Seawater exchange type breakwater management method based on intelligent control

Abstract

The invention discloses a seawater exchange type breakwater management method based on intelligent control, and particularly relates to the technical field of breakwater management. By constructing a multi-parameter correlation model, environmental data such as dissolved oxygen, turbidity and water temperature are collected in real time, the coupling relation is identified, and a risk factor value is calculated, so that a water change control decision is dynamically corrected; by controlling execution units such as a water gate, a water pump or a valve, water changing operation is accurately implemented, feedback data is collected after operation, the control effect is evaluated, then self-adaptive optimization of a next round of strategy is achieved, and the judgment accuracy, control flexibility and ecological safety of the system in a complex environment are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of breakwater management, and particularly relates to a breakwater management method of seawater exchange type based on intelligent control. Background Art

[0002] The management of seawater exchange type breakwater based on intelligent control refers to a system that uses intelligent technologies (such as sensors, automatic control systems and algorithms) to dynamically regulate and manage a breakwater with the function of seawater exchange. This type of breakwater not only has the traditional wave prevention function, but also can realize the automatic exchange of internal and external seawater, thereby improving the water quality of the port basin. The intelligent control system can optimize the opening and closing mechanism and exchange frequency according to real-time data such as tides, water quality, and wave heights, so as to achieve energy-saving, efficient and environmentally friendly sea area management.

[0003] The existing technologies have the following deficiencies: In the management of seawater exchange type breakwater based on intelligent control, when complex changes occur in the external environment (such as high temperature, heavy rain, and sediment entering the sea), resulting in conflicting trends in multiple water quality parameters, the static threshold strategy may trigger incorrect decisions. For example, when the dissolved oxygen is too low and water replacement is required, the turbidity of the outer sea is too high and water replacement is not appropriate, which may lead to secondary problems such as red tides, water quality deterioration or equipment damage. Summary of the Invention

[0004] The purpose of the present invention is to provide a breakwater management method of seawater exchange type based on intelligent control to solve the deficiencies in the background art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: A breakwater management method of seawater exchange type based on intelligent control, including: Real-time collection of environmental data through a number of sensors arranged inside and outside the breakwater and the port basin, the environmental data including dissolved oxygen, turbidity and water temperature; Construct a multi-parameter correlation model to identify and analyze the coupling relationship between dissolved oxygen, turbidity and water temperature indicators, calculate the risk factor value, and compare the score value with the set threshold. If the score value exceeds the threshold, the control decision is corrected; Based on the corrected control parameters, control the seawater exchange execution unit, including adjustable sluices, water pumps or valves, to implement seawater exchange operations; After the operation is completed, continue to collect feedback data to evaluate the correction effect of the control decision, and dynamically optimize the next round of control strategy according to the evaluation results.

[0006] Preferably, the sensors include dissolved oxygen sensors, turbidity sensors and water temperature sensors; the sensors are respectively arranged at the surface, middle and bottom water levels of the port basin, and are set at the port basin entrance, dead water area and outer sea exchange port.

[0007] Preferably, align the data from different sensors in time sequence, construct a multi-dimensional time series sample set composed of the triple of DO, turbidity, and water temperature, and establish a coupling relationship model between multiple parameters: input the historical monitoring data into the model, train to generate a long-term coupling baseline, and compare the real-time monitoring data with the expected coupling state of the model; if a certain parameter is lower than the set threshold and its coupling index indicates a trend of prohibiting water replacement, it is determined that there is a coupling conflict. The calculation expression of the risk factor value CCSI is: ; where i is the index of the i-th water quality index; n is the number of all relevant water quality parameters; the weight represents the importance of the water quality parameter in the system control decision, and the deviation represents the difference between the current value and the set ideal value; represents the grade score of the ecological or structural risk caused by the deviation; When , the system will prevent the water replacement operation and give an alarm; when CCSI is between 1 and 2.5, the system enables a cautious exchange strategy; when CCSI < 1 point, the system normally executes the water replacement control decision.

[0008] Preferably, the risk grade score represents the grade score of the ecological or structural risk caused by the deviation. The determination method is: for each out-of-bounds index, calculate its change trend and directional offset degree, and calculate the hypoxia index, sediment disturbance index, and heat stress index; convert the hypoxia index, sediment disturbance index, and heat stress index into a comprehensive feature vector, use the comprehensive feature vector as the input of the machine learning model, use the machine learning model to predict the risk grade score label for each group of comprehensive feature vectors as the prediction target, and use minimizing the sum of the prediction errors for all risk grade score labels as the training target to train the machine learning model until the sum of the prediction errors reaches convergence and stop the model training, and determine the risk grade score according to the model output result, where the machine learning model is a polynomial regression model.

[0009] Preferably, among them, the calculation expression of the hypoxia index is: ; where: is the set minimum safety threshold, is the current real-time dissolved oxygen value, represents the change rate within the past 15 minutes; represents the weighted factor of DO on ecological stability in the region; The calculation method of the sediment disturbance index is: collect the turbidity sequence within the past 24 hours, calculate the average value and the standard deviation , and collect the current turbidity value Estimate the mutation trend SAD based on the derivative change rate within the past T minutes. The expression is as follows: Obtain the tidal current or wind speed data for a period, and set the disturbance factor weight , and calculate the sediment disturbance index . The expression is as follows: ; Heat stress index The acquisition method is as follows: Collect the surface water temperature of the current port basin ; Calculate the temperature change rate in the past S hours ; Estimate the nutrient enrichment weight factor , look up the table / calculate the fuzzy membership degree according to the temperature value , and calculate the heat stress index ; .

[0010] Preferably, after the operation is completed, continue to collect the risk factor CCSI within a fixed time period and establish a data set, calculate the mean and standard deviation of the data set, analyze it, and evaluate the correction effect of the control decision according to the analysis result, and dynamically optimize the next round of control strategy according to the evaluation result.

[0011] Preferably, if the mean value of the risk factor in the data set is greater than or equal to the reference threshold of the risk factor mean value, and the standard deviation of the risk factor is less than the reference threshold of the risk factor standard deviation, perform a fine-tuning of the directional control strategy; If the mean value of the risk factor is greater than or equal to the reference threshold of the risk factor mean value, and the standard deviation of the risk factor is greater than or equal to the reference threshold of the risk factor standard deviation, trigger a full parameter recalibration mechanism or enter the manual intervention mode; If the mean value of the risk factor is less than the reference threshold of the risk factor mean value, and the standard deviation of the risk factor is greater than or equal to the reference threshold of the risk factor standard deviation, enhance the local feedback monitoring and dynamically optimize by region; If the mean value of the risk factor is less than the reference threshold of the risk factor mean value, and the standard deviation of the risk factor is less than the reference threshold of the risk factor standard deviation, the strategy is in the optimal state, and record the current parameter set as the subsequent control benchmark.

[0012] In the above technical solution, the technical effects and advantages provided by the present invention are as follows: 1. The present invention provides a management method for a seawater exchange type breakwater based on intelligent control, which overcomes the defect that the traditional static threshold strategy is prone to misjudgment in the face of complex environmental changes. By deploying multiple types of sensors to collect key water quality parameters such as dissolved oxygen, turbidity, and water temperature in real time, constructing a multi-parameter correlation model to identify the coupling relationship, and using a machine learning model to predict the risk level score, it realizes the accurate evaluation of the multi-index conflict state and the dynamic correction of control decisions, thereby effectively avoiding ecological and structural risks such as red tides, water quality deterioration, or equipment damage caused by inappropriate water exchange operations.

[0013] 2. Through a closed-loop feedback mechanism, the present invention collects risk factor data and evaluates the control effect after the operation is completed, and combines the dual criteria of mean and standard deviation to realize the dynamic analysis and hierarchical response of the optimization effect of the control strategy, thereby continuously improving the stability, self-adaptability, and operation efficiency of the system. The overall system has technical advantages such as high environmental perception accuracy, flexible response control, strong risk identification ability, and excellent ecological protection effect, and is applicable to various application scenarios such as port basins, hydraulic structures, and offshore ecological management. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0015] Figure 1 It is a method mind map of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Embodiment, please refer to Figure 1 As shown, a management method for a seawater exchange type breakwater based on intelligent control in this embodiment includes: Collecting environmental data in real time through a number of sensors arranged inside and outside the breakwater and the port basin, where the environmental data includes dissolved oxygen, turbidity, and water temperature; Build a multi-parameter correlation model to identify and analyze the coupling relationship among dissolved oxygen, turbidity, and water temperature indicators, calculate the risk factor value, and compare the scoring value with the set threshold. If the scoring value exceeds the threshold, correct the control decision; Based on the corrected control parameters, control the seawater exchange execution unit, including adjustable sluices, water pumps, or valves, to implement seawater exchange operations; After the operation is completed, continue to collect feedback data, evaluate the correction effect of the control decision, and dynamically optimize the next round of control strategies according to the evaluation results.

[0018] In the present invention, to achieve precise perception and dynamic response to the water environment state of the harbor basin, several environmental monitoring sensors are deployed inside and outside the breakwater and in key areas of the harbor basin. The types of sensors include, but are not limited to, dissolved oxygen sensors, turbidity sensors, and water temperature sensors. The specific deployment methods and data collection processes are as follows: Dissolved oxygen sensor: Select an electrochemical or optical fluorescence dissolved oxygen sensor with high sensitivity and anti-interference ability; the sensor is deployed in different water layers (such as the surface layer, middle layer, and bottom layer) to monitor the vertical oxygen distribution; continuously collect the dissolved oxygen concentration in the unit water body (unit: mg / L) to determine whether the water body is in an anoxic or eutrophic state; the sensor has an automatic calibration function and can perform self-check and correction regularly through the built-in standard solution or temperature compensation module.

[0019] Turbidity sensor: Adopt an infrared scattering or laser refraction optical sensor to monitor the concentration of suspended particulate matter in the water body; the unit is NTU (Nephelometric Turbidity Units, scattering turbidity unit), which reflects the clarity of the water body and the sediment content; the sensor is deployed at the outer sea exchange inlet and the dead water area of the harbor basin to facilitate the judgment of the risk of sediment backflow or pollutant retention; a dynamic threshold can be set to identify short-term turbidity surges caused by rainstorms or tides.

[0020] Water temperature sensor: Deploy a high-precision thermistor temperature sensor (such as PT1000) to measure the water temperature at different depths in the harbor basin and the outer sea; the unit of the collected data is degrees Celsius (°C), which is used to assist in identifying thermal stratification, water body exchange trends, and the risk of algal growth; conduct a joint analysis with the dissolved oxygen data to judge the trend of dissolved oxygen decline caused by rising water temperature; identify the "thermal pollution area" or "thermocline" existing in the harbor basin through the horizontal distribution sensor group.

[0021] The sensors are respectively deployed at the surface (0.5m), middle (2m), and bottom (5m) water levels of the harbor basin, and are set at the harbor basin entrance, dead water area, and outer sea exchange port. The data sampling frequency is once every 2 minutes and is synchronously uploaded to the central control system through the LoRa wireless network.

[0022] After the central system receives the original sensing data, it performs data integrity checks, extreme value elimination, and smoothing filtering; uses algorithms such as the moving window averaging method and median filter to correct instantaneous noise; for abrupt data, it is marked as an abnormal event and enters the abnormal mode assessment.

[0023] Through the above multi-point layout and multi-parameter synchronous acquisition method, the all-round and real-time monitoring of the water environment inside and outside the port basin is realized, providing accurate and reliable data support for subsequent seawater exchange decision-making based on intelligent control, and effectively improving the system's perception ability and response accuracy to complex environmental changes.

[0024] Align the data from different sensors in time series, and construct a multi-dimensional time series sample set composed of a triple of DO, turbidity, and water temperature to ensure the time synchronization of the data.

[0025] Establish a coupling relationship model between multi-parameters by using any one or a combination of the following methods: Pearson / Spearman correlation analysis: Identify linear or monotonic non-linear relationships; Principal component analysis (PCA): Extract the main causes of joint changes; Granger causality test: Judge the causal coupling direction between variables; Bayesian network model: Establish the probabilistic dependence relationship between indicators; Fuzzy entropy weight algorithm: Assign weights to the fluctuation degree of each indicator to form a multi-index weight model.

[0026] Input the historical monitoring data into the above model to train and generate a long-term coupling baseline. At the same time, the system regularly retrains the model in a sliding window manner to adapt to seasonal or sudden environmental changes.

[0027] Compare the real-time monitoring data with the expected coupling state of the model; If a certain indicator (such as DO) is in a severely abnormal state (such as below the set threshold), and its coupled indicators (such as high water temperature and high turbidity) indicate a trend of prohibiting water exchange, then a coupling conflict is determined.

[0028] The calculation expression of the risk factor value CCSI is: ; where, i is the index of the i-th water quality indicator (such as dissolved oxygen, turbidity, water temperature, etc.); n is the number of all relevant water quality parameters considered by the system; the weight represents the importance of the water quality parameter in the system control decision-making, and the deviation represents the difference between the current value and the set ideal value; represents the risk level score of the ecology or structure caused by the deviation; When the CCSI ≥ the high-risk threshold (set at 2.5 points for example), the system will block the water exchange operation and alarm; when the CCSI is in the intermediate range (such as 1–2.5 points), the system enables a cautious exchange strategy (such as delaying water exchange and reducing the water volume); when the CCSI < the safety threshold (such as 1 point), the system normally executes the water exchange control decision.

[0029] Among them, the weights are determined by the entropy weight method, specifically including: collecting historical data over a period of time; calculating the information entropy of each index, that is, the greater the change degree of the index value, the richer the information it contains; the weight is proportional to the information entropy of the index, indicating its "influence" or "contribution of fluctuations to the decision-making".

[0030] Risk level score represents the level score of the ecological or structural risk caused by the deviation, and the determination method is as follows: The system continuously compares the real-time collected data with the threshold to judge whether any of the following conditions are met: Dissolved oxygen DO < threshold (such as 4.5 mg / L); turbidity > threshold (such as 100 NTU); water temperature > threshold (such as 30°C); once any index exceeds the limit, the system enters the abnormal state discrimination mode.

[0031] For each out-of-limit index, calculate its change trend (derivative) and the degree of directional deviation, and introduce a certain environmental weight parameter to define the directionality of its abnormal behavior: : Hypoxia index, reflecting the rate and amplitude of oxygen decline; : Sediment disturbance index, reflecting whether the increase in turbidity is due to external strong disturbances (such as heavy rain, pump flow); : Heat stress index, reflecting the possible ecological stress response (such as red tide) caused by the increase in water temperature.

[0032] Among them, the hypoxia index The calculation expression is: ; where: is the set minimum safety threshold, is the current real-time dissolved oxygen value, represents the change rate in the past 15 minutes; represents the weighted factor of DO on ecological stability in the region.

[0033] Sediment disturbance index The calculation method is: collect the turbidity sequence in the past 24 hours and calculate the average value and the standard deviation , and collect the current turbidity value in real time ; Estimate the mutation trend SAD according to the derivative change rate within the past T minutes. The expression is as follows: ; Obtain the tide or wind speed data during this period and set the disturbance factor weight , which depends on the tide speed at the port pool entrance, wind speed, drainage flow rate, etc. The typical value is 0.5–1.5; Calculate the sediment disturbance index , which is used to judge whether the abnormal turbidity is caused by typical disturbances. The expression is as follows: .

[0034] Heat stress index The acquisition method is as follows: Collect the current surface water temperature of the port pool ; Calculate the temperature change rate in the past S hours (which can be smoothed) ; Query the current total nitrogen, total phosphorus or algae density, and estimate the nutrient enrichment weight factor , which is set according to the nitrogen and phosphorus concentration / algae density. The typical value is 0.8–1.6; Look up the table / calculate the fuzzy membership degree according to the temperature value , and calculate the heat stress index ; .

[0035] Convert the hypoxia index, sediment disturbance index and heat stress index into a comprehensive feature vector, and use the comprehensive feature vector as the input of the machine learning model. The machine learning model takes predicting the risk level score label for each group of comprehensive feature vectors as the prediction target, and takes minimizing the sum of the prediction errors for all risk level score labels as the training target. Train the machine learning model until the sum of the prediction errors reaches convergence and then stop the model training. Determine the risk level score according to the model output result, where the machine learning model is a polynomial regression model.

[0036] For example, the system automatically collects the following real-time data: Dissolved oxygen (DO) in the middle layer of the port pool: 2.3 mg / L (the set safety threshold is 4.5 mg / L); Outer seawater temperature: 31.5°C (the set reference threshold is 30°C); Outer sea turbidity: 128 NTU (the set water exchange safety threshold is 100 NTU); The system calls the trained Bayesian network model to identify the following coupling relationships: High temperature → Decrease in DO (confidence = 0.84); High turbidity → Increase in the risk of algal bloom (confidence = 0.91); Combination of high temperature + high turbidity → Increase in the probability of red tide after aeration or water exchange (confidence = 0.78); Calculate the risk factor value CCSI: ; The correlation strength W between turbidity and temperature is 0.85; The combined risk factor R is 1.4 (evaluated according to the ecological consequence level); Calculation result: CCSI = 0.85 × 2.2 × 1.4 = 2.618; The system maps CCSI to a 0–5 scale, where ≥2.5 is the high-risk area.

[0037] Since the risk factor is higher than the high-risk threshold (2.5), the system determines that changing water at this time may exacerbate ecological deterioration, so the following strategies are adopted: Suspend the water-changing operation; Start the bottom aeration system in the harbor basin to relieve the too low dissolved oxygen; Record this coupling event as a dynamic learning sample for the subsequent model.

[0038] If the DO does not show obvious improvement within 30 minutes and the turbidity in the open sea drops to the safe range at the same time, the system will re-evaluate whether to resume the water-changing operation.

[0039] In the method described in the present invention, after the central control system completes the real-time acquisition and intelligent analysis of multiple water quality parameters such as dissolved oxygen, turbidity, and water temperature, and outputs the corrected control parameters (including the water-changing requirement level, water-changing volume, duration, execution period, etc.) through the machine learning model, it automatically controls the seawater exchange execution unit according to the control parameters, so as to achieve accurate, efficient, and eco-friendly seawater exchange operations.

[0040] The corrected control parameters include but are not limited to: Water-changing operation instruction (execute / suspend); Water-changing flow rate (unit: L / s or m³ / h); Water-changing duration (unit: minute); Gate opening angle (0°–90°); Valve opening and closing state (open / close / regulate); Water pump start-stop state and power level.

[0041] Adjustable sluice structure: Electric screw gate or hydraulic-driven gate body; Control method: Set the opening angle according to the instruction (such as opening 40% and lasting for 15 minutes); Function: Regulate the water exchange speed and capacity between the harbor basin and the open sea; Feature: Fast response speed, fine adjustment available, suitable for tide-level linkage control.

[0042] Variable-frequency water pump structure: Centrifugal pump or axial flow pump with adjustable speed; Control method: Adjust the operating frequency by the central system to achieve flow rate change (such as operating at 60% power and lasting for 20 minutes); Function: Assist water body flow, especially for dead water areas and deep water exchange areas; Feature: Strong controllability, suitable for long-time and quantitative water changing.

[0043] Intelligent valve types: solenoid valves, electric control valves; Control method: automatically adjust the valve opening according to pressure feedback; Function: control unidirectional flow, avoid backflow or overshoot; Feature: can be linked with the water quality feedback system.

[0044] The control process includes: the central control system receives the control parameters output by the intelligent model; calculates the optimal execution time according to the control priority and operation window period (such as tidal periods); issues control commands to the execution unit; the controller drives the corresponding actuators (such as water pump frequency converters, gate motors) to adjust the state; collects the execution status feedback in real time to determine whether the preset water exchange target is completed; if the control effect deviates from the expectation, the system triggers an adjustment mechanism for parameter compensation or manual intervention prompt. After the operation is completed, continue to collect feedback data to evaluate the correction effect of the control decision, and dynamically optimize the next round of control strategies according to the evaluation results.

[0045] After the operation is completed, continue to collect the risk factor CCSI within a fixed time period and establish a data set, calculate the mean and standard deviation of the data set, analyze it, and then evaluate the correction effect of the control decision according to the analysis results, and dynamically optimize the next round of control strategies according to the evaluation results.

[0046] If the mean value of the risk factors in the data set is greater than or equal to the reference threshold of the risk factor mean value, and the standard deviation of the risk factors is less than the reference threshold of the risk factor standard deviation, it is determined that although the current strategy has a relatively high risk level, the system response is stable and the risks are concentrated on specific factors. The system performs targeted fine-tuning of the control strategy, such as optimizing a certain type of parameter (such as the DO water exchange threshold).

[0047] If the mean value of the risk factors is greater than or equal to the reference threshold of the risk factor mean value, and the standard deviation of the risk factors is greater than or equal to the reference threshold of the risk factor standard deviation, it is determined that the current strategy has a high risk and the system fluctuates greatly, and there is a risk of chaotic control logic or model deviation in the system. The system will trigger a comprehensive parameter recalibration mechanism or enter the manual intervention mode.

[0048] If the mean value of the risk factors is less than the reference threshold of the risk factor mean value, and the standard deviation of the risk factors is greater than or equal to the reference threshold of the risk factor standard deviation, it is determined that the current control strategy is overall effective, but there are unstable points in some areas or parameters. The system enhances local feedback monitoring and dynamically optimizes by region (such as local water exchange / zoned regulation).

[0049] If the mean value of the risk factors is less than the reference threshold of the risk factor mean value, and the standard deviation of the risk factors is less than the reference threshold of the risk factor standard deviation, it is determined that the current control strategy is effective and the system response is stable. The system confirms that this strategy is the optimal state and records the current parameter set as the subsequent control benchmark.

[0050] The present invention realizes the post-evaluation of control based on the mean score + volatility (stability); it does not rely on single-point judgment, improving the fault tolerance of the system for abnormal control performance; it supports the switching among three modes: automatic optimization, adaptive learning, and expert intervention; and it enhances the closed-loop ability, evolution ability, and operational robustness of the system.

[0051] All the above formulas are dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0052] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other arbitrary combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0053] It should be understood that the term "and / or" in this text is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this text generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, and the specific meaning can be understood by referring to the context before and after. Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this text can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0054] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, and all should be covered within the protection scope of this application.

Claims

1. A management method for a seawater-exchange type breakwater based on intelligent control, characterized in that: Including: Collecting environmental data in real time through a number of sensors installed inside and outside the breakwater and the harbor basin. The environmental data includes dissolved oxygen, turbidity, and water temperature; Constructing a multi-parameter correlation model to identify and analyze the coupling relationship between dissolved oxygen, turbidity, and water temperature indicators, calculating the risk factor value, and comparing the score value with the set threshold. If the score value exceeds the threshold, the control decision is corrected; Controlling the seawater exchange execution unit based on the corrected control parameters, including adjustable sluices, water pumps, or valves, to implement seawater exchange operations; After the operation is completed, continue to collect feedback data to evaluate the correction effect of the control decision, and dynamically optimize the next round of control strategies according to the evaluation results.

2. The management method of a seawater exchange type breakwater based on intelligent control according to claim 1, characterized in that: The sensors include dissolved oxygen sensors, turbidity sensors, and water temperature sensors; the sensors are respectively arranged at the surface, middle, and bottom water levels of the harbor basin, and are set at the entrance of the harbor basin, the dead water area, and the offshore exchange port.

3. A seawater exchange type breakwater management method based on intelligent control according to claim 1, characterized in that: Align the data from different sensors in time series, construct a multi-dimensional time series sample set composed of DO, turbidity, and water temperature triples, and establish a coupling relationship model between multi-parameters: input historical monitoring data into the model, train to generate a long-term coupling baseline, and compare the real-time monitoring data with the expected coupling state of the model; if a certain parameter is lower than the set threshold, and its coupling index indicates a trend of prohibiting water exchange, it is determined that there is a coupling conflict; The calculation expression of the risk factor value CCSI is as follows: ; where i is the index of the i-th water quality indicator; n is the number of all relevant water quality parameters; the weight represents the importance of the water quality parameter in the system control decision-making, and the deviation represents the difference between the current value and the set ideal value; represents the level score of the ecological or structural risk caused by the deviation; When CCSI≥2.5, the system will prevent the water exchange operation and alarm; when CCSI is between 1 and 2.5, the system enables a cautious exchange strategy; when CCSI<1 point, the system normally executes the water exchange control decision.

4. A management method for a seawater exchange type breakwater based on intelligent control according to claim 3, characterized in that: Risk level score It represents the level score of ecological or structural risks caused by deviations. The determination method is as follows: for each out-of-bounds indicator, calculate its change trend and directional deviation degree, and calculate the hypoxia index, sediment disturbance index, and heat stress index; convert the hypoxia index, sediment disturbance index, and heat stress index into a comprehensive feature vector, use the comprehensive feature vector as the input of the machine learning model, take the prediction of the risk level score label for each group of comprehensive feature vectors by the machine learning model as the prediction target, take minimizing the sum of prediction errors for all risk level score labels as the training target, train the machine learning model until the sum of prediction errors reaches convergence and then stop the model training, and determine the risk level score according to the model output result. Among them, the machine learning model is a polynomial regression model.

5. A management method for a seawater exchange type breakwater based on intelligent control according to claim 4, characterized in that: Wherein, Hypoxia index The calculation expression is as follows: Where: is the set minimum safety threshold, is the current real-time dissolved oxygen value, represents the change rate within the past 15 minutes; represents the weighted factor of DO on ecological stability within the region; Sediment disturbance index The calculation method is as follows: collect the turbidity sequence within the past 24 hours and calculate the average value and the standard deviation , and collect the current turbidity value in real time ; estimate the mutation trend SAD according to the derivative change rate within the past T minutes, and the expression is: ; obtain the tidal current or wind speed data of the period, set the disturbance factor weight , and calculate the sediment disturbance index , and the expression is: ; Method for obtaining heat stress index is as follows: Collect the surface water temperature of the current harbor basin ; Calculate the temperature change rate in the past S hours ; Estimate the nutrient enrichment weight factor , look up the table / calculate the fuzzy membership degree according to the temperature value , calculate the heat stress index ; .

6. A management method for a seawater-exchange type breakwater based on intelligent control according to claim 5, characterized in that: After the operation is completed, continue to collect the risk factor CCSI within a fixed time period and establish a data set, calculate the mean and standard deviation of the data set, analyze it, and then evaluate the correction effect of the control decision according to the analysis results, and dynamically optimize the next round of control strategies according to the evaluation results.

7. A management method for a seawater exchange type breakwater based on intelligent control according to claim 6, characterized in that: If the mean value of the risk factor in the data set is greater than or equal to the reference threshold of the mean value of the risk factor, and the standard deviation of the risk factor is less than the reference threshold of the standard deviation of the risk factor, perform targeted fine-tuning of the control strategy; If the mean value of the risk factor is greater than or equal to the reference threshold of the mean value of the risk factor, and the standard deviation of the risk factor is greater than or equal to the reference threshold of the standard deviation of the risk factor, a full parameter recalibration mechanism will be triggered or the manual intervention mode will be entered; If the mean value of the risk factor is less than the reference threshold of the mean value of the risk factor, and the standard deviation of the risk factor is greater than or equal to the reference threshold of the standard deviation of the risk factor, enhance local feedback monitoring and dynamically optimize by region; If the mean value of the risk factor is less than the reference threshold of the mean value of the risk factor, and the standard deviation of the risk factor is less than the reference threshold of the standard deviation of the risk factor, the strategy is in the optimal state, and record the current parameter set as the subsequent control benchmark.

Images