the pollutant adsorption and migration data, and the filter tank operation feedback data, and identifying the morphology transformation nodes and transformation amplitudes of the volcanic rock filter material pore evolution data, the pollutant adsorption and migration data, and the filter tank operation feedback data in the same time process;

[0007] construct a dynamic correlation network based on the morphology transition node and the transition amplitude, the dynamic correlation network taking a time flow as an axis to connect the interaction relationship among the filter material pore morphology change, the pollutant migration path change and the operation parameter adjustment;

[0008] generate a filter material state adaptive adjustment rule according to the dynamic correlation network, the adaptive adjustment rule being used to describe an optimal matching mode of the filter material pore morphology and the operation parameter under different pollutant migration states;

[0009] generate an operation parameter dynamic correction instruction of the volcanic rock water purification system according to the filter material state adaptive adjustment rule, the operation parameter dynamic correction instruction including a filter speed real-time adjustment parameter and a backwashing cycle adjustment parameter.

[0010] In still another aspect, an embodiment of the present application further provides a data fusion processing system applied to a volcanic rock water purification system, characterized by comprising:

[0011] a processor; a machine readable storage medium for storing machine executable instructions of the processor; wherein the processor is configured to execute the above-mentioned data fusion processing method applied to the volcanic rock water purification system by executing the machine executable instructions.

[0012] In still another aspect, an embodiment of the present application further provides a computer program product, the computer program product comprising machine executable instructions stored in a computer readable storage medium, a processor of a computer device reading the machine executable instructions from the computer readable storage medium, and the processor executing the machine executable instructions, so that the computer device executes the above-mentioned data fusion processing method applied to the volcanic rock water purification system.

[0013] Based on the above aspects, by acquiring the volcanic rock filter material pore evolution data, the pollutant adsorption and migration data and the filter pool operation feedback data, the key information in the operation process of the volcanic rock water purification system is comprehensively covered. The time sequence correlation capture processing of the three types of data can accurately identify the morphological change node and the change amplitude in the same time process, and deeply reveals the dynamic correlation between the elements of the system. The dynamic correlation network constructed based on the morphological change node and the change amplitude clearly connects the interaction relationship of the filter material pore morphology change, the pollutant migration path change and the operation parameter adjustment with the time flow as the axis. According to the filter material state self-adaptive adjustment rule generated by the dynamic correlation network, the optimal matching mode of the filter material pore morphology and the operation parameter under different pollutant migration states can be described. The operation parameter dynamic correction instruction generated according to the self-adaptive adjustment rule contains the filter speed real-time adjustment parameter and the backwashing cycle adjustment parameter, which can realize the precise dynamic adjustment of the operation parameters of the volcanic rock water purification system, effectively improve the purification efficiency, stability and adaptability of the system, reduce the operation cost, and meet the sewage treatment demand under different water quality and water quantity conditions. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is the execution flow schematic diagram of the data fusion processing method applied to the volcanic rock water purification system provided by the embodiment of the present application.

[0015] Figure 2 is the schematic diagram of exemplary hardware and software components of the data fusion processing system applied to the volcanic rock water purification system provided by the embodiment of the present application. DETAILED DESCRIPTION

[0016] The present application will be described in detail below with reference to the accompanying drawings, Figure 1 is the flow schematic diagram of the data fusion processing method applied to the volcanic rock water purification system provided by an embodiment of the present application, and the data fusion processing method applied to the volcanic rock water purification system will be described in detail below.

[0017] Step S110: acquiring volcanic rock filter material pore evolution data, pollutant adsorption and migration data and filter pool operation feedback data, the volcanic rock filter material pore evolution data records the time sequence change of the pore morphology between filter particles with the water purification process, the pollutant adsorption and migration data records the position change and morphological transformation of the pollutants in the filter layer, and the filter pool operation feedback data records the filter material state change caused by the filter speed adjustment and backwashing operation.

[0018] In the application scenario of a municipal waterworks' volcanic rock filter depth purification system, the volcanic rock filter material pore evolution data is collected by a microscopic imaging device arranged at different depths of the filter. The microscopic imaging device scans the filter material layer at a preset time interval to obtain a sequence of images containing pore structure, which records the dynamic change information of the pore geometry, connectivity path, etc. with the purification process. The pollutant adsorption and migration data is collected by multi-parameter water quality sensors set at the inlet, different depths of the filter material layer, and the outlet of the filter. The sensors monitor the spatial distribution of pollutants, concentration gradient, and chemical form transformation in real time, such as the functional group change of organic pollutants and the valence state conversion of heavy metal ions. The filter operation feedback data is automatically recorded by the filter automatic control system, including real-time operation parameters of the filter speed adjustment device, start and stop signals of the backwashing equipment, pressure loss changes of the filter material layer before and after backwashing, and filter material layer height changes, etc.

[0019] Step S120: Time sequence correlation capture processing is performed on the volcanic rock filter material pore evolution data, the pollutant adsorption and migration data, and the filter operation feedback data to identify the morphological transformation nodes and transformation amplitudes in the same time process.

[0020] In the above-mentioned municipal waterworks' volcanic rock filter depth purification system scenario, the time sequence correlation capture processing needs to unify the time base of the three types of data. Through timestamp alignment technology, the time records of the data collected by the microscopic imaging device, water quality sensor, and automatic control system are adjusted to the same time reference system, ensuring the comparability of the data in the time dimension. Subsequently, for each data sequence, a trend analysis method is used to identify the time point at which the data change trend significantly changes, i.e., the morphological transformation node, and the transformation amplitude is determined by calculating the difference between the feature values before and after the transformation node, such as the change amount of filter material pore volume, the change ratio of pollutant concentration, and the change rate of filter speed adjustment, etc.

[0021] Step S121: Extracting the cross-sectional morphology time sequence of the filter material pore and the pore connectivity path time sequence from the volcanic rock filter material pore evolution data. The cross-sectional morphology time sequence records the geometric shape data of the pore at different times, and the pore connectivity path time sequence records the connection mode data between pores at different times.

[0022] In the above scenario, the image sequence in the volcanic rock filter pore evolution data is processed. First, the image is grayed and preprocessed to enhance the contrast between the pore area and the background. Then, the edge detection algorithm is used to identify the cross-sectional profile of the pore, and the geometric parameters of the profile are extracted, including the perimeter, the diameter of the inscribed circle, and the shape factor. These parameters at different times are arranged in chronological order to form the cross-sectional shape time sequence. For the pore connectivity path time sequence, the image is regionally segmented to mark each independent pore region, and then the region adjacency analysis method is used to identify the connection channels between adjacent pores, record the number, distribution, and connection mode (such as direct connection, indirect connection, and isolated state) of the connection channels, and arrange the connection mode data at different times in chronological order to form the pore connectivity path time sequence.

[0023] Step S1211: Analyzing the volcanic rock filter pore evolution data, which includes a pore image sequence of different depths of the filter layer collected by a microscopic imaging technique.

[0024] In the above scenario, the volcanic rock filter pore evolution data is stored as a set of image files, each corresponding to a specific filter layer depth and collection time. The analysis process includes reading the metadata of the image file, obtaining the collection time, corresponding filter layer depth, resolution, and other information of the image, then decoding the image data, and converting the image from a compressed format to an original pixel matrix, preparing for subsequent image segmentation and feature extraction.

[0025] Step S1212: Image segmentation processing is performed on the pore image sequence to separate the pore area and non-pore area in each pore image.

[0026] In the above scenario, the image segmentation processing uses a combination of threshold segmentation and region growing. First, the pore image is analyzed by gray histogram to determine the initial threshold value for distinguishing the pore area and the non-pore area. The threshold value is used for binary processing of the image to preliminarily separate the pore area. Then, the pore area in the binary image is subjected to morphological filtering to remove noise interference and small-area false pore areas. Next, the region growing algorithm is used with the preliminarily segmented pore area as the seed point to further grow and merge adjacent pore areas based on the similarity of pixel gray values and spatial continuity, and finally obtain the segmentation results of the complete pore area and non-pore area.

[0027] Step S1213: For each segmented pore area, the geometric shape parameters of its cross-section are extracted, including the profile perimeter, the inscribed circle diameter, and the shape factor, which together constitute the geometric shape data of the pore.

[0028] In the above scenario, for each segmented pore region, the boundary contour of the pore region is extracted by a contour tracking algorithm. The sum of the distances between all consecutive pixel points on the contour is calculated to obtain the contour perimeter. The largest inscribed circle that can be completely contained in the pore region is found inside the contour, and the diameter of the largest inscribed circle is the inscribed circle diameter. The shape factor is calculated by the ratio of the square of the contour perimeter to the area of the pore region, which is used to describe the irregularity of the pore shape. The smaller the ratio, the closer the pore shape is to a circle. The contour perimeter, inscribed circle diameter and shape factor of each pore region are recorded as geometric shape data.

[0029] Step S1214: Arrange the geometric shape data in the order of the acquisition time of the image sequence to form a time sequence of cross-sectional morphologies of the filter pores.

[0030] In the above scenario, the geometric shape data extracted from the segmented pore images of different acquisition times are arranged in the order of the acquisition time of the images, and each time point corresponds to a data set containing the geometric shape parameters of all pore regions, thereby forming a time sequence of cross-sectional morphologies that changes over time.

[0031] Step S1215: Perform connectivity analysis on the pore image sequence to identify the connecting channels between adjacent pores and the width and length of the connecting channels.

[0032] In the above scenario, the connectivity analysis is performed based on image segmentation. First, distance transformation is performed on the segmented binary image to calculate the distance value of each non-pore region pixel to the nearest pore region. Then, according to the distance transformation result, a distance threshold is set, and non-pore regions with a distance less than the threshold to two or more pore regions are identified as potential connecting channel regions. Morphological dilation operation is performed on the potential connecting channel regions to fill small gaps in the channels, and then a skeleton extraction algorithm is used to extract the center axis of the connecting channel region, and the length of the center axis is the length of the connecting channel. At the start and end positions of the connecting channel, the width of the channel is measured perpendicular to the center axis direction, and the average value of multiple measurement points is taken as the width of the connecting channel.

[0033] Step S1215-1: Perform binaryzation processing on each image in the segmented pore image sequence, and mark the pore regions as a first value and the non-pore regions as a second value.

[0034] In the above scenario, the segmented pore image is a grayscale image, and during binaryzation processing, the pixel grayscale values corresponding to the pore regions are set to the first value, and the pixel grayscale values corresponding to the non-pore regions are set to the second value, forming a binary image containing only two values, which facilitates subsequent connectivity analysis and feature extraction.

[0035] Step S1215-2: Pre-process the binary image using a morphological erosion operation, analyze the pre-processed image using a connected component labeling algorithm, identify each independent pore region and assign a unique identifier.

[0036] In the above scenario, the morphological erosion operation scans the binary image using a preset size of structural elements, removes small protruding parts and noise points on the boundary of the pore region, and makes the outline of the pore region smoother. The pre-processed image is analyzed by the connected component labeling algorithm, which traverses all pixels in the image, groups pixels in mutually connected pore regions into a connected component, and assigns a unique identifier to each connected component, achieving the differentiation and labeling of each independent pore region.

[0037] Step S1215-3: Perform distance transform processing on the labeled pore region, calculate the distance from each non-pore region pixel to the nearest pore region.

[0038] In the above scenario, the distance transform processing takes the labeled pore region as the foreground and the non-pore region as the background. For each non-pore region pixel, the Euclidean distance from the nearest pore region pixel is calculated, and the distance value is assigned to the non-pore region pixel, resulting in a distance transform image. The gray value of each pixel in the image represents the distance from the nearest pore region.

[0039] Step S1215-4: Identify potential connection channel regions based on the distance transform results. The potential connection channel region is a non-pore region that is close to two or more pore regions.

[0040] In the above scenario, the distance transform image is analyzed, a distance threshold is set, and non-pore region pixels with a distance value less than the threshold are selected. For these pixels, further judgment is made as to whether they are close to two or more different identified pore regions, i.e., the distance from the pixel to two or more pore regions is less than the distance threshold. The region formed by the collection of pixels that meet the conditions is identified as a potential connection channel region.

[0041] Step S1215-5: Perform morphological dilation operation on the potential connection channel region, use skeleton extraction algorithm to extract the center axis of the dilated connection channel region as the path of the connection channel, and measure the length of the connection channel center axis as the length parameter of the connection channel.

[0042] In the above scenario, the morphological dilation operation uses a preset size of structural elements to expand the potential connection channel region, filling the small breaks and gaps in the channel, so that the connection channel region forms a continuous region. Skeleton extraction is performed on the expanded connection channel region, and by continuously eroding the boundary pixels of the connection channel region, the region is reduced to a single-pixel-wide center axis, which is the path of the connection channel. Along the center axis from the starting point to the end point, the pixel points are traversed, and the cumulative distance between the pixel points is calculated to obtain the length of the connection channel center axis as the length parameter of the connection channel.

[0043] Step S1215-6: At the starting point and the end point of the connection channel, the channel width perpendicular to the center axis direction is measured, and the average value is taken as the width parameter of the connection channel. The starting point pore identification, end point pore identification, length parameter and width parameter of each connection channel are associated and stored, and the identification of the connection channel between adjacent pores is completed.

[0044] In the above scenario, the starting point and the end point of the connection channel correspond to two different pore regions respectively. At the starting point, the distance from the center axis to the connection channel boundary is measured on both sides perpendicular to the center axis direction of the connection channel, and the sum of the two is the channel width at the starting point. The same method is used to measure the channel width at the end point. The average value of the channel width at the starting point and the end point is taken as the width parameter of the connection channel. The starting point pore identification, end point pore identification, length parameter and width parameter of each connection channel are associated and stored in the database, realizing the identification and recording of the connection channel between adjacent pores.

[0045] Step S1216: Determine the connection mode between pores according to the distribution of the connection channel, including direct connection, indirect connection and isolated state.

[0046] In the above scenario, according to the identification result of the connection channel, the connection mode between pores is determined. If there is a direct connection channel between two pore regions, i.e. directly connected through a connection channel, then the two pore regions are in a direct connection state. If two pore regions are indirectly connected through multiple intermediate pore regions and connection channels, then the two pore regions are in an indirect connection state. If a pore region does not have a connection channel with any other pore region, then the pore region is in an isolated state.

[0047] Step S1217: Record the connection mode of each pore and the corresponding connection channel parameter at each time, and arrange them in time sequence to form a time sequence of pore connection paths.

[0048] In the above scenario, for each acquisition time of the pore image, the connection mode (directly connected, indirectly connected, isolated state) and the corresponding connection channel parameters (length parameter, width parameter) of all pore regions at that time are recorded. These data at different times are arranged in time sequence to form a pore connection path time sequence, which reflects the dynamic changes of the connection mode and connection channel parameters between pores over time.

[0049] Step S1218: Add the same time marker to the cross-sectional morphology time sequence and the pore connection path time sequence.

[0050] In the above scenario, the cross-sectional morphology time sequence and the pore connection path time sequence are both derived from the pore image sequence, and the data in each sequence corresponds to a specific acquisition time. The acquisition time of the pore image sequence is used as a time marker and added to the cross-sectional morphology time sequence and the pore connection path time sequence, respectively, so that the two time sequences maintain a corresponding relationship in the time dimension, facilitating subsequent time sequence correlation analysis.

[0051] Step S122: Extract a pollutant spatial distribution time sequence and a pollutant chemical form transformation time sequence from the pollutant adsorption and migration data, the spatial distribution time sequence records the position coordinate data of the pollutant at different times, and the chemical form transformation time sequence records the molecular structure data of the pollutant at different times.

[0052] In the above scenario, the pollutant adsorption and migration data includes pollutant concentration data collected at different times and different positions in the filter material layer. The spatial distribution time sequence is extracted by first determining the spatial coordinate system of the filter material layer, taking the bottom of the filter as the origin, the vertical upward direction as the height direction, and the horizontal direction as the radial direction. According to the installation position of the water quality sensor in the filter material layer, the spatial coordinates corresponding to each sensor are obtained. The pollutant concentration data collected by each sensor at different times is associated with the corresponding spatial coordinates and arranged in time sequence to form a spatial distribution time sequence, which reflects the change of the spatial position of the pollutant in the filter material layer over time. The pollutant chemical form transformation time sequence is extracted by analyzing the pollutant molecular structure detection data collected by the water quality sensor, extracting feature parameters representing the molecular structure, such as functional group type, chemical bond type, and molecular configuration, and arranging the feature parameters at different times in time sequence to form a pollutant chemical form transformation time sequence.

[0053] Step S123: Extract a filter speed change time sequence and a backwash effect time sequence from the filter operation feedback data, the filter speed change time sequence records the water flow velocity data at different times, and the backwash effect time sequence records the filter material state recovery data after backwashing.

[0054] In the above scenario, the filtration rate data in the filter tank operation feedback data is collected by a filtration rate sensor, recording the water flow rate through the filter material layer at different times. The filtration rate time sequence is extracted by arranging the filtration rate data at different times in chronological order to form a sequence of filtration rate changes over time. The backwashing effect time sequence is extracted by collecting filter material state data at different times after the backwashing operation is completed, such as the porosity of the filter material layer, pressure loss, uniformity of filter material particles, etc. These data reflect the recovery of the filter material state after backwashing. The above data is arranged in chronological order after the backwashing is completed to form a backwashing effect time sequence.

[0055] Step S124: The time bases of the cross-sectional morphology time sequence, the pore connectivity path time sequence, the spatial distribution time sequence, the chemical morphology conversion time sequence, the filtration rate change time sequence, and the backwashing effect time sequence are unified into the same time unit. With the unified time base as the axis, the mutation positions of the pore geometry in the cross-sectional morphology time sequence are compared and marked as filter pore morphology transition nodes.

[0056] In the above scenario, the time base is unified by using hours as the time unit to convert the time markers in each time sequence, ensuring that the time units of all sequences are consistent. The cross-sectional morphology time sequence is divided into multiple time periods based on the unified time axis. The trend of the pore geometry parameters (contour perimeter, inscribed circle diameter, shape factor) in each time period is analyzed, and the change rate of the parameters is calculated. When the change rate of a parameter at a certain time point exceeds a preset mutation threshold, the time point is marked as a filter pore morphology transition node, indicating that the pore geometry has changed significantly at that time.

[0057] Step S125: The mutation positions of the connection mode in the pore connectivity path time sequence are compared and marked as filter pore connection transition nodes; the mutation positions of the pollutant position coordinates in the spatial distribution time sequence are compared and marked as pollutant migration transition nodes; and the mutation positions of the molecular structure in the chemical morphology conversion time sequence are compared and marked as pollutant morphology transition nodes.

[0058] In the above scenario, for the pore connectivity path time sequence, the change of the connection mode (direct connection, indirect connection, isolated state) between pores at different time is analyzed. When the connection mode at a certain time point changes significantly compared with the previous time, such as direct connection changing to indirect connection, isolated state changing to connected state, etc., the time point is marked as a filter pore connection transition node. For the spatial distribution time sequence, the change trend of the position coordinates of the pollutants is analyzed, and the moving speed of the position coordinates in the horizontal and vertical directions is calculated. When the moving speed at a certain time point exceeds the preset migration mutation threshold, the time point is marked as a pollutant migration transition node. For the chemical form transformation time sequence, the change of the molecular structure feature parameters is analyzed, and when the feature parameters at a certain time point change significantly, such as the disappearance or appearance of functional groups, the change of chemical bond type, etc., the time point is marked as a pollutant form transition node.

[0059] Step S126: Compare the mutation position of the water flow speed in the filter speed change time sequence, and mark it as a filter speed adjustment transition node; compare the starting position of the filter material state recovery in the backwashing effect time sequence, and mark it as a backwashing operation transition node.

[0060] In the above scenario, for the filter speed change time sequence, the change of the water flow speed with time is analyzed. The difference value of the water flow speed at adjacent time points is calculated, and when the difference value exceeds the preset filter speed mutation threshold, the time point is marked as a filter speed adjustment transition node, indicating that the filter speed has been significantly adjusted at that time. For the backwashing effect time sequence, the starting time of the backwashing operation is the starting position of the filter material state recovery, and the time point corresponding to the starting position is marked as a backwashing operation transition node. The backwashing operation transition node marks the beginning of the backwashing process, and the filter material state enters the recovery stage.

[0061] Step S127: Calculate the relative change rate or standardized change amount of the values of the corresponding data sequence before and after each transition node in its own dimension as the transition amplitude of the transition node in its own data dimension.

[0062] In the above scenario, for each transition node, the values of the corresponding data sequence within a certain time window before and after the node are obtained. For example, for the filter pore form transition node, the average value of the pore geometric shape parameters within a time window before the node and the average value of the pore geometric shape parameters within a time window after the node are obtained, and the relative change rate or standardized change amount of the values before and after the node in its own dimension is calculated, which is the transition amplitude of the filter pore form transition node. Using the same method, the relative change rate or standardized change amount of the values before and after the corresponding data sequence at the filter pore connection transition node, the pollutant migration transition node, the pollutant form transition node, the filter speed adjustment transition node and the backwashing operation transition node in its own dimension is calculated, as the transition amplitude of each transition node.

[0063] Step S128: Arrange all the morphology transition nodes in chronological order to form a sequence of transition nodes on a unified time axis, and associate the transition amplitude of each transition node.

[0064] In the above scenario, all identified morphology transition nodes (including filter material pore morphology transition nodes, filter material pore connection transition nodes, pollutant migration transition nodes, pollutant morphology transition nodes, filter speed adjustment transition nodes, and backwash operation transition nodes) are collected, the time markers of each node are extracted, and the above nodes are sorted in chronological order to form a sequence of transition nodes on a unified time axis. In this sequence, each transition node is associated with its corresponding transition amplitude to facilitate subsequent analysis of the association and influence between different transition nodes.

[0065] Step S130: Construct a dynamic association network based on the morphology transition nodes and the transition amplitudes, with the time flow as the axis to link the interaction relationships of filter material pore morphology changes, pollutant migration path changes, and operation parameter adjustments.

[0066] In the above scenario of the municipal tap water plant volcanic rock filter depth purification system, the construction of the dynamic association network takes the time flow as the core clue, organically integrates the morphology transition nodes and the transition amplitudes, and intuitively displays the dynamic interaction between the filter material, the pollutant, and the operation parameters. Through the network structure, it can be clearly observed that at different time points, how the change of the filter material pore morphology affects the migration path of the pollutant, how the adjustment of the operation parameters affects the filter material pore morphology and the pollutant migration, and how the change of the pollutant migration in turn prompts the adjustment of the operation parameters.

[0067] Step S131: Construct a network framework with the time flow as the vertical axis, with the vertical dimension of the network framework being the time process and the horizontal dimension being the filter material feature dimension, the pollutant feature dimension, and the operation feature dimension.

[0068] In the above scenario, the network framework adopts a two-dimensional coordinate system form, with the vertical axis representing the time process, corresponding to the chronological order of time from bottom to top, and the time interval being determined according to the distribution of the transition nodes to ensure that all transition nodes can be clearly displayed. The horizontal dimension is divided into three parallel feature dimensions, namely the filter material feature dimension, the pollutant feature dimension, and the operation feature dimension, each of which is used to display the corresponding type of transition node and related data.

[0069] Step S132: Embed the morphology transition nodes in the corresponding horizontal dimension position of the network framework according to the feature dimension they belong to, and associate the different feature dimension transition nodes at the same time point through horizontal connection lines.

[0070] In the above scenario, according to the type of the morphology transition node, it is embedded into the corresponding transverse feature dimension position. For example, the filter material pore morphology transition node and the filter material pore connection transition node are embedded into the filter material feature dimension, the pollutant migration transition node and the pollutant morphology transition node are embedded into the pollutant feature dimension, and the filter speed adjustment transition node and the backwashing operation transition node are embedded into the operation feature dimension. For the transition nodes of different feature dimensions occurring at the same time point, they are connected by a transverse connection line in the network framework, indicating the synchronization of these transition nodes in time.

[0071] Step S133: For the transition nodes of different feature dimensions at the same time point, their transition directions (increase / decrease) are respectively judged and their standardized transition amplitude levels are recorded, and the line width of the transverse connection line is adjusted according to the correlation of the transition directions and the matching degree of the amplitude levels. The higher the correlation and the more matched the amplitude are, the wider the line width is.

[0072] In the above scenario, for the transition nodes of different feature dimensions associated by the transverse connection line at the same time point, the transition amplitude ratio of them is calculated. For example, the ratio of the transition amplitude of the filter material feature dimension transition node to the transition amplitude of the pollutant feature dimension transition node, or the ratio of the transition amplitude of the operation feature dimension transition node to the transition amplitude of the filter material feature dimension transition node, etc. According to the size of the ratio, the line width of the transverse connection line is set. The larger the ratio is, the wider the line width is set, to intuitively reflect the relative influence degree of the transition nodes of different feature dimensions at the same time point.

[0073] Step S134: In the network framework, a longitudinal connection line is established between the filter material feature dimension transition nodes of adjacent time points, and the pore morphology change direction corresponding to the connection line is marked.

[0074] In the above scenario, in the filter material feature dimension, for the two filter material feature dimension transition nodes of adjacent time points, they are connected by a longitudinal connection line. According to the change of the pore morphology parameter from the former node to the latter node, the pore morphology change direction is marked. If the pore morphology parameter increases (such as the pore volume increases, the connectivity enhances), the positive change direction is marked; if the pore morphology parameter decreases (such as the pore volume decreases, the connectivity weakens), the negative change direction is marked.

[0075] Step S135: A longitudinal connection line is established between the pollutant feature dimension transition nodes of adjacent time points, and the migration path change direction corresponding to the connection line is marked.

[0076] In the above scenario, in the pollutant feature dimension, for two pollutant feature dimension transition nodes of adjacent time points, connect them with a longitudinal connection line. According to the change of the pollutant migration path from the previous node to the next node, mark the migration path change direction. If the pollutant migrates in the depth direction of the filter material layer, mark it as the downward migration direction; if the pollutant migrates in the surface direction of the filter material layer, mark it as the upward migration direction; if the pollutant diffuses in the horizontal direction, mark it as the horizontal migration direction.

[0077] Step S136: Establish a longitudinal connection line between the operation feature dimension transition nodes of adjacent time points, and mark the parameter adjustment direction corresponding to the connection line.

[0078] In the above scenario, in the operation feature dimension, for two operation feature dimension transition nodes of adjacent time points, connect them with a longitudinal connection line. According to the change of the operation parameter from the previous node to the next node, mark the parameter adjustment direction. For the filter speed adjustment parameter, if the filter speed increases, mark it as the speed-up direction; if the filter speed decreases, mark it as the speed-down direction. For the backwash cycle parameter, if the backwash cycle is extended, mark it as the cycle extension direction; if the backwash cycle is shortened, mark it as the cycle shortening direction.

[0079] Step S137: For the same feature dimension transition nodes of adjacent time points, calculate the standardized transition amplitude level difference, and adjust the color depth of the longitudinal connection line according to the size of the transition amplitude level difference. The larger the level difference, the deeper the color.

[0080] In the above scenario, for the transition nodes of the same feature dimension in adjacent time points, calculate their transition amplitude level difference. For example, the difference between the transition amplitude level of the previous transition node and the transition amplitude level of the next transition node in the filter material feature dimension. According to the size of the difference, adjust the color depth of the longitudinal connection line. The larger the difference, the deeper the color depth, to intuitively reflect the transition amplitude change of the same feature dimension transition nodes in adjacent time points.

[0081] Step S138: Identify the time interval in which the filter material feature dimension and the pollutant feature dimension transition nodes appear simultaneously in the network framework, and add a bidirectional arrow to identify the interaction relationship between the two in the time interval.

[0082] In the above scenario, traverse the timeline of the network framework to find the time interval in which both the filter characteristic dimension transition node and the pollutant characteristic dimension transition node exist. In this time interval, add a bidirectional arrow between the filter characteristic dimension and the pollutant characteristic dimension, with the arrow pointing to each of the two characteristic dimensions, indicating that there is an interaction relationship between the change in filter pore morphology and the change in pollutant migration path, i.e., the change in filter pore morphology will affect the migration path of the pollutant, and at the same time, the migration of the pollutant will also affect the pore morphology of the filter.

[0083] Step S1381: Traverse the timeline of the network framework and check whether there is a filter characteristic dimension transition node and a pollutant characteristic dimension transition node in the same time interval.

[0084] In the above scenario, according to the order of the timeline, divide the timeline into multiple continuous time intervals, and the length of each time interval is determined according to the distribution density of the transition nodes. For each time interval, check whether it contains both the filter characteristic dimension transition node and the pollutant characteristic dimension transition node. If so, record the start time and end time of the time interval.

[0085] Step S1382: For each time interval in which both types of transition nodes exist, determine the start time and end time of the time interval and mark it as an interaction interval.

[0086] In the above scenario, for the time interval found to contain both the filter characteristic dimension transition node and the pollutant characteristic dimension transition node, determine the start time and end time of the interval as the boundaries of the interaction interval, and mark the interval to facilitate the addition of interaction relationship identification in the future.

[0087] Step S1383: Extract the transition amplitude of the filter characteristic dimension transition node and the transition amplitude of the pollutant characteristic dimension transition node in the interaction interval.

[0088] In the above scenario, in the marked interaction interval, find all filter characteristic dimension transition nodes and pollutant characteristic dimension transition nodes, extract the transition amplitude data of each node, and calculate the sum of the filter characteristic dimension transition amplitude and the sum of the pollutant characteristic dimension transition amplitude, or calculate the average transition amplitude of the two.

[0089] Step S1384: Calculate the ratio of the filter characteristic transition amplitude to the pollutant characteristic transition amplitude, which is used to represent the strength ratio of their interaction.

[0090] In the above scenario, the ratio of the sum (or average transition amplitude) of the filter material feature dimension transition amplitude to the sum (or average transition amplitude) of the pollutant feature dimension transition amplitude is obtained by dividing the sum (or average transition amplitude) of the filter material feature dimension transition amplitude in the interaction interval by the sum (or average transition amplitude) of the pollutant feature dimension transition amplitude, which reflects the proportion of the strength of the interaction between the filter material feature and the pollutant feature in the interaction interval.

[0091] Step S1385: Determine the arrow size of the bidirectional arrow according to the strength ratio. The closer the ratio is to 1, the more consistent the arrow size.

[0092] In the above scenario, the two arrows of the bidirectional arrow respectively correspond to the action direction of the filter material feature dimension to the pollutant feature dimension and the action direction of the pollutant feature dimension to the filter material feature dimension. According to the strength ratio, the size of the two arrows is adjusted. If the strength ratio is close to 1, it means that the interaction strength of the two is comparable, and the size of the two arrows is set to be consistent; if the strength ratio is greater than 1, it means that the action strength of the filter material feature on the pollutant feature is greater, and the arrow in the corresponding direction is set to be larger; if the strength ratio is less than 1, it means that the action strength of the pollutant feature on the filter material feature is greater, and the arrow in the corresponding direction is set to be larger.

[0093] Step S1386: Analyze the change direction of the filter material feature dimension data and the change direction of the pollutant feature dimension data in the interaction interval to determine the action direction relationship between the two; label the action direction relationship on the bidirectional arrow, including the promoting relationship and the inhibiting relationship. The promoting relationship means that the change of one side aggravates the change of the other side, and the inhibiting relationship means that the change of one side alleviates the change of the other side.

[0094] In the above scenario, the change direction of the filter material feature dimension data (such as pore enlargement or reduction) and the change direction of the pollutant feature dimension data (such as migration speed acceleration or deceleration) are analyzed. If the filter material feature dimension data changes in a certain direction, and the pollutant feature dimension data also changes in the same direction, and the change degree is aggravated, then the two are in a promoting relationship; if the filter material feature dimension data changes in a certain direction, and the pollutant feature dimension data changes in the opposite direction, or the change degree is alleviated, then the two are in an inhibiting relationship. The promoting relationship or the inhibiting relationship is labeled in text on the bidirectional arrow to clearly indicate the action direction relationship between the two.

[0095] Step S1387: Statistically analyze the change synchronization rate of the two types of feature dimension data in the interaction interval. The higher the synchronization rate, the thicker the line of the bidirectional arrow.

[0096] In the above scenario, the synchronization rate of the change of the two types of feature dimension data in the interaction interval is determined by calculating the time coincidence degree and the change trend consistency of the filter feature dimension data change and the pollutant feature dimension data change. The higher the time coincidence degree and the change trend consistency, the higher the synchronization rate. According to the high and low of the synchronization rate, the line thickness of the bidirectional arrow is adjusted, the higher the synchronization rate, the thicker the line is set, to reflect the synchronization degree of the interaction of the two.

[0097] Step S1388: Add the labeled bidirectional arrow to the interaction interval of the network framework, connecting the corresponding filter feature dimension transition node and the pollutant feature dimension transition node.

[0098] In the above scenario, according to the position and range of the interaction interval, a bidirectional arrow labeled with the direction of action relationship, arrow size and line thickness is added to the network framework, and the two ends of the arrow are connected to the corresponding filter feature dimension transition node and the pollutant feature dimension transition node in the interaction interval, to intuitively display the interaction relationship between the two in the interval.

[0099] Step S1389: Number all the added bidirectional arrows and establish a correspondence table of arrows and interaction intervals.

[0100] In the above scenario, each bidirectional arrow added to the network framework is assigned a unique number, and then a correspondence table is established, which records the arrow number, the start time and end time of the interaction interval, the direction of action relationship and other information, to facilitate the management and query of the interaction relationship represented by the bidirectional arrow.

[0101] Step S139: After identifying the running feature dimension transition node in the network framework, the time interval when the filter feature dimension or the pollutant feature dimension appears the transition node is identified, and a one-way arrow is added in the time interval to identify the influence of the running parameter adjustment on the filter or the pollutant.

[0102] In the above scenario, the time axis of the network framework is traversed to find the time interval after the running feature dimension transition node appears. In the time interval, if a filter feature dimension transition node or a pollutant feature dimension transition node appears subsequently, it is considered that the adjustment of the running parameter has an impact on the filter pore morphology or the pollutant migration. A one-way arrow is added between the running feature dimension transition node and the subsequently appearing filter feature dimension transition node or pollutant feature dimension transition node, and the arrow direction points from the running feature dimension to the filter feature dimension or the pollutant feature dimension, to identify the influence of the running parameter adjustment on the filter or the pollutant.

[0103] Step S1310: For the same feature dimension transition nodes of adjacent time points, calculate the standardized transition amplitude level difference, and adjust the color depth of the longitudinal connection line according to the size of the transition amplitude level difference. The larger the level difference is, the deeper the color is.

[0104] In the above scenario, this step is repeated with step S137, which will not be repeated here.

[0105] Step S1311: Count the occurrence frequency of each type of connection line in the network framework, and mark the connection mode with the highest occurrence frequency as the core association mode.

[0106] In the above scenario, each type of connection line in the network framework, such as horizontal connection line, longitudinal connection line and bidirectional arrow, is counted, and the occurrence frequency of each connection mode (such as horizontal connection between filter material feature dimension and pollutant feature dimension within a certain transition amplitude ratio range, longitudinal connection of filter material feature dimension within a certain transition amplitude difference range, etc.) is calculated. The connection mode with the highest occurrence frequency is marked as the core association mode, which reflects the most important feature association relationship in the system.

[0107] Step S140: Generating filter state adaptive adjustment rules from the dynamic association network, the adaptive adjustment rules are used to describe the optimal matching mode of filter pore morphology and operating parameters under different pollutant migration states.

[0108] In the above scenario of municipal tap water plant volcanic rock filter depth water purification system, the dynamic association network clearly shows the association relationship between filter, pollutant and operating parameter at different time points. Based on these association relationships, how the filter pore morphology changes with the adjustment of operating parameters under different pollutant migration states is analyzed, and which combination of filter pore morphology and operating parameters can achieve the best water purification effect, so as to generate filter state adaptive adjustment rules.

[0109] Step S141: Extract all sub-network segments containing pollutant migration transition nodes from the dynamic association network, each sub-network segment containing filter feature dimension data and operating feature dimension data corresponding to the pollutant migration transition node.

[0110] In the above scenario, the dynamic association network is traversed to find all network parts containing pollutant migration transition nodes, and each pollutant migration transition node and its related filter feature dimension data (such as pore morphology parameters, connection mode) and operating feature dimension data (such as filter speed adjustment parameters, backwashing cycle parameters) within a certain time range before and after the node are extracted to form a sub-network segment. Each sub-network segment revolves around a pollutant migration transition node, reflecting the filter state and operating state of the node.

[0111] Step S142: divide the sub-network segments according to the transition amplitude of the pollutant migration transition node into multiple pollutant migration state categories, each corresponding to a range of pollutant migration intensity.

[0112] In the above scenario, statistical analysis is performed on the transition amplitudes of the pollutant migration transition nodes in all extracted sub-network segments to determine the distribution range of the transition amplitudes. According to the size of the transition amplitudes, the distribution range is divided into multiple consecutive intervals, each corresponding to a pollutant migration state category representing a range of pollutant migration intensity, such as low-intensity migration, medium-intensity migration, high-intensity migration, etc. Each sub-network segment is classified into the corresponding pollutant migration state category according to the interval to which the transition amplitude of its pollutant migration transition node belongs.

[0113] Step S143: for each pollutant migration state category, extract the filter material pore morphology data in all sub-network segments under that category, including cross-sectional morphology parameters and pore connectivity path parameters.

[0114] In the above scenario, for each pollutant migration state category, the filter material pore morphology data in all sub-network segments under that category is collected. The cross-sectional morphology parameters include the contour perimeter, inscribed circle diameter, shape factor, etc. of the pores; the pore connectivity path parameters include the length, width, and connection mode (direct connection, indirect connection) of the connecting channels, etc. The above parameters are summarized to form the filter material pore morphology data set corresponding to the pollutant migration state category.

[0115] Step S144: analyze the correlation between the filter material pore morphology data and the water purification effect under the same pollutant migration state category, and filter out the filter material pore morphology parameter range that meets the preset standard of water purification effect as the optimal filter material pore morphology range under that pollutant migration state category.

[0116] In the above scenario, historical water purification effect data corresponding to each pollutant migration state category is retrieved, including water quality compliance (such as whether turbidity, COD, heavy metal ion concentration, etc. meet the standards) and purification efficiency (such as pollutant removal rate). Correlation analysis is performed between the filter material pore morphology data and the water purification effect data, and correlation analysis methods are used to determine which filter material pore morphology parameters have a significant impact on the water purification effect. Then, according to the preset water purification effect standard (such as water quality compliance and purification efficiency reaching a certain value), the value range of the filter material pore morphology parameters that can make the water purification effect meet the standard is filtered out, and this range is determined as the optimal filter material pore morphology range under the pollutant migration state category.

[0117] Step S1441: retrieve historical water purification effect data corresponding to each pollutant migration state category, the historical water purification effect data including water quality compliance and purification efficiency data.

[0118] In the above scenario, data is retrieved from the historical database of the water purification system, and for each pollutant migration state category, the water purification effect data in the corresponding time period is found according to the time marker of the sub-network segment. The water quality compliance is determined by comparing the water quality indicators with the national standards or design standards; the purification efficiency data is obtained by calculating the removal rate of pollutants (such as (influent pollutant concentration - effluent pollutant concentration) / influent pollutant concentration).

[0119] Step S1442: divide the historical water purification effect data into a compliance data group and a non-compliance data group according to a preset standard, the preset standard being determined based on the effluent water quality requirements of the water purification system.

[0120] In the above scenario, the preset standard is formulated according to the design effluent water quality requirements of the water purification system, including the compliance threshold of each water quality indicator and the minimum requirement of purification efficiency. The data in the historical water purification effect data that meets the conditions of the compliance threshold of each water quality indicator and the minimum requirement of purification efficiency is divided into the compliance data group; the data that does not meet the above conditions is divided into the non-compliance data group.

[0121] Step S1443: extract the filter media pore morphology data corresponding to the compliance data group, including cross-sectional morphology parameters and pore connectivity path parameters.

[0122] In the above scenario, the filter media pore morphology data of the corresponding time period is extracted from the filter media pore morphology data set according to the time marker of the compliance data group, including cross-sectional morphology parameters (contour perimeter, inscribed circle diameter, shape factor) and pore connectivity path parameters (connection channel length, width, connection mode).

[0123] Step S1444: statistically analyze each type of cross-sectional morphology parameter of the compliance data group, calculate the numerical distribution characteristics (such as mean, median and standard deviation) of each type of parameter (such as contour perimeter, inscribed circle diameter, shape factor), and determine the compliance numerical range of each type of parameter.

[0124] In the above scenario, for each cross-sectional morphology parameter (such as contour perimeter) in the compliance data group, the numerical distribution characteristics of all data of the parameter are calculated, such as mean, median and standard deviation. The mean reflects the average level of the data, the median reflects the intermediate level of the data, and the standard deviation reflects the dispersion degree of the data. According to the mean and the standard deviation, the concentrated distribution range of the cross-sectional morphology parameter is determined, which is usually the compliance numerical range covered by the mean plus or minus a certain multiple of the standard deviation, which contains most of the cross-sectional morphology parameter values of the compliance data group.

[0125] Step S1445: Perform statistical analysis on each type of pore connectivity path parameter of the qualified data set, calculate the numerical distribution characteristics of each type of parameter, and determine the qualified numerical value range of each type of parameter.

[0126] In the above scenario, the same method as step S1444 is used to perform statistical analysis on the pore connectivity path parameters (such as connection channel length, width) of the qualified data set, calculate the mean, median and standard deviation, and determine the qualified numerical value range of each pore connectivity path parameter.

[0127] Step S1446: Combine the qualified numerical value range of each type of cross-sectional shape parameter and the qualified numerical value range of each type of pore connectivity path parameter to form an initial candidate filter pore morphology parameter range set.

[0128] In the above scenario, the central distribution range of the cross-sectional shape parameter and the central distribution range of the pore connectivity path parameter are integrated, and the central distribution range of each parameter is taken as the value interval of the parameter in the initial candidate filter pore morphology range. For example, the central distribution range of the contour perimeter is [L1, L2], the central distribution range of the inscribed circle diameter is [D1, D2], and the central distribution range of the connection channel length is [C1, C2], etc. The above intervals are combined to form the initial candidate filter pore morphology range.

[0129] Step S1447: Extract the filter pore morphology data corresponding to the unqualified data set, analyze the differences between it and the initial candidate filter pore morphology range, and determine the parameter value interval that needs to be excluded.

[0130] In the above scenario, the filter pore morphology data of the corresponding time period is extracted according to the time mark of the unqualified data set. The above data is compared with the initial candidate filter pore morphology range to find the value range of the filter pore morphology parameter in the unqualified data set. If some parameter value intervals frequently appear in the unqualified data set but less frequently appear in the qualified data set, the above parameter value intervals are determined as the parameter value intervals that need to be excluded.

[0131] Step S1448: Remove the parameter value intervals that need to be excluded from the initial candidate filter pore morphology range to form an optimized candidate filter pore morphology range.

[0132] In the above scenario, the parameter value intervals that need to be excluded in the initial candidate filter pore morphology range are removed. For each filter pore morphology parameter, subtract the parameter value intervals that need to be excluded from the central distribution range to obtain the optimized parameter value range. Combine the optimized value ranges of all parameters to form the optimized candidate filter pore morphology range.

[0133] Step S1449: verifying the optimized candidate filter media pore morphology range, and calculating a historical water purification effect compliance rate corresponding to the filter media pore morphology data in the optimized candidate filter media pore morphology range.

[0134] In the above scenario, all time periods in which the filter media pore morphology data extracted from the historical database falls within the optimized candidate filter media pore morphology range are extracted, and the ratio of the number of times of meeting the water purification effect compliance to the total number of times in these time periods is calculated to obtain the historical water purification effect compliance rate.

[0135] Step S14410: if the historical water purification effect compliance rate reaches a preset threshold, the optimized candidate filter media pore morphology range is determined as the optimal filter media pore morphology range under the pollutant migration state category; if the historical water purification effect compliance rate does not reach the preset threshold, the statistical analysis method of the parameters is adjusted, and the concentrated distribution range is re-determined until the historical water purification effect compliance rate meets the requirements.

[0136] In the above scenario, the preset threshold is determined according to the reliability requirements of the water purification system. If the calculated historical water purification effect compliance rate reaches or exceeds the preset threshold, it is considered that the optimized candidate filter media pore morphology range can stably ensure the water purification effect, and it is determined as the optimal filter media pore morphology range under the pollutant migration state category. If the preset threshold is not reached, the statistical analysis method of the parameters is adjusted, such as changing the multiple of the mean plus or minus the standard deviation when determining the concentrated distribution range, using other statistical quantities (such as the quartile range) to determine the concentrated distribution range, etc., and the operations of steps S1444 to S1449 are re-performed until the historical water purification effect compliance rate meets the preset threshold requirements.

[0137] Step S145: extracting the operation parameter data in all sub-network segments under each pollutant migration state category, including filter speed adjustment parameters and backwashing cycle parameters.

[0138] In the above scenario, for each pollutant migration state category, the operation parameter data in all sub-network segments under the category is collected. The filter speed adjustment parameters include filter speed values at different times, filter speed adjustment rates, etc.; the backwashing cycle parameters include time intervals of backwashing, backwashing durations, etc. The above parameters are summarized to form an operation parameter data set corresponding to the pollutant migration state category.

[0139] Step S146: analyzing the correlation between the operation parameter data and the filter media pore morphology parameters under the same pollutant migration state category, determining an operation parameter combination that can maintain the filter media pore morphology in the optimal filter media pore morphology range as a candidate operation parameter combination under the pollutant migration state category.

[0140] In the above scenario, a multiple regression analysis or a machine learning method is used to analyze the correlation between the operating parameter data and the filter media pore morphology parameters. The goal is to maintain the filter media pore morphology parameters within the optimal filter media pore morphology range, and to find the value combination of the operating parameters that can achieve this goal. For example, the trend of the filter media pore morphology parameters under different filtration rates and backwashing cycles is analyzed, and the combination of filtration rates and backwashing cycles that can keep the filter media pore morphology parameters within the optimal range is selected as the candidate operating parameter combination for the pollutant migration state category.

[0141] Step S147: The candidate operating parameter combination for each pollutant migration state category is input into the dynamic correlation network for simulation verification, and whether the filter media pore morphology remains within the optimal filter media pore morphology range during the simulation process is observed.

[0142] In the above scenario, a dynamic correlation network is used to build a simulation model, and the candidate operating parameter combination for each pollutant migration state category is input into the simulation model. The simulation model simulates the evolution process of the filter media pore morphology under the action of the operating parameter combination according to the characteristic correlation recorded in the dynamic correlation network. During the simulation process, it is continuously monitored whether the filter media pore morphology parameters fluctuate within the optimal filter media pore morphology range. If the filter media pore morphology parameters can be maintained within the optimal range throughout the simulation period, the candidate operating parameter combination is considered effective.

[0143] Step S148: According to the verification result, the value range of the candidate operating parameter combination is adjusted, and the parameter values that cause the filter media pore morphology to deviate from the optimal filter media pore morphology range are eliminated, forming the optimal operating parameter combination for the pollutant migration state category.

[0144] In the above scenario, for the candidate operating parameter combination that causes the filter media pore morphology to deviate from the optimal filter media pore morphology range during the simulation verification process, the deviation reason is analyzed to determine which operating parameter values cause the deviation. Then the value range of these operating parameters is adjusted, and the parameter values that cause the deviation are eliminated. For the adjusted candidate operating parameter combination, simulation verification is performed again, and the process is repeated until the candidate operating parameter combination can keep the filter media pore morphology within the optimal filter media pore morphology range. At this time, the candidate operating parameter combination is the optimal operating parameter combination for the pollutant migration state category.

[0145] Step S149: The pollutant migration state category, the optimal filter media pore morphology range, and the optimal operating parameter combination are bound to form the basic adjustment rule.

[0146] In the above scenario, each pollutant migration state category is associated with a corresponding optimal filter pore morphology range and optimal operating parameter combination to form a basic adjustment rule. The form of the basic adjustment rule can be "when the pollutant migration state is a certain category, the filter pore morphology should be maintained in a certain optimal range, and the corresponding operating parameter combination is a certain combination".

[0147] Step S1410: Analyze the transition interval between different pollutant migration state categories, supplement the corresponding transition adjustment rule for each transition interval, and ensure that the adjustment rules of adjacent pollutant migration state categories can be smoothly connected.

[0148] In the above scenario, there is a transition interval between different pollutant migration state categories, i.e. the process of pollutant migration state transition from one category to another. By analyzing the variation characteristics of the pollutant migration state and the evolution law of the filter pore morphology in the transition interval, and according to the basic adjustment rules of the adjacent two pollutant migration state categories, the transition adjustment rule is formulated. In the transition adjustment rule, the operating parameter combination should gradually transition from the optimal operating parameter combination corresponding to the current category to the optimal operating parameter combination corresponding to the adjacent category, avoiding sudden changes in operating parameters that cause drastic changes in filter pore morphology, and ensuring smooth connection of the adjustment rules of adjacent categories.

[0149] Step S1411: Integrate all basic adjustment rules and transition adjustment rules to form filter state adaptive adjustment rules, each of which describes the optimal matching mode of filter pore morphology and operating parameters under a specific pollutant migration state.

[0150] In the above scenario, all basic adjustment rules of different pollutant migration state categories and transition adjustment rules of transition intervals are integrated and arranged in the order of pollutant migration state variation to form a complete set of filter state adaptive adjustment rules. Each rule in the set of filter state adaptive adjustment rules is for a specific pollutant migration state (including stable state and transition state), and describes the optimal range of filter pore morphology that should be maintained under that state and the matching operating parameter combination, achieving the optimal matching of filter pore morphology and operating parameters under different pollutant migration states.

[0151] Step S150: Generating the operating parameter dynamic correction instruction of the volcanic rock water purification system according to the filter state adaptive adjustment rule, the operating parameter dynamic correction instruction includes filter speed real-time adjustment parameter and backwashing cycle adjustment parameter.

[0152] In the above scenario of the municipal waterworks' volcanic rock filter depth purification system, the current pollutant migration state is monitored in real time, matched with the pollutant migration state categories in the filter material state adaptive adjustment rule, and the corresponding optimal operation parameter combination is determined. According to the deviation of the current filter material pore morphology and operation parameters from the optimal values, the operation parameters are adjusted, dynamic correction instructions containing real-time filter speed adjustment parameters and backwashing cycle adjustment parameters are generated and sent to the control system of the purification system, realizing automatic adjustment of the operation parameters.

[0153] Step S151: Real-time collection of current pollutant migration data of the volcanic rock purification system, extraction of pollutant spatial distribution parameters and chemical form parameters.

[0154] In the above scenario, pollutant migration data in the volcanic rock purification system is collected in real time by water quality sensors, including spatial distribution parameters such as spatial distribution position and concentration gradient of pollutants in the filter material layer, and chemical form parameters such as molecular structure characteristics, functional group types, and chemical bond types of pollutants. The data collected by the sensors are transmitted in real time to the data processing unit, which analyzes and extracts the data to obtain the current pollutant spatial distribution parameters and chemical form parameters.

[0155] Step S152: Matching the pollutant spatial distribution parameters and the chemical form parameters with the pollutant migration state categories in the filter material state adaptive adjustment rule to determine the current pollutant migration state category.

[0156] In the above scenario, the real-time extracted pollutant spatial distribution parameters and chemical form parameters are compared with the characteristic parameters of each pollutant migration state category defined in the filter material state adaptive adjustment rule. A pattern recognition method is used to calculate the similarity of the current parameters with the characteristic parameters of each category, and the category with the highest similarity is determined as the current pollutant migration state category. If the current parameters are in the transition interval between two adjacent categories, it is determined as a transition state.

[0157] Step S153: According to the matching result, the optimal filter material pore morphology range and the optimal operation parameter combination corresponding to the pollutant migration state category are retrieved.

[0158] In the above scenario, according to the determined current pollutant migration state category, the corresponding rule entry is searched from the filter material state adaptive adjustment rule set, and the optimal filter material pore morphology range and optimal operation parameter combination (including filter speed adjustment parameters and backwashing cycle parameters) recorded in the rule entry are retrieved.

[0159] Step S154: Collect the current filter material pore morphology data of the volcanic rock water purification system in real time, compare each type of parameter contained therein with the corresponding standard value range of each type of parameter in the optimal filter material pore morphology parameter range set, and calculate the deviation parameters of both.

[0160] In the above scenario, the current filter material pore image is collected in real time by the microscopic imaging device, and the current filter material pore morphology data (cross-sectional morphology parameters and pore connectivity path parameters) are obtained through image segmentation and feature extraction. The current filter material pore morphology data are compared with the retrieved optimal filter material pore morphology range. For each pore morphology parameter, the difference between the current parameter value and the center value or boundary value of the optimal range is calculated, and the above difference is integrated into a deviation parameter for representing the deviation degree of the current filter material pore morphology from the optimal range.

[0161] Step S155: Judge the deviation degree of the overall state of the filter material pores according to the deviation parameters of each type of parameter, and adjust the filter speed adjustment parameter and the backwashing cycle parameter in the optimal operating parameter combination according to the comprehensive deviation degree. The larger the comprehensive deviation degree, the greater the adjustment amplitude.

[0162] In the above scenario, a mapping relationship between the deviation parameter and the operating parameter adjustment amplitude is established. When the deviation parameter is zero, it indicates that the current filter material pore morphology is within the optimal range, and there is no need to adjust the operating parameter. When the deviation parameter is not zero, the adjustment amplitude is determined according to the size of the deviation parameter. The larger the deviation parameter, the greater the adjustment amplitude. For the filter speed adjustment parameter, if the current filter material pore morphology parameter is less than the lower limit of the optimal range (e.g., the pore is too small), the filter speed is increased; if it is greater than the upper limit of the optimal range (e.g., the pore is too large), the filter speed is decreased. For the backwashing cycle parameter, if the current filter material pore morphology parameter is less than the lower limit of the optimal range, the backwashing cycle is shortened; if it is greater than the upper limit of the optimal range, the backwashing cycle is extended.

[0163] For example, step S1551: determine the numerical range of the deviation parameter, divide the numerical range into multiple deviation levels, and each deviation level corresponds to a different adjustment amplitude coefficient.

[0164] In the above scenario, the possible numerical range of the deviation parameter is determined according to historical data and experimental results. The numerical range is divided into multiple continuous intervals, and each interval corresponds to a deviation level, such as slight deviation, moderate deviation, and severe deviation. An adjustment amplitude coefficient is assigned to each deviation level. The higher the deviation level, the greater the adjustment amplitude coefficient, indicating that a greater adjustment of the operating parameter is needed.

[0165] Step S1552: Assign a corresponding filter speed adjustment amplitude coefficient and backwashing cycle adjustment amplitude coefficient to each deviation level. The higher the deviation level, the greater the adjustment amplitude coefficient.

[0166] In the above scenario, the filter speed adjustment amplitude coefficient and the backwashing cycle adjustment amplitude coefficient are respectively assigned to each deviation level according to the degree of influence of the filter material pore morphology under different deviation levels. For example, a slight deviation level corresponds to a smaller adjustment amplitude coefficient, and a severe deviation level corresponds to a larger adjustment amplitude coefficient.

[0167] Step S1553: Extract the reference filter speed parameter and the reference backwashing cycle parameter in the optimal operating parameter combination.

[0168] In the above scenario, the filter speed adjustment parameter and the backwashing cycle parameter in the optimal operating parameter combination are reference values, respectively referred to as the reference filter speed parameter and the reference backwashing cycle parameter. These two reference parameters are extracted from the optimal operating parameter combination.

[0169] Step S1554: Multiply the reference filter speed parameter by the filter speed adjustment amplitude coefficient corresponding to the current deviation level to obtain the adjustment amount of the filter speed parameter.

[0170] In the above scenario, the deviation level to which the current deviation parameter belongs is determined according to the current deviation parameter, and the filter speed adjustment amplitude coefficient corresponding to the deviation level is obtained. The reference filter speed parameter is multiplied by the filter speed adjustment amplitude coefficient to obtain the adjustment amount of the filter speed parameter.

[0171] Step S1555: Determine the adjustment direction of the filter speed parameter according to the deviation direction of the filter material pore morphology data from the optimal filter material pore morphology range. If the current pore morphology parameter is less than the lower limit of the optimal filter material pore morphology range, increase the filter speed; if it is greater than the upper limit of the optimal filter material pore morphology range, decrease the filter speed.

[0172] In the above scenario, the current filter material pore morphology parameter is compared with the lower limit and the upper limit of the optimal filter material pore morphology range. If the current parameter is less than the lower limit, it indicates that the filter material pore may be at risk of clogging, and the filter speed needs to be increased to enhance the flushing effect of water flow on the pore; if the current parameter is greater than the upper limit, it indicates that the filter material pore may be too loose, and the filter speed needs to be decreased to prolong the residence time of pollutants in the filter material layer and improve the adsorption effect. The adjustment direction of the filter speed parameter is determined according to the comparison result.

[0173] Step S1556: Multiply the reference backwashing cycle parameter by the backwashing cycle adjustment amplitude coefficient corresponding to the current deviation level to obtain the adjustment amount of the backwashing cycle parameter.

[0174] In the above scenario, the reference backwashing cycle parameter is multiplied by the backwashing cycle adjustment amplitude coefficient corresponding to the current deviation level to obtain the adjustment amount of the backwashing cycle parameter, using a method similar to step S1554.

[0175] Step S1557: Determine the adjustment direction of the backwashing cycle according to the deviation direction. If the current pore morphology parameter is less than the lower limit of the optimal filter material pore morphology range, shorten the backwashing cycle. If it is greater than the upper limit of the optimal filter material pore morphology range, lengthen the backwashing cycle.

[0176] In the above scenario, if the current filter material pore morphology parameter is less than the lower limit of the optimal range, it indicates that the filter material pore is blocked to a high degree, and the backwashing cycle needs to be shortened to restore the pore morphology more frequently. If the current parameter is greater than the upper limit of the optimal range, it indicates that the filter material pore is relatively unobstructed, and the backwashing cycle can be appropriately lengthened to reduce the disturbance of backwashing to the filter material layer structure. The adjustment direction of the backwashing cycle is determined according to the deviation direction.

[0177] Step S1558: Calculate the adjusted filter speed parameter value, which is the result of the operation of the reference filter speed parameter and the filter speed adjustment amount in the adjustment direction.

[0178] In the above scenario, according to the determined filter speed adjustment direction, if the filter speed needs to be increased, the adjusted filter speed parameter value is the reference filter speed parameter plus the filter speed adjustment amount; if the filter speed needs to be reduced, the adjusted filter speed parameter value is the reference filter speed parameter minus the filter speed adjustment amount.

[0179] Step S1559: Calculate the adjusted backwashing cycle parameter value, which is the result of the operation of the reference backwashing cycle parameter and the backwashing cycle adjustment amount in the adjustment direction.

[0180] In the above scenario, according to the determined backwashing cycle adjustment direction, if the backwashing cycle needs to be shortened, the adjusted backwashing cycle parameter value is the reference backwashing cycle parameter minus the backwashing cycle adjustment amount; if the backwashing cycle needs to be lengthened, the adjusted backwashing cycle parameter value is the reference backwashing cycle parameter plus the backwashing cycle adjustment amount.

[0181] Step S15510: Check whether the adjusted parameter value exceeds the allowed maximum and minimum range, and limit it within the allowed range if it does.

[0182] In the above scenario, the filter speed and backwashing cycle parameters of the water purification system have designed maximum and minimum ranges to ensure the safe and stable operation of the system. The adjusted filter speed parameter value and backwashing cycle parameter value are compared with the corresponding maximum and minimum ranges, respectively. If it exceeds the range, the parameter value is limited to the maximum or minimum value within the allowed range.

[0183] Step S15511: Record the parameter values before and after adjustment and the corresponding deviation parameters and adjustment amplitude coefficients to form a parameter adjustment record.

[0184] In the above scenario, the reference filter speed parameter before adjustment, the reference backwashing period parameter, the adjusted filter speed parameter value, the backwashing period parameter value, the corresponding deviation parameter and the adjustment amplitude coefficient and other information are recorded to form a parameter adjustment record and stored in the log database of the system, facilitating subsequent system operation analysis and parameter optimization.

[0185] Step S156: Extract the historical associated segment similar to the current pollutant migration state category and filter material pore morphology deviation parameter in the dynamic association network, and obtain the operation parameter adjustment effect data in the historical associated segment.

[0186] In the above scenario, search in the dynamic association network to find a historical associated segment that is the same or similar to the current pollutant migration state category and has a similar filter material pore morphology deviation parameter. For the found historical associated segment, extract the filter material pore morphology recovery condition after adjustment of the recorded operation parameters, the change of water purification effect and other operation parameter adjustment effect data.

[0187] Step S157: Further optimize the adjusted filter speed adjustment parameter and backwashing period parameter based on the operation parameter adjustment effect data, and convert the optimized filter speed adjustment parameter into a specific filter speed change rate and change duration to form a filter speed real-time adjustment parameter.

[0188] In the above scenario, analyze the operation parameter adjustment effect data of the historical associated segment. If the similar adjustment in history leads to good filter material pore morphology recovery effect and stable water purification effect, the current adjustment parameter is maintained. If the historical adjustment effect is not good, fine-tune the adjusted filter speed adjustment parameter and backwashing period parameter according to the feedback of the effect data. Convert the optimized filter speed adjustment parameter into a filter speed change rate per unit time (such as an increased or decreased filter speed value per minute) and a change duration required to complete the adjustment. These two parameters together constitute a filter speed real-time adjustment parameter.

[0189] Step S158: Convert the optimized backwashing period parameter into a specific period shortening or lengthening duration to form a backwashing period adjustment parameter.

[0190] In the above scenario, according to the difference between the optimized backwashing period parameter and the current backwashing period, the duration of shortening or lengthening required is calculated. If the optimized backwashing period is less than the current period, it is the period shortening duration; if it is greater than the current period, it is the period lengthening duration. The shortening or lengthening duration is determined as the backwashing period adjustment parameter.

[0191] Step S159: Arrange the filter speed real-time adjustment parameter and the backwashing period adjustment parameter in chronological order and add corresponding execution time markers.

[0192] In the above scenario, the filter rate real-time adjustment parameter and the backwash cycle adjustment parameter are time-sequenced according to the priority of the system operation and the urgency of the adjustment. Generally, the filter rate real-time adjustment parameter needs to be executed immediately, while the backwash cycle adjustment parameter is executed after the current backwash cycle ends. A corresponding execution time tag is added to each adjustment parameter to specify the time point when the parameter takes effect.

[0193] Step S1510: The arranged parameters and time tags are integrated to generate a dynamic correction instruction of the operation parameters of the volcanic rock water purification system.

[0194] In the above scenario, the filter rate real-time adjustment parameter, the backwash cycle adjustment parameter, and the corresponding execution time tag arranged in time sequence are integrated to form a formatted dynamic correction instruction of the operation parameters. The instruction contains information such as parameter type (filter rate adjustment or backwash cycle adjustment), adjustment value (change rate, change duration, shortened or lengthened duration), and execution time. The instruction is sent to the control system of the volcanic rock water purification system, and the control system executes the corresponding operation parameter adjustment operation according to the instruction content.

[0195] Based on the same inventive concept, please refer to Figure 2 , which shows a structural schematic block diagram of a data fusion processing system 100 applied to a volcanic rock water purification system for executing the above-mentioned video stream processing method, which can include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0196] In this embodiment, the machine-readable storage medium 120 and the processor 130 are both located in the data fusion processing system 100 applied to the volcanic rock water purification system and are separately arranged. However, it should be understood that the machine-readable storage medium 120 can also be independent of the data fusion processing system 100 applied to the volcanic rock water purification system, and can be accessed by the processor 130 through a bus interface. Alternatively, the machine-readable storage medium 120 can also be integrated into the processor 130 and can communicate and interact with external systems through the communication unit 110.

[0197] The processor 130 is a control center of the data fusion processing system 100 applied to the volcanic rock water purification system, connects each part of the data fusion processing system 100 applied to the volcanic rock water purification system through various interfaces and lines, executes the software programs and / or modules stored in the machine readable storage medium 120 and calls the data stored in the machine readable storage medium 120, executes various functions and processes data of the data fusion processing system 100 applied to the volcanic rock water purification system, and thus monitors the data fusion processing system 100 applied to the volcanic rock water purification system as a whole. Optionally, the processor 130 can include one or more processing cores; for example, the processor 130 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, the user interface and the application program, and the modem processor mainly processes the wireless communication. It can be understood that the above-mentioned modem processor can also not be integrated into the processor. The machine readable storage medium 120 is used to store machine executable instructions for executing the scheme of the present application, and the processor 130 is used to execute the machine executable instructions stored in the machine readable storage medium 120 to realize the method for processing the inspection video stream provided by the foregoing method embodiment.

[0198] It should be noted that, in order to simplify the expression of the present disclosure and to help the understanding of one or more embodiments of the present application, in the foregoing description of the embodiments of the present application, various features are sometimes combined into one embodiment, figure or description thereof.

Claims

1. A data fusion processing method applied to a volcanic rock water purification system, characterized in that, The method comprises: acquiring volcanic rock filter material pore evolution data, pollutant adsorption and migration data, and filter pool operation feedback data, the volcanic rock filter material pore evolution data recording the change in pore morphology between filter material particles over time during the water purification process, the pollutant adsorption and migration data recording the change in position and morphology of pollutants in the filter material layer, and the filter pool operation feedback data recording the change in filter material state caused by filter speed adjustment and backwashing operation; performing time sequence correlation capture processing on the volcanic rock filter material pore evolution data, the pollutant adsorption and migration data, and the filter pool operation feedback data, and identifying the morphology transition nodes and transition amplitudes of the volcanic rock filter material pore evolution data, the pollutant adsorption and migration data, and the filter pool operation feedback data in the same time process; the morphology transition nodes include filter material pore morphology transition nodes, filter material pore connection transition nodes, pollutant migration transition nodes, pollutant morphology transition nodes, filter speed adjustment transition nodes, and backwashing operation transition nodes; constructing a dynamic correlation network based on the morphology transition nodes and the transition amplitudes, the dynamic correlation network being a series connection of the interaction relationship between filter material pore morphology change, pollutant migration path change, and operation parameter adjustment along the time flow axis; generating a filter material state self-adaptive adjustment rule according to the dynamic correlation network, the self-adaptive adjustment rule being used to describe the optimal matching mode of filter material pore morphology and operation parameters under different pollutant migration states; generating an operation parameter dynamic correction instruction of a volcanic rock water purification system according to the filter material state self-adaptive adjustment rule, the operation parameter dynamic correction instruction including filter speed real-time adjustment parameters and backwashing cycle adjustment parameters; extracting a cross-sectional morphology time sequence and a pore connection path time sequence of filter material pores from the volcanic rock filter material pore evolution data, the cross-sectional morphology time sequence recording the geometric shape data of pores at different times, and the pore connection path time sequence recording the connection mode data between pores at different times; extracting a spatial distribution time sequence and a pollutant chemical morphology conversion time sequence of pollutants in the filter material layer from the pollutant adsorption and migration data, the spatial distribution time sequence recording the position coordinate data of pollutants at different times, and the chemical morphology conversion time sequence recording the molecular structure data of pollutants at different times; extracting a filter speed change time sequence and a backwashing effect time sequence from the filter pool operation feedback data, the filter speed change time sequence recording the water flow speed data at different times, and the backwashing effect time sequence recording the filter material state recovery data after backwashing; unifying the time bases of the cross-sectional morphology time sequence, the pore connection path time sequence, the spatial distribution time sequence, the chemical morphology conversion time sequence, the filter speed change time sequence, and the backwashing effect time sequence into the same time unit, and comparing the mutation positions of pore geometric shapes in the cross-sectional morphology time sequence in segments with the unified time base, and marking the filter material pore morphology transition nodes. The mutation position of the connection mode in the pore connection path time sequence is marked as a filter pore connection transition node; the mutation position of the pollutant position coordinate in the spatial distribution time sequence is marked as a pollutant migration transition node; and the mutation position of the molecular structure in the chemical form transformation time sequence is marked as a pollutant form transition node; The mutation position of the water flow velocity in the filter speed change time sequence is marked as a filter speed adjustment transition node; and the starting position of the filter material state recovery in the backwashing effect time sequence is marked as a backwashing operation transition node; The relative change rate or the standardized change amount of the corresponding data sequence at each transition node is calculated as the transition amplitude of the transition node in its own data dimension. All the form transition nodes are arranged in chronological order to form a transition node sequence on a unified time axis, and the transition amplitude of each transition node is associated.

2. The data fusion processing method applied to the volcanic rock water purification system according to claim 1, characterized in that, The cross-sectional form time sequence and the pore connection path time sequence of the filter material pores are extracted from the volcanic rock filter material pore evolution data, including: The volcanic rock filter material pore evolution data is analyzed, and the volcanic rock filter material pore evolution data includes a pore image sequence of different depths of the filter material layer collected by a microscopic imaging technology; Image segmentation processing is performed on the pore image sequence to separate the pore region and the non-pore region in each pore image; For each segmented pore region, the geometric shape parameters of the cross section thereof are extracted, including the contour perimeter, the inscribed circle diameter and the shape factor, which together constitute the geometric shape data of the pore; The geometric shape data is arranged in the time sequence of the image sequence to form the cross-sectional form time sequence of the filter material pores; The connectivity analysis is performed on the pore image sequence to identify the connection channels between adjacent pores and the width and length of the connection channels; The connection mode between the pores is determined according to the distribution of the connection channels, including direct connection, indirect connection and isolated state; The connection mode and the corresponding connection channel parameters of all the pores at each time are recorded to form the pore connection path time sequence in chronological order; The cross-sectional form time sequence and the pore connection path time sequence are added with the same time mark.

3. The data fusion processing method applied to the volcanic rock water purification system according to claim 2, characterized in that, The connectivity analysis is performed on the pore image sequence to identify the connection channels between adjacent pores and the width and length of the connection channels, including: Binary processing is performed on each image in the segmented pore image sequence, and the pore region is marked as a first value and the non-pore region is marked as a second value; A morphological erosion operation is performed on the binary image for pretreatment, and a connected component labeling algorithm is used to analyze the pretreated image to identify each independent pore region and assign a unique identifier; Distance transformation processing is performed on the labeled pore region to calculate the distance from each non-pore region pixel to the nearest pore region; Potential connection channel regions are identified according to the distance transformation results, and the potential connection channel regions are non-pore regions close to two or more pore regions; A morphological dilation operation is performed on the potential connection channel region, a skeleton extraction algorithm is used to extract the center axis of the dilated connection channel region as the path of the connection channel, and the length of the center axis of the connection channel is measured as the length parameter of the connection channel; At the start and end positions of the connection channel, the channel width perpendicular to the center axis direction is measured, the average value is taken as the width parameter of the connection channel, and the start aperture identification, end aperture identification, length parameter and width parameter of each connection channel are associated and stored to complete the identification of the connection channel between adjacent pores.

4. The data fusion processing method applied to the volcanic rock water purification system according to claim 1, characterized in that, The dynamic association network is constructed based on the morphological transition node and the transition amplitude, including: A network framework is constructed with time flow as the longitudinal axis, the longitudinal dimension of the network framework is the time process, and the transverse dimension is the filter characteristics dimension, the pollutant characteristics dimension and the operation characteristics dimension; The morphological transition node is embedded in the corresponding transverse dimension position of the network framework according to the belonging characteristic dimension, and different characteristic dimension transition nodes at the same time point are associated through transverse connection lines; For different characteristic dimension transition nodes at the same time point, their transition directions are judged respectively and their standardized transition amplitude levels are recorded, and the line width of the transverse connection line is adjusted according to the correlation of the transition direction and the matching degree of the amplitude level, the higher the correlation and the more matched the amplitude, the wider the line width; In the network framework, longitudinal connection lines are established between filter characteristic dimension transition nodes at adjacent time points, and the pore morphology change direction corresponding to the connection line is marked; Longitudinal connection lines are established between pollutant characteristic dimension transition nodes at adjacent time points, and the migration path change direction corresponding to the connection line is marked; Longitudinal connection lines are established between operation characteristic dimension transition nodes at adjacent time points, and the parameter adjustment direction corresponding to the connection line is marked; For the same characteristic dimension transition nodes at adjacent time points, the difference between their standardized transition amplitude levels is calculated, and the color depth of the longitudinal connection line is adjusted according to the size of the transition amplitude level difference, the larger the level difference, the deeper the color; The time interval in which filter characteristic dimension and pollutant characteristic dimension transition nodes appear at the same time in the network framework is identified, and a bidirectional arrow mark is added to indicate the interaction relationship between the two in the time interval; After identifying the time interval in which filter characteristic dimension or pollutant characteristic dimension transition nodes appear after the appearance of operation characteristic dimension transition nodes in the network framework, a unidirectional arrow mark is added to indicate the influence relationship of the operation parameter adjustment on the filter or the pollutant in the time interval; The occurrence frequency of each type of connection line in the network framework is counted, and the connection mode with the highest occurrence frequency is marked as the core association mode; The core association mode is taken as the skeleton, and other connection modes are supplemented to form a dynamic association network containing time flow, characteristic dimension, transition node, transition amplitude and interaction relationship.

5. The data fusion processing method applied to the volcanic rock water purification system according to claim 4, characterized in that, The time interval in which filter characteristic dimension and pollutant characteristic dimension transition nodes appear at the same time in the network framework is identified, and a bidirectional arrow mark is added to indicate the interaction relationship between the two in the time interval, including: The time axis of the network framework is traversed, and whether filter characteristic dimension transition nodes and pollutant characteristic dimension transition nodes appear at the same time in the same time interval is checked section by section; For each time interval in which two types of transition nodes coexist, determine the start time and end time of the time interval, and mark it as an interaction interval; Extract the transition amplitude of the filter feature dimension transition node and the transition amplitude of the pollutant feature dimension transition node within the interaction interval; Calculate the ratio of the filter feature transition amplitude to the pollutant feature transition amplitude, which represents the proportion of the strength of their interaction; Determine the arrow size of the bidirectional arrow according to the strength ratio. The closer the ratio is to 1, the more consistent the arrow size is; Analyze the change direction of the filter feature dimension data and the change direction of the pollutant feature dimension data within the interaction interval to determine the direction relationship of their interaction; Label the direction relationship on the bidirectional arrow, including promotion relationship and inhibition relationship. The promotion relationship indicates that the change of one side intensifies the change of the other side, and the inhibition relationship indicates that the change of one side slows down the change of the other side; Statistical the change synchronization rate of the two types of feature dimension data within the interaction interval. The higher the synchronization rate, the thicker the line of the bidirectional arrow; Add the labeled bidirectional arrow to the interaction interval of the network framework, connecting the corresponding filter feature dimension transition node and pollutant feature dimension transition node; Number all the added bidirectional arrows to establish a correspondence table between the arrows and the interaction intervals.

6. The data fusion processing method for application to a volcanic rock water purification system according to claim 1, characterized by, The filter state adaptive adjustment rule generated according to the dynamic correlation network includes: Extract all sub-network segments containing pollutant migration transition nodes from the dynamic correlation network. Each sub-network segment contains filter feature dimension data and operating feature dimension data corresponding to the pollutant migration transition node; Divide the sub-network segments into multiple pollutant migration state categories according to the transition amplitude of the pollutant migration transition node. Each pollutant migration state category corresponds to a pollutant migration intensity range; For each pollutant migration state category, extract the filter pore morphology data from all sub-network segments in that category, including cross-sectional morphology parameters and pore connectivity path parameters; Analyze the correlation between filter pore morphology data and water purification effect under the same pollutant migration state category, and filter out the filter pore morphology parameter range that meets the preset standard as the optimal filter pore morphology range for that pollutant migration state category; Extract the operating parameter data from all sub-network segments in each pollutant migration state category, including filter speed adjustment parameters and backwashing cycle parameters; Analyze the correlation between operating parameter data and filter pore morphology parameters under the same pollutant migration state category, and determine the operating parameter combination that can maintain the filter pore morphology within the optimal filter pore morphology range as the candidate operating parameter combination for that pollutant migration state category; Substitute the candidate operating parameter combination of each pollutant migration state category into the dynamic correlation network for simulation verification, and observe whether the filter pore morphology remains within the optimal filter pore morphology range during the simulation process; Adjust the numerical range of the candidate operating parameter combination according to the verification result, and eliminate parameter values that cause the filter pore morphology to deviate from the optimal filter pore morphology range, forming the optimal operating parameter combination for that pollutant migration state category; The basic adjustment rule is formed by binding the pollutant migration state category, the optimal filter pore morphology range, and the optimal operation parameter combination; The transition interval between different pollutant migration state categories is analyzed, and the corresponding transition adjustment rule is supplemented for each transition interval to ensure that the adjustment rules of adjacent pollutant migration state categories can be smoothly connected; All the basic adjustment rules and the transition adjustment rules are integrated to form the filter state adaptive adjustment rule, and each filter state adaptive adjustment rule describes the optimal matching mode of the filter pore morphology and the operation parameter under a specific pollutant migration state.

7. The data fusion processing method applied to the volcanic rock water purification system according to claim 6, characterized in that, The correlation between the filter pore morphology data and the water purification effect under the same pollutant migration state category is analyzed, and the filter pore morphology parameter range with the water purification effect meeting the preset standard is selected as the optimal filter pore morphology range under the pollutant migration state category, including: The historical water purification effect data corresponding to each pollutant migration state category is retrieved, and the historical water purification effect data includes the water quality standard reaching condition and the purification efficiency data; The historical water purification effect data is divided into a standard reaching data group and a non-standard reaching data group according to a preset standard, and the preset standard is determined based on the water quality requirement of the water purification system; The filter pore morphology data corresponding to the standard reaching data group is extracted, including the cross-sectional morphology parameter and the pore connectivity path parameter; The statistical analysis is performed on each type of cross-sectional morphology parameter of the standard reaching data group, the numerical distribution characteristics of each type of parameter are calculated, and the standard reaching numerical range of each type of parameter is determined; The statistical analysis is performed on each type of pore connectivity path parameter of the standard reaching data group, the numerical distribution characteristics of each type of parameter are calculated, and the standard reaching numerical range of each type of parameter is determined; The standard reaching numerical range of each type of cross-sectional morphology parameter and the standard reaching numerical range of each type of pore connectivity path parameter are combined to form an initial candidate filter pore morphology parameter range set; The filter pore morphology data corresponding to the non-standard reaching data group is extracted, the difference between the initial candidate filter pore morphology range and the filter pore morphology data is analyzed, and the parameter value interval that needs to be excluded is determined; The parameter value interval that needs to be excluded is removed from the initial candidate filter pore morphology range to form an optimized candidate filter pore morphology range; The optimized candidate filter pore morphology range is verified, and the historical water purification effect reaching rate of the filter pore morphology data in the optimized candidate filter pore morphology range is calculated; If the historical water purification effect reaching rate reaches a preset threshold, the optimized candidate filter pore morphology range is determined as the optimal filter pore morphology range under the pollutant migration state category; If the historical water purification effect reaching rate does not reach the preset threshold, the statistical analysis method of the parameter is adjusted, the concentrated distribution range is re-determined, and the historical water purification effect reaching rate is satisfied until the requirement is met.

8. The data fusion processing method for application to a volcanic rock water purification system according to claim 1, characterized by, The operation parameter dynamic correction instruction of the volcanic rock water purification system is generated according to the filter state adaptive adjustment rule, including: The pollutant migration data of the volcanic rock water purification system is collected in real time, and the pollutant spatial distribution parameter and the chemical form parameter are extracted; The pollutant spatial distribution parameter and the chemical form parameter are matched with the pollutant migration state category in the filter material state adaptive adjustment rule to determine the pollutant migration state category to which the current belongs; According to the matching result, the optimal filter material pore shape range and the optimal operation parameter combination corresponding to the pollutant migration state category are called; The current filter material pore shape data of the volcanic rock water purification system is collected in real time, and each type of parameter contained therein is compared with the standard value range of each type of parameter corresponding to the optimal filter material pore shape parameter range set, and the deviation parameters of the two are calculated; The deviation degree of the overall state of the filter material pore is comprehensively judged according to the deviation parameters of each type of parameter, and the filter speed adjustment parameter and the backwashing cycle parameter in the optimal operation parameter combination are adjusted according to the comprehensive deviation degree, and the larger the comprehensive deviation degree is, the larger the adjustment range is; The historical correlation fragment similar to the current pollutant migration state category and the filter material pore shape deviation parameter is extracted from the dynamic correlation network, and the operation parameter adjustment effect data in the historical correlation fragment is obtained; The filter speed adjustment parameter and the backwashing cycle parameter after adjustment are further optimized based on the operation parameter adjustment effect data, the optimized filter speed adjustment parameter is converted into specific filter speed change rate and change time length, and the filter speed real-time adjustment parameter is formed; The optimized backwashing cycle parameter is converted into specific cycle shortening or lengthening time length, and the backwashing cycle adjustment parameter is formed; The filter speed real-time adjustment parameter and the backwashing cycle adjustment parameter are arranged in time sequence, and corresponding execution time marks are added; The integrated and arranged parameters and time marks are generated to form the operation parameter dynamic correction instruction of the volcanic rock water purification system.

9. A data fusion processing system applied to a volcanic rock water purification system, characterized in that, It comprises: a processor; a machine readable storage medium for storing machine executable instructions of the processor; wherein the processor is configured to execute the machine executable instructions to perform the data fusion processing method applied to the volcanic rock water purification system according to any one of claims 1 to 8.

Images