An ecological floating bed intelligent control system for water body eutrophication treatment and ecological restoration

Abstract

This invention discloses an intelligent control system for ecological floating beds used in the treatment and ecological restoration of eutrophication in water bodies, belonging to the field of aquatic ecological restoration technology. It includes a digital twin model, an ecological purification module, an electrodynamic enhancement module, a microbial remediation module, an intelligent monitoring module, a power control module, and an ecological environment assessment module. The intelligent control system constructs a comprehensive system from bottom-level data acquisition to top-level intelligent application technology. Based on an IoT-based multi-dimensional monitoring network and intelligent algorithms, it achieves a digital twin closed loop of "sensing-transmission-intelligent computing-prediction-decision-application-management" and intelligent control interaction. This invention integrates remote visualization, multiple purification processes, multi-dimensional monitoring, intelligent decision-making, and ecological management. Through the synergistic enhancement of multiple purification effects—plant absorption, electrochemical enrichment, and microbial degradation—it shifts from "end-of-pipe treatment" to proactive prevention and control through "early warning and intelligent control," dynamically adjusting purification strategies to efficiently and persistently remove pollutants and reduce the occurrence of eutrophication disasters.

Description

Technical Field

[0001] This invention belongs to the field of water ecological restoration technology, and in particular relates to an intelligent control system for ecological floating beds used for eutrophication treatment and ecological restoration of water bodies. Background Technology

[0002] Rivers and lakes are important urban hydrological ecosystems, serving functions such as flood control and drainage, climate regulation, ecological balance, and landscape recreation. With the acceleration of urbanization and population growth, a large amount of pollutants are discharged into surface water bodies, leading to eutrophication, water quality deterioration, and reduced biodiversity in rivers and lakes. This seriously affects residents' lives and urban development, making the ecological restoration of rivers and lakes an urgent need.

[0003] Ecological floating beds, with aquatic plants as the core, rely on the synergistic effect of plant roots and microorganisms to adsorb and degrade pollutants, and have advantages such as low cost, simple operation and maintenance, and eco-friendliness. However, traditional ecological floating beds have obvious defects: (1) Fixed installation, limited purification range, relying on only a single plant for purification, low removal efficiency of nitrogen, phosphorus and heavy metals, and easy rebound of treatment effect; (2) Unable to adapt to water level changes, the contact area of ​​plant roots with water body is unstable during dry and wet seasons, and the purification efficiency decreases; (3) Lack of real-time monitoring, water quality prediction and pollution early warning capabilities, unable to cope with sudden pollution, and difficult to dynamically adjust purification strategies; (4) No underwater ecological monitoring and remote visualization control, and insufficient intelligent decision-making and ecological management capabilities. Summary of the Invention

[0004] Technical Problem Solved: Addressing the technical problems existing in the background technology, this invention provides an intelligent control system for ecological floating beds that integrates digital twins, the Internet of Things, artificial intelligence, and multiple purification technologies. By constructing an integrated intelligent control system encompassing remote visualization, multiple purification, multi-dimensional monitoring, intelligent decision-making, and ecological management, it achieves a digital twin closed loop of "perception-transmission-intelligent computing-prediction-decision-application-management." Through synergistic enhancement of plant absorption, electrochemical enrichment, and microbial degradation, it efficiently removes pollutants such as COD, ammonia nitrogen, total phosphorus, and heavy metals, achieving proactive early warning, intelligent regulation, and efficient and sustained purification. This solves the problems of limited purification capacity, lack of intelligent regulation, and inability to dynamically adapt to changes in water quality and level in existing ecological floating beds, hindering the treatment of eutrophication in water bodies.

[0005] Technical solution: The present invention provides an intelligent control system for ecological floating beds used for eutrophication control and ecological restoration of water bodies, comprising a visualization layer, a physical layer, a sensing layer, a transmission layer, a decision-making layer, an application layer, and a management layer; The visualization layer is a digital twin model. A three-dimensional terrain digital model is constructed based on a terrain database and an API interface is set up. Information is exchanged in real time through the API interface to obtain a dynamic digital model. On the basis of the digital model, a hydrodynamic model, a water quality model, and a water ecology model are built to form a dynamic digital twin model for water environment visualization. The physical layer includes an ecological purification module, an electrodynamic enhancement module, and a microbial remediation module. The ecological purification module consists of a walking mechanism and an ecological floating bed, substrate, and aquatic plants located on the walking mechanism. The walking mechanism includes a track, a floating bed frame, walking wheels, a drive motor, and a hydraulic lifting mechanism, which drives the ecological floating bed to move in the water and adjust its height. The electrodynamic enhancement module includes anode and cathode pairs, wires, and a low-voltage DC power supply. The anode and cathode pairs are arranged in the ecological floating bed substrate and drive pollutants to migrate and accumulate under the action of a DC electric field. The microbial remediation module includes microbial biofilm packing material, a bacterial agent dosing device, and a micro-nano aeration device. The microbial biofilm packing material is located at the bottom of the ecological floating bed. The sensing layer is an intelligent monitoring module, which includes a multi-parameter water quality sensor, a hyperspectral multi-parameter water quality monitor, a water level sensor, and an underwater camera to monitor and predict water quality changes and assess the eutrophication status of the water body in real time. The transmission layer transmits the monitoring data from the LoRaWAN low-power wide area network to the control center workstation. The data processing module preprocesses the collected water level and water quality data and then inputs it into the digital twin model for real-time updates. The decision-making layer is an intelligent control module that performs real-time analysis and processing of pre-processed water quality data based on the LSTM neural network algorithm to predict water quality changes; the intelligent control module also assesses the eutrophication status of the water body based on an integrated neural network algorithm and issues early warning signals. The application layer is a power control module, including a solar power supply device, a walking mechanism, a microbial agent dosing device, a micro-nano aeration device, and an intelligent control module. The intelligent control module performs real-time analysis and processing of water quality data based on a deep reinforcement learning algorithm. The processed water level and water quality data are exchanged between the solar power supply device, the walking mechanism, the microbial agent dosing device, and the micro-nano aeration device. The intelligent algorithm is activated to perform distributed collaborative optimization to generate intelligent control commands and distribute them to each power execution module. The management layer is the ecological environment assessment module, which uses a multi-factor weighting method to evaluate water quality indicators and aquatic ecological indicators, thereby realizing intelligent assessment and management of aquatic ecological environment health. The digital twin model, ecological purification module, electrodynamic enhancement module, microbial remediation module, intelligent monitoring module, power control module, and ecological environment assessment module are integrated into one unit. They interact and coordinate through the control center workstation to form an intelligent management and control platform for the ecological floating bed purification system.

[0006] Preferably, the digital twin model, ecological purification module, electrodynamic enhancement module, microbial remediation module, intelligent monitoring module, power control module, and ecological environment assessment module can all be upgraded by independently replacing their physical entities.

[0007] Preferably, the control system further includes a support layer, which includes hardware, software, network communication, and a data center, providing fundamental support for the operation of the entire system.

[0008] Preferably, the steps for constructing the digital twin model include: Step 1: Collect river and lake cross-section design data and establish a topographic information database; Step 2: Construct a 3D terrain digital model based on the terrain database and set up an API interface. Use the API interface to exchange information in real time and obtain a dynamic digital model. Step 3: Construct hydrodynamic model, water quality model, and aquatic ecosystem model, and interface them with the 3D digital model to generate a dynamic digital twin model.

[0009] Preferably, the hydrodynamic model uses the MIKE 21 hydrodynamic module to simulate and predict changes in water level and flow velocity; the water quality model uses the MIKE 21 transport module to simulate and predict the transport paths, diffusion attenuation processes, and concentration changes of pollutants in the water body; and the aquatic ecosystem model uses the MIKE EcoLab ecological module to simulate and predict the growth and reproduction of organisms in the aquatic ecosystem.

[0010] Preferably, the walking mechanism drives the ecological purification module, the electrodynamic enhancement module, and the microbial remediation module to move synchronously, and adaptively adjusts the height of the ecological purification module to maintain the optimal contact state between the plant roots and the water body; the ecological purification module constructs a plant-microbe system to provide an attachment substrate for microorganisms, and the electrodynamic enhancement module enhances dissolved oxygen and pollutant migration.

[0011] Preferably, the multi-parameter water quality sensor detects pH, conductivity, turbidity, dissolved oxygen, ammonia nitrogen, total phosphorus, COD, BOD, chlorophyll a, and heavy metals; the water level sensor detects water level and flow velocity; the hyperspectral multi-parameter water quality monitor detects the concentration of dissolved substances in the water and algae growth; the underwater camera monitors aquatic biological activity and algae growth in real time; the monitoring data is transmitted to the control center workstation via LoRaWAN low-power wide-area network, and then filtered and preprocessed by the data processing module.

[0012] Preferably, the intelligent control module is based on LSTM neural network, ensemble neural network and deep reinforcement learning algorithm; the LSTM neural network performs real-time analysis and processing of pre-processed water quality data to predict water quality changes; the ensemble neural network assesses the eutrophication state of the water body and issues early warning signals; the deep reinforcement learning algorithm outputs optimized control commands, and the intelligent control equipment operates.

[0013] Preferably, the deep reinforcement learning algorithm constructs an objective function to solve a combinatorial optimization problem, the objective function including water quality compliance and energy conservation; the water purification efficiency is calculated based on water quality parameters, plant growth status, and water level change information, and the mathematical model is as follows: ; In the formula, E t Let T be the water purification efficiency at time t, with a value ranging from 0 to 1; t For runtime; P t Pollutant concentration (mg / L); P max Maximum treatable pollutant concentration (mg / L); D t The water body disturbance degree, with a value ranging from 0 to 1; C t Plant coverage area / (m 2 );C ref For reference, the vegetation cover area is / (m²) 2 ); H t Water level / m; H ref The reference water level is given in m; k1 is the basic purification efficiency coefficient, ranging from 0.8 to 0.95; α is the time decay coefficient, ranging from 0.01 to 0.05; β is the disturbance influence coefficient, ranging from 0.1 to 0.3; γ is the light cycle influence coefficient, ranging from 0.1 to 0.2; and λ is the water level influence coefficient, ranging from 0.2 to 0.4.

[0014] Preferably, the water quality index factors include water area, flow velocity, water depth, hydraulic loading, and eutrophication level; the aquatic ecological index factors include water area, flow velocity, water depth, and aquatic plant abundance; the weights of the index factors are determined using the analytic hierarchy process or expert evaluation. The water quality evaluation formula is as follows: ; In the formula, WQ is the water quality sensitivity index, ω i X represents the weight of the i-th water quality indicator; i Let be the standardized value of the i-th water quality indicator; The formula for evaluating the water ecological indicators is as follows: ; In the formula, WE represents the water ecological index; ε i X represents the weight of the i-th water ecological indicator; i Let be the standardized value of the i-th water ecological indicator.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: 1. This invention integrates a digital twin model, an ecological purification module, an electrodynamic enhancement module, a microbial remediation module, an intelligent monitoring module, a power control module, and an ecological environment assessment module, forming a river and lake ecological floating bed control method that integrates remote visualization, multiple purification, multi-dimensional monitoring, intelligent decision-making, and ecological management. Through the synergistic effect of multiple purification processes such as plant absorption, electrochemical enrichment, and microbial degradation, it can efficiently and persistently remove pollutants such as COD, ammonia nitrogen, and heavy metals. 2. This invention constructs a system architecture consisting of a visualization layer, a physical layer, a perception layer, a transmission layer, a decision-making layer, an application layer, and a management layer, realizing a digital twin closed loop of "perception-transmission-intelligent computing-prediction-decision-application-management"; 3. This invention deeply integrates MIKE model simulation, hydrodynamics and water quality diffusion mechanisms, Internet of Things and artificial intelligence technologies to construct a three-dimensional digital twin model for river and lake water bodies, and integrates numerical simulation with monitoring data in real time to realize simulation based on mechanism models and prediction, optimization, intelligent decision-making and application management functions of multi-dimensional monitoring. 4. The intelligent monitoring module of this invention constructs a multi-dimensional monitoring network based on the Internet of Things and intelligent algorithms. It uses LSTM neural network to predict water quality changes, integrates neural network to assess the eutrophication status of water bodies and issue early warning signals, and uses deep reinforcement learning to optimize control commands. This realizes a proactive prevention and control mode from "end-of-pipe treatment" to "early warning and intelligent control", dynamically adjusts purification strategies, improves the effectiveness of ecological restoration, and reduces the occurrence of eutrophication disasters.

[0016] The present invention also has other beneficial effects, which are described in the embodiments section of the specification and will not be repeated here. Attached Figure Description

[0017] Figure 1 This is a structural block diagram of the ecological floating bed intelligent control system according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the construction of a water quality monitoring, digital twin model, and intelligent control system according to an embodiment of the present invention. Figure 3 This is a cross-sectional layout diagram of the ecological floating bed according to an embodiment of the present invention; Figure 4 This is a plan view of the ecological floating bed according to an embodiment of the present invention.

[0018] Attached label: 1. Visualization layer; 2. Physical layer; 3. Perception layer; 4. Transmission layer; 5. Decision layer; 6. Application layer; 7. Management layer; 8. Support layer; 9. Ecological purification module; 10. Electrodynamic enhancement module; 11. Microbial remediation module; 12. Walking mechanism; 121. Track; 122. Floating bed frame; 123. Walking wheels; 124. Drive motor; 125. Hydraulic lifting mechanism; 13. Ecological floating bed; 14. Aquatic plants; 15. Anode and cathode pairs; 16. Wire; 17. Low-voltage DC power supply; 18. Bacterial agent dosing device; 19. Micro-nano aeration device; 20. Intelligent monitoring module; 201. Multi-parameter water quality sensor; 202. Hyperspectral multi-parameter water quality monitor; 203. Water level sensor; 204. Underwater camera; 21. Control center workstation; 22. Data processing module; 23. Intelligent control module; 24. Solar power supply device; 25. Power control module; 26. Solenoid valve. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will be described in conjunction with the accompanying drawings. Figures 1-4 The technical solutions of the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0020] Example 1: As Figures 1-4 As shown, this invention provides an intelligent control system for ecological floating beds used in the treatment and ecological restoration of eutrophication in water bodies. The system includes a visualization layer 1, a physical layer 2, a perception layer 3, a transmission layer 4, a decision-making layer 5, an application layer 6, a management layer 7, and a support layer 8. It is used for the ecological restoration, water quality purification, and ecological environment health assessment and management of natural water bodies in rivers and lakes. The structure and function of each layer are as follows. (I) Visualization layer 1 is a digital twin model, which provides a three-dimensional visualization of the water environment and equipment operation status.

[0021] A 3D digital terrain model is constructed based on a terrain database, and an API interface is set up to enable real-time information interaction and obtain dynamic digital models. Based on the digital model, a hydrodynamic model, a water quality model, and a water ecology model are built to form a dynamic digital twin model for water environment visualization.

[0022] The specific steps for constructing a digital twin model include: Step 1: Collect river and lake cross-section design data and establish a topographic information database; Step 2: Construct a 3D terrain digital model based on the terrain database and set up an API interface. Use the API interface to exchange information in real time and obtain a dynamic digital model. Step 3: Construct a hydrodynamic model, a water quality model, and an aquatic ecosystem model, and interface them with the 3D digital model to generate a dynamic digital twin model. The hydrodynamic model uses the MIKE 21 hydrodynamic module to simulate and predict changes in water level and flow velocity. The water quality model uses the MIKE 21 transport module to simulate and predict the transport paths, diffusion and decay processes, and concentration changes of pollutants in the water body. The aquatic ecosystem model uses the MIKE EcoLab ecological module to simulate and predict the growth and reproduction of organisms in the aquatic ecosystem.

[0023] (ii) The physical layer 2 includes the ecological purification module 9, the electrodynamic enhancement module 10 and the microbial repair module 11, which perform water purification functions.

[0024] The ecological purification module 9 consists of a walking mechanism 12 and an ecological floating bed 13, substrate, and aquatic plants 14 located on the walking mechanism 12. The aquatic plants are one or more combinations of reeds, cattails, irises, loosestrife, and canna lilies, and the plant spacing can be set to 10cm to ensure sufficient space for plant growth. The walking mechanism 12 includes a track 121, a floating bed frame 122, walking wheels 123, a drive motor 124, and a hydraulic lifting mechanism 125; Figure 3 As shown, concrete foundations are set on both sides of the water body. Multiple hydraulic lifting mechanisms 125 are set on the concrete foundations along the water flow direction. Tracks 121 are mounted on the hydraulic lifting mechanisms 125. Ecological floating beds 13 are correspondingly set on floating bed frames 122. Multiple traveling wheels 123 are set on both sides of the floating bed frame 122 and are mounted on the tracks 121 through the traveling wheels 123. Drive motors 124 are set in cooperation with the traveling wheels. The traveling mechanism 12 drives the ecological floating bed 13 to move in the water body and adjust its height.

[0025] like Figures 3-4 As shown, the electrodynamic enhancement module 10 includes multiple pairs of anode and cathode electrodes 15, wires 16, and a low-voltage DC power supply 17. The anode and cathode electrodes 15 are arranged in the substrate of the ecological floating bed 13, and the low-voltage DC power supply 17 is electrically connected to the corresponding anode and cathode electrodes 15 through the wires 16. The distance between the anode and cathode electrodes is less than 30 cm. The cathode is a graphite rod, and the anode is titanium, stainless steel, or a conductive metal-organic framework material. The low-voltage DC power supply is a solar-powered device. When a DC voltage is applied to the electrodes, a DC electric field will be formed between the electrodes, driving the electrodynamic remediation within a voltage range of 0-60V. After the power is turned on, under the action of the DC electric field, pollutants such as heavy metals and organic matter are driven to migrate and accumulate towards the electrodes with opposite charges, enhancing the degradation of pollutants by plants and microorganisms and improving the remediation efficiency.

[0026] The microbial remediation module 11 includes a microbial biofilm packing material, a microbial agent dosing device 18, and a micro / nano aeration device 19. The microbial biofilm packing material is located at the bottom of the ecological floating bed 13, providing an attachment carrier for microorganisms. The microbial biofilm packing material is artificial aquatic plants, 0.5m in length, with a suspension density of 15 bundles / m. 2 As a biological biofilm carrier, the microbial agent dosing device 18 and the micro-nano aeration device 19 are suspended in the water body through a chain structure to ensure full contact with the water body. Both the microbial agent dosing device 18 and the micro-nano aeration device 19 are connected to a delivery pipe and a solenoid valve 26 on the pipe to deliver microbial agents, nutrients and dissolved oxygen to the biofilm packing, ensuring the activity of microorganisms and continuously degrading pollutants.

[0027] The walking mechanism 12 drives the ecological purification module 9, the electro-dynamic enhancement module 10, and the microbial remediation module 11 to move synchronously along the water body, and adaptively adjusts the height of the ecological purification module 9 to maintain the optimal contact state between the plant roots and the water body; the ecological purification module 9 constructs a plant-microbe system to provide an attachment substrate for microorganisms and improve the dissolved oxygen level, while the electro-dynamic enhancement module 10 enhances dissolved oxygen and pollutant migration.

[0028] (III) The perception layer 3 is the intelligent monitoring module 20, which collects water environment data in all dimensions.

[0029] like Figures 1-2 As shown, the intelligent monitoring module 20 includes a multi-parameter water quality sensor 201, a hyperspectral multi-parameter water quality monitor 202, a water level sensor 203, and an underwater camera 204. The multi-parameter water quality sensor 201 detects pH, conductivity, turbidity, dissolved oxygen, ammonia nitrogen, total phosphorus, COD, BOD, chlorophyll a, and heavy metals. The water level sensor 203 detects water level and flow velocity. The hyperspectral multi-parameter water quality monitor 202 detects the concentration of dissolved substances in the water and algal growth. The underwater camera 204 monitors aquatic biological activity and algal growth in real time. The intelligent monitoring module 20 monitors and predicts water quality changes and assesses the eutrophication status of the water body in real time. Each monitoring module is deployed and fixed at the bottom of the floating bed frame. Simultaneously, a water quality sensor network is arranged in the ecological floating bed activity area and fixed on a stainless steel support, spaced 2*2m apart, extending 0.5m underwater, with a sampling frequency set to 10min / time. The monitoring data is transmitted to the control center workstation 21 via LoRaWAN low-power wide area network, and is filtered and preprocessed by the data processing module 22.

[0030] (iv) The transmission layer 4 transmits the monitoring data from the LoRaWAN low-power wide area network to the control center workstation 21, and transmits the monitoring data stably; the data processing module 22 preprocesses the collected water level and water quality data and then inputs it into the digital twin model for real-time updates.

[0031] (v) The decision-making layer 5 is the intelligent control module 23, which completes water quality prediction, eutrophication early warning and intelligent decision-making.

[0032] The pre-treated water quality data is analyzed and processed in real time based on the LSTM neural network algorithm to predict water quality changes; the intelligent control module 23 assesses the eutrophication status of the water body based on the integrated neural network algorithm and issues an early warning signal; for example, when the concentration of chlorophyll a is detected to exceed the set threshold (10 μg / L), or when the LSTM neural network predicts that the water body is at risk of eutrophication, the control center issues an early warning signal and the management personnel activate the emergency response plan.

[0033] Specifically, the intelligent control module 23 is based on LSTM neural networks, ensemble neural networks, and deep reinforcement learning algorithms. The LSTM neural network performs real-time analysis and processing of pre-treated water quality data to predict water quality changes. The ensemble neural network assesses the eutrophication status of the water body and issues early warning signals. The deep reinforcement learning algorithm outputs optimized control commands, enabling the intelligent control equipment to operate. The deep reinforcement learning algorithm constructs an objective function to solve a combinatorial optimization problem, with the objective function including water quality compliance and energy saving. The water purification efficiency is calculated based on water quality parameters, plant growth status, and water level changes. The mathematical model is as follows: ; In the formula, E t Let T be the water purification efficiency at time t, with a value ranging from 0 to 1; t For runtime h; P t Pollutant concentration (mg / L); P max Maximum treatable pollutant concentration (mg / L); D t The water body disturbance degree, with a value ranging from 0 to 1; C t Plant coverage area / (m 2 );C ref For reference, the vegetation cover area is / (m²) 2 ); H t Water level / m; H ref The reference water level is given in m; k1 is the basic purification efficiency coefficient, ranging from 0.8 to 0.95; α is the time decay coefficient, ranging from 0.01 to 0.05; β is the disturbance influence coefficient, ranging from 0.1 to 0.3; γ is the light cycle influence coefficient, ranging from 0.1 to 0.2; and λ is the water level influence coefficient, ranging from 0.2 to 0.4.

[0034] (vi) Application layer 6 is the power control module 25, which executes intelligent control commands.

[0035] Application layer 6 serves as the specific execution system, which includes a solar power supply device 24, a walking mechanism 12, a microbial agent dosing device 18, a micro-nano aeration device 19, and an intelligent control module 23. It is powered by solar energy and integrates the walking mechanism, microbial agent dosing device, micro-nano aeration device, and intelligent control module into one unit.

[0036] Specifically, the independent switch of the solar power supply device 24, the drive motor and hydraulic lifting mechanism of the walking mechanism, the dosing pump and solenoid valve of the microbial agent dosing device, and the solenoid valve of the micro-nano aeration device are defined as intelligent entities. Each intelligent entity is equipped with a microcontroller and control rules are set. The microcontroller is installed in the control cabinet.

[0037] The intelligent control module 23 uses a deep reinforcement learning algorithm to analyze and process water quality data in real time. The processed water level and quality data interacts between the solar power supply device 24, the walking mechanism 12, the microbial agent dosing device 18, and the micro-nano aeration device 19. It then activates an intelligent algorithm to perform distributed collaborative optimization, generating intelligent control commands and distributing them to each power execution module (intelligent agent). This dynamically adjusts the purification plan, enabling intelligent operation of the equipment. Priorities are ranked according to the degree of impact on water quality, ensuring that critical tasks are executed first. The walking mechanism has the highest priority (level 1); the dosing pump and solenoid valve, and the solenoid valve of the micro-nano aeration device are level 2; and the others are level 3.

[0038] (vii) Management 7 is the ecological environment assessment module, which carries out water ecological health assessment and management.

[0039] A multi-factor weighting method is used to evaluate water quality and aquatic ecological indicators, enabling intelligent assessment and management of aquatic ecological environment health. Water quality indicators include water area, flow velocity, water depth, hydraulic loading, and eutrophication level; aquatic ecological indicators include water area, flow velocity, water depth, and aquatic plant abundance. The weights of the indicator factors are determined using the analytic hierarchy process (AHP) or expert evaluation. The water quality indicator evaluation formula is as follows: ; In the formula, WQ is the water quality sensitivity index, ω i X represents the weight of the i-th water quality indicator; i Let be the standardized value of the i-th water quality indicator.

[0040] The formula for evaluating water ecological indicators is: ; In the formula, WE represents the water ecological index; ε i X represents the weight of the i-th water ecological indicator; i Let be the standardized value of the i-th water ecological indicator.

[0041] (viii) Support layer 8 includes hardware (workstations, servers), software (operating systems, software, basic service framework), network communication (communication protocols, networks) and data center, which serve as the basic guarantee for the operation of the entire system and provide support for other layers.

[0042] The digital twin model, ecological purification module 9, electrodynamic enhancement module 10, microbial remediation module 11, intelligent monitoring module 20, power control module 25, and ecological environment assessment module of this invention are integrated into one unit. Information exchange and collaboration are achieved through a control center workstation 21, forming a smart management and control platform for the ecological floating bed 13 purification system. Notably, the digital twin model, ecological purification module 9, electrodynamic enhancement module 10, microbial remediation module 11, intelligent monitoring module 20, power control module 25, and ecological environment assessment module can all be independently upgraded by replacing their physical entities.

[0043] Example 2: The ecological floating bed intelligent control system from Example 1 was deployed at the river / lake inlet, middle, and end points, with progressively enhanced treatments. The operating length of the ecological floating bed was 30m. The equipment was regularly inspected and maintained according to production specifications. (1) Monthly inspection and maintenance of solar photovoltaic panels, batteries and operating mechanisms, cleaning dust from the surface of photovoltaic panels, checking the performance of batteries, and repairing or replacing damaged equipment parts; (2) Monthly calibration and maintenance of water quality sensors, hyperspectral water quality multi-parameter monitors, water level sensors, and underwater cameras are performed to ensure the accuracy of the monitoring instruments, and the position and stability of the instruments are checked. (3) The growth of aquatic plants is assessed quarterly, and the types and density of plants are adjusted and the substrate is replaced in a timely manner based on the growth status and purification effect. (4) Clean the microbial biofilm packing material, replace the electrodes and change their positions every quarter; (5) Evaluate the entire purification system annually, analyze the synergistic effect and operating efficiency between modules, optimize and upgrade the system, and improve the system's intelligent control performance and ecological restoration efficiency.

[0044] The above are preferred embodiments of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent control system for ecological floating beds used in eutrophication control and ecological restoration of water bodies, characterized in that, It includes a visualization layer (1), a physical layer (2), a perception layer (3), a transmission layer (4), a decision layer (5), an application layer (6), and a management layer (7); The visualization layer (1) is a digital twin model. A three-dimensional terrain digital model is constructed based on the terrain database and an API interface is set up. Information is exchanged in real time through the API interface to obtain a dynamic digital model. A hydrodynamic model, a water quality model and a water ecology model are built on the digital model to form a dynamic digital twin model for water environment visualization. The physical layer (2) includes an ecological purification module (9), an electrodynamic enhancement module (10), and a microbial remediation module (11); the ecological purification module (9) consists of a walking mechanism (12) and an ecological floating bed (13), substrate, and aquatic plants (14) located on the walking mechanism (12); the walking mechanism (12) includes a track (121), a floating bed frame (122), walking wheels (123), a drive motor (124), and a hydraulic lifting mechanism (125), and the walking mechanism (12) drives the ecological purification module (9) to perform ecological purification. The floating bed (13) moves and adjusts its height in the water; the electrodynamic enhancement module (10) includes a cathode and anode pairs (15), wires (16) and a low-voltage DC power supply (17). The cathode and anode pairs (15) are arranged in the substrate of the ecological floating bed (13) and drive pollutants to migrate and accumulate under the action of DC electric field; the microbial remediation module (11) includes microbial biofilm packing, bacterial agent dosing device (18) and micro-nano aeration device (19). The microbial biofilm packing is located at the bottom of the ecological floating bed (13); The sensing layer (3) is an intelligent monitoring module (20), which includes a multi-parameter water quality sensor (201), a hyperspectral water quality multi-parameter monitor (202), a water level sensor (203), and an underwater camera (204) to monitor and predict water quality changes and assess the eutrophication status of water bodies in real time. The transmission layer (4) transmits the monitoring data from the LoRaWAN low-power wide area network to the control center workstation (21). The data processing module (22) preprocesses the collected water level and water quality data and then inputs it into the digital twin model for real-time updates. The decision-making layer (5) is an intelligent control module (23), which performs real-time analysis and processing of pre-processed water quality data based on the LSTM neural network algorithm to predict water quality changes; the intelligent control module (23) assesses the eutrophication status of water bodies and issues early warning signals based on the integrated neural network algorithm; The application layer (6) is a power control module (25), including a solar power supply device (24), a walking mechanism (12), a microbial agent dosing device (18), a micro-nano aeration device (19), and an intelligent control module (23). The intelligent control module (23) performs real-time analysis and processing of water quality data based on a deep reinforcement learning algorithm. The processed water level and water quality data are interacted between the solar power supply device (24), the walking mechanism (12), the microbial agent dosing device (18), and the micro-nano aeration device (19). The intelligent algorithm is activated to perform distributed collaborative optimization to generate intelligent control instructions and distribute them to each power execution module. The management layer (7) is an ecological environment assessment module, which uses a multi-factor weighting method to evaluate water quality indicators and aquatic ecological indicators, thereby realizing intelligent assessment and management of aquatic ecological environment health. The digital twin model, ecological purification module (9), electrodynamic enhancement module (10), microbial remediation module (11), intelligent monitoring module (20), power control module (25) and ecological environment assessment module are integrated into one, and information interaction and collaboration are carried out through the control center workstation (21) to form an intelligent management and control platform for the ecological floating bed (13) purification system.

2. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 1, characterized in that, The digital twin model, ecological purification module (9), electrodynamic enhancement module (10), microbial remediation module (11), intelligent monitoring module (20), power control module (25) and ecological environment assessment module can all be upgraded by independently replacing their physical entities.

3. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 1, characterized in that, The control system also includes a support layer (8), which includes hardware, software, network communication and data center, and serves as the basic guarantee for the operation of the entire system.

4. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 1, characterized in that, The steps for constructing the digital twin model include: Step 1: Collect river and lake cross-section design data and establish a topographic information database; Step 2: Construct a 3D terrain digital model based on the terrain database and set up an API interface. Use the API interface to exchange information in real time and obtain a dynamic digital model. Step 3: Construct hydrodynamic model, water quality model, and aquatic ecosystem model, and interface them with the 3D digital model to generate a dynamic digital twin model.

5. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 4, characterized in that, The hydrodynamic model uses the MIKE 21 hydrodynamic module to simulate and predict changes in water level and flow velocity; the water quality model uses the MIKE 21 transport module to simulate and predict the transport paths, diffusion and decay processes, and concentration changes of pollutants in the water body; and the aquatic ecosystem model uses the MIKE EcoLab ecological module to simulate and predict the growth and reproduction of organisms in the aquatic ecosystem.

6. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 1, characterized in that, The walking mechanism (12) drives the ecological purification module (9), the electro-power enhancement module (10) and the microbial repair module (11) to move synchronously and adaptively adjust the height of the ecological purification module (9) to maintain the optimal contact state between plant roots and water. The ecological purification module (9) constructs a plant-microbe system to provide an attachment substrate for microorganisms, and the electro-power enhancement module (10) enhances dissolved oxygen and pollutant migration.

7. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 1, characterized in that, The multi-parameter water quality sensor (201) detects pH, conductivity, turbidity, dissolved oxygen, ammonia nitrogen, total phosphorus, COD, BOD, chlorophyll a, and heavy metals; the water level sensor (203) detects water level and flow rate; the hyperspectral water quality multi-parameter monitor (202) detects the concentration of dissolved substances in the water and algae growth; the underwater camera (204) monitors aquatic biological activities and algae growth in real time; the monitoring data is transmitted to the control center workstation (21) through the LoRaWAN low-power wide area network, and is preprocessed by the data processing module (22) through screening and interpolation.

8. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 1, characterized in that, The intelligent control module (23) is based on LSTM neural network, ensemble neural network and deep reinforcement learning algorithm; the LSTM neural network performs real-time analysis and processing of pre-processed water quality data and predicts water quality changes; the ensemble neural network assesses the eutrophication status of the water body and issues early warning signals; the deep reinforcement learning algorithm outputs optimized control commands and the intelligent control equipment operates.

9. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 8, characterized in that, The deep reinforcement learning algorithm constructs an objective function to solve a combinatorial optimization problem. The objective function includes achieving water quality standards and saving energy. Based on water quality parameters, plant growth status, and water level changes, the water purification efficiency is calculated using the following mathematical model: ； In the formula, E t Let T be the water purification efficiency at time t, with a value ranging from 0 to 1; t Runtime (h); P t Pollutant concentration (mg / L); P max Maximum treatable pollutant concentration (mg / L); D t The water body disturbance degree, with a value ranging from 0 to 1; C t Plant coverage area / (m 2 );C ref For reference, the vegetation cover area is / (m²) 2 ); H t Water level / m; H ref The reference water level is given in m; k1 is the basic purification efficiency coefficient, ranging from 0.8 to 0.95; α is the time decay coefficient, ranging from 0.01 to 0.05; β is the disturbance influence coefficient, ranging from 0.1 to 0.3; γ is the light cycle influence coefficient, ranging from 0.1 to 0.2; and λ is the water level influence coefficient, ranging from 0.2 to 0.

4.

10. The ecological floating bed intelligent control system for eutrophication treatment and ecological restoration of water bodies according to claim 1, characterized in that, The water quality indicators include water area, flow velocity, water depth, hydraulic loading, and eutrophication level; the aquatic ecological indicators include water area, flow velocity, water depth, and aquatic plant abundance; the weights of the indicators are determined using the analytic hierarchy process or expert evaluation. The water quality evaluation formula is as follows: ； In the formula, WQ is the water quality sensitivity index, ω i X represents the weight of the i-th water quality indicator; i Let be the standardized value of the i-th water quality indicator; The formula for evaluating the water ecological indicators is as follows: ； In the formula, WE represents the water ecological index; ε i X represents the weight of the i-th water ecological indicator; i Let be the standardized value of the i-th water ecological indicator.

Images