Water ecological environment monitoring ball

Abstract

The invention discloses a water ecological environment monitoring ball, which comprises a composite multilayer ball body and an intelligent analysis and decision support system, and is characterized in that the intelligent analysis and decision support system is used for deeply analyzing and processing water ecological environment monitoring data; the composite multi-layer sphere comprises a core function layer, a protection buffer layer and a shell protection layer from inside to outside, a carrying platform is installed at the sphere middle ring of the shell protection layer, the carrying platform is of an annular structure, a plurality of installation cylinders are arranged on the carrying platform in a penetrating mode, and electric lifting rods are installed in the installation cylinders; an ion selective electrode array is installed at the bottom end of the electric lifting rod, and a plurality of floating balls are connected to the periphery of the carrying platform. The composite multi-layer sphere structure can adapt to various complex water area environments, protect normal operation of internal equipment and maintain long-term underwater operation performance, and is compact and light in design, convenient to put and deploy, capable of working independently or in a networking mode and wide in application scene.

Description

Technical Field

[0001] The invention relates to the technical field of environmental monitoring, and in particular to a water ecological environment monitoring ball. Background Art

[0002] With the acceleration of industrialization and urbanization, the water ecological environment is facing increasingly severe challenges, such as frequent industrial wastewater discharge, agricultural non-point source pollution, and domestic sewage inflow, which have led to the deterioration of water quality and the imbalance of the water ecosystem. Traditional water ecological environment monitoring methods have many limitations. Manual monitoring methods consume a lot of manpower and material resources, and the monitoring frequency is low and the efficiency is poor, making it difficult to achieve real-time and continuous monitoring of the water ecological environment; fixed monitoring stations are expensive to build and have limited layout, and cannot fully cover complex and diverse water environments. There are often gaps in monitoring of remote areas or special waters; although satellite remote sensing monitoring can obtain information on a large area of ​​water, it is greatly interfered by factors such as clouds, water depth and water optical properties, and has obvious deficiencies in monitoring accuracy and details; although drone monitoring has certain flexibility, its endurance is limited, it is significantly affected by weather conditions, and its operation stability is poor in complex terrain and harsh climate, making it difficult to meet long-term and stable monitoring needs. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a water ecological environment monitoring ball.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] The present invention provides a water ecological environment monitoring ball, including a composite multi-layer ball and an intelligent analysis and decision support system, wherein the intelligent analysis and decision support system is used to perform in-depth analysis and processing on water ecological environment monitoring data, and the composite multi-layer ball comprises a core functional layer, a protective buffer layer and an outer shell protective layer from inside to outside:

[0006] The core functional layer has a multi-architecture microprocessor, a storage module and a control unit installed inside through a bracket, and they are connected to each other through a high-speed data bus;

[0007] The protective buffer layer is formed by combining nano rubber material and carbon fiber reinforced composite material, and the ratio of the nano rubber material to the carbon fiber reinforced composite material is 70%:30%;

[0008] The outer shell protective layer is made of transparent polyetheretherketone material, and its surface has special treatments such as special texture and nano titanium dioxide coating;

[0009] A carrying platform is installed at the middle circle of the sphere of the outer shell protective layer. The carrying platform is a ring structure. A plurality of mounting tubes are arranged through the carrying platform. An electric lifting rod is installed inside the mounting tube. An ion selective electrode array is installed at the bottom end of the electric lifting rod. A plurality of floating balls are connected to the periphery of the carrying platform.

[0010] A T-shaped bracket is installed at the bottom end of the outer shell protective layer, and a high-resolution underwater camera and a multi-band sonar sensor are installed at both ends of the T-shaped bracket.

[0011] As a preferred technical solution of the present invention, a flexible solar cell film is installed on the top of the outer shell protective layer, and the flexible solar cell film is an arc-shaped structure.

[0012] As a preferred technical solution of the present invention, a plurality of micro-turbine water flow power generation devices are installed at the bottom end of the carrying platform.

[0013] As a preferred technical solution of the present invention, a water quality spectrometer is installed inside the sphere, a flip switch motor is provided on the top of the water quality spectrometer, a flip sealing window is provided on the top of the front end of the composite multi-layer sphere, and the sealing window is opened and closed by the flip switch motor.

[0014] As a preferred technical solution of the present invention, the intelligent analysis and decision support system includes an intelligent analysis module and a decision support module. The decision support module automatically generates a targeted monitoring strategy adjustment plan based on the results of the intelligent analysis module. The intelligent analysis module is composed of a water quality assessment unit, an ecosystem health assessment unit, and a water ecological environment risk assessment unit.

[0015] The water quality assessment unit adopts a water quality assessment model that combines the long short-term memory network and the convolutional neural network. Through learning and training of massive historical water quality data and real-time monitoring data, it predicts the trend of water quality changes, identifies potential pollution sources, and generates an assessment report based on water quality standards. The long short-term memory network is used to process time series data. The calculation formula is as follows:

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] In the above formula, is at the time step The input data is a time series of water quality data points; is the hidden state of the previous time step; is the cell state at the previous time step; W and b are learnable weights and bias parameters, and different subscripts indicate their roles in different gating and calculations; is the sigmoid activation function, which is used to compress values ​​between 0 and 1 to determine the degree of information retention or discarding; tanh is the hyperbolic tangent activation function, which is used to transform the candidate values ​​of the cell state;

[0023] The convolutional neural network unit processes water quality spectrum data, and the calculation formula is as follows:

[0024] ;

[0025] In the above formula, It is The input feature map of the layer; It is The convolution kernel of the layer extracts the characteristic patterns in the spectrum through convolution operation; It is The bias term of the layer; It is The output feature map of the layer is processed by convolution and activation function and then passed to the next layer for feature extraction and analysis.

[0026] The ecosystem health assessment unit integrates biological monitoring data, hydrological data and environmental factors, and uses the ecological dynamics principle and the Bayesian network model of the ecosystem health model to assess the health status of the water ecosystem, warn of the risk of ecological imbalance, and provide cause analysis and response plans;

[0027] The water ecological environment risk assessment unit adopts a random forest algorithm, takes water quality data, biological monitoring data, hydrological data and surrounding human activity data as input features, predicts the direction of water ecological community succession and quantitatively assesses water ecological environment risks.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1: The present invention provides key data and reliable basis for water pollution assessment, biodiversity assessment and comprehensive water quality evaluation through all-round and accurate monitoring of the chemical, biological and physical dimensions of the water ecological environment, and has high measurement accuracy.

[0030] 2: The present invention uses intelligent analysis and decision support systems to accurately evaluate, warn and provide suggestions. The decision support module automatically generates monitoring strategy adjustment plans, realizes intelligent and automated monitoring, and improves monitoring efficiency and management level.

[0031] 3: The present invention uses solar energy and water flow energy generation devices in conjunction with supercapacitor groups to ensure continuous and stable power supply in different environments, solving the energy and endurance problems of traditional equipment.

[0032] 4: The composite multi-layer spherical structure of the present invention enables it to adapt to a variety of complex water environments, protect the normal operation of internal equipment, maintain long-term underwater operation performance, and has a compact and lightweight design, which is easy to deploy. It can work independently or in a network, has a wide range of application scenarios, and is suitable for routine and emergency monitoring, providing a convenient and efficient solution for water ecological protection and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0034] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0035] Figure 2 It is a schematic diagram of the internal structure of the installation tube of the present invention;

[0036] Figure 3 It is a schematic diagram of the internal structure of the composite multi-layer sphere of the present invention;

[0037] Figure 4 It is a schematic diagram of the intelligent analysis and decision support system of the present invention;

[0038] In the figure: 10, composite multi-layer sphere; 11, core functional layer; 12, protective buffer layer; 13, outer shell protective layer; 14, carrying platform; 15, installation tube; 16, electric lifting rod; 17, ion selective electrode array; 18, floating ball; 19, T-shaped bracket; 110, underwater camera; 111, multi-band sonar sensor; 112, flexible solar cell film; 113, water flow power generation device; 114, water quality spectrometer; 115, flip switch motor; 116, sealing window. DETAILED DESCRIPTION

[0039] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0040] Example 1

[0041] like Figure 1-4As shown, the present invention provides a water ecological environment monitoring ball, including a composite multi-layer ball 10 and an intelligent analysis and decision support system, the intelligent analysis and decision support system is used to perform in-depth analysis and processing on water ecological environment monitoring data, and the composite multi-layer ball 10 comprises a core functional layer 11, a protective buffer layer 12 and an outer shell protective layer 13 from inside to outside:

[0042] The core function layer 11 has a multi-architecture microprocessor, a storage module and a control unit installed inside through a bracket, and they are connected to each other through a high-speed data bus;

[0043] The protective buffer layer 12 is formed by combining nano rubber material and carbon fiber reinforced composite material, and the ratio of nano rubber material to carbon fiber reinforced composite material is 70%:30%;

[0044] The outer shell protective layer 13 is made of a transparent polyetheretherketone material, and its surface has special treatments such as special texture and nano titanium dioxide coating;

[0045] A carrying platform 14 is installed at the middle circle of the sphere of the outer shell protective layer 13. The carrying platform 14 is a ring structure. A plurality of mounting tubes 15 are arranged through the carrying platform 14. An electric lifting rod 16 is installed inside the mounting tube 15. An ion selective electrode array 17 is installed at the bottom end of the electric lifting rod 16. The ion selective electrode array 17 is mainly used to accurately detect specific ion components in water, such as heavy metal ions such as lead, mercury, and cadmium, nutrients such as nitrogen and phosphorus, and trace organic pollutants. The detection principle is based on the selective response of the ion selective electrode array to specific ions. A plurality of floating balls 18 are connected to the periphery of the carrying platform 14.

[0046] A T-shaped bracket 19 is installed at the bottom end of the outer shell protective layer 13 , and a high-resolution underwater camera 110 and a multi-band sonar sensor 111 are installed at both ends of the T-shaped bracket 19 .

[0047] Furthermore, a flexible solar cell film 112 is installed on the top of the outer shell protective layer 13, and the flexible solar cell film 112 is an arc-shaped structure. The flexible solar cell film 112 converts light energy into electrical energy. The circuit is connected to the built-in supercapacitor group to achieve efficient utilization and stable storage of solar energy, providing energy support for the continuous operation of the monitoring ball.

[0048] Furthermore, a plurality of micro-turbine water flow power generation devices 113 are installed at the bottom of the carrying platform 14. Water flows through the water flow power generation devices 113 to generate electricity, which cooperates with the flexible solar cell film 112 to ensure that the monitoring ball has sufficient power.

[0049] Furthermore, a water quality spectrometer 114 is installed inside the sphere. The water quality spectrometer 114 adopts a combination of a high-sensitivity photomultiplier tube and an array detector. A flip switch motor 115 is arranged on the top of the water quality spectrometer 114. A flippable sealing window 116 is opened on the top of the front end of the composite multi-layer sphere 10, and the sealing window 116 is controlled to open and close by the flip switch motor 115. The flip switch motor 115 opens the sealing window 116. The quantum cascade laser inside the water quality spectrometer 114 generates a high-power, narrow-linewidth laser beam to scan the water body in a wide spectral range to measure key indicators such as dissolved oxygen, chemical oxygen demand (COD), and biochemical oxygen demand (BOD) in the water.

[0050] Furthermore, the intelligent analysis and decision support system includes an intelligent analysis module and a decision support module. The decision support module automatically generates a targeted monitoring strategy adjustment plan based on the results of the intelligent analysis module. The intelligent analysis module consists of a water quality assessment unit, an ecosystem health assessment unit, and a water ecological environment risk assessment unit.

[0051] The water quality assessment unit uses a water quality assessment model that combines the long short-term memory network and the convolutional neural network. Through learning and training of massive historical water quality data and real-time monitoring data, it predicts water quality change trends, identifies potential pollution sources, and generates assessment reports based on water quality standards. The long short-term memory network is used to process time series data. The calculation formula is as follows:

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] ;

[0057] ;

[0058] In the above formula, is at the time step The input data is a time series of water quality data points; is the hidden state of the previous time step; is the cell state at the previous time step; W and b are learnable weights and bias parameters, and different subscripts indicate their roles in different gating and calculations; is the sigmoid activation function, which is used to compress values ​​between 0 and 1 to determine the degree of information retention or discarding; tanh is the hyperbolic tangent activation function, which is used to transform the candidate values ​​of the cell state;

[0059] The convolutional neural network unit processes water quality spectrum data, and the calculation formula is as follows:

[0060] ;

[0061] In the above formula, It is The input feature map of the layer; It is The convolution kernel of the layer extracts the characteristic patterns in the spectrum through convolution operation; It is The bias term of the layer;

[0062] It is The output feature map of the layer is processed by convolution and activation functions (such as ReLU, etc.) and then passed to the next layer for feature extraction and analysis.

[0063] In the water quality assessment model, the convolutional neural network uses a combination of multiple such convolutional layers and pooling layers to extract and analyze the characteristics of spatial information such as water quality spectral data, accurately identify the characteristic patterns of different substances in the spectrum, and assist in judging water quality conditions and potential pollution sources; the long short-term memory network unit is responsible for processing time series water quality data and capturing the dynamic changes of water quality parameters over time. The convolutional neural network unit extracts and analyzes the characteristics of spatial information such as water quality spectral data, and uses its combination of convolutional layers and pooling layers to accurately identify the characteristic patterns of different substances in the spectrum. Through deep learning training of massive historical water quality data and real-time monitoring data, it can accurately predict water quality change trends, identify potential pollution sources, and automatically generate assessment reports based on pre-set water quality standards;

[0064] The ecosystem health assessment unit integrates biological monitoring data, hydrological data and environmental factors, and uses the principles of ecological dynamics and the ecosystem health model of the Bayesian network model to assess the health of the water ecosystem, warn of ecological imbalance risks, and provide cause analysis and response plans; the Bayesian network model consists of multiple nodes and directed edges. The nodes represent different ecological variables (such as biological population size, water temperature, dissolved oxygen, etc.), and the directed edges represent the probabilistic dependency relationship between variables. Through a large amount of field monitoring data and expert knowledge, the parameters of the Bayesian network are learned and trained, so that multi-source uncertain information can be effectively integrated to infer and predict the health status of the ecosystem. It can timely warn of ecological imbalance risks, such as algae outbreaks, fish deaths and other abnormal events, and provide detailed cause analysis and response suggestions;

[0065] Bayesian networks are based on Bayes' theorem:

[0066] ;

[0067] In the ecosystem health assessment unit, for ecological variables A (such as biological population) and B (such as water temperature), the Bayesian network determines the conditional probability P(B / A), prior probability P(A) and P(B) through a large amount of field monitoring data and expert knowledge;

[0068] Assume there are n variables in the ecosystem , the joint probability distribution of the Bayesian network can be expressed as:

[0069] ;

[0070] In the above formula, Representation variables In this way, the Bayesian network model can effectively integrate multi-source uncertain information, infer and predict the health status of the ecosystem, timely warn of ecological imbalance risks, and provide cause analysis and response suggestions;

[0071] The water ecological environment risk assessment unit uses the random forest algorithm to take water quality data, biological monitoring data, hydrological data, and surrounding human activity data (such as industrial emissions, agricultural non-point source pollution, urban sewage discharge, etc.) as input features to predict the direction of water ecological community succession and quantitatively assess water ecological environment risks; the random forest algorithm predicts and evaluates by constructing multiple decision trees. For the construction process of a decision tree, taking the classification problem (such as judging whether an area is high-risk) as an example, the best split feature is selected at each node based on information gain or other splitting criteria (such as the Gini index):

[0072] ;

[0073] In the above formula, D is the data set of the current node; a is the split feature to be selected; V is the number of values ​​of feature a; is the subset of D with value v on feature a; Ent(D) is the information entropy of data set D, which is used to measure the uncertainty of the data set. The calculation formula is:

[0074] ;

[0075] In the above formula, K is the number of categories, is the proportion of samples belonging to the kth class in the dataset.

[0076] Specific application of the present invention:

[0077] The electric lifting rod 16 is accurately installed in the installation tube 15 of the carrying platform 14, and the ion selective electrode array 17 is connected to the bottom of the electric lifting rod 16 to ensure that the electrode is stable during the lifting process and can accurately collect water samples. A plurality of floating balls 18 are firmly connected to the periphery of the carrying platform 14 to ensure that they are evenly distributed so that the monitoring ball maintains a suitable floating posture and depth in the water. The high-resolution underwater camera 110 and the multi-band sonar sensor 111 are installed at both ends of the T-shaped bracket 19, and their angles and positions are adjusted so that they can fully cover the monitoring area and obtain clear underwater images and accurate acoustic signals. The water quality spectrometer 114 is installed at a suitable position inside the composite multi-layer sphere 10, and it is ensured that the flip switch motor 115 on the top is smoothly connected to the sealing window 116, and the sealing window 116 can be normally controlled to open and close, ensuring that the water quality spectrometer 114 can work normally after the sealing window 116 is opened. The arc structure of the flexible solar cell film 112 ensures that it can efficiently absorb light energy and convert it into electrical energy, and stably store it in the supercapacitor group. The micro-turbine water flow power generation device 113 adjusts the blade angle and the installation position so that it can work in coordination with the solar cell film 112 to meet the power demand of the monitoring ball.

[0078] After that, select the target water area and determine the placement location and method based on the characteristics of the water area (such as water flow speed, water depth, surrounding environment, etc.). You can use professional placement equipment or manual assistance to slowly place the monitoring ball into the water to avoid violent collisions that damage the equipment.

[0079] After the monitoring ball enters the water, the intelligent control unit automatically starts all sensors and devices. The electric lifting rod 16 lowers the ion selective electrode array 17 to a predetermined depth, collects water samples at preset time intervals (e.g., every 30 minutes), detects the concentration of heavy metal ions such as lead, mercury, and cadmium, nutrients such as nitrogen and phosphorus, and trace organic pollutants in the water, and transmits the data to the storage module of the core functional layer through a high-speed data bus. The high-resolution underwater camera 110 and the multi-band sonar sensor 111 work continuously. The underwater camera 110 captures images and videos of aquatic organisms at a certain frame rate (e.g., 5 frames per second), and the sonar sensor 111 transmits and receives sound wave signals to monitor the activity trajectory and distribution range of aquatic animals and transmit data in real time.

[0080] When water quality spectral analysis is required, the switch motor 115 is flipped to open the sealed window 116, and the quantum cascade laser inside the water quality spectrometer 114 generates a high-power, narrow-linewidth laser beam to scan the water body over a wide spectral range. The detector uses a combination of a high-sensitivity photomultiplier tube and an array detector to accurately measure key indicators such as dissolved oxygen, chemical oxygen demand (COD), and biochemical oxygen demand (BOD) in the water. After the measurement is completed, the sealed window 116 is closed and the data is transmitted to the storage module for storage and subsequent processing.

[0081] The water quality assessment unit of the intelligent analysis module processes data using a combination of long-term and short-term memory networks and convolutional neural networks. The long-term and short-term memory network unit processes time series water quality data according to formulas, such as hourly water temperature, pH value and other data points, to capture the dynamic changes of water quality parameters over time. The convolutional neural network unit extracts and analyzes spatial information such as water quality spectrum data according to formulas, and accurately identifies the characteristic patterns of different substances in the spectrum. Through deep learning training of massive historical water quality data and real-time monitoring data, it predicts water quality change trends, identifies potential pollution sources, and automatically generates assessment reports based on pre-set water quality standards. The ecosystem health assessment unit integrates biological monitoring data (such as biological images taken by underwater cameras, animal activities monitored by sonar), hydrological data (such as water temperature, water flow speed, etc.) and environmental factors (such as surrounding land use types), and uses ecological dynamics principles and Bayesian network models to assess the health of water ecosystems. Based on Bayes' theorem, the Bayesian network determines parameters such as conditional probability and prior probability through a large amount of field monitoring data and expert knowledge to infer and predict the health status of the ecosystem. For example, when a sudden increase in algae biomass and an increase in water temperature in a certain area are detected, timely warning of ecological imbalance risks (such as algae outbreaks) is issued, and detailed cause analysis and response suggestions are provided. The water ecological environment risk assessment unit uses a random forest algorithm, taking water quality data, biological monitoring data, hydrological data, and surrounding human activity data as input features. In the process of decision tree construction, the best split feature is selected according to splitting criteria such as information gain or Gini index, such as using industrial emission data as splitting features to judge the regional water ecological environment risk level, predict the direction of water ecological community succession, and quantitatively evaluate water ecological environment risks. The decision support module automatically generates targeted monitoring strategy adjustment plans based on the results of intelligent analysis. If water pollution worsens, such as the concentration of heavy metal ions in a certain area exceeds the standard value by 50%, the decision support module automatically increases the monitoring frequency of the area from the original monitoring every 2 hours to monitoring every 30 minutes. When the water ecological environment risk assessment determines that a certain area is a high-risk area, the decision support module automatically sends early warning information to the relevant management departments and provides detailed risk analysis reports and response measures, such as strengthening supervision of surrounding pollution sources, formulating emergency plans, and implementing ecological restoration projects, etc., to help management departments take effective prevention and control measures in a timely manner and reduce water ecological environment risks.

[0082] During the entire working process, the monitoring ball runs continuously, constantly collecting, processing and transmitting data, realizing all-round, dynamic monitoring and intelligent analysis of the water ecological environment, providing strong support for water ecological protection and management.

[0083] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A water ecological environment monitoring ball, characterized in that: The invention comprises a composite multi-layer sphere (10) and an intelligent analysis and decision support system, wherein the intelligent analysis and decision support system is used to perform in-depth analysis and processing on water ecological environment monitoring data. The composite multi-layer sphere (10) comprises, from inside to outside, a core functional layer (11), a protective buffer layer (12), and an outer shell protective layer (13): The core functional layer (11) has a multi-architecture microprocessor, a storage module and a control unit installed inside via a bracket, and they are connected to each other via a high-speed data bus; The protective buffer layer (12) is formed by combining a nano rubber material and a carbon fiber reinforced composite material, wherein the ratio of the nano rubber material to the carbon fiber reinforced composite material is 70%:30%; The outer shell protective layer (13) is made of a transparent polyetheretherketone material, and its surface has special treatments such as special texture and nano titanium dioxide coating; A carrying platform (14) is installed at the center circle of the sphere of the outer shell protective layer (13); the carrying platform (14) is an annular structure; a plurality of mounting tubes (15) are provided through the carrying platform (14); an electric lifting rod (16) is installed inside the mounting tube (15); an ion selective electrode array (17) is installed at the bottom end of the electric lifting rod (16); and a plurality of floating balls (18) are connected to the periphery of the carrying platform (14); A T-shaped bracket (19) is installed at the bottom end of the outer shell protective layer (13), and a high-resolution underwater camera (110) and a multi-band sonar sensor (111) are installed at both ends of the T-shaped bracket (19).

2. A water ecological environment monitoring ball according to claim 1, characterized in that: A flexible solar cell film (112) is installed on the top of the outer shell protective layer (13), and the flexible solar cell film (112) is an arc-shaped structure.

3. A water ecological environment monitoring ball according to claim 1, characterized in that: A plurality of micro-turbine water flow power generation devices (113) are installed at the bottom end of the carrying platform (14).

4. The water ecological environment monitoring ball according to claim 1, characterized in that: A water quality spectrometer (114) is installed inside the sphere, a flip switch motor (115) is arranged on the top of the water quality spectrometer (114), a flippable sealing window (116) is provided on the top of the front end of the composite multi-layer sphere (10), and the sealing window (116) is controlled to open and close by the flip switch motor (115).

5. The water ecological environment monitoring ball according to claim 1, characterized in that: The intelligent analysis and decision support system includes an intelligent analysis module and a decision support module. The decision support module automatically generates a targeted monitoring strategy adjustment plan based on the results of the intelligent analysis module. The intelligent analysis module consists of a water quality assessment unit, an ecosystem health assessment unit, and a water ecological environment risk assessment unit. The water quality assessment unit adopts a water quality assessment model that combines the long short-term memory network and the convolutional neural network. Through learning and training of massive historical water quality data and real-time monitoring data, it predicts the trend of water quality changes, identifies potential pollution sources, and generates an assessment report based on water quality standards. The long short-term memory network is used to process time series data. The calculation formula is as follows: ； ； ； ； ； ； In the above formula, is at the time step The input data is a time series of water quality data points; is the hidden state of the previous time step; is the cell state at the previous time step; W and b are learnable weights and bias parameters, and different subscripts indicate their roles in different gating and calculations; is the sigmoid activation function, which is used to compress values ​​between 0 and 1 to determine the degree of information retention or discarding; tanh is the hyperbolic tangent activation function, which is used to transform the candidate values ​​of the cell state; The convolutional neural network unit processes water quality spectrum data, and the calculation formula is as follows: ； In the above formula, It is The input feature map of the layer; It is The convolution kernel of the layer extracts the characteristic patterns in the spectrum through convolution operation; It is The bias term of the layer; It is The output feature map of the layer is processed by convolution and activation function and then passed to the next layer for feature extraction and analysis. The ecosystem health assessment unit integrates biological monitoring data, hydrological data and environmental factors, and uses the ecological dynamics principle and the Bayesian network model of the ecosystem health model to assess the health status of the water ecosystem, warn of the risk of ecological imbalance, and provide cause analysis and response plans; The water ecological environment risk assessment unit adopts a random forest algorithm, takes water quality data, biological monitoring data, hydrological data and surrounding human activity data as input features, predicts the direction of water ecological community succession and quantitatively assesses water ecological environment risks.

Images