Mother-son ship system and collaborative deployment method thereof

Abstract

The invention provides a son-mother ship system and a collaborative deployment method thereof, and relates to the technical field of water conservancy projects. According to the method, an initial diffusion candidate region is identified by extracting pollutant spatial texture features; analyzing local disturbance characteristics of a flow field according to water area flow field data to determine an initial deployment position of a sub-ship; collecting pollutant component data in real time, extracting interaction characteristics among pollutant components, inputting the interaction characteristics into the pollutant fine diffusion model, and determining a pollutant enrichment area; and establishing an intersection feature set of the pollutant enrichment area and the stable diffusion path of the flow field, planning a cooperative cruise trajectory of the primary ship and the secondary ship, and implementing cooperative control, so that accurate identification and efficient cooperative monitoring of the pollutant diffusion area under the complex flow field condition are realized.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering technology, and in particular to a mother-daughter ship system and its collaborative deployment method. Background Technology

[0002] With the rapid development of industrialization and urbanization in my country, the environmental pollution problems faced by rivers, lakes and other water bodies are becoming increasingly severe. Affected by the complex hydrological and hydrodynamic conditions within the basin, the diffusion and migration process of pollutants exhibits spatiotemporal heterogeneity and multi-scale characteristics. Traditional monitoring methods are difficult to accurately capture the diffusion trends and enrichment patterns of pollutants, making it difficult to implement pollution control measures precisely and hindering the effective implementation of watershed water environment protection and management.

[0003] Existing water pollution monitoring technologies typically rely on fixed stations or single-vehicle patrol methods, which have drawbacks such as limited monitoring range, insufficient pollutant identification accuracy, and untimely response to complex water flow field characteristics. In addition, they lack the ability to coordinate the analysis of pollutant component interaction characteristics and local disturbance characteristics of water flow field, resulting in low efficiency in identifying polluted areas and failing to meet the urgent needs of current refined water environment management.

[0004] Therefore, there is an urgent need to develop an efficient monitoring method that can simultaneously take into account the spatial diffusion characteristics of pollutants and the characteristics of water flow fields, so as to improve the accurate identification and response capabilities of pollutant diffusion trends under complex flow field conditions, thereby providing strong technical support for the refined prevention and control of water pollution and the precise management of the water environment. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a mother ship system and a method for its coordinated deployment.

[0006] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a mother-daughter ship system, comprising a mother ship and multiple daughter ships. The mother ship is equipped with a path planning module, as well as an image acquisition module and a flow field acquisition module communicatively connected to the path planning module. Each daughter ship is equipped with a component acquisition module and a spatial analysis module communicatively connected to the component acquisition module. The spatial analysis module and the path planning module perform task coordination through a communication connection when performing tasks.

[0007] Secondly, the present invention provides a method for collaborative deployment of mother and daughter ships applied to the system, comprising the following steps: Based on the pollutant image acquired by the image acquisition module, the spatial texture features of the pollutants are extracted, and the initial diffusion candidate regions of the pollutants are identified based on the spatial texture features. Based on the water flow field data acquired by the flow field acquisition module, the set of local flow field disturbance features between the initial diffusion candidate region and the water flow field is identified, and the initial deployment position of each sub-vessel is determined by the path planning module using the set of local flow field disturbance features. The component acquisition module acquires pollutant component data at each initial deployment location, and the spatial analysis module extracts the set of interaction features between pollutant components; the set of interaction features is then input into a pre-constructed fine-grained pollutant diffusion model to determine pollutant enrichment areas. Establish a set of flow field stable diffusion path intersection features between the pollutant enrichment area and the water flow field data, and use the path planning module to generate a mother ship and daughter ship cooperative cruise trajectory covering the pollutant enrichment area; perform cooperative control of the daughter ship and mother ship based on the cooperative cruise trajectory.

[0008] Furthermore, the extraction process of the spatial texture features of the pollutants includes: Multi-scale texture decomposition processing with embedded gradient orientation sensitive constraints is performed on pollutant images to generate multi-scale pollutant spatial gradient feature maps. The multi-scale pollutant spatial gradient feature map is subjected to region adaptive fusion processing to obtain pollutant spatial texture features that can reflect the spatial diffusion trend of pollutants. .

[0009] Furthermore, the method for identifying the initial diffusion candidate region of the pollutant includes: Based on the spatial texture characteristics of pollutants The Otsu's method was used to establish a set of diffusion-sensitive candidate regions; A clustering and fusion method based on spatial continuity constraints is used to perform regional fusion on the set of diffusion-sensitive candidate regions to determine the initial candidate regions for pollutant diffusion. .

[0010] Furthermore, the method for identifying the set of local disturbance features in the flow field includes: Based on the water flow field data, a disturbance region identification method with embedded velocity vector space gradient constraints is used to determine the set of local disturbance regions. ; Utilizing the initial diffusion candidate region of pollutants Set of local disturbance regions Spatial correlation matching is performed to obtain the set of local perturbation features in the flow field. .

[0011] Furthermore, the process for determining the initial deployment positions of each sub-ship is as follows: Based on the set of local disturbance characteristics of the flow field With the initial diffusion candidate region of pollutants The degree of spatial correlation between them, for the set of local disturbance regions Sensitivity levels are classified, and the sensitivity level classification results are obtained; Based on the sensitivity level classification results, combined with the preset number of sub-ships and the preset deployment area priority conditions, the initial deployment location of each sub-ship is determined.

[0012] Furthermore, the process of generating the set of interaction features among the pollutant components includes: Based on the pollutant component data, calculate the spatial concentration gradient of each pollutant component. ; Based on the spatial concentration gradient of each pollutant component Identify the concentration gradient variation regions of each pollutant component. ; For the concentration gradient change region Spatial statistical analysis was conducted to obtain a set of interaction characteristics reflecting the degree of mutual influence among pollutant components. .

[0013] Furthermore, the process of determining the pollutant enrichment zone includes: Set of interaction characteristics among pollutant components The pre-constructed fine-grained pollutant diffusion model is input into a pre-built model for multiple iterative prediction calculations to obtain the spatial concentration distribution of pollutants after the concentration reaches a steady state. Based on the spatial concentration distribution, spatial peak regions are identified to obtain pollutant enrichment areas. .

[0014] Furthermore, the method for constructing the set of features of stable diffusion paths in the flow field includes: Determine the stable spatial distribution area of ​​water flow based on water flow field data. ; According to pollutant enrichment areas Determine the stable distribution area of ​​water flow spatial trend. With pollutant-rich areas Spatial sensitive intersection points; By analyzing the spatial accumulation indicators of pollutants at spatially sensitive intersection locations, a set of characteristics of stable diffusion paths in the flow field is obtained. .

[0015] Furthermore, the method for generating the cooperative cruising trajectory of the mother and daughter ships includes: Based on the set of characteristics of the convergence of stable diffusion paths in the flow field A path optimization algorithm is used to determine the mother ship's main cruising trajectory. ; Based on pollutant enrichment areas The spatial gridding method, combined with the path planning algorithm, is used to determine the auxiliary cruise trajectory of the sub-ship. ; A trajectory fusion method with embedded spatial position constraints is used to analyze the main cruise trajectory of the mother ship. With the auxiliary cruise trajectory of the sub-ship By integrating the data, a coordinated cruise trajectory of the mother and daughter ships covering pollutant-rich areas can be obtained. .

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention effectively solves the problem that traditional monitoring methods cannot accurately identify the pollutant diffusion area by accurately extracting and correlating the spatial texture features of pollutants and the local disturbance features of the flow field. It achieves rapid locking of the initial area of ​​pollution diffusion and improves the monitoring accuracy and rapid response capability of pollution sources.

[0017] This invention improves the accuracy of predicting pollutant enrichment areas by establishing a set of interaction features of pollutant components and combining it with a fine-grained pollutant diffusion model. It effectively overcomes the problem of poor targeting of pollution identification and treatment measures caused by insufficient analysis of component interaction relationships in existing technologies.

[0018] This invention achieves comprehensive coverage of the monitoring range and efficient implementation of collaborative tasks by optimizing the planning of the collaborative cruise trajectory of the mother and daughter ships, thereby improving the real-time performance, accuracy, and overall efficiency of water pollution monitoring and treatment under complex hydrological and hydrodynamic conditions. Attached Figure Description

[0019] Figure 1 This is an architectural diagram of a mother-daughter ship system according to Example 1.

[0020] Figure 2 This is a flowchart of a mother-daughter ship collaborative deployment method in Example 2. Detailed Implementation

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1 Please see Figure 1This invention provides a mother-daughter ship system, including a mother ship and multiple daughter ships. The mother ship is equipped with a path planning module, as well as an image acquisition module and a flow field acquisition module that are communicatively connected to the path planning module. Each daughter ship is equipped with a component acquisition module and a spatial analysis module that is communicatively connected to the component acquisition module. The spatial analysis module and the path planning module coordinate their tasks through a communication connection when performing tasks.

[0023] It should be noted that, in this embodiment, the path planning module on the mother ship is the core unit of the collaborative operation. Specifically, it is used to plan the initial position deployment of the sub-ship, the dynamic cruise trajectory planning, and the allocation of collaborative operation tasks between the mother ship and the sub-ship based on the data obtained by the image acquisition module and the flow field acquisition module.

[0024] In specific implementation, the image acquisition module on the mother ship preferably adopts a high-resolution optical imaging sensor installed on the top or side of the mother ship, such as a visible light camera or multispectral imager with a resolution of not less than 0.1 meters, to acquire pollutant images of the water area to be monitored in real time. The pollutant images specifically include the spatial distribution characteristics of pollutants, such as the spatial outline and texture features of the pollutant diffusion area, so as to be used by the path planning module to identify and analyze the initial area of ​​pollutant diffusion.

[0025] The flow field acquisition module of the mother ship is preferably located at the bottom of the mother ship's hull, specifically including an acoustic Doppler velocity profiler ( ADCP (or multi-point ultrasonic flow velocity measurement equipment) to acquire flow field data of the water area to be monitored in real time, including but not limited to flow velocity magnitude, flow direction and local flow field disturbance characteristics. The flow field data is provided to the path planning module in real time to realize the initial planning and optimization adjustment of the deployment position of the sub-vessel.

[0026] The component acquisition module on each sub-boat is specifically a water quality multi-parameter online monitoring device or pollutant component on-site analysis equipment installed on the bottom or side of the sub-boat. Preferably, the component acquisition module can measure the component concentration information of multiple pollutants in the water in real time, such as typical water pollution parameters such as ammonia nitrogen, nitrate, phosphate and organic pollutant concentrations.

[0027] The spatial analysis module configured in each sub-ship is specifically used to extract the interaction characteristics between pollutant components based on the pollutant component concentration data obtained in real time by the component acquisition module. Specifically, this includes the spatial gradient distribution characteristics of pollutant component concentration, the spatial variation trend of component concentration, and the interaction relationship between different pollutant components (e.g., mutual promotion and mutual inhibition relationships between components). The pollutant interaction characteristics are then transmitted in real time to the path planning module of the mother ship so that the path planning module can further plan the coordinated cruise mission of the mother and sub-ships.

[0028] To achieve task collaboration between the spatial analysis module and the path planning module, real-time data transmission between the mother ship and the liner ship is preferably achieved via wireless communication. This wireless communication method specifically includes, but is not limited to, 5... G Wireless communication networks, satellite communication links, or self-organizing network communication links ( Ad-hoc This ensures that the path planning module can receive pollutant interaction characteristic information from each sub-ship in real time and update the collaborative cruise path of the mother and daughter ships in real time, thereby achieving collaborative control of the mother and daughter ship system.

[0029] Example 2 like Figure 2 As shown, this embodiment discloses a collaborative deployment method for mother and daughter ships as described in Embodiment 1, including the following steps: Based on the pollutant image acquired by the image acquisition module, the spatial texture features of the pollutants are extracted, and the initial diffusion candidate regions of the pollutants are identified based on the spatial texture features. Based on the water flow field data acquired by the flow field acquisition module, the set of local flow field disturbance features between the initial diffusion candidate region and the water flow field is identified, and the initial deployment position of each sub-vessel is determined by the path planning module using the set of local flow field disturbance features. The component acquisition module acquires pollutant component data at each initial deployment location, and the spatial analysis module extracts the set of interaction features between pollutant components; the set of interaction features is then input into a pre-constructed fine-grained pollutant diffusion model to determine pollutant enrichment areas. Establish a set of flow field stable diffusion path intersection features between the pollutant enrichment area and the water flow field data, and use the path planning module to generate a mother ship and daughter ship cooperative cruise trajectory covering the pollutant enrichment area; perform cooperative control of the daughter ship and mother ship based on the cooperative cruise trajectory.

[0030] Specifically, the extraction process of the spatial texture features of the pollutants includes: Multi-scale texture decomposition processing with embedded gradient orientation sensitive constraints is performed on pollutant images to generate multi-scale pollutant spatial gradient feature maps. During implementation, this embodiment acquires image data of pollutants in the monitored water area in real time through the image acquisition module as described in Embodiment 1; the pollutant image data is specifically a grayscale image or RGB Color image; to obtain texture features that accurately reflect the diffusion trend of pollutants, this embodiment further constructs a multi-scale texture decomposition structure embedded with gradient direction sensitive constraints, the implementation details of which are as follows: First, the acquired original image of pollutants is input into a multi-scale texture decomposition structure, which includes multiple parallel filtering branches with convolution kernels of different sizes. The preferred kernel sizes of the specific filtering branches are, but not limited to, 3×3, 5×5, and 7×7. Furthermore, to accurately extract direction-sensitive gradient features from pollutant images, each filter branch embeds gradient direction-sensitive constraints, specifically expressed as follows: ; In the formula, Indicates the pollutant image at the orientation angle The gradient feature map generated after filtering is shown below. This represents the original image of the contaminants input. This represents a direction-sensitive gradient filter. This represents a two-dimensional convolution operation.

[0031] It should be understood that the direction-sensitive gradient filter described in this embodiment Using a Gaussian derivative filter, specifically expressed as: ; In the formula, This represents the scaling parameter of the Gaussian derivative filter, used to adjust the scaling range of the filter, with a value ranging from 0.5 to 2.0; This indicates the gradient filtering direction angle, such as 0°, 45°, 90°, 135°, etc. This represents the local coordinates of the filter after the coordinate transformation.

[0032] It should be noted that the local coordinates of the filter after rotational coordinate transformation The rotation method is as follows: ; in, This indicates the coordinate position in the original image coordinate system.

[0033] It is understandable that by using the above-mentioned multi-scale filtering and orientation-sensitive constraint processing, pollutant spatial gradient feature maps at multiple scales can be obtained, which can fully reflect the changes in spatial gradient features at different scales and in different directions in the pollutant diffusion area, thereby improving the accuracy of subsequent pollutant diffusion area identification.

[0034] The multi-scale pollutant spatial gradient feature map is subjected to region adaptive fusion processing to obtain pollutant spatial texture features that can reflect the spatial diffusion trend of pollutants. .

[0035] In practical implementation, firstly, for each scale For the spatial gradient feature maps, calculate the region adaptive fusion weights respectively. The fusion weights are defined as follows: ; In the formula, Indicates the first Each scale in image coordinates The fusion weight at the point, Indicates the first Each scale feature map in image coordinates Local variance of gradient magnitude within the neighborhood window This indicates the total number of scales used in multi-scale texture decomposition processing.

[0036] It should be noted that the fusion weight Local variance is used to characterize the contribution of this scale to the fusion result at this location. The specific calculation formula is as follows: ; In the formula, Indicates the first The first in the scale region Gradient magnitude of each pixel, This represents the average gradient magnitude within the region. This indicates the number of pixels within a local area.

[0037] Furthermore, the fusion weights The pollutant spatial texture features are obtained by weighted fusion with gradient feature maps at the corresponding scale, specifically represented as follows: ; In the formula, This represents the spatial texture features of the pollutants obtained through the final fusion.

[0038] It is understandable that the above-mentioned regional adaptive fusion method comprehensively considers the differences in the contribution of multi-scale gradient information in local areas, thereby obtaining stable and accurate pollutant diffusion trend characteristics.

[0039] Based on the above-mentioned spatial texture features of pollutants, this embodiment further identifies candidate regions for initial diffusion of pollutants; the method for identifying candidate regions for initial diffusion of pollutants includes: Utilizing the spatial texture features of pollutants Establish a set of diffusion-sensitive candidate regions; In specific implementation, this embodiment is based on the spatial texture features of pollutants. The gradient magnitude distribution in the data is adopted. OTSU Method to determine the optimal threshold The binary candidate region map is obtained, specifically represented as follows: ; in, A binary label map representing the set of diffusion-sensitive candidate regions, where a value of 1 indicates that the region belongs to the diffusion-sensitive candidate region and a value of 0 indicates that the region is not diffusion-sensitive.

[0040] Through the above binarization process, regions with diffusion trends in the spatial texture feature map of pollutants are marked as diffusion-sensitive candidate regions, thus forming a set of diffusion-sensitive candidate regions.

[0041] A clustering and fusion method based on spatial continuity constraints is used to perform regional fusion on the set of diffusion-sensitive candidate regions to determine the initial candidate regions for pollutant diffusion. .

[0042] It should be noted that, to improve the coherence and effectiveness of identifying initial candidate regions for pollutant diffusion, this embodiment further employs a clustering fusion method based on spatial continuity constraints. Specifically, it adopts... DBSCAN (Density clustering) algorithm implementation; specifically, the spatial coordinates of pixels in the initially obtained diffusion-sensitive candidate region set are used as clustering input, and the neighborhood radius parameter of the clustering algorithm is set within the range of 10 to 50 pixels; minimum number of samples for clustering. MinPts The optimal value is between 50 and 200. Multiple connected regions are obtained through density clustering. Subsequently, the connected regions obtained by cluster fusion are defined as candidate regions for initial diffusion of pollutants. The specific output format is the coordinates of the regional boundary or the coordinates of the regional center and the regional radius, which facilitates the subsequent deployment of sub-vessels and the planning of collaborative monitoring tasks.

[0043] In one embodiment, the set of local perturbation features of the flow field The identification methods include: Based on the water flow field data, a disturbance region identification method with embedded velocity vector space gradient constraints is used to determine the set of local disturbance regions. ; It should be noted that, in this embodiment, the flow field data specifically includes: flow velocity data. This indicates the magnitude of the flow velocity at each measurement point within the water body; flow direction data. This indicates the angle of the flow velocity direction at the horizontal plane at each measurement point; water depth data. This indicates the water depth at the measurement point; the coordinates of the sampling location. , representing the two-dimensional spatial location of the measurement point; in the calculation process of this embodiment, in order to achieve the adaptation of planar path planning, the collected three-dimensional spatial flow field data is first subjected to mean projection processing in the depth direction (or the target depth layer data of pollutant distribution is selected) to obtain the velocity component of the two-dimensional planar attribute.

[0044] Furthermore, to achieve accurate identification of local disturbance regions in the flow field data, this embodiment employs a disturbance region identification method that embeds gradient constraints in the velocity vector space, as detailed below: First, the acquired flow field data is constructed into a velocity vector field. Specifically, it is expressed as: ; Secondly, based on the velocity vector field Calculate the velocity space gradient tensor Specifically, it means: ; in, Represents the velocity vector field along x The components of the axis, i.e. ; Represents the velocity vector field along y The components of the axis, i.e. .

[0045] Subsequently, the local disturbance intensity index is calculated to quantitatively describe the degree of flow field disturbance in the local region, specifically expressed as: ; Then, through the preset disturbance intensity threshold Determine the set of local disturbance regions in the flow direction Specifically, it is expressed as: .

[0046] It should be noted that the preset disturbance intensity threshold The value is determined based on statistical values ​​from historical monitoring data, and is taken as 20% to 30% of the local maximum disturbance value.

[0047] Utilizing the initial diffusion candidate region of pollutants Set of local disturbance regions Spatial correlation matching is performed to obtain the set of local perturbation features in the flow field. .

[0048] In specific implementation, this embodiment utilizes the aforementioned set of initial candidate regions for pollutant diffusion. Spatial association matching is then performed, and the specific process is as follows: Calculate spatial correlation index To characterize candidate regions for initial diffusion of pollutants With local disturbance area The degree of spatial matching between them is specifically expressed as: ; In the formula, Indicates the initial diffusion candidate region of pollutants center coordinates With disturbance area center coordinates The Euclidean distance between them This represents the normalized reference distance, which is the average of the distances between the centers of the perturbation region set and the diffusion candidate region set.

[0049] Through the above steps, the spatial matching degree between the pollutant diffusion area and the local disturbance area of ​​the flow field is obtained, and the set of local disturbance characteristics of the flow field is further obtained. This includes the spatial correlation strength between different regions, which serves as an important basis for the path planning module to determine the initial deployment location of sub-ships.

[0050] Furthermore, the process for determining the initial deployment positions of each sub-ship is as follows: Based on the set of local disturbance characteristics of the flow field With the initial diffusion candidate region of pollutants The degree of spatial correlation between them, for the set of local disturbance regions Sensitivity levels are classified, and the sensitivity level classification results are obtained; In practice, for each group of initial candidate areas for pollutant diffusion With local disturbance area Spatial correlation calculated between them Set a correlation threshold .

[0051] It should be noted that the correlation threshold is... The values ​​are obtained from the analysis of historical experimental or field monitoring data, and range from 0.7 to 0.9.

[0052] Specifically, the rules for classifying sensitivity levels are as follows: when At that time, the corresponding candidate areas for initial deployment of sub-ships will be marked as high-sensitivity areas; when At that time, the corresponding candidate areas for initial deployment of sub-ships will be marked as medium sensitivity. when At that time, the corresponding candidate areas for initial deployment of sub-ships will be marked as low sensitivity level.

[0053] Based on the sensitivity level classification results, combined with the preset number of sub-ships and the preset deployment area priority conditions, the initial deployment location of each sub-ship is determined.

[0054] In specific implementation, this embodiment determines the initial deployment position of the sub-ship through the following steps: First, based on the sensitivity level classification results, sets of high-sensitivity regions, medium-sensitivity regions, and low-sensitivity regions are constructed respectively. Subsequently, the number of sub-ships deployed was determined by the pre-set parameters in the mission. Regions are selected sequentially based on their sensitivity level, from highest to lowest: Prioritize selecting initial deployment areas for sub-ships from the set of high-sensitivity areas; in other words, when the number of areas in the set of high-sensitivity areas is greater than or equal to the number of sub-ships to be deployed. At that time, from the set of highly sensitive areas, the optimal selection is determined based on the principles of pollutant diffusion trend intensity and regional spatial distribution uniformity. Each deployment area serves as the initial deployment location for the sub-ships; When the number of areas within the high-sensitivity zone set is insufficient to meet the number of sub-ships to be deployed. At the same time, the set of medium-sensitivity areas will be further included, and the initial deployment areas of the sub-ships will be supplemented by the same area selection principle until the deployment quantity requirements are met; When the total number of areas within the set of highly sensitive and moderately sensitive areas is still insufficient to meet the deployment requirements of sub-ships At the same time, continue to include low-sensitivity areas in the set to complete the selection process for the deployment areas of sub-ships, so as to ensure that all sub-ships obtain an initial deployment location.

[0055] Understandably, high-sensitivity areas indicate a higher risk of pollutant spread, requiring priority deployment of sub-vessels for intensive monitoring to accurately capture and rapidly respond to pollutant spread trends. Medium-sensitivity areas indicate a moderate risk of pollutant spread. When the number of high-sensitivity areas is insufficient to meet the deployment needs of sub-vessels, medium-sensitivity areas will be further included as secondary priority deployment areas. Low-sensitivity areas indicate a relatively low risk of pollutant spread. Low-sensitivity areas are only included when high-sensitivity and medium-sensitivity areas are insufficient to meet the deployment needs of all sub-vessels, in order to ensure that the sub-vessel deployment area fully covers all potential pollutant spread areas.

[0056] After determining the initial deployment position of the sub-ships, the mother ship's path planning module generates a deployment instruction containing the target position coordinates and navigation parameters, and sends it to each sub-ship via a wireless communication link; each sub-ship receives the instruction and autonomously navigates to the corresponding initial deployment position, and then activates the component acquisition module.

[0057] In another embodiment, the set of interaction features between the pollutant components The generation process includes: Based on the pollutant component data, calculate the spatial concentration gradient of each pollutant component. ; It should be noted that the pollutant component data includes, but is not limited to, the concentration values ​​of various typical pollutants (such as ammonia nitrogen, phosphate, nitrate, organic matter, etc.) and the spatial coordinates of the corresponding component data. .

[0058] When calculating the spatial concentration gradient of each pollutant component, the spatial concentration distribution function of each pollutant component must first be determined. ;in, Indicates the type number of pollutant components (e.g., ammonia nitrogen is denoted as...). Phosphate is denoted as (and so on) The spatial concentration gradient of each pollutant component is calculated based on the concentration distribution function, and is specifically expressed as follows: ; In the formula, Indicates the first The concentration gradient of pollutant components in the horizontal direction is used to characterize the concentration along the [horizontal axis]. x Rate of change of direction; Indicates the first The concentration gradient of pollutant components in the vertical direction is used to characterize the concentration along the [concentration range]. y Rate of change of direction.

[0059] By calculating this concentration gradient, the spatial distribution of pollutant components at different locations within the monitored water area can be obtained, providing important data support for the accurate identification of subsequent regions with varying component concentration gradients.

[0060] Based on the spatial concentration gradient of each pollutant component Identify the concentration gradient variation regions of each pollutant component. ; In specific implementation, this embodiment focuses on the spatial concentration gradient of each pollutant component. Determine the intensity index of the concentration gradient change of each pollutant component. To characterize the degree of local variation in the concentration gradient of each pollutant component, the calculation formula is as follows: .

[0061] Gradient change intensity index Compared with the preset gradient change intensity threshold The comparison will be performed, and the gradient change intensity will be greater than or equal to the preset threshold. Corresponding position The regions marked as having significant changes in component concentration gradients are thus obtained. .

[0062] For the concentration gradient change region Spatial statistical analysis was conducted to obtain a set of interaction characteristics reflecting the degree of mutual influence among pollutant components. .

[0063] In practical implementation, this embodiment addresses the concentration gradient variation region of each pollutant component. Spatial statistical characteristics analysis was conducted, as follows: First, for any two pollutant components (e.g., the first... Components and the first Correlation analysis was performed on the spatial distribution of the concentrations of the components within the corresponding gradient change regions, using the spatial correlation coefficient. express: ; In the formula, , They represent the first At the spatial location of the first , The concentration of pollutant components, , They represent the first , Spatial mean of pollutant component concentrations This indicates the total number of sampling locations within the corresponding area.

[0064] After obtaining the spatial correlation coefficients among the pollutant components through the above calculations, the Pearson correlation analysis method is used to extract the significant spatial correlation coefficients between all pairwise pollutant components, forming a set of interaction characteristics among pollutant components. This set specifically includes the spatial correlation coefficients between each pollutant component. This is used to quantitatively describe the intensity of interactions between pollutant components.

[0065] Specifically, the pollutant enrichment area The determination process includes: Set of interaction characteristics among pollutant components The pre-constructed fine-grained pollutant diffusion model is input into a pre-built model for multiple iterative prediction calculations to obtain the spatial concentration distribution of pollutants after the concentration reaches a steady state. It should be noted that, in the specific execution process of this embodiment, the space analysis module of the sub-ship will extract the set of interaction features. The data is transmitted in real time to the mother ship via a wireless communication link. The mother ship's path planning module then calls a pre-built fine-grained pollutant diffusion model for collaborative prediction calculations. In this embodiment, a fine-grained pollutant diffusion model is pre-built based on historical measured data and numerical simulation results. This model is specifically a multi-component convection-diffusion model embedding the interaction characteristics of pollutant components, and its representation is as follows: ; in, Indicates the first The concentration of pollutant components, , These represent the average flow velocities in the horizontal and vertical directions, respectively. Indicates the first Diffusion coefficient of pollutant components This represents the interaction terms between the various pollutant components. This indicates the total number of pollutant components.

[0066] It should also be noted that the diffusion coefficient of pollutant components The value was determined through laboratory measurements in conjunction with aquatic environmental conditions (such as temperature, viscosity, and flow characteristics), and the range was [value range missing]. .

[0067] It should be understood that, in order to achieve deep embedding of statistical features into the physical model, the interaction terms... The specific construction method is as follows: ; In the formula, Represents the set of interaction features among pollutant components The Middle Components and the first Spatial correlation coefficient between components This represents a predefined inter-component interaction influence factor, used to characterize the synergistic or antagonistic strength exhibited by different chemical components through spatial correlation, with a value range of [value missing]. .

[0068] Set of interaction characteristics among pollutant components Input the fine-grained dispersion model of the pollutant and perform multiple iterative prediction calculations until the pollutant concentration distribution reaches a stable state, i.e., satisfies: ; In the formula, This represents the threshold for determining concentration stability, with a value ranging from 0.01 to 0.05. , The first Second and third After the nth iteration Calculated concentrations of pollutant components.

[0069] It should be understood that the initial pollutant concentration The component concentrations measured by each sub-ship at its initial deployment position were obtained through spatial interpolation.

[0070] After the above iterative calculations, once the pollutant concentration reaches a stable state, a stable spatial distribution of pollutant concentration is obtained, which serves as the basic data input for subsequent identification of pollutant enrichment areas.

[0071] Based on the spatial concentration distribution, spatial peak regions are identified to obtain pollutant enrichment areas. .

[0072] It should be understood that, after the pollutant concentration stabilizes, this embodiment further identifies the spatial peak of the stabilized spatial concentration distribution and determines the spatial peak point of the concentration through a spatial extreme value search algorithm. The area around the peak point (e.g., 10-50 meters from the peak) is designated as the pollutant enrichment zone.

[0073] In another embodiment, the set of features of the convergence of stable diffusion paths in the flow field The construction methods include: Determine the stable spatial distribution area of ​​water flow based on water flow field data. ; It should be noted that, in this embodiment, the region with stable spatial distribution of water flow trends is the location region that meets specific stability conditions, specifically determined by calculating the stability coefficient of the flow field time series. Sure: ; In the formula, The standard deviation of the time series representing the flow velocity at the measurement location. It represents the average time series value of the flow velocity at the measurement location.

[0074] By setting a stability threshold for the stability coefficient The value ranges from 0.8 to 0.95, identifying and marking areas of stable spatial distribution of water flow trends. The specific judgment rules are as follows: When the stability coefficient Greater than or equal to the preset stability threshold When, then the position This is marked as a region with a stable spatial trend in water flow; conversely, when the stability coefficient is low... Less than the preset stability threshold If so, no marking is performed.

[0075] According to pollutant enrichment areas Determine the stable distribution area of ​​water flow spatial trend. With pollutant-rich areas Spatial sensitive intersection points; In specific implementation, the pollutant enrichment area is utilized. The method for identifying spatially sensitive intersection locations is as follows: First, determine the spatial sensitivity indicators. , is represented as: ; In the formula, Indicates the stable distribution area of ​​water flow Middle position With pollutant-rich areas The shortest distance to the nearest point on the boundary. This represents the sensitivity normalization distance constant, with a value ranging from 10 to 50 meters, used to ensure the normalization of sensitivity indicators.

[0076] By setting a sensitivity threshold for spatial sensitivity indicators The value ranges from 0.7 to 0.9, identifying and labeling a set of spatially sensitive intersection locations. The specific judgment rules are as follows: When sensitivity index Greater than or equal to the preset sensitivity threshold When, then the position Marked as spatially sensitive intersection locations; conversely, when the sensitivity index... Less than the preset sensitivity threshold If so, no marking is performed.

[0077] By analyzing the spatial accumulation indicators of pollutants at spatially sensitive intersection locations, a set of characteristics of stable diffusion paths in the flow field is obtained. .

[0078] In specific implementation, this embodiment targets a set of spatially sensitive intersection locations. Determine the spatial accumulation index of pollutants at each location point. Specifically, it is expressed as: ; In the formula, Represents position coordinates Spatial accumulation indicators of pollutant concentration This represents the first [predicted] by the fine-grained pollutant diffusion model. Various pollutant components in The spatiotemporal distribution concentration at time t, This indicates the initial moment when the model enters a stable predictive state. This indicates the preset prediction evolution window length, with a value ranging from 10 minutes to 30 minutes. This indicates that the time variable is used during the integration process. The infinitesimal increment.

[0079] It should be noted that the spatial accumulation index is calculated by the integral mean of the total concentration of each component within the prediction period, and is used to characterize the pollutant retention and enrichment potential at that location during the flow field transport process.

[0080] Understandably, this involves setting up spatially sensitive intersection locations. Geometric intersection constraints and cumulative index of model predictions By combining these methods, we can accurately identify key pathway points in the flow field that meet both the hydraulically stable confluence conditions and have high concentrations of pollutants.

[0081] Based on the magnitude of the spatial accumulation index, determine the set of characteristics of the convergence of stable diffusion paths in the flow field. Specifically, this means obtaining spatial cumulative indicators based on historical field data or numerical simulation results. Based on the distribution characteristics, determine an intersection feature threshold. The threshold value can be set to 70% to 90% of the maximum value of the spatial cumulative index within the observation area, and the specific value can be determined according to the actual application scenario; when the spatial cumulative index Greater than or equal to the intersection feature threshold At that time, the corresponding position Marked as the characteristic region of the convergence of stable diffusion paths in the flow field; set of all spatially sensitive convergence locations. The above threshold judgment is performed on each location in the sequence, and the set of spatial locations that meet the conditions is determined as the set of convergence features of stable diffusion paths in the flow field. .

[0082] Specifically, the method for generating the cooperative cruising trajectory of the mother and daughter ships includes: Based on the set of characteristics of the convergence of stable diffusion paths in the flow field A path optimization algorithm is used to determine the mother ship's main cruising trajectory. ; It should be noted that this embodiment focuses on the set of characteristics of the convergence of stable diffusion paths in the flow field. By monitoring pollutant accumulation indicators Use path optimization algorithms (such as ant colony optimization algorithm) ACO The main cruise trajectory of the mother ship, covering key intersection areas, is determined. Specifically, in the ant colony optimization algorithm, the objective function of the main cruise trajectory is defined as minimizing the total path length while covering all key intersection areas. Solving this algorithm yields the main cruise trajectory of the mother ship. .

[0083] Based on pollutant enrichment areas The spatial gridding method, combined with the path planning algorithm, is used to determine the auxiliary cruise trajectory of the sub-ship. ; It should be understood that this embodiment is based on the pollutant enrichment area. Based on spatial morphological characteristics, the enriched area is divided into multiple sub-regions (e.g., grids with side lengths of 10–50 meters) using a spatial gridding method, and the auxiliary cruising trajectory of the sub-ship is determined using path planning algorithms (e.g., local greedy algorithms or space-filling curves). This allows for precise coverage of each sub-region.

[0084] A trajectory fusion method with embedded spatial position constraints is used to analyze the main cruise trajectory of the mother ship. With the auxiliary cruise trajectory of the sub-ship By integrating the data, a coordinated cruise trajectory of the mother and daughter ships covering pollutant-rich areas can be obtained. .

[0085] In specific implementation, this embodiment defines a spatial location constraint fusion strategy, as follows: First, the mother ship's main cruise trajectory Defined as the coordinated cruise trajectory of mother and daughter ships The dominant trajectory; Secondly, determine the spatial location constraint function. Specifically, it is expressed as: ; In the formula, This represents the average distance between each point on the child ship's trajectory and the centerline of the mother ship's main cruising trajectory. This indicates the spatial deviation of the sub-ship's trajectory covering areas not covered by pollutant-rich regions. This represents the fusion weighting coefficient, with a value ranging from 0.4 to 0.6.

[0086] It is worth mentioning that the fusion weighting coefficient The path planning module adaptively and dynamically adjusts its approach based on the pollutant diffusion rate. Specifically, when the path planning module detects that the pollutant diffusion rate is greater than or equal to a preset speed threshold, it automatically reduces the speed. The value (e.g., the value of) (Adjusted from 0.6 to 0.4) to increase the coverage deviation of pollutant-rich areas. The weighting of the data ensures that the collaborative trajectory has higher monitoring coverage and response time in scenarios of rapid pollutant diffusion; wherein, the pollutant diffusion rate is determined by comparing the pollutant enrichment areas in adjacent sampling periods. The centroid displacement or area change rate is obtained.

[0087] By optimizing the aforementioned fusion function, the auxiliary cruise trajectory of the sub-ship can be adjusted. This makes it consistent with the mother ship's cruising trajectory. Cooperative trajectories Fully cover areas rich in pollutants.

[0088] Based on the generated mother-daughter ship cooperative cruise trajectory The path planning module sends cruise control commands to the mother ship and each vessel in real time, including information such as speed, heading angle, and position coordinates. The mother ship executes the main cruise path according to these control commands, while simultaneously monitoring the positions and assigning tasks to the vessels. Each vessel, based on received auxiliary cruise commands, performs refined cruise tasks within a designated trajectory range and reports its position and monitoring data back to the mother ship in real time. This achieves precise coordination and dynamic task adjustment between the mother ship and the vessels, ensuring efficient coverage of pollutant-rich areas.

[0089] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0090] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A mother-daughter ship system, comprising a mother ship and multiple daughter ships, characterized in that, The mother ship is equipped with a path planning module, as well as an image acquisition module and a flow field acquisition module that are communicatively connected to the path planning module; each of the sub-ships is equipped with a component acquisition module and a spatial analysis module that is communicatively connected to the component acquisition module; wherein, the spatial analysis module and the path planning module coordinate their tasks through a communication connection when performing tasks.

2. A method for coordinated deployment of mother and daughter ships using the system as described in claim 1, characterized in that, Includes the following steps: Based on the pollutant image acquired by the image acquisition module, the spatial texture features of the pollutants are extracted, and the initial diffusion candidate regions of the pollutants are identified based on the spatial texture features. Based on the water flow field data acquired by the flow field acquisition module, the set of local flow field disturbance features between the initial diffusion candidate region and the water flow field is identified, and the initial deployment position of each sub-vessel is determined by the path planning module using the set of local flow field disturbance features. The component acquisition module acquires pollutant component data at each initial deployment location, and the spatial analysis module extracts the set of interaction features between pollutant components; the set of interaction features is then input into a pre-constructed fine-grained pollutant diffusion model to determine pollutant enrichment areas. Establish a set of flow field stable diffusion path intersection features between the pollutant enrichment area and the water flow field data, and use the path planning module to generate a collaborative cruise trajectory of the mother and daughter ships covering the pollutant enrichment area; perform collaborative control of the daughter ship and the mother ship based on the collaborative cruise trajectory.

3. The method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The extraction process of the spatial texture features of the pollutants includes: Multi-scale texture decomposition processing with embedded gradient orientation sensitive constraints is performed on pollutant images to generate multi-scale pollutant spatial gradient feature maps. The multi-scale pollutant spatial gradient feature map is subjected to region adaptive fusion processing to obtain pollutant spatial texture features that can reflect the spatial diffusion trend of pollutants. .

4. The method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The method for identifying the initial diffusion candidate region of the pollutant includes: Based on the spatial texture characteristics of pollutants The Otsu's method was used to establish a set of diffusion-sensitive candidate regions; A clustering and fusion method based on spatial continuity constraints is used to perform regional fusion on the set of diffusion-sensitive candidate regions to determine the initial candidate regions for pollutant diffusion. .

5. The method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The method for identifying the set of local disturbance features in the flow field includes: Based on the water flow field data, a disturbance region identification method with embedded velocity vector space gradient constraints is used to determine the set of local disturbance regions. ; Utilizing the initial diffusion candidate region of pollutants Set of local disturbance regions Spatial correlation matching is performed to obtain the set of local perturbation features in the flow field. .

6. The method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The process for determining the initial deployment positions of each sub-ship is as follows: Based on the set of local disturbance characteristics of the flow field With the initial diffusion candidate region of pollutants The degree of spatial correlation between them, for the set of local disturbance regions Sensitivity levels are classified, and the sensitivity level classification results are obtained; Based on the sensitivity level classification results, combined with the preset number of sub-ships and the preset deployment area priority conditions, the initial deployment location of each sub-ship is determined.

7. The method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The process of generating the set of interaction features among the pollutant components includes: Based on the pollutant component data, calculate the spatial concentration gradient of each pollutant component. ; Based on the spatial concentration gradient of each pollutant component Identify the concentration gradient variation regions of each pollutant component. ; For the region of concentration gradient change Spatial statistical analysis was conducted to obtain a set of interaction characteristics reflecting the degree of mutual influence among pollutant components. .

8. The method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The process for determining the pollutant enrichment zone includes: Set of interaction characteristics among pollutant components The pre-constructed fine-grained pollutant diffusion model is input into a pre-built model for multiple iterative prediction calculations to obtain the spatial concentration distribution of pollutants after the concentration reaches a steady state. Based on the spatial concentration distribution, spatial peak regions are identified to obtain pollutant enrichment areas. .

9. A method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The method for constructing the set of features of the intersection of stable diffusion paths in the flow field includes: Determine the stable spatial distribution area of ​​water flow based on water flow field data. ; According to pollutant enrichment areas Determine the stable distribution area of ​​water flow spatial trend. With pollutant-rich areas Spatial sensitive intersection points between them; By analyzing the spatial accumulation indicators of pollutants at spatially sensitive intersection locations, a set of characteristics of stable diffusion paths in the flow field is obtained. .

10. A method for coordinated deployment of mother and daughter ships according to claim 2, characterized in that, The method for generating the collaborative cruise trajectory of the mother and daughter ships includes: Based on the set of characteristics of the convergence of stable diffusion paths in the flow field A path optimization algorithm is used to determine the mother ship's main cruising trajectory. ; Based on pollutant enrichment areas The spatial gridding method, combined with the path planning algorithm, is used to determine the auxiliary cruise trajectory of the sub-ship. ; A trajectory fusion method with embedded spatial position constraints is used to analyze the main cruise trajectory of the mother ship. With the auxiliary cruise trajectory of the sub-ship By integrating the data, a coordinated cruise trajectory of the mother and daughter ships covering pollutant-rich areas can be obtained. .

Images