Bionic fish school cooperative control method integrating underwater monitoring and ecological restoration

Abstract

The invention relates to a bionic fish school cooperative control method integrating underwater monitoring and ecological restoration, which comprises the following steps: in a monitoring process, if a water quality index at a certain point exceeds a threshold value, firstly starting high-frequency monitoring and sample collection by bionic fish; and analyzing the position of the pollution source according to the water quality data of the monitored point in the monitoring route, planning and changing the route by taking the pollution source as an end point, and performing ecological restoration on the pollution source if restoration is needed. And if the water quality index of each point location in the monitoring process is normal, cruising normally, and returning monitoring data. According to the invention, efficient monitoring and accurate restoration of the underwater environment are realized, and the system is of great significance to underwater monitoring and restoration work.

Description

TECHNICAL FIELD

[0001] The present application relates to a bionic fish swarm cooperative control method integrating underwater monitoring and ecological restoration, and belongs to the technical field of underwater robots, environmental monitoring and ecological restoration. BACKGROUND

[0002] The importance of water quality monitoring and ecological restoration technology is increasingly prominent. Existing water quality monitoring equipment is mostly fixed equipment or single-function mobile devices, with limited monitoring range and no real-time pollution repair. For example, the patent with publication number CN222379392U proposes a water quality sampling device for river and lake water environment monitoring, which can be moved to different areas for water quality monitoring. However, this device relies on manual pushing and transportation, and does not achieve full automation of movement. Moreover, the monitoring range is limited, and it can only monitor the water quality of a certain depth area on the shore of a river or lake with good shore conditions. The patent with publication number CN111289712A proposes a water quality in-situ monitoring device based on an unmanned ship. This device can move to different points for water quality monitoring and water sampling. However, this device cannot achieve simultaneous repair during monitoring and sampling, and the single sampling strategy is also difficult to cope with unexpected situations during sampling in natural rivers and lakes. The patent application with publication number CN116239234A proposes a water body ecological restoration device that combines aeration and dosing. However, this device can only operate in a small range and cannot achieve monitoring and repair coordination.

[0003] The above-mentioned defects result in the difficulty of existing water quality monitoring devices in achieving multi-range, stable water quality monitoring and timely repair measures in river and lake water bodies. Therefore, there is an urgent need for a method that can integrate underwater monitoring and ecological restoration and achieve intelligence. SUMMARY

[0004] In order to solve the above-mentioned problems, the present application discloses a bionic fish swarm cooperative control method integrating underwater monitoring and ecological restoration, and the specific technical solution is as follows: A bionic fish swarm cooperative control method integrating underwater monitoring and ecological restoration includes the following processes: Step 1: Determine whether there is a monitoring task. When there is no need to monitor water quality, go to Step 2. If water quality needs to be monitored, go to Step 3. Step 2: The bionic fish swarm directly follows the set route for regular cruising, detects and records the terrain during cruising, and collects water samples at regular intervals. Step 3: Real-time monitoring according to the set monitoring route, the sensor real-time monitors water quality information, and compares the water quality information with the pre-set corresponding water quality index threshold value. If it is abnormal, go to Step 4. If it is normal, go directly to Step 5. Step 4: Analyze the pollution source location based on the water quality data of the monitored points in the monitoring route, and plan and change the route with the pollution source as the terminal point. Verify and confirm the location of the pollution source. Step 5: The control terminal generates a new work path by Q-Learning optimization, and adjusts the bionic fish dynamically based on the CNN-LSTM; Step 6: Determine whether the pollution needs to be repaired. If it needs to be repaired, go to step 7. The bionic fish implements in-situ repair. If it does not need to be repaired, repeat steps 1-5. Step 7: Determine whether the bionic fish needs to continue working. If it needs to continue working, repeat steps 1-6. If it does not need to continue working, execute the return command.

[0005] Further, the bionic fish group includes a tail fish bionic fish and several tail ordinary bionic fish. Each bionic fish is equipped with a data acquisition system, a data processing and command center, a navigation and communication module, a driving device and a power supply device. It also has a sampling device and a repair module, which includes a probiotic slow-release capsule warehouse, a chitosan adsorption particle warehouse and an electrolytic aeration unit. The data acquisition system includes a camera, an ultrasonic sensor and a water quality sensor. The power supply device includes a main power supply device and a backup power supply device.

[0006] Further, the data processing and command center of all bionic fish is connected to the control center through the communication module. The control center records and analyzes the position coordinates of each bionic fish and the data transmitted by it, and sends action instructions to each bionic fish. The driving device adjusts the swimming direction and speed of the bionic fish in a timely and accurate manner according to the instructions of the control center.

[0007] Further, during the operation of the bionic fish group, every 2 minutes, all bionic fish broadcast their heartbeat packets, which include id code, battery voltage and power. The bionic fish is detected by the following method: Heartbeat packet monitoring: every 2 minutes, the state code is broadcast through ZigBee to check, and if it does not respond for 3 times in a row, it is judged to be abnormal. Neighbor node detection: ultrasonic ranging is used to scan the neighboring fish within a range of 10m to detect whether the physical displacement is normal. Data stream analysis: the data processing system carried by each bionic fish checks the continuity of the data packet in real time. If the data jump interval is greater than 5 minutes, it is determined that the transmission is interrupted.

[0008] Further, the specific process of step 3 is as follows: the control center first calculates and distributes Voronoi units according to the work area information and the state of the bionic fish device executing the task, and generates a monitoring route. The bionic fish device monitors regularly or at fixed points according to the set monitoring route, and the AI system carried by it processes the terrain and water quality data in real time. According to the monitored data and path optimization algorithm, the monitoring route is optimized and adjusted. In the monitoring process, if a water quality index at a certain point exceeds the threshold value, the bionic fish device first starts high-frequency monitoring and sample collection, the control center uses the CNN-LSTM algorithm to analyze the location of the pollution source according to the water quality data of the monitored points in the monitoring route and the historical data of the region, and plans and changes the route with the pollution source as the terminal, verifies and confirms the location of the pollution source, and then the control center generates a new operation path through Q-Learning optimization and adjusts the bionic fish group dynamically through the deep learning algorithm CNN-LSTM. If the water quality indexes at all points are normal during the monitoring process, the control center optimizes the operation path and adjusts the bionic fish group dynamically according to the data obtained during the operation process.

[0009] Further, during the underwater operation process, when a single bionic fish device fails, the bionic fish device and its neighbor first determine the type of failure: if it is a non-repairable failure, including bionic fish device body damage, communication loss or low battery, the failure is reported to the control terminal, and after obtaining the failure information and positioning, the failure fish is salvaged in time; if a local failure that does not affect navigation and driving occurs, the failure fish directly returns to the set homing point and reports the failure to the terminal, and the failure fish is recovered at the homing point; Upon receiving the information that the bionic fish device has failed, the head fish immediately recalculates the optimal adjustment scheme using the Voronoi algorithm based on the number and location of existing normal bionic fish devices, and issues instructions, and all normal running bionic fish devices change the operation path and monitoring range according to the optimal adjustment scheme.

[0010] Further, when the pollution type is dissolved oxygen that is too low, the repair method is to start aeration, and the multi-fish ring array synchronously aerates; When the pollution type is heavy metal exceeding the standard, the repair method is to release magnetic nano adsorbent, and the fish group surrounds the pollution source to form a dynamic barrier; When the pollution type is organic matter pollution, the repair method is to release probiotic slow-release capsules, and the regional responsibility is allocated based on the Voronoi algorithm.

[0011] Further, during the monitoring process, if water quality abnormalities are found, the bionic fish device reports the point information and water quality information to the head fish and the control center, the control center calculates and analyzes the pollution source point based on historical and real-time data using deep learning algorithms, the head fish calculates and redistributes the operation units, and allocates more bionic fish devices to the pollution source and nearby to verify and densely monitor and sample, the control center determines in real time whether the monitoring point needs in-situ repair according to the water quality index monitoring data, if repair is needed, the control center issues corresponding repair instructions according to the needs, and after repair, the bionic fish group continues operation or executes the homing command.

[0012] Further, the navigation and communication module adopts a water acoustic positioning system to realize navigation, and adopts a ZigBee / LoRa dual-mode adaptive module to realize communication.

[0013] Further, the control center carries multiple machine learning models, and inputs real fish school moving videos, a Shapley value-based game theory model, double Q-Learning and a plurality of emergency cases before use.

[0014] The present application has the following advantages: The present application realizes efficient monitoring and accurate repair of underwater environment, and is of great significance to underwater monitoring and repair work.

[0015] The bionic fish device designed in the present application is bionic in shape, can reduce the disturbance to aquatic organisms, improve the accuracy of monitoring data, and improve the concealment of the system underwater.

[0016] In the present application, the bionic fish device and the control center carry and use multiple algorithms, the application of Voronoi algorithm can realize the optimized division and dynamic adjustment of the task area, ensure that the whole area is covered without omission, CNN-LSTM can efficiently analyze and locate the pollution source combined with historical data, and the operation efficiency of the bionic fish school can be improved by combining the Q-Learning method to optimize and adjust the bionic fish school operation line.

[0017] When the bionic fish school is operating, the bionic fish device reports its own state and judges the state of the adjacent fish, after the bionic fish device fails, the bionic fish school can dynamically adjust the operation area distribution, and the operator can obtain relevant information in time and process it.

[0018] After the bionic fish school detects water quality abnormalities, the bionic fish school can locate and verify the pollution source, densely monitor water quality and collect samples, and improve the accuracy of data.

[0019] The control center can judge whether accurate repair instructions are needed according to real-time water quality indicators, the bionic fish school implements in-situ repair measures, realizes accurate in-situ repair and monitors and judges the repair effect. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a structural schematic diagram of the present application, LIST OF REFERENCE NUMBERS: 01 - ultrasonic probe, 02 - water quality sensor, 03 - camera, 04 - electrolytic aeration unit, 05 - chitosan adsorption particle warehouse, 06 - probiotic slow-release capsule warehouse, 07 - data processing and instruction center, 08 - navigation and communication module, 09 - power supply device, 10 - water sample collection unit, 11 - tail driving device.

[0021] Figure 2 is a conventional operation flowchart of the bionic fish school system of the present application.

[0022] Figure 3 is a schematic diagram of the initial distribution of the embodiment 1 bionic fish group of the present application. DETAILED DESCRIPTION

[0023] The present application will be further clarified by the following examples and figures. It should be understood that the following examples are intended to illustrate the present application and are not intended to limit the scope of the present application.

[0024] In combination with the figures and specific embodiments, the present application is further illustrated. It should be understood that the following specific embodiments are intended to illustrate the present application and are not intended to limit the scope of the present application. Figure 1 The bionic fish system integrating underwater monitoring and ecological restoration of the present embodiment is composed of multiple bionic fish devices including a leader fish and ordinary member fish. The bionic fish device is designed in shape and structure as follows: (1) the bionic fish device is similar in shape to common fish in rivers and lakes to reduce interference with underwater organisms and improve its concealment in the natural environment. The shell is made of thermoplastic polyurethane elastomer rubber material with good waterproofness and firmness, which can effectively protect the internal components and also simulate the flexibility and elasticity of real fish; (2) multiple sensors are installed at the head and torso of the bionic fish device, among which the ultrasonic sensor at the head can detect the terrain, and the water quality sensor is used to monitor temperature, pH, turbidity, dissolved oxygen, heavy metal content, and organic matter content. The camera obtains underwater images for analyzing aquatic organisms and assisting in judging the terrain; (3) the bionic fish device has a retractable sampling device at the abdomen, which can complete the extension and retraction operation according to the instructions of the control module. When reaching the designated position, the sampling device extends and opens the sampling port to collect water samples into the storage bottle. After a portion of the water sample is collected, a new storage bottle is automatically replaced. The collected water sample is sealed to ensure its stability and purity during transmission; (4) three restoration modules, i.e. electrolytic aeration unit, chitosan adsorption particle bin, and probiotic slow-release bin, are installed on the back of the bionic fish device, which operate according to the restoration requirements; (5) the tail of the bionic fish device is a driving device, which adopts a multi-joint driving system to achieve efficient underwater propulsion by simulating the swing of the fish tail fin; (6) the power supply equipment provides power support for the entire device, which is located in the middle and rear position of the bionic fish torso. It adopts high-performance lithium batteries to ensure its long-term endurance capability underwater. At the same time, it is also equipped with an energy management system to optimize the energy use efficiency and prolong the service life; (7) the bionic fish device is loaded with a data processing and command center inside to simply process and judge the water quality and terrain information obtained by the bionic fish device; (8) the bionic fish is also provided with a navigation module and a communication module. The navigation module adopts a hydroacoustic positioning system to achieve accurate underwater positioning and path planning. The communication module adopts ZigBee / LoRa dual-mode adaptive technology to transmit the data and device status obtained by the bionic fish device to the ground control center in real time, while receiving the instructions from the control center. At the same time, the status information of the bionic fish device is also transmitted to the adjacent bionic fish through the communication module.

[0025] The ground control center is loaded with multiple machine learning models, and before use, real fish school swimming videos, double Q-Learning, regional historical monitoring data, and several emergency events and emergency cases are input to improve the data processing efficiency and decision accuracy of the control center.

[0026] A set of sampling devices, probiotic slow-release capsule warehouses, chitosan adsorption particle warehouses, and electrolytic aeration units are provided below: The specific structure of the sampling device is as follows: The sampling device includes a retractable micro sampling head, a micro corrosion-resistant peristaltic pump, and a set of fluid pipelines. The retractable micro sampling head is provided with a one-way valve. The sampling head is stored in the bionic fish shell in the non-working state and is driven out by a micro steering engine when working. The sample storage unit is a rotatable sample disc with multiple independent and sealed sample bottles evenly distributed on it. The sample disc is driven by a stepping motor. A sealed cover automatic opening and closing mechanism is also provided in the unit to open the sealed cover of the sampling bottle when the sampling bottle is rotated to the sampling station and close the cover immediately after sampling is completed.

[0027] After receiving the sampling instruction, the stepping motor drives the sample disc to rotate an unused empty sample bottle to the sampling station, and the sealed cover opening mechanism opens the sealed cover of the sample bottle. The micro steering engine pushes out the sampling head, and the peristaltic pump starts to extract a specific volume of target water sample through the pipeline and inject it into the sample bottle. After sampling is completed, the peristaltic pump stops, the sealed cover is tightly closed, and the sampling head is retracted. The sample disc rotates one grid to prepare the next empty bottle to the sampling station.

[0028] Beneficial effects: automated and orderly sampling is achieved, and the sample bottles are in a sealed state during non-sampling periods, effectively preventing sample evaporation, external contamination, or cross-infection between samples, improving the accuracy and reliability of the sampling data.

[0029] The specific structure of the probiotic slow-release capsule warehouse and the chitosan adsorption particle warehouse is as follows: The probiotic slow-release capsule warehouse and the chitosan adsorption particle warehouse are used to store and release slow-release capsules / chitosan particles. The essence is a regulated storage warehouse, which is mainly characterized by isolating drugs and water. The probiotic slow-release capsule warehouse and the chitosan adsorption particle warehouse have the same structure, both using a double-warehouse structure of a main storage warehouse + a buffer release warehouse. The main storage warehouse is a closed dry warehouse for storing unused capsules or particles. Its bottom outlet is connected to the buffer release warehouse below through an electric sealing butterfly valve. The buffer release warehouse is a small cavity with a bottom opening type door driven by a micro motor. The door has good sealing performance and only opens for a moment during release.

[0030] Take the probiotic slow-release capsule as an example: after receiving the release instruction, first control the electric butterfly valve at the bottom of the main storage bin to open a certain degree or time, and drop a certain number of capsules into the buffer release bin; then the electric butterfly valve is immediately closed, completely isolating the main storage bin from the outside environment; the double-door at the bottom of the buffer release bin opens quickly, and the probiotic slow-release capsule is released into the water body, and the door is closed immediately after the release is completed.

[0031] Beneficial effects: Ensure that only the buffer release bin is in direct contact with the external water body, greatly reducing the risk of drug moisture, hardening and failure due to water entering the main storage bin, and ensuring the reliability of long-term work.

[0032] The specific structure of the electrolytic aeration unit is: The electrolytic aeration unit includes an electrolytic tank, a power supply and management area, a gas-liquid separation and discharge system. The electrolytic tank adopts a compact plate or honeycomb design, and the electrode material is a base metal oxide coated anode and a stainless steel cathode. The power supply and management module integrates a high-voltage micro DC-DC converter, which can raise the voltage of the main power supply of the bionic fish to the working voltage required for electrolytic water; the built-in intelligent control chip can adjust the output power according to the instruction to control the oxygen production rate. The mixture of hydrogen and oxygen generated by electrolysis is naturally separated at the top of the electrolytic tank and collected in the micro gas collection chamber above the unit; the gas collection chamber guides the gas to the micro-porous aeration head on the surface of the bionic fish through a small aeration pipeline.

[0033] Beneficial effects: When the bionic fish detects that the dissolved oxygen in a small area is too low, it can immediately improve the dissolved oxygen condition of the local water body around the bionic fish by generating and releasing micro-bubbles rich in oxygen, achieving "precise aeration". The design of the electrolytic tank can maximize the electrode surface area and reduce the volume, and the electrode material has good catalytic efficiency and corrosion resistance, which can prolong its service life.

[0034] In combination with the attached Figure 2The main operation process of the bionic fish swarm cooperative control method integrating underwater monitoring and ecological restoration is as follows: during normal operation, if the water quality does not need to be monitored, the bionic fish swarm directly performs normal cruising according to the route set by the operator, detects and records the terrain during the cruising process, and collects water samples at regular time intervals; if the water quality needs to be monitored, the control center first calculates and distributes Voronoi cells and generates a monitoring route according to the operation area information and the state of the bionic fish device performing the task, the bionic fish device monitors at regular time intervals or fixed points according to the set monitoring route, and the terrain and water quality data are processed in real time by the AI system carried on the bionic fish device, and the monitoring route is optimized and adjusted according to the monitored data and the path optimization algorithm. During the monitoring process, if the water quality index at a certain point exceeds the threshold value, the bionic fish device first starts high-frequency monitoring and sample collection, the control center analyzes the location of the pollution source according to the water quality data of the monitored points in the monitoring route and the historical data of the region by using the CNN-LSTM algorithm, and plans and changes the route with the pollution source as the terminal point, verifies and confirms the location of the pollution source, and then the control center generates a new operation path by optimizing through Q-Learning, and dynamically adjusts the bionic fish swarm through the deep learning algorithm CNN-LSTM. If the water quality indexes of all points are normal during the monitoring process, the control center optimizes the operation path and adjusts the dynamics of the bionic fish swarm according to the obtained data during the operation process. During the operation process, the leader fish in the bionic fish swarm system uses the Voronoi algorithm to process and generate the operation path and responsibility range of all bionic fish devices in real time with the input sampling range as the constraint boundary, and all members of the bionic fish swarm adjust the route and dynamics according to the planning and distribution of the leader fish. At the same time, during the monitoring process, the control center judges in real time whether the monitoring point needs to be repaired in situ according to the water quality index monitoring data, and if the repair is needed, the control center issues the corresponding repair instruction, and after the repair is completed, the bionic fish swarm continues to operate or executes the return command.

[0035] Embodiment 1: An embodiment of fault handling during actual operation of the present application is given as follows: The bionic fish swarm cooperative control method integrating underwater monitoring and ecological restoration can timely handle the fault fish and re-plan the operation path and range of the normal bionic fish device when a single bionic fish device fails during operation. The present embodiment gives an embodiment of bionic fish swarm system processing and redistribution when multiple bionic fish devices fail during zoned operation.

[0036] The bionic fish device realizes perception and response through a three-level linkage mechanism: heartbeat packet monitoring, neighbor node monitoring and data flow analysis, judges whether other bionic fish devices are normally operated, and the corresponding specific operation and judgment method is shown in Table 1. During the operation of the bionic fish swarm, every 2 minutes, all bionic fish devices broadcast their own heartbeat packets, which contain id codes, battery voltages and power.

[0037] Table 1: Judgment of bionic fish device running state

[0038] In a task with the core goal of regional terrain exploration and water quality monitoring, 14 bionic fish devices form a monitoring network covering a rectangular water area of about 3 km x 2 km, which is divided into three regions: ecological protection area, industrial drainage area, and shipping lane area. The IDs of the 14 bionic fish devices are F01~F14, with F7 as the head fish.

[0039] In this monitoring task, the monitoring value, environmental risk, and node capability of different regions differ significantly. To obtain more monitoring resources in high-ecological-value areas, increase coverage redundancy in high-pollution-risk areas, and assign more responsibilities to high-performance nodes, a weighted Voronoi diagram is used to divide the monitoring area. The weight is calculated to quantify the monitoring capability or priority of the node during the initial pattern distribution process. Nodes with higher weights will have larger Voronoi cells, thus assuming more monitoring responsibilities. The following is the calculation method of the weight: weight =α*Vi +β*Ci + γ*Ri - δ*Ei Where α, β, γ, δ are weighting coefficients, and satisfy α+β+γ+δ=1. V is the regional value coefficient, C is the node capability index, R is the environmental risk index, and E is the energy consumption rate. i represents each fish, such as each fish has its own number, i represents the number of each fish.

[0040] V is the regional value coefficient, automatically assigned based on the region type, ranging from 0.8 to 1.5, and the calculation method is (python): # Based on the automatic assignment of region type def get_value_factor(zone_type): mapping = { "RARE_SPECIES_HABITAT": 1.5, "WATER_INTAKE": 1.3, "INDUSTRIAL_DRAIN": 1.2, "SHIPPING_LANE": 1.0, "GENERAL_WATER": 0.8 } return mapping.get(zone_type, 1.0) C is the node capability index, which is calculated by assigning different weights to sensor status, battery status, CPU performance, and reliability, ranging from 0 to 1, and the calculation method is: Ci = 0.4 \times \frac{S_{sensor}}{S_{max}} + 0.3 \times \frac{B_{battery}}{B_{max}} + 0.2 \times P_{cpu} + 0.1 \times R_{reliability} R is the environmental risk index, calculated based on historical pollution data, ranging from 1 to 2, and the calculation method is: risk_factor = min(2.0, 1.0 + log10(pollution_events + 1)) E is the energy consumption rate, calculated by the relative value of distance and power, ranging from 0.1 to 0.5, and the calculation method is: Ei = 0.3 \times \frac{D_{distance}}{D_{max}} + 0.7 \times \frac{P_{power}}{P_{max}} By multi-objective optimization, α = 0.5, β = 0.3, γ = 0.15, δ = 0.05, the weight of different biomimetic fish devices is calculated and adjusted to determine the Voronoi unit distribution, and Figure 3 The initial distribution pattern of the biomimetic fish group task execution is shown. Under this distribution pattern, F02, F03 and F04 have the highest weight, F07 and F08 have higher weight, F01 and F11 have general weight, and F05, F06, F09, F10, F12, F13 and F14 have relatively low weight. During the operation, the weight is updated according to the monitoring situation and the running state of the biomimetic fish device. If the node weight changes greatly, the Voronoi unit is recalculated and adjusted.

[0041] After reaching the set point, the biomimetic fish device starts cruising and monitoring. During the process, the biomimetic fish device F04 is damaged and cannot operate normally and communicate. F03 and F08 find that F04 does not respond for three times in a row, immediately send a signal to the head fish and the control center, and report the position of the last communication before the heart of F04 is lost. After receiving the information of F04 abnormality, the head fish immediately performs Voronoi redistribution, and finally F03 and F08 take over 60% and 40% of the remaining monitoring area of F04 respectively and allocate part of the original detection area to the neighbor fish. After redistribution, the operation area of all members of the biomimetic fish group increases by 5% to 15%.

[0042] Example 2: The following gives an embodiment of the actual operation of the present application for pollution treatment: The bionic fish swarm cooperative control method integrated with underwater monitoring and ecological restoration triggers intensive monitoring and sampling when water quality abnormalities are detected, analyzes and verifies the location of pollution sources, and implements in-situ restoration when the environment needs it. Table 2 lists the restoration methods and system cooperative mechanisms of the bionic fish swarm under three pollution conditions. This embodiment gives the response process after detecting water quality abnormalities.

[0043] Table 2 In-situ restoration methods and cooperative mechanisms

[0044] In a water quality monitoring task, ten bionic fish devices form a monitoring network and operate in a strip-like equidistant distribution mode. The IDs of the bionic fish devices are F01-F10, among which the head fish is F05. The distance between adjacent bionic fish devices is 100 m, the river width is 20 m, the speed of the bionic fish devices during operation is about 2 m / min, the monitoring frequency is 4 min / time, and the sampling frequency is 10 min / time. During the operation, F08 detects that the dissolved oxygen content at a certain point is 1 / 3 lower than the threshold value, immediately packs its own coordinates, sensor readings, and time stamp, uploads them to the control center through the communication module, and broadcasts a simple alarm signal to the adjacent units F07 and F09.

[0045] After the control center receives the detailed data of F08, it first calls the historical time series data of F07, F08, F09 in the recent period, the water quality data in the historical database of the region, the diffusion model and the water flow direction data, the shoreline, water depth, and sewage outlet location information provided by the electronic map. The system automatically creates a 200m x 20m analysis grid centered on the abnormal point F08, aligns and normalizes the real-time and historical time series data of F07, F08, F09, and forms a time series tensor X_temporal with dimensions (3 observation points, 5 time steps, 4 water quality parameters). The spatial coordinates of the three robotic fish and their current dissolved oxygen detection values are subjected to spatial Kriging interpolation to generate a 100 x 10 grid "water quality spatial snapshot" and form a spatial tensor X_spatial with dimensions (20, 20, 1). X_spatial is input into a lightweight CNN with the structure Conv2D(filters=8, kernel_size=3, activation='relu') -> MaxPooling2D() -> Flatten(), which identifies the concentration gradient change from F08 to F07 and outputs a spatial feature vector. X_temporal is input into an LSTM with the structure LSTM(units=16, return_sequences=False), which analyzes the dynamic change rule of pollutant concentration and outputs a time series feature vector. The two feature vectors output by CNN and LSTM are spliced and input into a fully connected layer for regression prediction: Dense(units=32, activation='relu') -> Dense(units=2, activation='linear'), which outputs the most likely pollution source coordinates (40, 10) with a confidence of 85%. The control center issues an instruction, and F07, F08, F09 move to the predicted point at a speed of 10m / min for confirmation and intensive monitoring.

[0046] F07, F08, F09 reach the target point, densely detect water quality at the target point and nearby, and confirm that there is a hidden sewage outlet at the location. The control center starts the Voronoi and Q-Learning algorithms to regenerate the global operation path. The new operation area of each bionic fish and the movement instructions to be assigned to each bionic fish are calculated based on the current position of all bionic fish, the position of the pollution source, and the priority of each region's uncompleted monitoring points. The reward function is the reward for covering high-priority areas + the reward for saving energy + the penalty for avoiding collision. After the algorithm runs, the output is the globally optimal new operation scheme: F07, F08, F09 form a surrounding array around the pollution source and execute aeration, material feeding, and effect monitoring; the operation range of other bionic fish is expanded by 20%-40%, and the path is readjusted to ensure full coverage of the original area with the highest efficiency.

[0047] Those skilled in the art can understand that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have meanings consistent with those in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless otherwise defined.

[0048] Based on the above ideal embodiments according to the present application, through the above description, relevant personnel can make various changes and modifications without deviating from the technical idea of the present application. The technical scope of the present application is not limited to the contents of the specification, and must be determined by the scope of the claims.

Claims

1. A bionic fish school collaborative control method integrating underwater monitoring and ecological restoration, characterized by: The following processes are included: Step 1: Determine whether there is a monitoring task. If water quality monitoring is not required, proceed to step 2. If water quality monitoring is required, proceed to step 3. Step 2: The bionic fish swarm cruises along the set route, detecting and recording the terrain and collecting water samples at regular intervals. Step 3: Real-time monitoring is performed according to the set monitoring route. The sensor monitors the water quality information in real time and compares the water quality information with the preset corresponding water quality index threshold. If the water quality information is abnormal, it will go to step 4; if it is normal, it will go directly to step 5. Step 4: Analyze the location of the pollution source based on the water quality data of the monitored points along the monitoring route, plan and change the route with the pollution source as the end point, and verify and confirm the location of the pollution source; Step 5: The control terminal generates a new operation path through Q-Learning optimization and performs dynamic adjustments based on the bionic fish school based on CNN-LSTM; Step 6: Determine whether the pollution needs to be repaired. If it needs to be repaired, proceed to step 7, where the bionic fish school performs in-situ repair. If it does not need to be repaired, repeat steps 1-5. Step 7: Determine whether the bionic fish school needs to continue operating. If so, loop through steps 1-6. If not, execute the return command.

2. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: The bionic fish school includes a bionic fish and several ordinary bionic fish. Each bionic fish is equipped with a data acquisition system, a data processing and command center, a navigation and communication module, a drive device and a power supply device. It is also equipped with a sampling device and a repair module. The repair module includes a probiotic slow-release capsule warehouse, a chitosan adsorption particle warehouse, and an electrolytic aeration unit. The data acquisition system includes a camera, an ultrasonic sensor and a water quality sensor; The power supply equipment includes a main power supply equipment and a backup power supply equipment.

3. The bionic fish school collaborative control method for integrated underwater monitoring and ecological restoration according to claim 2 is characterized in that: The data processing and command centers of all bionic fish are connected to the control center through a communication module. The control center records and analyzes the position coordinates of each bionic fish and the data it transmits, and sends motion instructions to each bionic fish. The drive device adjusts the bionic fish's swimming direction and speed in a timely and accurate manner according to the instructions of the control center.

4. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: During the bionic fish school operation, every 2 minutes, all bionic fish broadcast their own heartbeat packets, which include ID code, battery voltage, and power level; Detect whether the bionic fish has any abnormality by the following methods: Heartbeat packet monitoring: Check the ZigBee broadcast status code every 2 minutes, and judge it as abnormal if there is no response for 3 consecutive times; Neighbor node detection: Use ultrasonic ranging to scan nearby fish within a 10m range to detect whether their physical displacement is normal; Data flow analysis: The data processing system on board each bionic fish verifies the continuity of data packets in real time. If the data jump interval is greater than 5 minutes, the transmission is considered to be interrupted.

5. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: The specific process of step 3 is as follows: the control center first calculates and allocates Voronoi cells based on the operation area information and the status of the bionic fish device performing the task and generates a monitoring route. The bionic fish device performs regular or fixed-point monitoring according to the set monitoring route, and the onboard AI system processes the terrain and water quality data in real time, and optimizes and adjusts the monitoring route based on the monitored data and the path optimization algorithm; During the monitoring process, if a water quality indicator at a certain point exceeds the threshold, the bionic fish device first starts high-frequency monitoring and sample collection. The control center uses the CNN-LSTM algorithm to analyze the location of the pollution source based on the water quality data of the monitored points along the monitoring route and historical data of the area, and plans and changes the route with the pollution source as the end point. After verifying and confirming the location of the pollution source, the control center uses Q-Learning optimization to generate a new operation path and uses the deep learning algorithm CNN-LSTM to dynamically adjust the bionic fish school. If the water quality indicators at each point are normal during the monitoring process, the control center will optimize the operation path and adjust the dynamics of the bionic fish school based on the acquired data during the operation.

6. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: During underwater operations, when a single bionic fish device malfunctions, the device and its neighbors first determine the type of fault. If it is an irreparable fault, which includes damage to the device, loss of communication, or insufficient power, the fault is reported to the control terminal. After obtaining fault information and positioning, the faulty fish is promptly salvaged. If a local fault occurs that does not affect navigation and drive, the faulty fish returns directly to the set return point and reports the fault to the terminal, where it is recovered. Upon receiving the information that the bionic fish device has malfunctioned, the head fish immediately recalculates using the Voronoi algorithm based on the number and position of existing normal bionic fish devices, generates the optimal adjustment plan and issues instructions. All normally operating bionic fish devices change their operating paths and monitoring ranges according to the optimal adjustment plan.

7. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: When the pollution type is low dissolved oxygen, the remediation method is to start aeration and synchronize aeration in a multi-fish ring array; When the pollution type is excessive heavy metals, the remediation method is to release magnetic nano-adsorbents, and the fish school surrounds the pollution source to form a dynamic barrier; When the pollution type is organic pollution, the remediation method is to release probiotics sustained-release capsules and allocate regional responsibilities based on the Voronoi algorithm.

8. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: During the monitoring process, if abnormal water quality is found, the bionic fish device will report the location information and water quality information to the head fish and the control center. The control center will combine historical and real-time data and use deep learning algorithms to calculate and analyze the location of the pollution source. The head fish will calculate and reallocate the operation unit and assign more bionic fish devices to the pollution source and nearby areas for verification and intensive monitoring and sampling. The control center will determine in real time whether the monitoring point needs to be repaired in situ based on the water quality index monitoring data. If repair is required, the corresponding repair instructions will be issued according to the needs. After the repair is completed, the bionic fish will continue to operate or execute the return command.

9. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: The navigation and communication module uses an underwater acoustic positioning system to achieve navigation and a ZigBee / LoRa dual-mode adaptive module to achieve communication.

10. The bionic fish school collaborative control method integrating underwater monitoring and ecological restoration according to claim 1 is characterized in that: The control center is equipped with multiple machine learning models, and before use, it inputs real fish swimming videos, game theory models based on Shapley values, dual Q-Learning, and several emergencies and emergency cases.

Images