Subsea dosing system

Abstract

This invention discloses an underwater chemical dosing system, comprising a coagulation reaction tank, a baffle plate, an underwater image acquisition module, a metering and dosing unit, a housing, and an edge computing unit. The metering and dosing unit is located on one side of the coagulation reaction tank and has a chemical outlet connected to an inlet for adding chemicals into the coagulation reaction tank. The housing is located on one side of the coagulation reaction tank, and the edge computing unit is located within the housing. The edge computing unit is electrically connected between the metering and dosing unit and the underwater image acquisition module, enabling the edge computing unit to receive image data transmitted from the underwater image acquisition module. The edge computing unit calculates the dosage based on the image data and can transmit the dosage data to the metering and dosing unit. The metering and dosing unit can control the amount of chemicals added to the coagulation reaction tank based on the dosage data. This embodiment of the underwater chemical dosing system upgrades to feedback control based on real-time image data, improving dosing accuracy.

Description

Technical Field

[0001] This invention relates to the field of water treatment chemical dosing technology, and more specifically to an underwater chemical dosing system. Background Technology

[0002] In industrial production systems such as thermal power plants, nuclear power plants, and chemical plants, water treatment is a core component ensuring safe and stable operation. With the intelligent transformation of the power industry and the upgrading of environmental protection requirements, industrial water treatment is evolving from traditional manual monitoring and experience-based dosing to automation, precision, and data-driven approaches. Edge computing technology, due to its low latency and local data processing advantages, is widely used in industrial control. Meanwhile, continuous improvements in industrial-grade waterproof imaging technology provide a data foundation for real-time feedback on the effects of chemical reactions.

[0003] The relevant dosing technologies mainly include the following three categories: Automatic dosing systems based on online sensors: These systems control the dosage through data from sensors such as turbidity and pH, but the sensors are easily contaminated and cannot directly reflect the state of the flocs, leading to a delay in dosing; Semi-automatic dosing devices based on manual inspection: These devices rely on operators to observe the morphology of the flocs and adjust the dosage accordingly, resulting in strong subjectivity, large errors, and difficulty in adapting to fluctuations in water quality; Cloud-based intelligent dosing systems: These systems collect images through cameras and upload them to the cloud for analysis, but they are limited by network bandwidth, have high latency, rely on the cloud, and cannot operate when the network is down.

[0004] However, in related technologies, the system relies on indirect data or manual observation, which cannot capture core indicators such as floc particle size, morphology, and density. Manual inspection and calibration are required, which increases operation and maintenance costs and makes it impossible to achieve 24-hour uninterrupted and precise control. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the technical problems in the related art.

[0006] Therefore, embodiments of the present invention propose an underwater dosing system that upgrades the traditional dosing method, which relies on indirect data from sensors or human experience, to feedback control based on real-time image data, thereby improving the accuracy of dosing.

[0007] The underwater chemical dosing system of this invention includes a coagulation reaction tank, a baffle plate, an underwater image acquisition module, a metering and dosing unit, a housing, an edge computing unit, a communication module, and a cloud platform. The coagulation reaction tank has an inlet and an outlet. The baffle plate is disposed within the coagulation reaction tank. The underwater image acquisition module is disposed on the baffle plate for capturing images of suspended solids in the water and transmitting the captured image data externally. The metering and dosing unit is disposed on one side of the coagulation reaction tank and has a dosing outlet connected to the inlet for dosing chemicals into the coagulation reaction tank. The housing is disposed on one side of the coagulation reaction tank, and the edge computing unit is disposed within the housing. The edge computing unit is electrically connected between the metering and dosing unit and the underwater image acquisition module, so that the edge computing unit receives the image data transmitted outward by the underwater image acquisition module. The edge computing unit calculates the dosage based on the image data and can transmit the dosage data to the metering and dosing unit. The metering and dosing unit can control the dosage of chemicals injected into the coagulation reaction tank based on the dosage data. The communication module is located in the housing. The cloud platform and the metering and dosing unit realize bidirectional data transmission through the communication module. The metering and dosing unit can upload image data and dosage data to the cloud platform. The cloud platform can at least send control commands to the metering and dosing unit.

[0008] The underwater dosing system of this invention fixes an underwater image acquisition module to the guide plate of the coagulation reaction tank. The underwater image acquisition module captures the morphology of flocs in the water in real time and transmits the image data to the edge computing unit. The edge computing unit analyzes the core parameters of floc particle size, morphology, and density to calculate the dosage. The edge computing unit transmits control signals to the metering and dosing unit to implement the dosing. At the same time, the communication module enables bidirectional data transmission with the cloud platform, forming a complete closed-loop control system. This upgrades the traditional dosing method, which relies on indirect data from sensors or human experience, to feedback control based on real-time image data, improving dosing accuracy, reducing reagent waste, and minimizing human intervention.

[0009] In some embodiments of the present invention, the edge computing unit includes an edge controller, which is disposed in the housing. The edge controller has a built-in AI flocculent identification algorithm, which can identify the flocculent particle size through the YOLOv8 model.

[0010] In some embodiments, the metering and dosing unit of the present invention includes a reagent storage tank, a pipeline, a metering pump, and an electromagnetic flow valve; the pipeline is connected between the reagent storage tank and the water inlet; the metering pump is located in the pipeline to control the delivery of the reagent, and the electromagnetic flow valve is located in the pipeline to monitor the dosage of the reagent.

[0011] In some embodiments of the present invention, the metering and dosing unit further includes a liquid level sensor, which is disposed in the drug storage tank to monitor the liquid level of the drug.

[0012] In some embodiments of the present invention, the pharmaceutical storage tank is made of PE or carbon steel lined with rubber.

[0013] In some embodiments of the present invention, the pipeline is made of UPVC material.

[0014] In some embodiments of the present invention, the metering and dosing unit further includes a stirring device, which is disposed in the drug storage tank for stirring the drug.

[0015] In some embodiments, the underwater image acquisition module of the present invention includes a bracket and a camera. The camera is mounted on the guide plate via the bracket, and the lens surface of the camera is coated with an anti-fog and water-repellent coating.

[0016] In some embodiments of the present invention, the bracket is made of stainless steel.

[0017] In some embodiments of the present invention, the communication module includes an Ethernet module and a 4G / 5G dual-mode communication module. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall structure of the underwater chemical dosing system according to an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of the image preprocessing and AI floc recognition process according to an embodiment of the present invention.

[0020] Figure label:

[0021] 100. Underwater chemical dosing system; 1. Coagulation reaction tank; 101. Inlet; 102. Outlet; 2. Baffle plate; 3. Underwater image acquisition module; 301. Camera; 4. Metering and dosing unit; 401. Chemical outlet; 402. Chemical storage tank; 403. Piping; 404. Metering pump; 405. Liquid level sensor; 406. Stirring device; 5. Tank; 6. Edge computing unit; 7. Communication module. Detailed Implementation

[0022] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0023] Reference Figure 1 and Figure 2 The underwater chemical dosing system 100 of this invention includes a coagulation reaction tank 1, a guide plate 2, an underwater image acquisition module 3, a metering and dosing unit 4, a housing 5, an edge computing unit 6, and a communication module 7. The coagulation reaction tank 1 has an inlet 101 and an outlet 102. The guide plate 2 is disposed on the coagulation reaction tank 1. The underwater image acquisition module 3 is disposed on the guide plate 2 for capturing images of suspended solids in the water and transmitting the captured image data externally. The metering and dosing unit 4 is disposed on one side of the coagulation reaction tank 1 and has a chemical outlet 401, which is connected to the inlet 101 for dosing chemicals into the coagulation reaction tank 1. The housing 5 is located on one side of the coagulation reaction tank 1. An edge computing unit 6 is located within the housing 5 and is electrically connected between the metering and dosing unit 4 and the underwater image acquisition module 3. This allows the edge computing unit 6 to receive image data transmitted from the underwater image acquisition module 3, calculate the dosage based on the image data, and transmit this dosage data to the metering and dosing unit 4. The metering and dosing unit 4 can then control the dosage of chemicals injected into the coagulation reaction tank 1 based on this dosage data. A communication module 7 is located within the housing 5 and enables bidirectional data transmission with the metering and dosing unit 4. The metering and dosing unit 4 can transmit image data and dosage data, and can also send control commands to the metering and dosing unit 4.

[0024] The underwater dosing system 100 of this invention fixes the underwater image acquisition module 3 to the guide plate 2 of the coagulation reaction tank 1. The underwater image acquisition module 3 captures the morphology of flocs in the water in real time. The underwater image acquisition module 3 can transmit the image data to the edge computing unit 6. The edge computing unit 6 analyzes the core parameters of floc particle size, morphology, and density to calculate the dosage. The edge computing unit 6 transmits the control signal to the metering and dosing unit 4 to realize the dosing. At the same time, the communication module 7 realizes bidirectional data transmission, forming a complete closed-loop control system. This upgrades the traditional dosing method, which relies on indirect data from sensors or human experience, to feedback control based on real-time image data, improving dosing accuracy, reducing chemical waste, and minimizing human intervention.

[0025] In some embodiments, such as Figure 1 As shown, the edge computing unit 6 of this embodiment includes an edge controller, which is located in the housing 5. The edge controller has a built-in AI floc recognition algorithm, which can identify the floc particle size through the YOLOv8 model.

[0026] The edge computing unit 6, through the edge controller built into the housing 5, is equipped with an AI floc recognition algorithm based on the YOLOv8 model. This enables the identification and analysis of floc size in real-time image data transmitted by the underwater image acquisition module 3. The algorithm can complete image preprocessing, feature extraction and parameter calculation, upgrading the traditional judgment method that relies on manual observation or indirect sensor data to direct quantitative analysis based on computer vision.

[0027] In some embodiments, such as Figure 1 As shown, the metering and dosing unit 4 of this embodiment includes a reagent storage tank 402, a pipeline 403, a metering pump 404, and an electromagnetic flow valve. The pipeline 403 connects the reagent storage tank 402 and the water inlet 101. The metering pump 404 is located in the pipeline 403 to control the delivery of the reagent, and the electromagnetic flow valve is located in the pipeline 403 to monitor the dosage of the reagent.

[0028] In this embodiment of the invention, the metering and dosing unit 4 achieves full-process control of the reagent from storage to precise dosing through the coordinated operation of the reagent storage tank 402, pipeline 403, metering pump 404, and electromagnetic flow valve. The reagent storage tank 402 is responsible for storing the reagent, and the pipeline 403 connects the storage tank to the inlet 101 of the coagulation reaction tank 1 to ensure unobstructed reagent delivery. The metering pump 404 adjusts the reagent delivery rate according to the 4-20mA control signal sent by the edge computing unit 6, while the electromagnetic flow valve monitors the actual dosage in real time and feeds the data back to the edge computing unit 6 to form a closed-loop calibration. This design not only ensures the accuracy of reagent dosing (error less than 1%), but also achieves dynamic adjustment through the frequency conversion control of the metering pump 404 and the real-time monitoring of the electromagnetic flow valve, enabling the system to respond quickly and optimize the dosage according to changes in the floc state.

[0029] In some embodiments, such as Figure 1 As shown, the metering and dosing unit 4 in this embodiment of the invention further includes a liquid level sensor 405, which is disposed in the drug storage tank 402 to monitor the liquid level of the drug. The drug storage tank 402 realizes real-time monitoring of the remaining amount of drug through the liquid level sensor 405.

[0030] In some embodiments, such as Figure 1As shown, the reagent storage tank 402 in this embodiment of the invention is made of PE or carbon steel lined with rubber. The reagent storage tank 402 in this embodiment of the invention uses PE or carbon steel lined with rubber. By selecting materials with excellent corrosion resistance and chemical stability, it is ensured that the storage tank will not experience material degradation or contamination of the reagents in environments where it is in long-term contact with coagulants, flocculants, or corrosion inhibitors. This ensures the original activity and treatment effect of the reagents. This material selection not only effectively prevents chemical reactions and contamination caused by direct contact between the reagents and the metal tank, but also provides sufficient mechanical strength through the carbon steel lined structure to withstand pressure changes in industrial environments.

[0031] In some embodiments, such as Figure 1 As shown, the pipe 403 in this embodiment of the invention is made of UPVC material. PVC material not only has excellent resistance to chemical corrosion and can withstand long-term erosion by common acid and alkali agents used in industrial water treatment, but also has a smooth inner wall, effectively reducing the risk of agent adhesion and scaling.

[0032] In some embodiments, such as Figure 1 As shown, the metering and dosing unit 4 in this embodiment of the invention also includes a stirring device 406, which is disposed in the drug storage tank 402 for stirring the drug. The stirring device 406 continuously or periodically stirs the stored drug, effectively preventing precipitation, stratification, or clumping of the drug due to gravity during storage, and ensuring the uniformity and consistency of the drug concentration.

[0033] In some embodiments, such as Figure 1 As shown, the underwater image acquisition module 3 of this embodiment includes a bracket and a camera 301. The camera 301 is mounted on the guide plate 2 via the bracket, and the lens surface of the camera 301 is coated with an anti-fog and hydrophobic coating. The anti-fog and hydrophobic coating on the lens surface can prevent water vapor condensation and dirt adhesion. This helps to ensure that clear and interference-free floc images can be continuously captured even in complex environments such as chemical corrosion and water flow disturbance, solving the problems of easy contamination and blurry lenses in traditional underwater cameras 301.

[0034] In some embodiments, such as Figure 1 As shown, the bracket in this embodiment of the invention is made of stainless steel. Stainless steel possesses excellent corrosion resistance and high strength, ensuring that the underwater image acquisition module 3 can be stably fixed in the harsh environment of the coagulation reaction tank 1, which is subject to chemical corrosion and water flow disturbance. Furthermore, the stainless steel material resists the erosion of various chemical agents, preventing the bracket from rusting or experiencing a decrease in structural strength.

[0035] In some embodiments, such as Figure 1As shown, the communication module 7 of this embodiment includes an Ethernet module and a 4G / 5G dual-mode communication module 7. By integrating the Ethernet module and the 4G / 5G dual-mode communication module 7, the communication module 7 of this embodiment enables flexible and reliable data transmission between the edge computing unit 6 and the device. The Ethernet module provides a high-bandwidth, low-latency data channel for wired connections, suitable for real-time data transmission in a stable network environment within the factory. The 4G / 5G dual-mode communication module 7 serves as a wireless backup channel, ensuring no data loss in scenarios of network interruption or lack of cabling. This dual-mode design not only guarantees efficient system operation under normal working conditions but also enables seamless switching in situations of network instability or outages.

[0036] Underwater image acquisition module 3: Core components: Industrial-grade waterproof HD camera 301 (IP68 protection rating, 316L stainless steel housing), 1920x1080 resolution, 25-30fps frame rate, 8mm lens focal length; Auxiliary components: Built-in anti-glare LED fill light (brightness 500-1500 lm), lens surface coated with anti-fog and water-repellent coating, equipped with a micro scraper cleaning device (cleaning frequency can be set to once / 60 minutes or triggered as needed); Installation method: Fixed to the middle of the guide plate 2 inside the coagulation reaction tank 1 by a stainless steel bracket, the collection area covers the floc formation area in the middle of the reaction tank (1 / 2 tank length from the inlet end), the axis of camera 301 is at a 45° angle to the water flow direction, and the shooting distance is 0.5-1.5m.

[0037] Edge Computing Unit 6: Hardware: Industrial-grade edge controller, supports wide temperature operation, IP54 protection rating; Software Algorithm: Built-in AI floc recognition algorithm and dosage calculation model.

[0038] like Figure 2 As shown, image preprocessing: image noise caused by water flow disturbance and chemical residue is eliminated through grayscale enhancement and median filtering, and correction is performed based on lens distortion parameters; AI floc identification: a lightweight YOLOv8 model is used to identify floc particle size (measurement range 50-500um, error ≤3%), morphology (filamentous / spherical / blocky), and density (calculated based on grayscale value distribution, range 0-100%) in real time; dosage calculation: based on a preset "floc state-dosage" mapping model (trained through historical data of water treatment in the power industry), combined with raw water flow and turbidity sensor auxiliary data, the optimal dosage is calculated and a 4-20mA control signal is output.

[0039] Metering and dosing unit 4: Chemical storage tank 402: made of PE or carbon steel with rubber lining, equipped with a liquid level sensor 405 (range 0-100%, accuracy ±1%) and a stirring device 406 (speed 85r / min, to prevent chemical sedimentation); Variable frequency metering pump 404: dosing accuracy ±1%, supports 4-20mA signal control; Piping system 403: UPVC piping 403, equipped with check valve, filter and electromagnetic flow valve, to monitor the actual amount of chemical added in real time.

[0040] Communication Module 7: Hardware: Industrial Ethernet module and 4G / 5G dual-mode communication module 7, equipped with antenna to enhance signal; Function: Enables bidirectional data transmission between edge computing unit 6 and cloud management platform, uploads real-time images, floc parameters, and drug dosing data, receives parameter configurations (such as floc compliance threshold and drug type parameters) and control commands from the cloud, and automatically switches to local independent operation mode when the network is disconnected.

[0041] Coagulation reactor 1 adaptable structure: Mounting bracket: stainless steel material, adjustable height to ensure that the underwater image acquisition module 3 is always aligned with the core area of ​​floc formation; Chemical dosing port: located at about 3 meters from the water inlet of coagulation reactor 1, using a static pipeline mixer to ensure that the chemical and raw water are mixed quickly.

[0042] The beneficial effects are mainly reflected in the following aspects: 1. Significantly improved dosing accuracy: Dosing decisions are made directly based on core indicators of floc images, with a dosing error of ≤1%. Compared with traditional sensor-based dosing systems, the utilization rate of chemicals is increased by more than 25%, reducing the cost of water treatment chemicals in the power industry. 2. Fast response speed: Edge computing local operation can quickly respond to sudden changes in raw water turbidity and flow rate, ensuring stable effluent water quality; Strong environmental adaptability: Industrial-grade waterproof and corrosion-resistant design, hydrophobic coating on lens + automatic cleaning device, can operate continuously and stably in the environment of chemical corrosion and water flow disturbance in coagulation reaction tank 1, with a low failure rate. 3. High degree of automation: The entire process is under closed-loop control, eliminating the need for manual inspection and adjustment, adapting to the unmanned management requirements of smart power plants, and reducing maintenance workload by more than 75%; 4. Strong data traceability: Complete images, floc parameters, and dosing data are stored in the cloud, supporting historical data traceability and trend analysis, providing data support for water treatment process optimization; 5. Good compatibility: Supports commonly used industrial protocols such as Modbus, and can be seamlessly integrated into existing DCS control systems in the power industry, with low modification difficulty and controllable cost.

[0043] The workflow is as follows: Deployment Phase: The underwater image acquisition module 3 is fixed on the guide plate 2 of the coagulation reaction tank 1, and the shooting angle is adjusted to cover the floc formation area; the metering and dosing unit 4 is installed in the equipment room next to the reaction tank, and connected to the reagent storage tank 402 and the dosing pipeline 403; the edge computing unit 6 and the communication module 7 are deployed in the control cabinet, and the parameters are configured; Data Acquisition: The underwater image acquisition module 3 captures real-time images of the flocs and uploads them synchronously to the edge computing unit 6, and the raw water flow and turbidity sensor (optional) data are synchronously accessed; Local computation: Edge computing unit 6 extracts floc particle size, morphology, and density parameters through AI floc identification algorithm. If the particle size is less than the threshold or the density is lower than the threshold, the incremental dosage is calculated through the dosage calculation model; if the particle size is greater than the threshold or the density is higher than the threshold, the reduced dosage is calculated. Precise dosing: The edge computing unit sends a 4-20mA control signal to the variable frequency metering pump 404 to adjust the pump motor speed, realize dynamic adjustment of the dosage, and the electromagnetic flow valve provides real-time feedback on the actual dosage to form a closed-loop calibration. Data transmission: Edge computing unit 6 uploads image screenshots, floc parameters and drug dosing data to the cloud every 10 seconds. The cloud can monitor in real time and issue parameter modification instructions. Anomaly Handling: When the liquid level sensor 405 detects that the remaining amount of the reagent is less than 10%, or when the underwater image acquisition module 3 fails to identify flocs for 3 consecutive seconds, the edge computing unit 6 issues an audible and visual alarm and pushes the alarm information to the management personnel through the communication module 7.

[0044] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0045] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0046] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0047] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0048] In this invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0049] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. An underwater chemical dosing system, characterized in that, include: A coagulation reaction tank (1) has an inlet (101) and an outlet (102). A guide plate (2) is provided in the coagulation reaction tank (1). The underwater image acquisition module (3) and the metering and dosing unit (4) are provided. The underwater image acquisition module (3) is located on the guide plate (2) for taking pictures of suspended matter in the water and transmitting the captured image data to the outside. The metering and dosing unit (4) is located on one side of the coagulation reaction tank (1). The metering and dosing unit (4) has a drug outlet (401). The drug outlet (401) is connected to the water inlet (101) for adding drugs into the coagulation reaction tank (1). The container (5) and the edge computing unit (6) are located on one side of the coagulation reaction tank (1). The edge computing unit (6) is located on the container (5). The edge computing unit (6) is electrically connected between the metering and dosing unit (4) and the underwater image acquisition module (3) so that the edge computing unit (6) receives the image data transmitted outward by the underwater image acquisition module (3). The edge computing unit (6) calculates the dosage based on the image data and can transmit the dosage data to the metering and dosing unit (4). The metering and dosing unit (4) can control the dosage of the chemical injected into the coagulation reaction tank (1) based on the dosage data. The communication module (7) and the cloud platform are located in the housing (5). The cloud platform and the metering and dosing unit (4) can achieve bidirectional data transmission through the communication module (7). The metering and dosing unit (4) can upload image data and dosage data to the cloud platform. The cloud platform can send control commands to the metering and dosing unit (4).

2. The underwater chemical dosing system according to claim 1, characterized in that, The edge computing unit (6) includes an edge controller, which is located in the housing (5). The edge controller has a built-in AI floc recognition algorithm, which can identify the floc particle size through the YOLOv8 model.

3. The underwater chemical dosing system according to claim 1, characterized in that, The metering and dosing unit (4) includes: Pharmaceutical storage tank (402); Pipeline (403), the pipeline (403) is connected between the medicine storage tank (402) and the water inlet (101); A metering pump (404) and an electromagnetic flow valve are provided, wherein the metering pump (404) is located in the pipeline (403) to control the delivery of the agent, and the electromagnetic flow valve is located in the pipeline (403) to monitor the dosage of the agent.

4. The underwater dosing system according to claim 3, characterized in that, The metering and dosing unit (4) also includes a liquid level sensor (405), which is located in the drug storage tank (402) to monitor the liquid level of the drug.

5. The underwater chemical dosing system according to claim 3, characterized in that, The medicine storage tank (402) is made of PE or carbon steel lined with rubber.

6. The underwater chemical dosing system according to claim 3, characterized in that, The pipeline (403) is made of UPVC.

7. The underwater dosing system according to claim 3, characterized in that, The metering and dosing unit (4) also includes a stirring device (406), which is located in the drug storage tank (402) for stirring the drug.

8. The underwater chemical dosing system according to claim 1, characterized in that, The underwater image acquisition module (3) includes a bracket and a camera (301). The camera (301) is mounted on the guide plate (2) via the bracket. The lens surface of the camera (301) is coated with an anti-fog and water-repellent coating.

9. The underwater chemical dosing system according to claim 8, characterized in that, The bracket is made of stainless steel.

10. The underwater chemical dosing system according to any one of claims 1-9, characterized in that, The communication module (7) includes an Ethernet module and a 4G / 5G dual-mode communication module (7).

Images