A method and system for targeted removal of phosphorus-containing contaminants in natural water bodies

Abstract

The present application provides a method and system for targeted removal of phosphorus-containing contaminants in natural water bodies. The method includes placing an electrolytic cell comprising aluminum anodes under the sediment in the area of the natural water body to be treated; forming magnetically tagged phosphorus-containing flocs by applying electricity to the aluminum anodes while adding magnetic particles; identifying a phosphorus-rich zone of the phosphorus-containing flocs by the reflected spectrum; establishing a spatial distribution map of the phosphorus-containing flocs in the phosphorus-rich zone by analyzing the echo intensity and attenuation characteristics; detecting magnetic field anomalies caused by the magnetically tagged phosphorus-containing flocs to determine specific locations; and performing suction based on the specific locations to target removal of phosphorus-containing contaminants in the area of the natural water body. The method and system of the present application can identify in situ and perform selective targeted suction without moving the sediment, thereby improving efficiency and effectiveness.

Description

Technical Field

[0001] This application relates to the field of water treatment technology, and more particularly to a method and system for the targeted removal of phosphorus-containing pollutants from natural water bodies. Background Technology

[0002] High concentrations of phosphorus pollutants in rivers and lakes lead to severe eutrophication in urban waterways, easily causing problems such as cyanobacterial blooms. Chemical-based phosphorus removal methods are prone to secondary pollution, while ecological phosphorus removal methods suffer from low efficiency.

[0003] Currently, there are in-situ phosphorus removal technologies for rivers and lakes based on electrocoagulation. These technologies use iron, aluminum, and other metals as electrodes to generate metal ions through electrolysis. These ions combine with phosphorus-containing pollutants to form water-insoluble compounds that settle in the sediment, thereby reducing the concentration of phosphorus pollutants in the water. However, these flocs are difficult to collect in a targeted manner, and their accumulation in the sediment still poses a risk of releasing phosphorus pollutants again.

[0004] Traditional sediment dredging primarily utilizes environmentally friendly cutter suction dredgers, which suffer from the problem of "indiscriminate removal": not only are pollutant-rich surface active sediments removed, but also large amounts of unpolluted or low-polluted lower clean sediments are swept away, resulting in large dredging volumes, high subsequent treatment costs, and slow aquatic ecosystem restoration. Current technologies lack methods for in-situ identification and selective separation of phosphorus-containing flocs generated by electrocoagulation in mixed sediments.

[0005] Therefore, there is a need for improved methods and systems for the targeted removal of phosphorus-containing pollutants from natural water bodies. Summary of the Invention

[0006] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0007] In one aspect, this application provides a method for targeted removal of phosphorus-containing pollutants from natural water bodies, the method comprising the following steps: S1: Place the electrolysis unit containing the aluminum anode under the bottom sediment in the treatment area of ​​the natural water body; S2: By energizing the aluminum anode to form phosphorus-containing flocs, magnetic particles are added to encapsulate the magnetic particles in the phosphorus-containing flocs, forming magnetically labeled phosphorus-containing flocs. S3: Emit a spectrum to the bottom sediment surface of the natural water body and receive the reflectance spectrum; based on the reflectance spectrum, identify the enrichment area of ​​phosphorus-containing flocs; S4: Emit sound waves to the identified phosphorus-containing floc enrichment area and receive the echoes; establish the spatial distribution map of phosphorus-containing flocs in the enrichment area based on the echo intensity and attenuation characteristics; S5: Detect magnetic field anomalies caused by magnetically tagged phosphorus-containing flocs and locate the specific locations of phosphorus-containing flocs in the enrichment region within areas with magnetic induction intensity greater than a preset threshold; and S6: Based on the specific location of the phosphorus-containing flocs, the phosphorus-containing flocs are aspirated to remove phosphorus-containing pollutants from the natural water body in a targeted manner.

[0008] In an exemplary embodiment, in step S2, the current density and voltage applied to the aluminum anode are 5-10 mA / cm², respectively. 2 and 15-20V.

[0009] In one exemplary embodiment, the magnetic particles are nano-Fe3O4.

[0010] In an exemplary embodiment, the method further includes: periodically updating the preset threshold in the following manner: After each aspiration of the aluminum phosphate flocs, the sample data for that time is extracted and stored in the database unit; the sample data includes spectral signal parameters, sonar echo signal parameters, and magnetic induction signal parameters, as well as the corresponding aspiration resistance parameters, aspirated floc concentration parameters, and aspirated floc turbidity parameters. Labeling the sample data in the database unit includes: for each group of sample data, at least based on whether the floc concentration parameter is greater than the concentration threshold, labeling the group of sample data as a positive sample or a negative sample; Machine learning is performed periodically using the labeled sample data in the database unit to update the preset threshold.

[0011] In an exemplary embodiment, the sonar echo signal parameters include sediment impedance and layer thickness; the spectral signal parameters are absorption peaks or reflection peaks at specific wavelengths corresponding to the characteristic spectra of Al-OH or Fe3O4; and the magnetic induction signal parameters are magnetic susceptibility anomalies.

[0012] In an exemplary embodiment, in step S3, the emission spectrum includes the emission spectrum in the 400-1000 nm band.

[0013] In one exemplary embodiment, the moisture content is greater than or equal to 90% and the density is 1.05-1.15 g / cm³. 3 The spectral characteristics of phosphorus-containing flocs are that there is a reflection peak at 550-580 nm.

[0014] In an exemplary embodiment, step S4, emitting sound waves includes emitting sound waves of 50-500 kHz.

[0015] In an exemplary embodiment, the spatial distribution map of the phosphorus-containing flocs is constructed through the following steps: Obtain the raw data, including: Obtain grid data of magnetic field intensity anomalies on the sediment plane; obtain acoustic impedance layering data and thickness information of the sediment; obtain the coordinates of each detection point; Constructing a spatial distribution map includes: The target water area is divided into multiple grids. Based on the known magnetic anomaly values ​​of discrete points, the continuous values ​​of the entire grid plane are calculated to generate a thermal map of the planar distribution of magnetic flocs. By utilizing the time and intensity of sonar echoes and through a pre-defined sound velocity model, the impedance differences at different depths of the sediment are inverted to construct a vertical distribution profile of flocs. By overlaying the planar thermal map with the vertical cross-sectional map, a three-dimensional spatial distribution model of flocs is generated.

[0016] In an exemplary embodiment, the method further includes dynamically updating the spatial distribution map; the dynamic update includes: after completing the suction at a certain location, recording the volume of aluminum phosphate flocs suctioned at that location, and then setting the value of that location on the spatial distribution map to zero or reducing it to avoid repeated operations; during the suction process, if a new magnetic anomaly signal is detected, a new target location is added to the spatial distribution map in real time.

[0017] In one exemplary embodiment, the method further includes obtaining phosphorus-containing flocs by magnetically separating and dehydrating the pumped mud-water mixture.

[0018] On the other hand, a system for targeted removal of phosphorus-containing pollutants from natural water bodies is provided, the system comprising an electrolysis unit, a target identification unit, and a suction unit; The electrolysis unit includes a magnetic particle inlet, and the electrolysis unit is configured to generate phosphorus-containing flocs and form magnetically tagged phosphorus-containing flocs with the added magnetic particles. The target identification unit includes an underwater hyperspectral imaging unit, a multi-frequency sonar unit, and a magnetic induction unit. The underwater hyperspectral imaging unit is configured to emit a spectrum to the bottom sediment surface of the natural water body and receive a reflected spectrum; based on the reflected spectrum, it identifies the enrichment area of ​​phosphorus-containing flocs; the multi-frequency sonar unit is configured to emit sound waves to the identified phosphorus-containing floc enrichment area and receive echoes; based on the echo intensity and attenuation characteristics, it establishes a spatial distribution map of phosphorus-containing flocs in the enrichment area; the magnetic induction unit is configured to detect magnetic field anomalies caused by magnetically tagged phosphorus-containing flocs and locate the specific location of phosphorus-containing flocs in the enrichment area in areas with magnetic induction intensity greater than a preset threshold. The suction unit includes a suction head with a variable diameter, and the suction unit is configured to perform directional suction based on the specific location of the determined phosphorus-containing flocs.

[0019] In one exemplary embodiment, a controller and a database unit are also included; The database unit is configured to store sample data extracted after each aspiration of the phosphorus-containing flocs, and to store the sample data as labels, including positive and negative samples; the sample data includes sonar echo signal parameters, spectral signal parameters, magnetic induction signal parameters, as well as corresponding aspiration resistance parameters, aspiration floc concentration parameters, and aspiration floc turbidity. The controller is configured to periodically perform machine learning using labeled sample data in the database to update the preset threshold.

[0020] In this application, the term "natural water body" may refer to surface water bodies with functional use, such as rivers, lakes, canals, channels, and reservoirs, which are neutral (pH approximately 7-8).

[0021] In this application, phosphorus-containing substances in natural water bodies can be classified according to conventional definitions in the art, such as dissolved phosphorus (DP) and particulate phosphorus (PP) according to physical form; and inorganic phosphorus and organic phosphorus according to chemical properties.

[0022] In this application, the aluminum anode is used to ionize aluminum ions when energized.

[0023] When an aluminum anode is energized, a dissolution reaction occurs: Al → Al 3+ + 3e - .

[0024] Al 3+ The following reaction can occur in phosphorus-containing solutions: Path 1 (Direct Precipitation): Al 3+ + PO4 3- → AlPO4↓ Pathway 2 (Hydrolysis Flocculation): Al 3+ + 3H₂O → Al(OH)₃ + 3H + mAl(OH)3 → Al m (OH) 3m (Polymer complex).

[0025] Al m (OH) 3m Through adsorption bridging and banding, fine AlPO4 particles and other phosphorus-containing particles in water can be aggregated into large phosphorus-containing flocs (AlPO4·Al). m (OH) 3m ).

[0026] This patented system for the targeted removal of phosphorus pollutants from rivers and lakes comprises an electrolysis unit, a target identification unit, a suction unit, and a controller and database unit. The system can be carried by a vessel hull, an ecological floating board, or any other material capable of floating on water.

[0027] This application enables in-situ identification and selective suction without moving the bottom sediment, thus distinguishing it from traditional dredgers and shore-based magnetic separation.

[0028] This application employs a multi-modal targeted identification method based on the differences in the physicochemical characteristics of phosphorus-containing flocs. This application uses the differences in physicochemical properties such as water content, density, and acoustic impedance between phosphorus-containing flocs generated by electrocoagulation and traditional sediment particles as the identification basis. It integrates underwater hyperspectral imaging (capturing reflectance spectral features), multi-frequency sonar echo analysis (detecting acoustic impedance differences), and an array-type magnetic sensor to determine the specific location of the phosphorus-containing flocs. This solves the problem of traditional dredging techniques being unable to distinguish between flocs and sediment, thereby achieving precise removal of phosphorus-containing flocs.

[0029] This application employs a labeling technology for phosphorus-containing flocs using magnetic tracing and inductive localization. During electrocoagulation, nano-magnetic particles (such as Fe3O4) are simultaneously added. The phosphorus-containing flocs are then adsorbed and encapsulated in situ, acquiring "magnetic tags" that endow the originally non-magnetic phosphorus-containing substances with detectable magnetic characteristics. This is combined with an array-type magnetic sensor (Hall element) for magnetic anomaly detection, forming an integrated technology chain of "electrocoagulation generation - magnetic labeling - magnetic inductive localization." This significantly improves the sensitivity and anti-interference ability for identifying low-concentration phosphorus-rich areas in complex sediment environments, thereby enhancing the removal efficiency of phosphorus pollutants in water.

[0030] This application employs a multi-source heterogeneous signal fusion and self-learning control decision system. It integrates and fuses three types of sensor signals with vastly different physical properties—spectral, sonar, and magnetic induction—within the same control framework, overcoming the limitations of single-sensor technologies in turbid water / mud environments where they are susceptible to interference. In one example of this application, spatial synchronization may include unifying the coordinates of the detection points of the multi-frequency sonar unit, underwater hyperspectral imaging unit, and array-type magnetic sensor into the same shipborne GPS-RTK (real-time dynamic positioning) coordinate system. Calibration determines the positional offset (translation vector and rotation matrix) of each sensor relative to the ship's center. Temporal synchronization may include establishing a unified clock reference for all sensors and controllers (e.g., using GPS-PTP (Precise Time Protocol) for timing). The controller extracts magnetic induction signals (magnetic susceptibility anomalies), sonar echo signals (sediment impedance, layer thickness), and hyperspectral signals (absorption / reflection peaks at specific wavelengths, corresponding to the characteristic spectra of Al-OH or Fe3O4), and performs feature fusion of these three signals.

[0031] In the example of this application, the controller can have a built-in updatable database unit that can automatically update thresholds and operating parameters based on real-time suction feedback (such as resistance and floc concentration), forming a closed-loop control of "detection-identification-extraction-verification-correction", and has environmental adaptability and continuous learning capabilities.

[0032] In the example of this application, the suction path and suction force can be continuously adjusted according to the spatial distribution map output by the target recognition unit. When the sensor detects a sudden increase in suction resistance (indicating that the suction material has changed from flocs to clean sediment), the system automatically performs operations such as reducing force, shutting down, or jumping to the next target point, thereby achieving precise operation of only suctioning phosphorus-containing flocs while retaining clean sediment, reducing the amount of ineffective dredging by more than 50%.

[0033] Compared with the prior art, this application has the following beneficial effects: 1. Targeted: By integrating the spectral, acoustic, and magnetic characteristics that distinguish "phosphorus-containing flocs" from ordinary mineral sediments, targeted removal of phosphorus can be achieved.

[0034] 2. Volume reduction: Only the phosphorus-rich floc area is extracted, reducing the amount of sludge by 50%-70%, which greatly reduces the costs of subsequent dewatering, transportation and disposal.

[0035] 3. Low disturbance: Avoid large-scale excavation, preserve clean bottom sediment and original benthic ecosystem, which is conducive to the rapid self-repair of aquatic ecosystem.

[0036] 4. Continuous operation: The integrated design of electrolytic flocculation-identification-suction allows for continuous operation on a mobile platform (ship / water excavator).

[0037] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. Other advantages of this application may be realized and obtained by means of the embodiments described in the description and drawings. Attached Figure Description

[0038] The accompanying drawings are used to provide an understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0039] Figure 1 A schematic diagram of the structure of a system for targeted removal of phosphorus-containing pollutants from natural water bodies according to an exemplary embodiment of this application; and Figure 2 This is a schematic flowchart illustrating a method for targeted removal of phosphorus-containing pollutants from natural water bodies according to an exemplary embodiment of this application. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application are described in detail below. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be arbitrarily combined with each other.

[0041] The following is for reference. Figure 1 and Figure 2 To describe systems and methods for the targeted removal of phosphorus-containing pollutants from natural water bodies.

[0042] The system used in this embodiment can float on the water surface with the help of an ecological floating plate 1. The system may include an electrolysis unit 3 and a magnetic particle feeding device 2 located at the front end of the system. The electrolysis unit 3 may include an aluminum anode and a cathode (such as a stainless steel cathode). By applying direct current to the aluminum anode, phosphorus-containing flocs are generated. The magnetic particle feeding device 2 can add nano-magnetic particles (Fe3O4, particle size 20-100nm), so that the Fe3O4 particles are encapsulated and adsorbed during the generation of aluminum electrolysis products, forming magnetically tagged phosphorus-containing flocs.

[0043] The system may include a target recognition unit. The target recognition unit may include an underwater hyperspectral imaging unit 4, a multi-frequency sonar unit 5, and a magnetic induction unit 6.

[0044] The underwater hyperspectral imaging unit 4 can emit spectra in the 400-1000 nm band towards the sediment surface and receive the reflected spectra. This is compared with a pre-stored database containing data on water content (typically greater than or equal to 90%) and density (1.05-1.15 g / cm³). 3 The spectral characteristics of phosphorus-containing flocs (with reflection peaks in specific wavelength ranges such as 550-580 nm) are used to identify enrichment regions.

[0045] The multi-frequency sonar unit 5 can emit 50-500kHz sound waves and receive echoes. The acoustic impedance of phosphorus-containing flocs differs from that of ordinary sediment (mainly quartz / clay). By analyzing the echo intensity and attenuation characteristics, a spatial distribution map of the flocs is constructed.

[0046] The magnetic induction unit 6 can be an array-type magnetic sensor (Hall element), which can detect local magnetic field anomalies caused by phosphorus-containing flocs labeled with Fe3O4, determine the specific location, and achieve a positioning accuracy of ±2cm.

[0047] In the example of this application, the suction unit has a variable-diameter suction head 7 (inner diameter adjustable from 5-20 cm) and a variable-frequency vacuum pump. The suction head 7 can have multiple built-in miniature pressure sensors. The controller can automatically adjust the position, orientation, suction port diameter, and negative pressure intensity of the suction head based on the three-dimensional spatial coordinates (X, Y, Z) and floc concentration values ​​output by the identification module. For example, when the sensor detects a sudden increase in suction resistance (indicating entry into a clean sediment layer), it automatically reduces the suction or switches to an adjacent grid point. This process helps to avoid the adverse effects on the aquatic flora and fauna habitat caused by the intake of clean sediment.

[0048] In the examples of this application, the database may include spectral / sonar / magnetic signal data of pure aluminum phosphate, pure sediment, and samples with different mixing ratios. A lightweight convolutional neural network (CNN) can be used to decode the acquired real-time signals and output the operation path.

[0049] In the examples of this application, three types of sensor signals with vastly different physical properties—spectral, sonar, and magnetic induction—can be fused. In these examples, the fusion may include constructing a confidence-weighted voting mechanism. For example, if the magnetic induction signal intensity is greater than a threshold T1 and a characteristic peak is present in the hyperspectral signal, it is determined to be a "high-probability target" (90% confidence), triggering a strong suction mode; if only a magnetic induction signal or only a sonar signal is present, it is determined to be a "suspected target," triggering a weak suction detection mode. In these examples, the threshold T1 can be set according to the characteristics of the products generated in different water bodies.

[0050] In the example of this application, the database unit can be updated. For example, during each operation, all sensor data (magnetic, acoustic, optical) along with feedback data from the suction head (resistance, concentration of suctioned flocs, turbidity) are packaged to generate an "operation sample" and stored in the database unit. Then, the sample data in the database unit is labeled, and the system can automatically or semi-automatically evaluate the results of this suction operation. For example: Automatic evaluation method: An online turbidity / phosphorus concentration sensor is installed on the suction pipeline. If the phosphorus concentration in the suctioned mud-water mixture is higher than a set value (e.g., 0.1 mg / L), the sensor characteristic data corresponding to this operation is labeled as a "positive sample"; conversely, if the suctioned material is clear water or low-phosphorus sand and gravel, it is labeled as a "negative sample"; Manual verification: The suctioned samples are periodically sent to a laboratory for testing, and the test results (such as TP removal rate, organic matter content of bottom sediment) are input into the system to correct the "positive / negative samples".

[0051] In the examples of this application, updating the database unit may also include relearning and threshold optimization. For example, weekly or monthly, the controller calls upon "positive and negative samples" from the database and runs a lightweight machine learning algorithm (such as SVM (Support Vector Machine), Random Forest, or a simple Bayesian classifier). For example, the system initially sets the magnetic induction threshold to 100 nT. After learning from a large number of "positive samples," the algorithm discovers that when "magnetic induction is between 80-100 nT + sonar impedance is within a specific range," the phosphorus content drawn is also extremely high. The system will then automatically lower the trigger threshold to 80 nT, thereby "learning" to identify previously ignored targets and improving recall.

[0052] In the examples of this application, the construction process of a database cell is typically divided into three stages: "data acquisition," "feature extraction," and "database entry calibration." Sample preparation and collection: Aluminum phosphate flocs are generated in a laboratory or pilot-scale plant using specific electrocoagulation conditions (such as Al electrodes, specific current densities, and pH values). These flocs are then mixed with clean sediment from the target river or lake at different ratios to simulate different pollution layers. The samples at different mixing ratios are scanned using the proposed sensors (such as multibeam sonar, hyperspectral imager, and magnetic gradiometer) to acquire raw signal data.

[0053] Feature parameter extraction: Extracting key quantitative indicators from the raw signal to distinguish flocs from sediment: Acoustic characteristics: acoustic impedance of the floc layer, layer thickness, and backscattering intensity.

[0054] Optical characteristics: reflectance and absorption peak depth in specific wavelength bands (related to Al-OH bonds or moisture).

[0055] Magnetic characteristics: Abnormal magnetic susceptibility (due to the inclusion of Fe3O4 magnetic particles).

[0056] Data calibration and storage: Through destructive sampling (analyzing the aluminum / phosphorus content of the collected samples) or online sensor data, the extracted features are correlated with the actual floc concentration to establish a "feature-attribute" mapping table and stored in the database.

[0057] Correction of preset threshold: Because the sediment background (such as particle size and organic matter content) varies in different rivers and lakes, the database has an "adaptive correction" mechanism, rather than remaining static.

[0058] Initial Library: A prototype database built based on laboratory simulations.

[0059] On-site calibration: In the initial stage of equipment deployment, threshold deviations caused by differences in on-site water quality are corrected by drilling and sampling or suction feedback.

[0060] Dynamically updated: During actual suction, the accuracy of identification is calculated based on sensor feedback (such as turbidity and TP concentration) on the suction line, and the preset threshold in the database is automatically adjusted.

[0061] In the examples of this application, the construction of the floc spatial distribution map may include: 1. Acquire raw data through sensors, including: Grid data of magnetic field strength anomalies on the sediment plane can be obtained using a magnetometer / magnetic gradient meter; Acoustic impedance stratification data and thickness information of bottom sediment can be obtained using sonar / shallow seismic profiler; The precise coordinates of each detection point can be recorded in real time using the GPS-RTK positioning system.

[0062] 2. Algorithm for constructing spatial distribution maps To transform discrete detection points into a continuous distribution map, the following methods can typically be used: Rasterization and interpolation algorithms: The target water area is divided into grids of a certain resolution. Using Kriging interpolation or inverse distance weighting, the continuous values ​​of the entire grid plane are calculated based on the known magnetic anomaly values ​​at discrete points, generating a heat map of the planar distribution of magnetic flocs. Tomographic inversion algorithm: Using sonar echo time and intensity, and through a preset sound velocity model, the impedance difference of different depth layers of bottom sediment is inverted to construct a vertical distribution profile of flocs; Data fusion into a map: Overlaying planar heat maps with vertical profile maps, and using GIS technology to generate a three-dimensional spatial distribution model of flocs.

[0063] Furthermore, the spatial distribution map can be updated dynamically. Dynamically updating the spatial distribution map can include: Correction of "Removed" Areas: After the suction head completes its work at a certain point, the controller records the volume of flocs removed at that point. The system will automatically reset or reduce the value of that point on the distribution map to zero to avoid duplicate work.

[0064] Supplementing "missing" areas: If the sensor detects an unexpected magnetic signal during the suction process, it will add a new target point to the distribution map in real time.

[0065] Example 1 Operations were conducted in a eutrophic lake (2-3m deep, TP concentration 0.15mg / L).

[0066] step: 1) Position the dredging platform to the target grid (5m×5m) and place the electrolysis unit containing the aluminum anode 10cm below the mud-water interface.

[0067] 2) Turn on the DC power supply and set the current density to 20A / m. 2 The voltage is 30V, and phosphorus-containing flocs are generated by electrolysis. At the same time, nano-Fe3O4 suspension is added at a concentration of 10mg / L. After 10min, the Fe3O4 magnetic particles are coated with phosphorus-containing flocs to form magnetically labeled phosphorus-containing flocs.

[0068] 3) Activate the target identification unit: The underwater hyperspectral camera emits a spectrum and scans to obtain the reflection spectrum. Based on the reflection spectrum, the enrichment area of ​​phosphorus-containing flocs is identified. The multi-frequency sonar emits sound waves and receives echoes. Based on the echo intensity and attenuation characteristics, the spatial distribution map of phosphorus-containing flocs in the enrichment area is established. At the same time, the magnetic sensor array confirms the magnetic anomaly peak point and confirms the specific location of the phosphorus-containing flocs.

[0069] 4) Controller plans suction path: Based on the specific location, the suction head diameter is adjusted to 12cm and the negative pressure is -30kPa; the resistance can be monitored in real time during the suction process. When the resistance exceeds the threshold (such as -55kPa), it is determined that the bottom mud has been cleaned, and the suction at that point is automatically stopped and the device moves to the next target point.

[0070] In addition, after magnetic separation pretreatment, the collected mud-water mixture is sent to a filter press for dewatering to obtain recovered phosphorus-containing flocs. These phosphorus-containing flocs can be directly recycled (e.g., as raw materials for slow-release phosphate fertilizers).

[0071] The above steps can be repeated until the total phosphorus (TP) concentration in the area to be treated meets the requirements of the "Surface Water Environmental Quality Standard" (GB 3838-2002): for lakes and reservoirs, the limit for total phosphorus (as P) in Class III water is ≤ 0.05 mg / L; and the limit for total phosphorus (as P) in Class II water is ≤ 0.025 mg / L.

[0072] Therefore, by using the phosphorus removal method of this application, the total phosphorus (TP) removal rate in the treatment area containing phosphorus-containing substances in natural water bodies can be higher than 90%, the bottom sediment of the water body is not disturbed, and the efficiency and economic benefits of the pumping operation are improved.

[0073] While the embodiments disclosed in this application are as described above, the content is merely for the purpose of facilitating understanding of this application and is not intended to limit this application. Any person skilled in the art to which this application pertains may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in this application; however, the scope of protection of this application shall still be determined by the scope defined in the appended claims.

Claims

1. A method for targeted removal of phosphorus-containing pollutants from natural water bodies, characterized in that, The method includes the following steps: S1: Place the electrolysis unit containing the aluminum anode under the bottom sediment in the treatment area of ​​the natural water body; S2: By energizing the aluminum anode to form phosphorus-containing flocs, magnetic particles are added to encapsulate the magnetic particles in the phosphorus-containing flocs, forming magnetically labeled phosphorus-containing flocs. S3: Emit a spectrum to the bottom sediment surface of the natural water body and receive the reflectance spectrum; based on the reflectance spectrum, identify the enrichment area of ​​phosphorus-containing flocs; S4: Emit sound waves to the identified phosphorus-containing floc enrichment area and receive the echo; The spatial distribution map of phosphorus-containing flocs in the enrichment region is determined based on echo intensity and attenuation characteristics. S5: Detect magnetic field anomalies caused by magnetically tagged phosphorus-containing flocs and locate the specific locations of phosphorus-containing flocs in the enrichment region within areas with magnetic induction intensity greater than a preset threshold; and S6: Based on the specific location of the phosphorus-containing flocs, the phosphorus-containing flocs are aspirated to remove phosphorus-containing pollutants from the natural water body in a targeted manner.

2. The method according to claim 1, characterized in that, In step S2, the current density and voltage applied to the aluminum anode are 5-10 mA / cm², respectively. 2 and 15-20V; and / or, The magnetic particles are nano-Fe3O4.

3. The method according to claim 1, characterized in that, Also includes: The preset threshold is updated periodically in the following manner: After each aspiration of the aluminum phosphate flocs, the sample data for that time is extracted and stored in the database unit; the sample data includes spectral signal parameters, sonar echo signal parameters, and magnetic induction signal parameters, as well as the corresponding aspiration resistance parameters, aspirated floc concentration parameters, and aspirated floc turbidity parameters. Labeling the sample data in the database unit includes: for each group of sample data, at least based on whether the floc concentration parameter is greater than the concentration threshold, labeling the group of sample data as a positive sample or a negative sample; Machine learning is performed periodically using the labeled sample data in the database unit to update the preset threshold.

4. The method according to claim 3, characterized in that, The sonar echo signal parameters include sediment impedance and layer thickness; the spectral signal parameters are absorption peaks or reflection peaks at specific wavelengths corresponding to the characteristic spectra of Al-OH or Fe3O4; and the magnetic induction signal parameters are magnetic susceptibility anomalies.

5. The method according to claim 3, characterized in that, In step S3, the emission spectrum includes the emission spectrum in the 400-1000 nm wavelength range; and / or, Moisture content greater than or equal to 90% and density 1.05-1.15 g / cm³ 3 The spectral characteristics of phosphorus-containing flocs include a reflection peak at 550-580 nm, and / or... In step S4, emitting sound waves includes emitting sound waves of 50-500 kHz.

6. The method according to claim 1, characterized in that, The spatial distribution map of the phosphorus-containing flocs is constructed through the following steps: Obtain the raw data, including: Obtain grid data of magnetic field intensity anomalies on the sediment plane; obtain acoustic impedance layering data and thickness information of the sediment; obtain the coordinates of each detection point; Constructing a spatial distribution map includes: The target water area is divided into multiple grids. Based on the known magnetic anomaly values ​​of discrete points, the continuous values ​​of the entire grid plane are calculated to generate a thermal map of the planar distribution of magnetic flocs. By utilizing the time and intensity of sonar echoes and through a pre-defined sound velocity model, the impedance differences at different depths of the sediment are inverted to construct a vertical distribution profile of flocs. By overlaying the planar thermal map with the vertical cross-sectional map, a three-dimensional spatial distribution model of flocs is generated.

7. The method according to claim 6, characterized in that, It also includes dynamically updating the spatial distribution map; the dynamic update includes: after completing the suction at a certain location, recording the volume of aluminum phosphate flocs suctioned at that location, and then setting the value of that location on the spatial distribution map to zero or reducing it to avoid repeated work; during the suction process, if a new magnetic anomaly signal is found, a new target location is added to the spatial distribution map in real time.

8. The method according to claim 1, characterized in that, The method also includes obtaining phosphorus-containing flocs by magnetically separating and dehydrating the mud-water mixture.

9. A system for targeted removal of phosphorus-containing pollutants from natural water bodies, characterized in that, The system includes an electrolysis unit, a target recognition unit, and a suction unit; The electrolysis unit includes a magnetic particle inlet, and the electrolysis unit is configured to generate phosphorus-containing flocs and form magnetically tagged phosphorus-containing flocs with the added magnetic particles. The target identification unit includes an underwater hyperspectral imaging unit, a multi-frequency sonar unit, and a magnetic induction unit. The underwater hyperspectral imaging unit is configured to emit a spectrum to the bottom sediment surface of the natural water body and receive a reflected spectrum; based on the reflected spectrum, it identifies the enrichment area of ​​phosphorus-containing flocs; the multi-frequency sonar unit is configured to emit sound waves to the identified enrichment area of ​​phosphorus-containing flocs and receive echoes. The spatial distribution map of phosphorus-containing flocs in the enrichment area is established based on echo intensity and attenuation characteristics; the magnetic induction unit is configured to detect magnetic field anomalies caused by magnetically tagged phosphorus-containing flocs and locate the specific location of phosphorus-containing flocs in the enrichment area in the region with magnetic induction intensity greater than a preset threshold. The suction unit includes a suction head with a variable diameter, and the suction unit is configured to perform directional suction based on the specific location of the determined phosphorus-containing flocs.

10. The system according to claim 9, characterized in that, It also includes controllers and database units; The database unit is configured to store sample data extracted after each aspiration of the phosphorus-containing flocs, and to store the sample data as labels, including positive and negative samples; the sample data includes sonar echo signal parameters, spectral signal parameters, magnetic induction signal parameters, as well as corresponding aspiration resistance parameters, aspiration floc concentration parameters, and aspiration floc turbidity. The controller is configured to periodically perform machine learning using labeled sample data in the database to update the preset threshold.

Images