A tail water wetland clogging regulation method, device, equipment and medium

Abstract

This invention relates to the technical field of wastewater treatment, and in particular to a method, apparatus, equipment, and medium for controlling clogging in wastewater wetlands. The method includes performing time-series correlation processing based on clogging characterization information to obtain a clogging evolution sequence corresponding to each filter layer; performing clogging prediction processing based on the clogging evolution sequence to obtain a target filter layer and its corresponding clogging prediction result information; performing strategy matching processing based on the target filter layer and the clogging prediction result information to obtain adaptive control strategy information; performing target filter path adjustment processing or target filter layer recovery processing based on the adaptive control strategy information; obtaining feedback monitoring information after the target filter path adjustment processing or target filter layer recovery processing; and correcting the clogging prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information. This invention improves the targeting of filter layer clogging identification and the adaptability of the control process.

Description

Technical Field

[0001] This invention relates to the technical field of wastewater treatment, and in particular to a method, apparatus, equipment and medium for controlling wastewater wetland blockage. Background Technology

[0002] During the long-term operation of constructed wetlands for wastewater treatment, suspended particles, colloidal substances, microbial metabolites, and plant residues carried in the wastewater tend to gradually deposit, adhere to, or accumulate within the filter layer. This causes the pores of the filter layer to shrink, become blocked, or experience localized poor flow, leading to decreased seepage capacity, altered hydraulic retention, and abnormal fluctuations in localized liquid levels. As the blockage worsens, the mass transfer conditions and flow states between different filter layers within the constructed wetland gradually become unbalanced, resulting in abnormal treatment load distribution and affecting the overall operational stability of the constructed wetland.

[0003] In existing technologies, the clogging problem of constructed wetland filter layers in wastewater is typically addressed through regular inspections, manual judgment based on experience, single-parameter monitoring, or flushing, cleaning, and path switching only after significant clogging has occurred. These methods rely heavily on manual observation or judgment based on single parameters such as flow rate and water level, making it difficult to accurately reflect the actual clogging status of different filter layers. Furthermore, most technologies only address the issue after clogging has become quite apparent, lacking continuous identification and early prediction of the clogging evolution process, resulting in a lag in the control process. Simultaneously, existing technologies often employ relatively simplistic post-clogging treatments, failing to develop corresponding control strategies based on the clogging status, development trend, and priority of different filter layers, thus hindering targeted dynamic control. Summary of the Invention

[0004] To improve the targeting of filter layer clogging identification and the adaptability of the control process, this application provides a tailwater wetland clogging control method, device, equipment and medium.

[0005] The above-mentioned objective of this application is achieved through the following technical solution:

[0006] A method for controlling tailwater wetland clogging, the method comprising:

[0007] Acquire information on the interlayer electrical response, infiltration flow rate, and liquid level change of each filter layer in the constructed wetland for tailwater.

[0008] Feature extraction processing is performed on the interlayer electrical response information, the permeation flow information and the liquid level change information to obtain the electrical response offset feature, permeation attenuation feature and liquid level hysteresis feature corresponding to each filter layer;

[0009] The electrical response offset feature, the seepage attenuation feature, and the liquid level hysteresis feature are fused and characterized to obtain the clogging characterization information corresponding to each filter layer;

[0010] Based on the blockage characterization information, a temporal correlation processing is performed to obtain the blockage evolution sequence corresponding to each of the filter layers;

[0011] Based on the clogging evolution sequence, clogging prediction processing is performed to obtain the target filter layer and the clogging prediction result information corresponding to the target filter layer;

[0012] Based on the target filtering layer and the blockage prediction result information, strategy matching processing is performed to obtain adaptive control strategy information;

[0013] Perform target filtering path adjustment processing or target filtering layer restoration processing based on the adaptive control strategy information;

[0014] The system obtains feedback monitoring information after the target filtering path adjustment process or the target filtering layer recovery process, and corrects the blockage prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information.

[0015] By adopting the above technical solution, the interlayer electrical response information, seepage flow information, and liquid level change information of each filter layer in the constructed wetland can be jointly acquired and feature extracted. Based on the extracted electrical response offset features, seepage attenuation features, and liquid level hysteresis features, clogging characterization information is generated, thereby distinguishing the clogging state of different filter layers. Furthermore, based on the clogging characterization information, a clogging evolution sequence is formed and clogging prediction processing is performed to determine the target filter layer and the corresponding clogging prediction result information. Then, adaptive control strategy information is generated according to the target filter layer and the clogging prediction result information, and target filter path adjustment processing or target filter layer recovery processing is performed. At the same time, the clogging prediction result information is corrected and the adaptive control strategy information is updated by combining feedback monitoring information, forming a continuous control process.

[0016] In a preferred embodiment, this application can be further configured as follows: the feature extraction processing of the interlayer electrical response information, the permeation flow information, and the liquid level change information to obtain the electrical response offset features, permeation attenuation features, and liquid level hysteresis features corresponding to each filter layer includes:

[0017] The interlayer electrical response information, the permeation flow information, and the liquid level change information are synchronously registered according to the preset sampling time sequence to obtain the time sequence monitoring unit corresponding to each filter layer;

[0018] Based on the interlayer electrical response value corresponding to the current sampling time and the interlayer electrical response value corresponding to the historical reference sampling time in the time-series monitoring unit, the offset difference is calculated to obtain the electrical response offset feature corresponding to each filter layer.

[0019] Based on the seepage flow values ​​corresponding to multiple consecutive sampling times in the time-series monitoring unit, attenuation trend fitting is performed to obtain the seepage attenuation characteristics corresponding to each filter layer.

[0020] Based on the first moment when the liquid level change reaches a preset change amplitude and the second moment when the permeation flow rate changes to a preset change amplitude in the time-series monitoring unit, the time difference between the first moment and the second moment is calculated to obtain the liquid level hysteresis characteristics corresponding to each filter layer.

[0021] By adopting the above technical solution, the interlayer electrical response information, permeation flow information, and liquid level change information can be synchronously registered according to the preset sampling time sequence to form a time-series monitoring unit corresponding to each filter layer. This allows different types of monitoring information to correspond on the same sampling basis. Furthermore, by calculating the offset difference between the interlayer electrical response values ​​corresponding to the current sampling time and the historical benchmark sampling time, the electrical response offset characteristics corresponding to each filter layer can be formed. At the same time, by fitting the attenuation trend of the permeation flow values ​​corresponding to multiple consecutive sampling times, the permeation attenuation characteristics corresponding to each filter layer can be formed. Moreover, by calculating the time difference between the first time when the liquid level change reaches the preset change amplitude and the second time when the permeation flow change reaches the preset change amplitude, the liquid level hysteresis characteristics corresponding to each filter layer can be formed, thus forming a multi-dimensional feature extraction result for each filter layer.

[0022] In a preferred embodiment, this application can be further configured to: fuse the electrical response offset feature, the seepage attenuation feature, and the liquid level hysteresis feature to obtain clogging characterization information corresponding to each of the filter layers, including:

[0023] The electrical response offset feature, the seepage attenuation feature, and the liquid level hysteresis feature are subjected to interval normalization to obtain standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature.

[0024] The standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature are used to extract information on the direction of synchronous change of features and the magnitude of feature deviation.

[0025] Based on the feature synchronization change direction information and the feature deviation magnitude information, feature coupling relationship information corresponding to each of the filter layers is generated;

[0026] Based on the feature coupling relationship information corresponding to each of the filter layers, the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature are subjected to layered fusion processing to obtain the fusion characterization vector corresponding to each of the filter layers.

[0027] The fusion representation vector corresponding to each of the filter layers is input into a preset blocking mapping rule to obtain the blocking representation information corresponding to each of the filter layers.

[0028] By adopting the above technical solution, the electrical response offset feature, seepage attenuation feature, and liquid level hysteresis feature can be first normalized to form standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature, so that different features are on a unified representation basis. Furthermore, based on the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature, the synchronous change direction information and feature deviation amplitude information are extracted, and feature coupling relationship information corresponding to each filter layer is generated, which can distinguish the correlation state between different features. On this basis, the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature are subjected to layered fusion processing according to the feature coupling relationship information to obtain the fusion representation vector corresponding to each filter layer. The fusion representation vector is then input into the preset blockage mapping rule to obtain the blockage representation information corresponding to each filter layer, thereby forming a unified blockage representation result for each filter layer.

[0029] In a preferred embodiment, this application can be further configured as follows: the step of performing time-series correlation processing based on the blockage characterization information to obtain the blockage evolution sequence corresponding to each of the filter layers includes:

[0030] The blockage characterization information is processed in a time sequence according to a preset monitoring time order to obtain a time sequence characterization group corresponding to each filter layer;

[0031] Based on the blockage characterization information corresponding to adjacent monitoring periods in the time-series characterization group corresponding to each filter layer, the adjacent change amplitude information and change direction information are calculated to obtain the characterization change result information corresponding to each filter layer.

[0032] Based on the characterization change results, continuous unidirectional change segments and fluctuation reversal segments are identified to obtain the stage division results information corresponding to each filter layer.

[0033] Based on the stage division results, the continuous unidirectional change segments and the fluctuation reversal segments are sequentially connected to obtain the blockage evolution sequence corresponding to each filter layer.

[0034] By adopting the above technical solution, the blockage characterization information can be processed in a time sequence according to the preset monitoring time order to form a time sequence characterization group corresponding to each filter layer. This ensures that the blockage characterization information of the same filter layer in different monitoring cycles maintains a corresponding relationship. Furthermore, based on the blockage characterization information of adjacent monitoring cycles in the time sequence characterization group corresponding to each filter layer, the adjacent change amplitude information and change direction information are calculated to form the characterization change result information corresponding to each filter layer. On this basis, the continuous same-direction change segment and the fluctuation reversal segment are identified according to the characterization change result information to obtain the stage division result information corresponding to each filter layer. Then, based on the stage division result information, the continuous same-direction change segment and the fluctuation reversal segment are sequentially connected to obtain the blockage evolution sequence corresponding to each filter layer, thereby forming the time sequence evolution result for each filter layer.

[0035] In a preferred embodiment, this application can be further configured as follows: the clogging prediction processing based on the clogging evolution sequence to obtain the target filter layer and the clogging prediction result information corresponding to the target filter layer includes:

[0036] Extract the growth slope information, node interval information, and continuous enhancement segment length information between adjacent evolution nodes in each of the blockage evolution sequences;

[0037] Based on the growth slope information, node interval information and continuous enhancement segment length information corresponding to each filter layer, the sequence extrapolation result information corresponding to each filter layer is generated.

[0038] Based on the sequence extrapolation result information corresponding to each of the filter layers, determine the predicted clogging level information and the predicted priority information corresponding to each of the filter layers;

[0039] The target filter layer is determined based on the predicted clogging level information and the predicted priority information corresponding to each filter layer.

[0040] Based on the predicted congestion level information, the predicted priority information, and the sequence extrapolation result information corresponding to the target filter layer, the congestion prediction result information corresponding to the target filter layer is generated.

[0041] By adopting the above technical solution, the growth slope information, node interval information, and continuous enhancement segment length information between adjacent evolution nodes can be extracted from each congestion evolution sequence. Based on the growth slope information, node interval information, and continuous enhancement segment length information, sequence extrapolation result information corresponding to each filter layer is generated. Further, the predicted congestion level information and prediction priority information corresponding to each filter layer are determined according to the sequence extrapolation result information corresponding to each filter layer. Based on the predicted congestion level information and prediction priority information, the target filter layer is determined. On this basis, the congestion prediction result information corresponding to the target filter layer is generated according to the predicted congestion level information, prediction priority information, and sequence extrapolation result information corresponding to the target filter layer, thereby forming a prediction result for the target filter layer.

[0042] In a preferred embodiment, this application can be further configured as follows: the step of performing strategy matching processing based on the target filtering layer and the blockage prediction result information to obtain adaptive control strategy information includes:

[0043] Obtain preset strategy rule information, which includes path adjustment matching rule information and recovery processing matching rule information;

[0044] Extract the predicted blockage level information and prediction priority information corresponding to the target filter layer from the blockage prediction result information;

[0045] Based on the target filtering layer, the predicted congestion level information, and the predicted priority information, a matching process is performed with the path adjustment matching rule information to obtain path adjustment strategy candidate information;

[0046] Based on the target filtering layer, the predicted blockage level information, and the predicted priority information, a matching process is performed with the recovery processing matching rule information to obtain recovery processing strategy candidate information;

[0047] The candidate information for path adjustment strategy and the candidate information for recovery processing strategy are combined and filtered to obtain the adaptive control strategy information.

[0048] By adopting the above technical solution, the preset strategy rule information can be obtained first, and the predicted congestion level information and prediction priority information corresponding to the target filtering layer can be extracted from the congestion prediction result information. Then, they are matched with the path adjustment matching rule information and the recovery processing matching rule information respectively to form path adjustment strategy candidate information and recovery processing strategy candidate information. The two types of candidate information are further combined and screened to obtain adaptive control strategy information corresponding to the current prediction state of the target filtering layer, thereby forming the corresponding control result for the target filtering layer.

[0049] In a preferred embodiment, this application can be further configured as follows: obtaining feedback monitoring information after the target filtering path adjustment process or the target filtering layer recovery process, and correcting the congestion prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information, includes:

[0050] Based on the feedback monitoring information, generate feedback blockage characterization information corresponding to the target filter layer;

[0051] The feedback congestion characterization information and the congestion prediction result information are subjected to deviation decomposition processing to obtain the level deviation direction information and the priority deviation direction information.

[0052] Based on the grade deviation direction information and the priority deviation direction information, the congestion prediction result information is corrected by itemization to obtain the corrected congestion prediction result information.

[0053] The adaptive control strategy information is reconstructed based on the corrected congestion prediction results to obtain the updated adaptive control strategy information.

[0054] By adopting the above technical solution, feedback blockage characterization information corresponding to the target filtering layer can be generated first based on feedback monitoring information. Then, the feedback blockage characterization information and the blockage prediction result information are decomposed to obtain the level deviation direction information and the priority deviation direction information. Further, the blockage prediction result information is modified item by item based on the level deviation direction information and the priority deviation direction information. Finally, the adaptive control strategy information is reconstructed based on the modified blockage prediction result information to obtain the updated adaptive control strategy information, thereby forming a dynamic correction process based on feedback results.

[0055] The second objective of this invention is achieved through the following technical solution:

[0056] A tailwater wetland clogging control device, the tailwater wetland clogging control device comprising:

[0057] The monitoring information acquisition module is used to acquire interlayer electrical response information, infiltration flow information, and liquid level change information for each filter layer in the wastewater constructed wetland.

[0058] The feature extraction module is used to perform feature extraction processing on the interlayer electrical response information, the permeation flow information and the liquid level change information to obtain the electrical response offset feature, permeation attenuation feature and liquid level hysteresis feature corresponding to each filter layer.

[0059] The fusion characterization module is used to fuse the electrical response offset feature, the seepage attenuation feature and the liquid level hysteresis feature to obtain the clogging characterization information corresponding to each filter layer.

[0060] The temporal correlation module is used to perform temporal correlation processing based on the blockage characterization information to obtain the blockage evolution sequence corresponding to each of the filter layers;

[0061] The clogging prediction module is used to perform clogging prediction processing based on the clogging evolution sequence to obtain the target filter layer and the clogging prediction result information corresponding to the target filter layer.

[0062] The strategy matching module is used to perform strategy matching processing based on the target filtering layer and the blockage prediction result information to obtain adaptive control strategy information.

[0063] The control execution module is used to perform target filtering path adjustment processing or target filtering layer recovery processing according to the adaptive control strategy information.

[0064] The feedback correction module is used to obtain feedback monitoring information after the target filtering path adjustment processing or the target filtering layer recovery processing, and to correct the blockage prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information.

[0065] By adopting the above technical solution, the interlayer electrical response information, seepage flow information, and liquid level change information of each filter layer in the constructed wetland can be jointly acquired and feature extracted. Based on the extracted electrical response offset features, seepage attenuation features, and liquid level hysteresis features, clogging characterization information is generated, thereby distinguishing the clogging state of different filter layers. Furthermore, based on the clogging characterization information, a clogging evolution sequence is formed and clogging prediction processing is performed to determine the target filter layer and the corresponding clogging prediction result information. Then, adaptive control strategy information is generated according to the target filter layer and the clogging prediction result information, and target filter path adjustment processing or target filter layer recovery processing is performed. At the same time, the clogging prediction result information is corrected and the adaptive control strategy information is updated by combining feedback monitoring information, forming a continuous control process.

[0066] The above-mentioned objective three of this application is achieved through the following technical solution:

[0067] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described tailwater wetland blockage control method.

[0068] The fourth objective of this application is achieved through the following technical solution:

[0069] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described tailwater wetland clogging control method.

[0070] In summary, this application includes at least one of the following beneficial technical effects:

[0071] 1. It can jointly acquire and extract features from the interlayer electrical response information, seepage flow information, and liquid level change information of each filter layer in the constructed wetland. Based on the extracted electrical response offset features, seepage attenuation features, and liquid level hysteresis features, it generates clogging characterization information to distinguish the clogging state of different filter layers. Furthermore, based on the clogging characterization information, it forms a clogging evolution sequence and performs clogging prediction processing to determine the target filter layer and its corresponding clogging prediction results. Then, it generates adaptive control strategy information based on the target filter layer and clogging prediction results, and performs target filter path adjustment processing or target filter layer recovery processing. At the same time, it combines feedback monitoring information to correct the clogging prediction results and update the adaptive control strategy information to form a continuous control process. Attached Figure Description

[0072] Figure 1 This is a flowchart of a tailwater wetland blockage control method in one embodiment of this application. Detailed Implementation

[0073] The present application will be further described in detail below with reference to the accompanying drawings.

[0074] In one embodiment, such as Figure 1 As shown, this application discloses a method for controlling tailwater wetland blockage, which specifically includes the following steps:

[0075] S10: Obtain information on the interlayer electrical response, infiltration flow rate, and liquid level change of each filter layer in the constructed wetland.

[0076] In this embodiment, the filter layer refers to different layered filter structures arranged sequentially along the flow direction of the effluent artificial wetland, which serve to intercept, infiltrate, and facilitate the flow of effluent. The interlayer electrical response information refers to the electrical signal changes between adjacent filter layers during the flow of effluent. The infiltration flow rate information refers to the change in the amount of effluent passing through the filter layer per unit time. The liquid level change information refers to the information formed by the change in the liquid level height of the effluent in the corresponding area of ​​the filter layer over time.

[0077] Specifically, when acquiring the interlayer electrical response information, infiltration flow information, and liquid level change information corresponding to each filter layer in the constructed wetland, a correspondence is first established according to the distribution location of each filter layer in the constructed wetland. Based on the correspondence, the monitoring area corresponding to each filter layer is distinguished. Then, for the monitoring area corresponding to each filter layer, the electrical signal value, flux value, and liquid level value generated at the preset acquisition time are read. Among them, the interlayer electrical response information is obtained by reading the electrical signal change value at the corresponding position between adjacent filter layers. The infiltration flow information is obtained by reading the flux change value during the process of the effluent passing through the filter layer. The liquid level change information is obtained by reading the change value of the liquid level height in the corresponding area of ​​the filter layer at different acquisition times. After completing the reading of the corresponding values ​​at each acquisition time, the data are sorted according to the filter layer, acquisition time, and information category to obtain the interlayer electrical response information, infiltration flow information, and liquid level change information corresponding to each filter layer.

[0078] S20: Perform feature extraction processing on the interlayer electrical response information, permeation flow information, and liquid level change information to obtain the electrical response offset features, permeation attenuation features, and liquid level hysteresis features corresponding to each filter layer.

[0079] In this embodiment, the electrical response offset feature refers to the offset characterization result formed by comparing the electrical signal change values ​​contained in the interlayer electrical response information; the seepage attenuation feature refers to the attenuation characterization result formed by continuously analyzing the flux change values ​​contained in the seepage flow information; and the liquid level hysteresis feature refers to the hysteresis characterization result formed by comparing the change time relationship between liquid level change information and seepage flow information.

[0080] Specifically, when performing feature extraction processing on interlayer electrical response information, permeation flow information, and liquid level change information, the following steps are taken: First, the electrical signal change values ​​at each moment in the interlayer electrical response information, the permeation flow change values ​​at each moment in the permeation flow information, and the liquid level change values ​​at each moment in the liquid level change information are read in the order of the acquisition time corresponding to each filter layer. Then, the three types of information at the corresponding time are organized on a per-filter-layer basis. For the extraction of electrical response offset features, the electrical signal change values ​​at the current acquisition time and the electrical signal change values ​​at the reference acquisition time are read for each filter layer. The difference between the electrical signal change values ​​at the current acquisition time and the electrical signal change values ​​at the reference acquisition time is calculated. The result of the difference calculation represents the degree of offset of the electrical signal relative to the reference state, thus obtaining the corresponding features of each filter layer. For the extraction of seepage attenuation characteristics, the flux change values ​​of each filter layer at multiple consecutive acquisition times are read sequentially according to the acquisition time sequence. The flux change values ​​corresponding to adjacent acquisition times are compared to determine the decrease range and trend of flux change values ​​during continuous acquisition. Then, the seepage attenuation characteristics corresponding to each filter layer are formed based on the decrease range and trend. For the extraction of liquid level hysteresis characteristics, the acquisition time when the liquid level value starts to change in the liquid level change information corresponding to each filter layer and the acquisition time when the flux change value starts to change in the seepage flow information are read respectively. The time interval between the two acquisition times is calculated and used as the hysteresis degree of liquid level change relative to seepage flow change to obtain the liquid level hysteresis characteristics corresponding to each filter layer.

[0081] S30: The electrical response offset feature, seepage attenuation feature and liquid level hysteresis feature are fused and characterized to obtain the clogging characterization information corresponding to each filter layer.

[0082] In this embodiment, the clogging characterization information refers to the information formed by jointly characterizing the electrical response offset characteristics, seepage attenuation characteristics, and liquid level hysteresis characteristics, which is used to indicate the current clogging status of each filter layer.

[0083] Specifically, when fusing and characterizing the electrical response offset features, seepage attenuation features, and liquid level hysteresis features, the corresponding electrical response offset features, seepage attenuation features, and liquid level hysteresis features are first read separately for each filter layer. Then, the numerical ranges of the electrical response offset features, seepage attenuation features, and liquid level hysteresis features are unified to transform them into feature values ​​within the same numerical range. Subsequently, the direction of change of the electrical response offset features, seepage attenuation features, and liquid level hysteresis features corresponding to each filter layer within the same acquisition phase is compared, and the electrical response offset features, seepage attenuation features, and liquid level hysteresis features corresponding to each filter layer are read. The numerical differences between the reduction feature and the liquid level hysteresis feature are used to form the feature coupling relationship information corresponding to each filter layer based on the direction of change and the degree of numerical difference. Then, based on the feature coupling relationship information corresponding to each filter layer, the electrical response offset feature, seepage attenuation feature and liquid level hysteresis feature are combined and calculated. The combined calculation includes aggregating the feature values ​​that change in the same direction and correcting the differences of the feature values ​​that change in opposite directions. After the combined calculation, the fusion characterization result corresponding to each filter layer is formed. The fusion characterization result corresponding to each filter layer is then converted according to the preset blockage mapping relationship to obtain the blockage characterization information corresponding to each filter layer.

[0084] S40: Perform time-series correlation processing based on the clogging characterization information to obtain the clogging evolution sequence corresponding to each filter layer.

[0085] In this embodiment, the clogging evolution sequence refers to the information sequence formed by associating the clogging characterization information of each filter layer at different acquisition times according to the chronological order, which is used to indicate the process of clogging status changing over time.

[0086] Specifically, when performing time-series correlation processing based on blockage characterization information, the blockage characterization information formed at different acquisition times is first read according to each filter layer. Then, the blockage characterization information corresponding to the same filter layer is arranged according to the order of acquisition times to form a time-series characterization group corresponding to each filter layer. Next, the blockage characterization information adjacent to each other in the time-series characterization group corresponding to each filter layer is compared to determine the numerical change between two adjacent acquisition times. The numerical change includes numerical increase, numerical decrease, and numerical unchanged. Subsequently, the continuous same-direction change segment and the fluctuation reversal segment corresponding to each filter layer are identified based on the numerical change between multiple consecutive acquisition times. The continuous same-direction change segment refers to the time segment in which the blockage characterization information maintains the same direction of change between multiple consecutive acquisition times, and the fluctuation reversal segment refers to the time segment in which the direction of change of the blockage characterization information changes between adjacent acquisition times. After identifying the continuous same-direction change segment and the fluctuation reversal segment, they are sequentially concatenated according to the start acquisition time, end acquisition time, and arrangement order of each segment to obtain the blockage evolution sequence corresponding to each filter layer.

[0087] S50: Perform clogging prediction processing based on the clogging evolution sequence to obtain the target filter layer and the clogging prediction result information corresponding to the target filter layer.

[0088] In this embodiment, the target filter layer refers to the filter layer with a more prominent degree of blockage change determined after blockage prediction processing based on the blockage evolution sequence among multiple filter layers. The blockage prediction result information refers to the information formed based on the blockage evolution sequence corresponding to the target filter layer, which is used to indicate the subsequent blockage change state of the target filter layer.

[0089] Specifically, when performing congestion prediction processing based on congestion evolution sequences, the congestion evolution sequences corresponding to each filter layer are first read, and the numerical change relationships between adjacent evolution nodes are extracted according to the arrangement order in each congestion evolution sequence. Then, the growth slope information, node interval information, and continuous enhancement segment length information corresponding to each filter layer are determined based on the numerical change relationships between adjacent evolution nodes. Specifically, the growth slope information is calculated by comparing the numerical change in congestion characteristics between adjacent evolution nodes with the corresponding time change; the node interval information is formed by reading the time difference between adjacent evolution nodes; and the continuous enhancement segment length information is formed by counting the number of evolution nodes that continuously maintain a growth state. After obtaining the growth slope information, node interval information, and continuous enhancement segment length information corresponding to each filter layer... Afterwards, the growth slope information, node interval information, and continuous enhancement segment length information corresponding to each filter layer are jointly analyzed to determine the subsequent blockage change trend corresponding to each filter layer. The joint analysis includes comparing the magnitude of the growth slope information, comparing the length of the node interval information, and comparing the amount of continuous enhancement segment length information. Based on the joint analysis results, the predicted blockage level information and prediction priority information corresponding to each filter layer are determined. From the predicted blockage level information and prediction priority information corresponding to multiple filter layers, the filter layer with higher predicted blockage level information and higher prediction priority information is selected as the target filter layer. Finally, the predicted blockage level information, prediction priority information, and subsequent blockage change trend corresponding to the target filter layer are sorted together to obtain the blockage prediction result information corresponding to the target filter layer.

[0090] S60: Perform strategy matching processing based on the target filtering layer and the blockage prediction results to obtain adaptive control strategy information.

[0091] In this embodiment, the adaptive control strategy information refers to the control information determined based on the target filter layer and the blockage prediction results, which corresponds to the current blockage change state of the target filter layer.

[0092] Specifically, when performing strategy matching based on the target filter layer and the blockage prediction results, the predicted blockage level information and prediction priority information corresponding to the target filter layer are first read. Then, the target filter layer, predicted blockage level information, and prediction priority information are used as matching conditions to compare with the preset strategy rules information. The preset strategy rules information is used to record the control content corresponding to different filter layers under different predicted blockage levels and different prediction priorities. When performing the comparison, the strategy range corresponding to the target filter layer is first determined based on the location of the target filter layer in the tailwater constructed wetland. Then, within the strategy range, candidate strategy content corresponding to the current blockage change state is screened based on the predicted blockage level information. Subsequently, the order of the screened candidate strategy content is corrected based on the prediction priority information to determine the order relationship between the candidate strategy content. After the order correction is completed, candidate strategy content with the same execution direction is merged and sorted, and candidate strategy content with execution conflicts is discarded. Finally, the control content corresponding to the target filter layer and the blockage prediction results information is retained, and the retained control content is sorted to obtain adaptive control strategy information.

[0093] S70: Perform target filtering path adjustment processing or target filtering layer restoration processing based on adaptive control strategy information.

[0094] In this embodiment, the target filtration path adjustment process refers to the process of adjusting the passage path of the effluent in the constructed wetland according to the adaptive control strategy information, and the target filtration layer restoration process refers to the process of restoring the blockage state of the target filtration layer according to the adaptive control strategy information.

[0095] Specifically, when performing target filter path adjustment or target filter layer restoration processing based on adaptive control strategy information, the control content contained in the adaptive control strategy information is first read, and the execution type of the control content is identified to distinguish whether the control content corresponds to target filter path adjustment processing or target filter layer restoration processing. If the control content corresponds to target filter path adjustment processing, the target passage path of the effluent in the constructed wetland is determined based on the path adjustment content recorded in the adaptive control strategy information. The target passage path is then compared with the original passage path to determine the original passage path segments that need to be stopped and the target passage path segments that need to be opened. Then, according to a preset switching sequence, the effluent throughput corresponding to the original passage path segments is first reduced, and then the effluent throughput is gradually switched to the target passage path segments, so that the effluent flows along the adjusted path. The target filter flows through the area corresponding to the target filter layer via the path. When the control content corresponds to the target filter layer recovery treatment, the recovery operation range and recovery operation sequence acting on the target filter layer are determined according to the recovery treatment content recorded in the adaptive control strategy information. After determining the recovery operation range and recovery operation sequence, the recovery treatment is performed on the treatment area corresponding to the target filter layer in sequence according to the recovery operation sequence. The recovery treatment includes at least one of applying vibration disturbance treatment to the treatment area corresponding to the target filter layer, reducing the amount of tailwater flowing through the treatment area corresponding to the target filter layer, and guiding the tailwater away from the treatment area corresponding to the target filter layer. After completing the target filter path adjustment treatment or the target filter layer recovery treatment, the corresponding execution content, execution sequence, and execution time are recorded to form the processing execution result corresponding to the adaptive control strategy information.

[0096] S80: Obtain feedback monitoring information after target filtering path adjustment processing or target filtering layer recovery processing, and correct the blockage prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information.

[0097] In this embodiment, the feedback monitoring information refers to the information collected again for the corresponding area of ​​the target filter layer after the target filter path adjustment processing or target filter layer recovery processing is completed, which is used to indicate the current state change. The updated adaptive control strategy information refers to the control information that is re-determined after the blockage prediction result information is corrected.

[0098] Specifically, when acquiring feedback monitoring information after target filtration path adjustment or target filtration layer restoration, and correcting the blockage prediction results based on this feedback monitoring information, the process begins by rereading the interlayer electrical response, permeation flow, and liquid level change information of the corresponding region of the target filtration layer at a preset feedback acquisition time after the target filtration path adjustment or target filtration layer restoration is completed. This reread information is then used as feedback monitoring information. Next, feedback blockage characterization information corresponding to the target filtration layer is extracted based on this feedback monitoring information. This feedback blockage characterization information refers to information formed based on the feedback monitoring information that indicates the blockage status after treatment. Finally, the feedback blockage characterization information is compared with the blockage prediction results to determine the correctness of the blockage. The direction and degree of difference between the feedback congestion characterization information and the congestion prediction result information are determined. The direction of difference refers to the directional relationship between the feedback congestion characterization information and the congestion prediction result information, which is increasing, decreasing, or consistent. The degree of difference refers to the magnitude of the numerical deviation between the feedback congestion characterization information and the congestion prediction result information. After determining the direction and degree of difference, the predicted congestion level information and prediction priority information in the congestion prediction result information are adjusted according to the direction and degree of difference to form the corrected congestion prediction result information. Finally, the control content corresponding to the current state is reread based on the corrected congestion prediction result information, and the control content formed by the rereading is organized to obtain the updated adaptive control strategy information.

[0099] In one embodiment, in step S20, feature extraction processing is performed on the interlayer electrical response information, permeation flow information, and liquid level change information to obtain the electrical response offset features, permeation attenuation features, and liquid level hysteresis features corresponding to each filter layer, including:

[0100] S201: Synchronously register the interlayer electrical response information, permeation flow information, and liquid level change information according to the preset sampling sequence to obtain the time-series monitoring unit corresponding to each filter layer.

[0101] In this embodiment, the time-series monitoring unit refers to the information set formed by correspondingly organizing the interlayer electrical response information, permeation flow information and liquid level change information according to the same preset sampling time sequence, which is used to indicate the monitoring status of the same filter layer at the same sampling time.

[0102] Specifically, when performing synchronous registration processing on interlayer electrical response information, permeation flow information, and liquid level change information according to a preset sampling time sequence, the time sequence corresponding to each sampling moment in the preset sampling time sequence is first read. Then, the electrical signal change value in the interlayer electrical response information, the flux change value in the permeation flow information, and the liquid level change value in the liquid level change information are extracted according to each sampling moment. Then, taking each filter layer as the corresponding object, the time correspondence verification and filter layer correspondence verification are performed on the electrical signal change value, flux change value, and liquid level change value extracted at the same sampling moment. The time correspondence verification is used to confirm that the sampling moments corresponding to the three types of information are consistent, and the filter layer correspondence verification is used to confirm that the three types of information correspond to the same filter layer. After completing the time correspondence verification and filter layer correspondence verification, the electrical signal change value, flux change value, and liquid level change value corresponding to the same filter layer at the same sampling moment are combined and organized to form an information set that corresponds one-to-one with the filter layer and the sampling moment. The information sets formed at different sampling moments are then arranged sequentially according to the preset sampling time sequence to obtain the time sequence monitoring unit corresponding to each filter layer.

[0103] S202: Based on the interlayer electrical response value corresponding to the current sampling time in the time-series monitoring unit and the interlayer electrical response value corresponding to the historical benchmark sampling time, the offset difference is calculated to obtain the electrical response offset characteristics corresponding to each filter layer.

[0104] In this embodiment, the current sampling time refers to the sampling time currently being processed in the timing monitoring unit, the historical reference sampling time refers to the pre-selected sampling time used as the basis for comparison in the timing monitoring unit, and the interlayer electrical response value refers to the electrical signal value formed by the interlayer electrical response information at the corresponding sampling time.

[0105] Specifically, when calculating the offset difference based on the interlayer electrical response value corresponding to the current sampling time and the interlayer electrical response value corresponding to the historical reference sampling time in the time-series monitoring unit, the interlayer electrical response value Et corresponding to the current sampling time and the interlayer electrical response value E0 corresponding to the historical reference sampling time are read separately for each filter layer, and then the interlayer electrical response difference is calculated. The formula for calculating the interlayer electrical response difference is: Subsequently, based on the interlayer electrical response difference The inter-layer electrical response offset ratio RE is calculated using the inter-layer electrical response value E0 corresponding to the historical reference sampling time. The formula for calculating the inter-layer electrical response offset ratio RE is as follows: ,in, A preset small correction value is used to avoid invalid calculations when the inter-layer electrical response value corresponding to the historical reference sampling time is zero, and then the inter-layer electrical response value corresponding to the previous sampling time is read. Calculate the direction coefficient DE of the electrical response change between adjacent sampling times, when When Et-E0 has the same sign, the direction coefficient DE of the change in electrical response is set to 1. When the sign is opposite to that of Et - E0, the direction coefficient of electrical response change DE is set to 0, and then the interlayer electrical response difference is used as the basis for further calculations. The electrical response offset characteristic FE is calculated using the inter-layer electrical response offset ratio RE and the electrical response change direction coefficient DE. The formula for calculating the electrical response offset characteristic FE is as follows: The electrical response offset features F_E corresponding to each filter layer are then organized according to the filter layer correspondence to obtain the electrical response offset features corresponding to each filter layer.

[0106] S203: Based on the seepage flow values ​​corresponding to multiple consecutive sampling times in the time-series monitoring unit, the attenuation trend is fitted to obtain the seepage attenuation characteristics of each filter layer.

[0107] In this embodiment, the attenuation trend fitting process refers to the process of calculating the change trend and characterizing the attenuation degree of the permeation flux values ​​corresponding to multiple consecutive sampling times in the time-series monitoring unit according to the sampling time sequence.

[0108] Specifically, when performing attenuation trend fitting based on the permeation flux values ​​corresponding to multiple consecutive sampling times in the time-series monitoring unit, the permeation flux values ​​corresponding to multiple consecutive sampling times in the time-series monitoring unit are first read according to each filter layer, and a permeation flux sequence is constructed according to the chronological order of the sampling times. Let the permeation flux value corresponding to the i-th sampling time be Qi, and the corresponding sampling time be ti, where... Let n be the number of sampling points corresponding to multiple consecutive sampling times. Then, calculate the local attenuation slope Ki between adjacent sampling times based on the percolation flow sequence. The formula for calculating the local attenuation slope Ki is as follows: Then, the consecutive local attenuation slopes Ki are sequentially arranged to obtain the slope change sequence of the corresponding filter layer, and the local attenuation slopes Ki and the corresponding sampling interval length are then used to determine the slope change sequence of each filter layer. Calculate the weighted attenuation coefficient Kw. The formula for calculating the weighted attenuation coefficient Kw is as follows: Then, read the first permeation flux value Q1 and the last permeation flux value Qn in the permeation flux sequence, and calculate the overall attenuation ratio Rq. The formula for calculating the overall attenuation ratio Rq is as follows: ,in, To pre-set a small correction value and avoid invalid calculations when the first seepage flow value is zero, after obtaining the weighted attenuation coefficient Kw and the overall attenuation ratio Rq, the weighted attenuation coefficient Kw and the overall attenuation ratio Rq are jointly fitted to obtain the seepage attenuation characteristic Fq. The formula for calculating the seepage attenuation characteristic Fq is as follows: ,in, and This is a preset scaling factor, and Finally, the seepage attenuation characteristics Fq corresponding to each filter layer are sorted according to the filter layer correspondence to obtain the seepage attenuation characteristics corresponding to each filter layer.

[0109] S204: Based on the first moment when the liquid level change in the time-series monitoring unit reaches the preset change amplitude and the second moment when the permeation flow rate changes to the preset change amplitude, calculate the time difference between the first moment and the second moment to obtain the liquid level hysteresis characteristics corresponding to each filter layer.

[0110] In this embodiment, the liquid level hysteresis characteristic refers to the hysteresis characterization result formed by the liquid level change relative to the seepage flow rate change in the time response relationship. The first moment refers to the sampling moment corresponding to when the liquid level change in the time-series monitoring unit reaches the preset change amplitude, and the second moment refers to the sampling moment corresponding to when the seepage flow rate change in the time-series monitoring unit reaches the preset change amplitude.

[0111] Specifically, when calculating the time difference between the first and second moments based on the first moment when the liquid level change reaches the preset amplitude and the second moment when the permeation flow rate reaches the preset amplitude in the time-series monitoring unit, the liquid level change value and permeation flow rate value corresponding to multiple consecutive sampling moments in the time-series monitoring unit are first read according to each filter layer. Then, the liquid level reference value and permeation flow rate reference value are determined respectively. The liquid level reference value is taken as the liquid level change value corresponding to the first sampling moment in the time-series monitoring unit, and the permeation flow rate reference value is taken as the permeation flow rate value corresponding to the first sampling moment in the time-series monitoring unit. Subsequently, the liquid level change amplitude and permeation flow rate change amplitude corresponding to each sampling moment are calculated respectively. The liquid level change amplitude corresponding to the i-th sampling moment is denoted as... The amplitude of the change in permeation flux at the i-th sampling time is denoted as . Liquid level change amplitude The calculation formula is Amplitude of change in permeation flow The calculation formula is Where Hi is the liquid level change value corresponding to the i-th sampling time, H1 is the liquid level change value corresponding to the first sampling time, Qi is the seepage flow rate value corresponding to the i-th sampling time, and Q1 is the seepage flow rate value corresponding to the first sampling time. After obtaining the liquid level change amplitude and seepage flow rate change amplitude corresponding to each sampling time, the liquid level change amplitude is compared sequentially with the preset change amplitude, and the value that is satisfied first is selected. The sampling time is determined as the first moment. The amplitude of the seepage flow change is compared sequentially with the preset amplitude, and the value that is satisfied first is selected. The sampling time is determined as the second time, where, This is the preset change amplitude corresponding to the liquid level change. The preset change amplitude corresponding to the change in seepage flux is then used to calculate the time difference between the first and second time points. ,jet lag The calculation formula is Where tH is the time value corresponding to the first moment, tQ is the time value corresponding to the second moment, and then based on the time difference... The liquid level hysteresis characteristic FH is constructed by combining the amplitude of liquid level change and the amplitude of seepage flow rate change. The formula for calculating the liquid level hysteresis characteristic FH is as follows: ,in, To preset a small correction value, As a preset proportional coefficient, the liquid level hysteresis characteristics F_H corresponding to each filter layer are finally sorted according to the correspondence of the filter layers to obtain the liquid level hysteresis characteristics corresponding to each filter layer.

[0112] In one embodiment, in step S30, the electrical response offset characteristics, seepage attenuation characteristics, and liquid level hysteresis characteristics are fused and characterized to obtain the clogging characterization information corresponding to each filter layer, including:

[0113] S301: The electrical response offset feature, seepage attenuation feature, and liquid level hysteresis feature are normalized by interval processing to obtain standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature.

[0114] In this embodiment, interval normalization refers to the process of converting the electrical response offset feature, seepage attenuation feature, and liquid level hysteresis feature into the same numerical interval. Standardized electrical response offset feature refers to the feature result formed after interval normalization of the electrical response offset feature, standardized seepage attenuation feature refers to the feature result formed after interval normalization of the seepage attenuation feature, and standardized liquid level hysteresis feature refers to the feature result formed after interval normalization of the liquid level hysteresis feature.

[0115] Specifically, when performing interval normalization on the electrical response offset feature, seepage attenuation feature, and level hysteresis feature, the characteristic values ​​corresponding to the electrical response offset feature, seepage attenuation feature, and level hysteresis feature are first read for each filter layer. Then, the maximum and minimum characteristic values ​​of the electrical response offset feature, seepage attenuation feature, and level hysteresis feature within a preset statistical interval are determined. The preset statistical interval refers to the range of values ​​for statistical analysis of the characteristic values ​​corresponding to the multiple filter layers currently being processed. After determining the maximum and minimum characteristic values, normalization calculation is performed on the electrical response offset feature. Let the electrical response offset feature be FE, and the maximum characteristic value corresponding to the electrical response offset feature be... The minimum feature value corresponding to the electrical response offset feature is If the normalized electrical response offset characteristic is NE, then the formula for calculating the normalized electrical response offset characteristic NE is: Normalization calculations were performed on the seepage attenuation characteristics. Let the seepage attenuation characteristic be FQ, and the maximum characteristic value corresponding to the seepage attenuation characteristic be... The minimum characteristic value corresponding to the seepage attenuation characteristic is If the standardized seepage attenuation characteristic is NQ, then the formula for calculating the standardized seepage attenuation characteristic NQ is: Normalization calculations are performed on the liquid level hysteresis feature. Let the liquid level hysteresis feature be FH, and the maximum feature value corresponding to the liquid level hysteresis feature be... The minimum characteristic value corresponding to the liquid level hysteresis characteristic is If the standardized liquid level hysteresis characteristic is NH, then the formula for calculating the standardized liquid level hysteresis characteristic NH is: ,in, To preset small correction values, after calculating the standardized electrical response offset feature NE, the standardized seepage attenuation feature NQ, and the standardized liquid level hysteresis feature NH, the values ​​are organized according to the corresponding relationship of the filter layers to obtain the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature.

[0116] S302: Based on standardized electrical response offset features, standardized seepage attenuation features, and standardized liquid level hysteresis features, extract information on the direction of synchronous change of features and the magnitude of feature deviation.

[0117] In this embodiment, the characteristic synchronous change direction information refers to the relationship between the change direction of the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature under the same filter layer relative to their respective reference states. The characteristic deviation magnitude information refers to the degree of deviation between the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature in terms of numerical distribution.

[0118] Specifically, when extracting the synchronous change direction information and feature deviation amplitude information based on the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature, the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature are first read according to each filter layer. Then, the baseline feature values ​​of the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature under the corresponding baseline state are determined. The baseline feature values ​​are taken as the standardized values ​​corresponding to each standardized feature during the baseline acquisition stage. Subsequently, the change direction values ​​of the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature relative to their respective baseline feature values ​​are calculated. Let the standardized electrical response offset feature be NE, the standardized seepage attenuation feature be NQ, the standardized liquid level hysteresis feature be NH, and the baseline feature value corresponding to the standardized electrical response offset feature be... The baseline characteristic value corresponding to the standardized seepage attenuation characteristic is: The baseline feature value corresponding to the standardized liquid level hysteresis feature is: Then the formula for calculating the direction value DE of the electrical response is: The formula for calculating the seepage direction value DQ is: The formula for calculating the liquid level direction value DH is: ,in, The sign judgment function takes a value of 1 when the corresponding difference is greater than zero, 0 when the corresponding difference is equal to zero, and -1 when the corresponding difference is less than zero. After obtaining the electrical response direction value DE, seepage direction value DQ, and liquid level direction value DH, a consistency judgment is performed on the electrical response direction value DE, seepage direction value DQ, and liquid level direction value DH. When the three values ​​are the same, the corresponding result is determined to be a completely synchronous change. When any two of the three values ​​are the same and the remaining value is different, the corresponding result is determined to be a partially synchronous change. When all three values ​​are different, the corresponding result is determined to be a asynchronous change. The consistency judgment result is used as the characteristic synchronous change direction information. At the same time, the numerical differences between each pairwise of the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature are calculated to obtain the electrical response-seepage deviation value. Electrical response - liquid level deviation and seepage-level deviation value Among them, the electrical response-permeation deviation value The calculation formula is Electrical response - liquid level deviation value The calculation formula is Seepage-level deviation The calculation formula is Then, the electrical response-permeation deviation value was analyzed. Electrical response - liquid level deviation and seepage-level deviation value Aggregate calculations are performed to obtain the feature deviation magnitude information AF. The formula for calculating the feature deviation magnitude information AF is as follows: Finally, the information on the synchronous change direction of features and the magnitude of feature deviation corresponding to each filter layer are organized according to the correspondence between the filter layers.

[0119] S303: Based on the information of the direction of synchronous change of features and the information of the magnitude of feature deviation, generate the feature coupling relationship information corresponding to each filter layer.

[0120] In this embodiment, the feature coupling relationship information refers to the information formed by jointly analyzing the feature synchronous change direction information and feature deviation magnitude information, which is used to indicate the correlation between the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature.

[0121] Specifically, when generating feature coupling relationship information for each filter layer based on feature synchronization change direction information and feature deviation magnitude information, the feature synchronization change direction information and feature deviation magnitude information for each filter layer are read first. Then, the directional coupling category for each filter layer is determined based on the feature synchronization change direction information. Specifically, when the feature synchronization change direction information corresponds to a completely synchronized change, the directional coupling category is determined as a strong synchronization category; when the feature synchronization change direction information corresponds to a partially synchronized change, the directional coupling category is determined as a transitional synchronization category; and when the feature synchronization change direction information corresponds to a non-synchronous change, the directional coupling category is determined as a discrete change category. Subsequently, the feature deviation magnitude information is matched with a preset deviation interval. The preset deviation interval is used to distinguish the magnitude categories corresponding to different degrees of numerical deviation. When the feature deviation magnitude information falls into a smaller deviation interval, the magnitude category is determined as a low deviation category; when the feature deviation magnitude information falls into a medium deviation interval, the magnitude category is determined as a low deviation category; and so on. When the deviation range is reached, the amplitude category is determined as the medium deviation category. When the feature deviation amplitude information falls into a large deviation range, the amplitude category is determined as the high deviation category. After determining the directional coupling category and amplitude category, a combined mapping process is performed on the directional coupling category and amplitude category to obtain the coupling state identifier corresponding to each filter layer. The combined mapping process is used to map the synchronization relationship and the degree of deviation to the same associated state. When the directional coupling category is the strong synchronization category and the amplitude category is the low deviation category, the coupling state identifier is determined as the tight coupling state. When the directional coupling category is the transitional synchronization category or the amplitude category is the medium deviation category, the coupling state identifier is determined as the intermediate coupling state. When the directional coupling category is the discrete change category and the amplitude category is the high deviation category, the coupling state identifier is determined as the weak coupling state. Finally, the directional coupling category, amplitude category, and coupling state identifier are organized according to the filter layer correspondence to obtain the feature coupling relationship information corresponding to each filter layer.

[0122] S304: Based on the characteristic coupling relationship information corresponding to each filter layer, the standardized electrical response offset feature, standardized seepage attenuation feature and standardized liquid level hysteresis feature are subjected to hierarchical fusion processing to obtain the fusion characterization vector corresponding to each filter layer.

[0123] In this embodiment, the fusion characterization vector refers to the vector result formed after performing hierarchical fusion processing on the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature, which is used to indicate the current comprehensive characterization state of each filter layer.

[0124] Specifically, when performing layered fusion processing on the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature based on the feature coupling relationship information corresponding to each filter layer, the corresponding feature coupling relationship information, standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature are first read according to each filter layer. Then, the directional coupling category, amplitude category, and coupling state identifier are extracted from the feature coupling relationship information, and the corresponding layered fusion coefficients are determined based on the directional coupling category, amplitude category, and coupling state identifier. Among them, the directional coupling category is used to determine the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature. The consistency of feature values ​​during the fusion process is considered. The amplitude category is used to determine the difference correction ratio of the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized level hysteresis feature during the fusion process. The coupling state identifier is used to determine the combination method of the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized level hysteresis feature during the fusion process. After determining the stratified fusion coefficients, a weighted combination calculation is performed on the standardized electrical response offset feature, standardized seepage attenuation feature, and standardized level hysteresis feature. Let the standardized electrical response offset feature be NE, the standardized seepage attenuation feature be NQ, and the standardized level hysteresis feature be NH. The corresponding stratified fusion coefficients are as follows: , and Then the formula for calculating the fused principal value VM is: Subsequently, the stratification deviations between the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature were calculated. Let the deviation between the standardized electrical response offset feature and the standardized seepage attenuation feature be denoted as . The deviation between the standardized electrical response offset characteristic and the standardized liquid level hysteresis characteristic is _____. The deviation between the standardized seepage attenuation characteristics and the standardized liquid level hysteresis characteristics is _____. The formula for calculating the stratification deviation value VB is as follows: ,in, , , Then, based on the coupling state identifier, the fusion principal value VM and the hierarchical deviation value VB are combined and corrected to obtain the fusion correction value VC. The formula for calculating the fusion correction value VC is as follows: ,in, To obtain the correction coefficients corresponding to the coupling state identifier, after obtaining the fusion principal value VM, the hierarchical deviation value VB, and the fusion correction value VC, the fusion principal value VM, the hierarchical deviation value VB, and the fusion correction value VC are arranged in a fixed order to obtain the fusion representation vector corresponding to each filter layer.

[0125] S305: Input the fusion representation vector corresponding to each filter layer into the preset blocking mapping rule to obtain the blocking representation information corresponding to each filter layer.

[0126] In this embodiment, the preset congestion mapping rule refers to the pre-set rule content used to establish the correspondence between the fusion representation vector and the congestion state.

[0127] Specifically, when inputting the fusion representation vectors corresponding to each filter layer into the preset blocking mapping rules, the corresponding fusion representation vectors for each filter layer are first read. Then, the fusion principal value, layer deviation value, and fusion correction value are extracted sequentially from the fusion representation vectors. The fusion principal value, layer deviation value, and fusion correction value are used as mapping input items and matched with the preset blocking mapping rules. The preset blocking mapping rules are used to determine the corresponding blocking state based on the degree of comprehensive change indicated by the fusion principal value, the degree of feature dispersion indicated by the layer deviation value, and the corrected comprehensive state indicated by the fusion correction value. When performing the matching, the basic range of blocking change is first determined based on the fusion principal value, then the deviation correction level within the range is determined based on the layer deviation value, and finally the mapping result corresponding to the final blocking state is determined based on the fusion correction value. After the mapping result is determined, the mapping result is organized according to the correspondence of the filter layers to obtain the blocking representation information corresponding to each filter layer.

[0128] In one embodiment, step S40, namely, performing time-series correlation processing based on the clogging characterization information to obtain the clogging evolution sequence corresponding to each filter layer, includes:

[0129] S401: The blockage characterization information is processed according to the preset monitoring time sequence to obtain the time sequence characterization group corresponding to each filter layer.

[0130] In this embodiment, the time-series characterization group refers to the information group formed by arranging the blockage characterization information of the same filter layer at different monitoring times according to a preset monitoring time sequence.

[0131] Specifically, when processing the blockage characterization information according to the preset monitoring time sequence, the blockage characterization information corresponding to each filter layer at different monitoring times is first read, and the monitoring time identifier corresponding to each blockage characterization information is also read. Then, the blockage characterization information is classified and organized according to the filter layer correspondence, so that the blockage characterization information corresponding to the same filter layer is grouped into the same set to be organized. After the classification and organization are completed, the monitoring time identifiers corresponding to each blockage characterization information in the same set to be organized are compared sequentially according to the preset monitoring time sequence, and the blockage characterization information is rearranged in order from front to back. During the rearrangement process, the blockage characterization information with earlier monitoring time is arranged first, and the blockage characterization information with later monitoring time is arranged last. Then, the blockage characterization information after the rearrangement is combined in sequence according to the arrangement result to obtain the time sequence characterization group corresponding to each filter layer.

[0132] S402: Based on the blockage characterization information corresponding to adjacent monitoring cycles in the time-series characterization group corresponding to each filter layer, calculate the adjacent change amplitude information and change direction information to obtain the characterization change result information corresponding to each filter layer.

[0133] In this embodiment, the adjacent change amplitude information refers to the numerical change amplitude information between the blockage characterization information of the same filter layer in adjacent monitoring cycles, the change direction information refers to the direction information of increase, decrease or no change shown by the blockage characterization information of the same filter layer in adjacent monitoring cycles, and the characterization change result information refers to the result information formed after correspondingly organizing the adjacent change amplitude information and change direction information.

[0134] Specifically, when calculating the adjacent change amplitude and change direction information based on the blockage characterization information corresponding to adjacent monitoring periods in the time series characterization group corresponding to each filter layer, the corresponding time series characterization group is read for each filter layer, and the blockage characterization information corresponding to two adjacent monitoring periods is extracted sequentially according to the arrangement order in the time series characterization group. Let the blockage characterization information value corresponding to the previous monitoring period be Ci, and the blockage characterization information value corresponding to the next monitoring period be... ,in, , where n is the number of monitoring periods included in the time series characterization group, and then... The difference between Ci and the adjacent change amplitude Ai is calculated. The formula for calculating the adjacent change amplitude Ai is as follows: Subsequently The sign is used to determine the direction of change. When the value is greater than 0, the direction of change is determined to increase. When <0, the direction of change is determined to decrease; when When =0, the change direction information is determined to remain unchanged. After obtaining the adjacent change amplitude information and change direction information corresponding to each adjacent monitoring cycle, the adjacent change amplitude information and change direction information are sorted in order according to the monitoring cycle correspondence to obtain the characterization change result information corresponding to each filter layer.

[0135] S403: Based on the information of the characterization change results, identify the continuous same-direction change segment and the fluctuation reversal segment, and obtain the stage division result information corresponding to each filter layer.

[0136] In this embodiment, a continuous unidirectional change segment refers to a continuous segment in which the change direction information remains consistent across multiple adjacent monitoring periods. A fluctuation reversal segment refers to a segment in which the change direction information changes direction between adjacent monitoring periods. The stage division result information refers to the information formed after determining the segment boundaries and segment categories of the continuous unidirectional change segment and the fluctuation reversal segment.

[0137] Specifically, when identifying continuous unidirectional change segments and fluctuating reversal segments based on the characterization change results, the corresponding characterization change results are first read according to each filter layer. Then, the change direction information and adjacent change amplitude information corresponding to each adjacent monitoring cycle are extracted sequentially according to the monitoring cycle order. Let the change direction value corresponding to the i-th adjacent monitoring cycle be di, and the adjacent change amplitude value corresponding to the i-th adjacent monitoring cycle be Ai. The change direction value di takes the direction value corresponding to increasing, decreasing, or remaining unchanged. Then, the change direction value sequence is continuously scanned. When multiple consecutive adjacent monitoring cycles satisfy... When the i-th adjacent monitoring period to the j-th adjacent monitoring period are consecutively changing in the same direction, the continuous segment is identified as a continuous segment changing in the same direction. When two adjacent adjacent monitoring periods satisfy d_i \neq At that time, the switching segment consisting of the i-th adjacent monitoring period to the (i+1)-th adjacent monitoring period is identified as the fluctuation reversal segment. After identifying the continuous unidirectional change segment and the fluctuation reversal segment, the starting position and ending position of each segment are determined respectively. The formula for calculating the segment length Ls corresponding to the continuous unidirectional change segment is Ls=j-i+1, and the formula for calculating the segment amplitude Rs corresponding to the fluctuation reversal segment is... Then, the start position, end position, length Ls, amplitude Rs, and corresponding segment categories of each segment are organized to obtain the stage division results for each filter layer.

[0138] S404: Based on the stage division results, the continuous same-direction change segments and the fluctuation reversal segments are sequentially connected to obtain the blockage evolution sequence corresponding to each filter layer.

[0139] In this embodiment, the congestion evolution sequence refers to an information sequence formed by sequentially connecting continuous unidirectional change segments and fluctuation reversal segments according to the monitoring cycle sequence, which is used to indicate the phased changes in congestion characterization information over time.

[0140] Specifically, based on the stage division results, when sequentially connecting continuous unidirectional change segments and fluctuation reversal segments, the stage division results are first read according to each filter layer. The starting position, ending position, length, amplitude, and category of each segment are extracted from these results. Then, the segments are sorted sequentially based on their starting and ending positions, ensuring that segments connecting the end of the previous monitoring cycle with the start of the next cycle are arranged chronologically. Subsequently, the continuous unidirectional change segments and fluctuation reversal segments are sequentially connected according to this arrangement. The category, length, and amplitude of each segment are written into the corresponding sequence position, forming stage nodes for each sequence position. The k-th stage node is denoted as Sk. The congestion evolution sequence can then be represented as... Where m is the number of stage nodes formed after concatenation, the sequence transition value is determined according to the connection relationship between adjacent stage nodes. Let the amplitude of the segment corresponding to the k-th stage node be Rk, and the amplitude of the segment corresponding to the (k+1)-th stage node be Rk. Then, the formula for calculating the sequence transition value Tk between the k-th stage node and the (k+1)-th stage node is: The sequence transition values ​​Tk are written into the blocking evolution sequence according to the stage node arrangement order. Finally, the stage nodes and corresponding sequence transition values ​​after the sequential connection are sorted out to obtain the blocking evolution sequence corresponding to each filter layer.

[0141] In one embodiment, step S50 involves performing clogging prediction processing based on the clogging evolution sequence to obtain the target filter layer and the corresponding clogging prediction result information, including:

[0142] S501: Extract the growth slope information, node interval information, and continuous enhancement segment length information between adjacent evolution nodes in each congestion evolution sequence.

[0143] In this embodiment, the growth slope information refers to the growth rate of the degree of congestion change between adjacent evolution nodes over time, the node interval information refers to the time interval information between adjacent evolution nodes, and the continuous enhancement segment length information refers to the length information corresponding to the evolution node segment that continuously maintains an enhanced state in the congestion evolution sequence.

[0144] Specifically, when extracting the growth slope information, node interval information, and continuously enhancing segment length information between adjacent evolution nodes in each congestion evolution sequence, the corresponding congestion evolution sequence is first read according to each filter layer, and then adjacent evolution nodes are extracted sequentially according to the node arrangement order in the congestion evolution sequence. Let the node representation value corresponding to the k-th evolution node be Vk, the corresponding node time value be tk, and the node representation value corresponding to the (k+1)-th evolution node be Vk. The corresponding node time value is t k+1 Then, the formula for calculating the growth slope information Gk between adjacent evolutionary nodes is: ,in, To pre-set a small correction value, after obtaining the growth slope information Gk, the node interval information Ik between adjacent evolutionary nodes is further calculated. The formula for calculating the node interval information Ik is as follows: Subsequently, evolutionary node segments that continuously maintain an enhanced state are identified according to the value direction of the growth slope information Gk. When multiple consecutive adjacent evolutionary nodes satisfy Gk>0, the corresponding node segment is determined as a continuously enhanced segment. The number of node connections that continuously satisfy Gk>0 in the continuously enhanced segment is counted to obtain the length information Lk of the continuously enhanced segment. The formula for calculating the length information Lk of the continuously enhanced segment is Lk=q-p+1, where p is the starting node position corresponding to the continuously enhanced segment, and q is the ending node position corresponding to the continuously enhanced segment. Finally, the growth slope information, node interval information, and continuously enhanced segment length information corresponding to each filter layer are organized according to the correspondence of evolutionary nodes.

[0145] S502: Based on the growth slope information, node interval information, and continuous enhancement segment length information of each filter layer, generate the sequence extrapolation result information corresponding to each filter layer.

[0146] In this embodiment, the sequence extrapolation result information refers to the information formed by estimating the subsequent change state of the blockage evolution sequence corresponding to each filter layer based on the growth slope information, node interval information, and continuous enhancement segment length information.

[0147] Specifically, when generating the sequence extrapolation results for each filter layer based on the growth slope information, node interval information, and sustained enhancement segment length information corresponding to each filter layer, the growth slope information, node interval information, and sustained enhancement segment length information are first read separately for each filter layer. Then, the growth slope information corresponding to each filter layer is sorted sequentially to obtain the slope sequence, and the average growth slope of the slope sequence is calculated. Average growth slope The calculation formula is Where Gk represents the growth slope information corresponding to the k-th adjacent evolution node, and n represents the number of growth slope information. Subsequently, the node interval information corresponding to each filter layer is sorted sequentially to obtain the interval sequence, and the average node interval is calculated. Average node interval The calculation formula is Where Ik represents the node interval information corresponding to the kth adjacent evolutionary node, after obtaining the average growth slope and average node interval Then, the characterization value Vm of the current end-evolution node and the time value tm of the current end-evolution node corresponding to each filter layer are read, and the average growth slope is used as the basis for the calculation. Average node interval And continuously enhance the segment length information L to calculate the extrapolated node characterization value Vp. The formula for calculating the extrapolated node characterization value Vp is as follows: Where L represents the length of the sustained enhancement segment corresponding to the current terminal evolution node. This represents the natural logarithm operation, followed by the calculation of the extrapolated node time value tp. The formula for calculating the extrapolated node time value tp is: Then, the extrapolation increment is calculated based on the extrapolated node characterization value Vp and the current end-evolution node characterization value Vm. Extrapolation increment The calculation formula is The extrapolated node representation value Vp, the extrapolated node time value tp, and the extrapolated increment are then used to calculate the extrapolated node value Vp, the extrapolated node time value tp, and the extrapolated increment. The corresponding information of the filter layers is organized according to their correspondence to obtain the sequence extrapolation results for each filter layer.

[0148] S503: Based on the sequence extrapolation results of each filter layer, determine the predicted clogging level and prediction priority information of each filter layer.

[0149] In this embodiment, the predicted clogging level information refers to the information determined based on the sequence extrapolation results of each filter layer, which indicates the level of clogging development. The predicted priority information refers to the information determined based on the sequence extrapolation results of each filter layer, which indicates the order of different filter layers in subsequent control processes.

[0150] Specifically, when determining the predicted congestion level and priority information for each filter layer based on the sequence extrapolation results, the sequence extrapolation results for each filter layer are read first, and the extrapolation node representation value, extrapolation node time value, and extrapolation increment are extracted from the sequence extrapolation results. Then, interval matching is performed between the extrapolation node representation value and the preset congestion level interval to determine the predicted congestion level information for each filter layer. The preset congestion level interval refers to multiple numerical intervals pre-divided according to the congestion status from low to high. When the extrapolation node representation value falls into a lower numerical interval, the predicted congestion level information for the corresponding filter layer is determined to be a lower level; when the extrapolation node representation value falls into an intermediate numerical interval, the predicted congestion level information for that filter layer is determined to be lower. The predicted blockage level information of the corresponding filter layer is determined as the intermediate level. When the extrapolation node characterization value falls into the higher value range, the predicted blockage level information of the corresponding filter layer is determined as the higher level. After determining the predicted blockage level information of each filter layer, the extrapolation increments of each filter layer are compared according to their size, and the order is determined by combining the time relationship of the extrapolation node values. Among them, the filter layer with a larger extrapolation increment and an earlier extrapolation node time value is determined as the higher prediction priority information, and the filter layer with a smaller extrapolation increment or a later extrapolation node time value is determined as the relatively lower prediction priority information. Finally, the predicted blockage level information and prediction priority information of each filter layer are sorted according to the correspondence of the filter layers.

[0151] S504: Determine the target filter layer based on the predicted clogging level information and prediction priority information corresponding to each filter layer.

[0152] In this embodiment, the target filter layer refers to the filter layer that is currently the target for subsequent processing, determined after filtering multiple filter layers based on predicted congestion level information and predicted priority information.

[0153] Specifically, when determining the target filter layer based on the predicted congestion level information and prediction priority information corresponding to each filter layer, the predicted congestion level information and prediction priority information corresponding to each filter layer are read first. Then, the predicted congestion level information is used as the primary screening criterion to compare the levels of multiple filter layers. The filter layer with the highest predicted congestion level information is determined as the candidate filter layer set. If there is only one filter layer in the candidate filter layer set, the corresponding filter layer is directly determined as the target filter layer. If there are multiple filter layers in the candidate filter layer set, the prediction priority information is used as a further screening criterion to compare the order of multiple filter layers in the candidate filter layer set. The filter layer with the highest prediction priority information is determined as the target filter layer. After the target filter layer is determined, the filter layer identifier corresponding to the target filter layer is extracted from the results corresponding to multiple filter layers and recorded as the final determination result.

[0154] S505: Generate the blockage prediction result information corresponding to the target filter layer based on the predicted blockage level information, prediction priority information, and sequence extrapolation result information corresponding to the target filter layer.

[0155] In this embodiment, the blockage prediction result information refers to the information formed by correspondingly organizing the predicted blockage level information, prediction priority information, and sequence extrapolation result information corresponding to the target filter layer, which is used to indicate the predicted blockage status of the target filter layer.

[0156] Specifically, when generating the congestion prediction result information corresponding to the target filter layer based on the predicted congestion level information, prediction priority information, and sequence extrapolation result information corresponding to the target filter layer, the process first reads the predicted congestion level information and prediction priority information corresponding to the target filter layer, then reads the sequence extrapolation result information corresponding to the target filter layer, and extracts the extrapolation node representation value, extrapolation node time value, and extrapolation increment from the sequence extrapolation result information. Subsequently, the predicted congestion level information, prediction priority information, extrapolation node representation value, extrapolation node time value, and extrapolation increment are processed according to a unified correspondence. The field processing refers to writing information content from different sources into the same result structure according to a preset record order. After the field processing is completed, the consistency of each piece of information after processing is checked to confirm that the predicted congestion level information, prediction priority information, and sequence extrapolation result information all correspond to the same target filter layer. If the consistency check passes, the processed information content is combined sequentially to generate the congestion prediction result information corresponding to the target filter layer.

[0157] In one embodiment, step S60 involves performing strategy matching based on the target filtering layer and the clogging prediction result information to obtain adaptive control strategy information, including:

[0158] S601: Obtain preset strategy rule information, which includes path adjustment matching rule information and recovery processing matching rule information.

[0159] In this embodiment, the preset strategy rule information refers to the information that is pre-set and used to establish a matching relationship between the target filtering layer and the corresponding processing content; the path adjustment matching rule information refers to the rule information used to determine the target filtering path adjustment processing content; and the recovery processing matching rule information refers to the rule information used to determine the target filtering layer recovery processing content.

[0160] Specifically, when obtaining the preset strategy rule information, the rule record content corresponding to the tailwater constructed wetland regulation process is first read, and the rule record content is classified and identified according to rule category. The rule content describing the correspondence between the target filter path adjustment conditions and the path adjustment content is determined as the path adjustment matching rule information, and the rule content describing the correspondence between the target filter layer recovery conditions and the recovery treatment content is determined as the recovery treatment matching rule information. Subsequently, the path adjustment matching rule information and the recovery treatment matching rule information are parsed to extract the filter layer conditions, clogging level conditions, priority conditions, and treatment content corresponding to each rule. Among them, the filter layer condition refers to the range of filter layers to which the rule applies, the clogging level condition refers to the range of predicted clogging level information to which the rule applies, the priority condition refers to the range of predicted priority information to which the rule applies, and the treatment content refers to the regulation content that needs to be executed when the corresponding conditions are met. After completing the field parsing process, the path adjustment matching rule information and the recovery treatment matching rule information are sorted according to rule category to obtain the preset strategy rule information.

[0161] S602: Extract the predicted blockage level information and prediction priority information corresponding to the target filtering layer from the blockage prediction result information.

[0162] In this embodiment, the clogging prediction result information refers to the information used to indicate the predicted clogging status of the target filter layer, the predicted clogging level information refers to the clogging development level information of the target filter layer, and the prediction priority information refers to the processing order information of the target filter layer.

[0163] Specifically, when extracting the predicted congestion level information and prediction priority information corresponding to the target filter layer from the congestion prediction result information, the process first reads the contents of each record in the congestion prediction result information and parses each record according to field category to distinguish the filter layer identifier field, the predicted congestion level field, the prediction priority field, and the sequence extrapolation result field. Then, the filter layer identifier corresponding to the target filter layer is read and compared with the filter layer identifier field in the congestion prediction result information. If the comparison is consistent, the contents corresponding to the predicted congestion level field are extracted from the records corresponding to the target filter layer as the predicted congestion level information of the target filter layer. At the same time, the contents corresponding to the prediction priority field are extracted from the records corresponding to the target filter layer as the prediction priority information of the target filter layer. After the extraction of the predicted congestion level information and prediction priority information is completed, the consistency of the correspondence between the extracted two pieces of information and the target filter layer is checked, and the checked predicted congestion level information and prediction priority information are used as input for subsequent matching processing.

[0164] S603: Based on the target filtering layer, predicted congestion level information, and predicted priority information, match them with the path adjustment matching rule information to obtain candidate information for path adjustment strategies.

[0165] In this embodiment, the path adjustment strategy candidate information refers to the candidate path adjustment content information selected after matching the target filtering layer, the predicted congestion level information, and the predicted priority information with the path adjustment matching rule information.

[0166] Specifically, when matching the target filter layer, predicted congestion level information, and predicted priority information with the path adjustment matching rule information, the process first reads the content of each path adjustment rule in the path adjustment matching rule information and splits the fields of each path adjustment rule content to extract the corresponding filter layer conditions, congestion level conditions, priority conditions, and path adjustment content. Then, the target filter layer is compared with the filter layer conditions in each path adjustment rule content one by one. The path adjustment rule content whose filter layer conditions are consistent with the target filter layer is retained as the first candidate rule content. Subsequently, the predicted congestion level... The information is compared with the congestion level conditions in each of the first candidate rules. The path adjustment rules whose congestion level conditions match the predicted congestion level information are retained as second candidate rules. Then, the predicted priority information is compared with the priority conditions in each of the second candidate rules. The path adjustment rules whose priority conditions match the predicted priority information are retained as target candidate rules. After the target candidate rules are selected, the corresponding path adjustment content is extracted from each target candidate rule. The extracted path adjustment content is then organized according to the order of the rule source to obtain the path adjustment strategy candidate information.

[0167] S604: Based on the target filtering layer, predicted blockage level information, and predicted priority information, match them with the recovery processing matching rule information to obtain candidate information for recovery processing strategies.

[0168] In this embodiment, the candidate information for recovery processing strategy refers to the candidate recovery processing content information selected after matching the target filtering layer, the predicted blocking level information, and the predicted priority information with the recovery processing matching rule information.

[0169] Specifically, when matching the target filter layer, predicted congestion level information, and predicted priority information with the recovery processing matching rule information, the process first reads the content of each recovery processing rule in the recovery processing matching rule information and splits the fields of each recovery processing rule content to extract the corresponding filter layer conditions, congestion level conditions, priority conditions, and recovery processing content. Then, the target filter layer is compared with the filter layer conditions in each recovery processing rule content one by one. The recovery processing rule content whose filter layer conditions are consistent with the target filter layer is retained as the first candidate rule content. Subsequently, the predicted congestion level... The information is compared with the congestion level conditions in each of the first candidate rules. The recovery processing rules whose congestion level conditions match the predicted congestion level information are retained as the second candidate rules. Then, the predicted priority information is compared with the priority conditions in each of the second candidate rules. The recovery processing rules whose priority conditions match the predicted priority information are retained as the target candidate rules. After the target candidate rules are selected, the corresponding recovery processing content is extracted from each target candidate rule. The extracted recovery processing content is then organized according to the order of the rule source to obtain the recovery processing strategy candidate information.

[0170] S605: Combine and filter the candidate information of path adjustment strategy and the candidate information of recovery processing strategy to obtain adaptive control strategy information.

[0171] In this embodiment, the adaptive control strategy information refers to the control information corresponding to the current state of the target filtering layer, which is formed by combining and filtering the candidate information of path adjustment strategy and the candidate information of recovery processing strategy.

[0172] Specifically, when performing combined screening of candidate information for path adjustment strategies and candidate information for recovery processing strategies, the following steps are taken: First, each path adjustment content in the candidate information for path adjustment strategies and each recovery processing content in the candidate information for recovery processing strategies are read separately. Then, according to the correspondence of the target filtering layer, each path adjustment content and each recovery processing content is organized to form a candidate combination set. Subsequently, a compatibility judgment is performed on each candidate combination in the candidate combination set to determine whether there is an execution conflict between each path adjustment content and each recovery processing content. Here, an execution conflict refers to the situation where simultaneous execution within the same time period would lead to inconsistencies in the control direction or a lack of connection in the processing order. After completing the compatibility judgment, candidate combinations without execution conflicts are retained as optional combinations. Each optional combination is then corrected in order according to the predicted congestion level and prediction priority information, so that the optional combinations corresponding to higher predicted congestion level and earlier prediction priority information are arranged first. Then, the optional combinations after order correction are merged, and the path adjustment content and recovery processing content that can be continuously implemented under the same execution direction are integrated into unified control content. After the content merging is completed, the retained and integrated control content is sorted according to the preset output order to obtain adaptive control strategy information.

[0173] In one embodiment, step S80 involves acquiring feedback monitoring information after target filter path adjustment processing or target filter layer recovery processing, and correcting the blockage prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information, including:

[0174] S801: Generate feedback blockage characterization information corresponding to the target filter layer based on feedback monitoring information.

[0175] In this embodiment, the feedback blockage characterization information refers to the information formed based on feedback monitoring information, which indicates the current blockage status of the target filter layer after completing the target filter path adjustment process or the target filter layer recovery process.

[0176] Specifically, when generating feedback blockage characterization information corresponding to the target filter layer based on feedback monitoring information, the interlayer electrical response information, permeation flow information, and liquid level change information corresponding to the target filter layer are first read from the feedback monitoring information. Then, the interlayer electrical response information, permeation flow information, and liquid level change information are organized according to the correspondence of the target filter layer, so that the interlayer electrical response information, permeation flow information, and liquid level change information formed at the same acquisition time maintain a time correspondence. After completing the organization, the electrical response change value in the interlayer electrical response information, the flux change value in the permeation flow information, and the liquid level change value in the liquid level change information are extracted respectively, and the extracted values ​​are unified. The data is processed to make the values ​​under different information categories comparable. Then, the unified interlayer electrical response information, permeation flow information, and liquid level change information are jointly analyzed. The joint analysis includes identifying the corresponding directions of numerical changes among the interlayer electrical response information, permeation flow information, and liquid level change information, and comparing the corresponding deviations among the numerical values ​​among the interlayer electrical response information, permeation flow information, and liquid level change information. After the joint analysis is completed, the joint analysis results corresponding to the interlayer electrical response information, permeation flow information, and liquid level change information are mapped according to the preset characterization rules to form feedback blockage characterization information corresponding to the current state of the target filter layer.

[0177] S802: Perform deviation decomposition processing on the feedback congestion characterization information and the congestion prediction result information to obtain the level deviation direction information and the priority deviation direction information.

[0178] In this embodiment, the grade deviation direction information refers to the deviation direction information shown by the feedback congestion characterization information relative to the predicted congestion grade information in the congestion prediction result information, the priority deviation direction information refers to the deviation direction information shown by the feedback congestion characterization information relative to the predicted priority information in the congestion prediction result information, and the deviation decomposition processing refers to the process of splitting the overall difference between the feedback congestion characterization information and the congestion prediction result information into the grade correspondence and priority correspondence respectively.

[0179] Specifically, when performing deviation decomposition processing on the feedback congestion characterization information and the congestion prediction result information, the feedback congestion characterization information is first read, and the predicted congestion level information and prediction priority information corresponding to the target filtering layer are read from the congestion prediction result information. Then, the feedback congestion characterization information is subjected to level mapping processing to determine the feedback level result corresponding to the feedback congestion characterization information, and priority mapping processing is performed on the feedback congestion characterization information to determine the feedback priority result corresponding to the feedback congestion characterization information. The level mapping processing refers to the process of mapping the feedback congestion characterization information to the corresponding level interval according to a preset level correspondence, and the priority mapping processing refers to the process of mapping the feedback congestion characterization information to the corresponding priority order according to a preset priority correspondence. After obtaining the feedback level result and the feedback priority result, the feedback level result is compared with the predicted congestion level information... The process involves a line-by-line comparison. When the feedback level is higher than the predicted congestion level, the level deviation direction is determined as the upward adjustment direction; when the feedback level is lower than the predicted congestion level, it is determined as the downward adjustment direction; and when the feedback level matches the predicted congestion level, it is determined as the hold direction. Next, the feedback priority is compared with the predicted priority. When the feedback priority is higher than the predicted priority, the priority deviation direction is determined as the forward shift direction; when the feedback priority is lower than the predicted priority, it is determined as the backward shift direction; and when the feedback priority matches the predicted priority, it is determined as the hold direction. Finally, the level deviation direction and priority deviation direction are organized according to the target filtering layer correspondence.

[0180] S803: Based on the grade deviation direction information and the priority deviation direction information, the congestion prediction result information is modified by sub-item correction to obtain the corrected congestion prediction result information.

[0181] In this embodiment, the corrected congestion prediction result information refers to the information formed by adjusting the predicted congestion level information and the predicted priority information in the congestion prediction result information respectively. The itemized correction process refers to the process of adjusting the different information items in the congestion prediction result information as independent correction objects.

[0182] Specifically, when performing itemized correction processing on the blockage prediction results based on the level deviation direction information and the priority deviation direction information, the process first reads the predicted blockage level information, predicted priority information, and sequence extrapolation result information corresponding to the target filtering layer from the blockage prediction results. Then, the level deviation direction information is matched with the predicted blockage level information. If the level deviation direction information is upward, the predicted blockage level information is adjusted to an adjacent level higher than the current level. If the level deviation direction information is downward, the predicted blockage level information is adjusted to an adjacent level lower than the current level. If the level deviation direction information is in a stable direction, the original level of the predicted blockage level information is retained unchanged. Subsequently, the priority is adjusted... The priority deviation direction information is matched with the prediction priority information. When the priority deviation direction information is forward, the prediction priority information is adjusted to the adjacent priority that is earlier. When the priority deviation direction information is backward, the prediction priority information is adjusted to the adjacent priority that is later. When the priority deviation direction information is in the same direction, the original order of the prediction priority information is retained. After the prediction congestion level information and prediction priority information are adjusted respectively, the adjusted prediction congestion level information, the adjusted prediction priority information, and the original sequence extrapolation result information are reorganized and reordered according to the correspondence of the target filtering layer to obtain the corrected congestion prediction result information.

[0183] S804: The adaptive control strategy information is reconstructed based on the corrected congestion prediction results to obtain the updated adaptive control strategy information.

[0184] In this embodiment, the updated adaptive control strategy information refers to the control information re-determined based on the corrected congestion prediction results. The reconstruction process refers to the process of re-matching, re-filtering, and recombining the original control content in the adaptive control strategy information.

[0185] Specifically, when reconstructing the adaptive control strategy information based on the corrected congestion prediction results, the process first reads the adjusted predicted congestion level information, adjusted predicted priority information, and sequence extrapolation result information corresponding to the target filtering layer from the corrected congestion prediction results. Then, it reads the original path adjustment content and the original recovery processing content from the adaptive control strategy information. Using the adjusted predicted congestion level information, adjusted predicted priority information, and sequence extrapolation result information as new matching criteria, the original path adjustment content and the original recovery processing content are verified item by item. This item-by-item verification determines whether each original control content still maintains a corresponding relationship with the corrected congestion state. During the verification process, the original control content that is inconsistent with the corrected congestion prediction results is identified as content to be replaced, while the original control content that is still consistent with the corrected congestion prediction results is identified as content to be retained. Then, based on the corrected congestion prediction results, the path adjustment content and recovery processing content corresponding to the current congestion state are reread, and the control content formed by the reread is identified as the replacement content. The retained content and the replacement content are then recombined according to the correspondence of the target filtering layer, and the execution order of each recombined control content is rearranged according to the adjusted prediction priority information. Finally, the control content after recombining and rearranging is sorted out to obtain the updated adaptive control strategy information.

[0186] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0187] In one embodiment, a tailwater wetland clogging control device is provided, which corresponds one-to-one with the tailwater wetland clogging control method described in the above embodiments. This tailwater wetland clogging control device includes a monitoring information acquisition module, a feature extraction module, a fusion characterization module, a time-series correlation module, a clogging prediction module, a strategy matching module, a control execution module, and a feedback correction module.

[0188] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

Claims

1. A method for regulating tailwater wetland blockage, characterized in that, The tailwater wetland clogging control method includes: Acquire information on the interlayer electrical response, infiltration flow rate, and liquid level change of each filter layer in the constructed wetland for tailwater. Feature extraction processing is performed on the interlayer electrical response information, the permeation flow information and the liquid level change information to obtain the electrical response offset feature, permeation attenuation feature and liquid level hysteresis feature corresponding to each filter layer; The electrical response offset feature, the seepage attenuation feature, and the liquid level hysteresis feature are fused and characterized to obtain the clogging characterization information corresponding to each filter layer; Based on the blockage characterization information, a temporal correlation processing is performed to obtain the blockage evolution sequence corresponding to each of the filter layers; Based on the clogging evolution sequence, clogging prediction processing is performed to obtain the target filter layer and the clogging prediction result information corresponding to the target filter layer; Based on the target filtering layer and the blockage prediction result information, strategy matching processing is performed to obtain adaptive control strategy information; Perform target filtering path adjustment processing or target filtering layer restoration processing based on the adaptive control strategy information; The system obtains feedback monitoring information after the target filtering path adjustment process or the target filtering layer recovery process, and corrects the blockage prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information.

2. The tailwater wetland clogging control method according to claim 1, characterized in that, The step of extracting features from the interlayer electrical response information, the permeation flow information, and the liquid level change information to obtain the electrical response offset features, permeation attenuation features, and liquid level hysteresis features corresponding to each filter layer includes: The interlayer electrical response information, the permeation flow information, and the liquid level change information are synchronously registered according to the preset sampling time sequence to obtain the time sequence monitoring unit corresponding to each filter layer; Based on the interlayer electrical response value corresponding to the current sampling time and the interlayer electrical response value corresponding to the historical reference sampling time in the time-series monitoring unit, the offset difference is calculated to obtain the electrical response offset feature corresponding to each filter layer. Based on the seepage flow values ​​corresponding to multiple consecutive sampling times in the time-series monitoring unit, attenuation trend fitting is performed to obtain the seepage attenuation characteristics corresponding to each filter layer. Based on the first moment when the liquid level change reaches a preset change amplitude and the second moment when the permeation flow rate changes to a preset change amplitude in the time-series monitoring unit, the time difference between the first moment and the second moment is calculated to obtain the liquid level hysteresis characteristics corresponding to each filter layer.

3. The tailwater wetland clogging control method according to claim 1, characterized in that, The process of fusing the electrical response offset feature, the seepage attenuation feature, and the liquid level hysteresis feature to obtain the clogging characterization information corresponding to each filter layer includes: The electrical response offset feature, the seepage attenuation feature, and the liquid level hysteresis feature are subjected to interval normalization to obtain standardized electrical response offset feature, standardized seepage attenuation feature, and standardized liquid level hysteresis feature. The standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature are used to extract information on the direction of synchronous change of features and the magnitude of feature deviation. Based on the feature synchronization change direction information and the feature deviation magnitude information, feature coupling relationship information corresponding to each of the filter layers is generated; Based on the feature coupling relationship information corresponding to each of the filter layers, the standardized electrical response offset feature, the standardized seepage attenuation feature, and the standardized liquid level hysteresis feature are subjected to layered fusion processing to obtain the fusion characterization vector corresponding to each of the filter layers. The fusion representation vector corresponding to each of the filter layers is input into a preset blocking mapping rule to obtain the blocking representation information corresponding to each of the filter layers.

4. The tailwater wetland clogging control method according to claim 1, characterized in that, The step of performing time-series correlation processing based on the blockage characterization information to obtain the blockage evolution sequence corresponding to each of the filter layers includes: The blockage characterization information is processed in a time sequence according to a preset monitoring time order to obtain a time sequence characterization group corresponding to each filter layer; Based on the blockage characterization information corresponding to adjacent monitoring periods in the time-series characterization group corresponding to each filter layer, the adjacent change amplitude information and change direction information are calculated to obtain the characterization change result information corresponding to each filter layer. Based on the characterization change results, continuous unidirectional change segments and fluctuation reversal segments are identified to obtain the stage division results information corresponding to each filter layer. Based on the stage division results, the continuous unidirectional change segments and the fluctuation reversal segments are sequentially connected to obtain the blockage evolution sequence corresponding to each filter layer.

5. The tailwater wetland clogging control method according to claim 1, characterized in that, The process of performing clogging prediction based on the clogging evolution sequence to obtain the target filter layer and the corresponding clogging prediction result information includes: Extract the growth slope information, node interval information, and continuous enhancement segment length information between adjacent evolution nodes in each of the blockage evolution sequences; Based on the growth slope information, node interval information and continuous enhancement segment length information corresponding to each filter layer, the sequence extrapolation result information corresponding to each filter layer is generated. Based on the sequence extrapolation result information corresponding to each of the filter layers, determine the predicted clogging level information and the predicted priority information corresponding to each of the filter layers; The target filter layer is determined based on the predicted clogging level information and the predicted priority information corresponding to each filter layer. Based on the predicted congestion level information, the predicted priority information, and the sequence extrapolation result information corresponding to the target filter layer, the congestion prediction result information corresponding to the target filter layer is generated.

6. The tailwater wetland clogging control method according to claim 1, characterized in that, The step of performing strategy matching processing based on the target filtering layer and the blockage prediction result information to obtain adaptive control strategy information includes: Obtain preset strategy rule information, which includes path adjustment matching rule information and recovery processing matching rule information; Extract the predicted blockage level information and prediction priority information corresponding to the target filter layer from the blockage prediction result information; Based on the target filtering layer, the predicted congestion level information, and the predicted priority information, a matching process is performed with the path adjustment matching rule information to obtain path adjustment strategy candidate information; Based on the target filtering layer, the predicted blockage level information, and the predicted priority information, a matching process is performed with the recovery processing matching rule information to obtain recovery processing strategy candidate information; The candidate information for path adjustment strategy and the candidate information for recovery processing strategy are combined and filtered to obtain the adaptive control strategy information.

7. The tailwater wetland clogging control method according to claim 1, characterized in that, The step of obtaining feedback monitoring information after the target filtering path adjustment process or the target filtering layer recovery process, and correcting the blockage prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information, includes: Based on the feedback monitoring information, generate feedback blockage characterization information corresponding to the target filter layer; The feedback congestion characterization information and the congestion prediction result information are subjected to deviation decomposition processing to obtain the level deviation direction information and the priority deviation direction information. Based on the grade deviation direction information and the priority deviation direction information, the congestion prediction result information is corrected by itemization to obtain the corrected congestion prediction result information. The adaptive control strategy information is reconstructed based on the corrected congestion prediction results to obtain the updated adaptive control strategy information.

8. A tailwater wetland clogging control device, characterized in that, The tailwater wetland blockage control device includes: The monitoring information acquisition module is used to acquire interlayer electrical response information, infiltration flow information, and liquid level change information for each filter layer in the wastewater constructed wetland. The feature extraction module is used to perform feature extraction processing on the interlayer electrical response information, the permeation flow information and the liquid level change information to obtain the electrical response offset feature, permeation attenuation feature and liquid level hysteresis feature corresponding to each filter layer. The fusion characterization module is used to fuse the electrical response offset feature, the seepage attenuation feature and the liquid level hysteresis feature to obtain the clogging characterization information corresponding to each filter layer. The temporal correlation module is used to perform temporal correlation processing based on the blockage characterization information to obtain the blockage evolution sequence corresponding to each of the filter layers; The clogging prediction module is used to perform clogging prediction processing based on the clogging evolution sequence to obtain the target filter layer and the clogging prediction result information corresponding to the target filter layer. The strategy matching module is used to perform strategy matching processing based on the target filtering layer and the blockage prediction result information to obtain adaptive control strategy information. The control execution module is used to perform target filtering path adjustment processing or target filtering layer recovery processing according to the adaptive control strategy information. The feedback correction module is used to obtain feedback monitoring information after the target filtering path adjustment processing or the target filtering layer recovery processing, and to correct the blockage prediction result information based on the feedback monitoring information to obtain updated adaptive control strategy information.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the tailwater wetland blockage control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the tailwater wetland blockage regulation method as described in any one of claims 1 to 7.

Images