Multi-agent cooperative pollutant traceability system, method and device

Abstract

The invention provides a multi-agent collaborative pollutant traceability system, method and device, and relates to the technical field of river and lake system pollution traceability, and the system comprises a multi-source heterogeneous data sensing layer, a multi-agent collaborative network layer, a chaotic decision module and a collaborative traceability execution layer. The multi-source heterogeneous data sensing layer is used for collecting river and lake water body information in real time through a sensor set; the multi-agent collaborative network layer is used for issuing task instructions according to the river and lake water body information and task requirements to execute intelligent scheduling on various agents, and obtaining target river and lake water body information collected by the various agents; the chaotic decision module is used for performing chaotic mapping processing and path optimization processing on the target river and lake water body information to obtain a decision result; and the collaborative traceability execution layer is used for executing a traceability task according to a decision result of the chaos decision module to perform pollutant intelligent tracking and monitoring processing on the target traceability area to obtain a target pollutant traceability result. According to the invention, the accuracy and dynamic adaptability of pollutant traceability can be improved.

Description

Technical Field

[0001] This invention relates to the technical field of pollution source tracing in river and lake systems, and in particular to a multi-agent collaborative pollutant source tracing system, method, and apparatus. Background Technology

[0002] Currently, related technologies suggest that traditional pollutant tracing technologies mainly rely on single centralized decision-making or fixed rules for reasoning. This approach is prone to getting stuck in local optima or causing decision oscillations, thus affecting the accuracy of pollutant tracing. Furthermore, if a centralized tracing system is used, single points of failure are likely to occur, while if a distributed system is used, insufficient coordination is likely to occur. In addition, when facing time-varying environments, traditional static rules cannot adjust the tracing strategy in real time, which may lead to tracing delays or misjudgments. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a multi-agent collaborative pollutant source tracing system, method and apparatus, which can significantly improve the accuracy and dynamic adaptability of pollutant source tracing.

[0004] In a first aspect, embodiments of the present invention provide a multi-agent collaborative pollutant tracing system, comprising: a multi-source heterogeneous data perception layer, a multi-agent collaborative network layer, a chaotic decision-making module, and a collaborative tracing execution layer; wherein, the multi-source heterogeneous data perception layer is used to collect river and lake water information in real time through a set of sensors, wherein the river and lake water information includes: water quality information, flow information, and meteorological information; the multi-agent collaborative network layer is used to issue task instructions according to the river and lake water information and task requirements, to perform intelligent scheduling of various agents, and to obtain the target river and lake water information collected by each agent; the chaotic decision-making module is used to perform chaotic mapping processing and path optimization processing on the target river and lake water information to obtain a decision result, wherein the decision result includes: path information of the target tracing area to be traced for pollutant tracing; the collaborative tracing execution layer is used to execute the tracing task according to the decision result of the chaotic decision-making module, to perform intelligent tracking and monitoring processing of pollutants in the target tracing area, and to obtain the target pollutant tracing result.

[0005] In one implementation, the multi-agent collaborative network layer includes: a detection agent, an analysis agent, and a tracking agent; wherein, the detection agent is used to send the collected river and lake water information to the analysis agent; the analysis agent is used to generate task instructions based on task requirements and river and lake water information, and send the task instructions to the tracking agent; the tracking agent is used to trace the source of pollution targets according to the task instructions.

[0006] In one implementation, the system further includes a feedback and adaptive optimization layer; wherein the feedback and adaptive optimization layer is used to verify the effectiveness of the current source tracing path based on the river and lake water information obtained in real time by the multi-source heterogeneous data sensing layer, and to adjust the control parameters of the chaotic decision module according to the verification results, so as to enhance the environmental change adaptability of the chaotic decision module.

[0007] Secondly, embodiments of the present invention also provide a multi-agent collaborative pollutant source tracing method. The method is applied to a multi-agent collaborative pollutant source tracing system. The method includes: acquiring river and lake water information and initial weight values ​​for river and lake areas, wherein the river and lake water information includes water quality information, flow information, and meteorological information; performing chaotic mapping processing on the target river and lake water information and initial weight values ​​to obtain target dynamic edge weights, and using a preset shortest path algorithm to perform path optimization processing on the target dynamic edge weights to obtain a decision result, wherein the decision result includes path information of the target source tracing area to be traced for pollutants; and performing intelligent pollutant tracking and monitoring processing on the target source tracing area based on the decision result to obtain the target pollutant source tracing result.

[0008] In one implementation, the steps of obtaining river and lake water information and initial weight values ​​for river and lake areas include: after obtaining river and lake water information, dividing the river and lake area into a set of nodes, and obtaining the initial weight values ​​corresponding to the edges between each node in the set of nodes, wherein each node is a fixed monitoring point, the starting point of the detection agent, and the ending point.

[0009] In one implementation, the step of performing chaotic mapping processing on the target river / lake water information and initial weight values ​​to obtain the target dynamic edge weights includes: performing chaotic mapping processing on the target river / lake water information to obtain a time-varying sequence of state decision parameters, wherein the state decision parameters include: the activity of pollutant diffusion, the complexity of source tracing, and the degree of environmental impact; mapping the time-varying sequence of state decision parameters to the dynamic weights of each side of the river channel to obtain the adjusted target dynamic edge weights.

[0010] In one implementation, the step of performing intelligent pollutant tracking and monitoring on the target source area based on the decision result to obtain the target pollutant source tracing result includes: executing a source tracing task based on the decision result, performing intelligent pollutant tracking and monitoring on the target source area to obtain the local source tracing results reported by each intelligent agent; and fusing the local source tracing results through a preset distributed Bayesian inference model to obtain the target pollutant source tracing result.

[0011] Thirdly, embodiments of the present invention also provide a multi-agent collaborative pollutant tracing device. The device is applied to a multi-agent collaborative pollutant tracing system and includes: an information acquisition module for acquiring river and lake water information and initial weight values ​​for river and lake areas, wherein the river and lake water information includes water quality information, flow information, and meteorological information; a chaotic decision-making module for performing chaotic mapping processing on the target river and lake water information and initial weight values ​​to obtain target dynamic edge weights, and using a preset shortest path algorithm to perform path optimization processing on the target dynamic edge weights to obtain a decision result, wherein the decision result includes path information of the target tracing area to be traced for pollutant tracing; and a pollutant tracing module for performing intelligent pollutant tracking and monitoring processing on the target tracing area based on the decision result to obtain the target pollutant tracing result.

[0012] Fourthly, embodiments of the present invention also provide a server, including a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement any of the methods provided in the second aspect.

[0013] Fifthly, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement any of the methods provided in the second aspect.

[0014] The embodiments of the present invention bring the following beneficial effects: This invention provides a multi-agent collaborative pollutant tracing system, method, and apparatus. The system includes: a multi-source heterogeneous data perception layer, a multi-agent collaborative network layer, a chaotic decision-making module, and a collaborative tracing execution layer. The multi-source heterogeneous data perception layer is used to collect river and lake water information in real time through a set of sensors. This river and lake water information includes water quality information, flow information, and meteorological information. The multi-agent collaborative network layer is used to issue task instructions based on the river and lake water information and task requirements to intelligently schedule various agents and acquire the target river and lake water information collected by each agent. The chaotic decision-making module is used to perform chaotic mapping processing and path optimization processing on the target river and lake water information to obtain decision results. These decision results include path information of the target tracing area for pollutant tracing. The collaborative tracing execution layer is used to execute the tracing task based on the decision results of the chaotic decision-making module, performing intelligent pollutant tracking and monitoring processing on the target tracing area to obtain the target pollutant tracing results. This invention can significantly improve the accuracy and dynamic adaptability of pollutant tracing.

[0015] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the structure of a multi-agent collaborative pollutant tracing system provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating a multi-agent collaborative pollutant source tracing method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a multi-agent collaborative pollutant tracing device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a server provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0020] Currently, traditional pollutant tracing technologies rely on single centralized decision-making or fixed rule reasoning, which have the following core defects: 1. Local optima and decision oscillations: In complex systems with multiple paths and multiple disturbances (such as river and lake pollution with multiple pollution sources), traditional greedy algorithms or dynamic programming are prone to getting trapped in local optima, or the decision oscillates repeatedly due to environmental disturbances (such as repeatedly correcting the tracing path); 2. Single point of failure and insufficient coordination: Once a core node of a centralized tracing system fails (such as a server crash), the global tracing capability collapses, while in distributed systems, there is a lack of effective cooperation mechanisms between agents, resulting in redundant task allocation or serious information silos; 3. Poor dynamic adaptability: Faced with time-varying environments (such as the sudden addition or disappearance of pollution sources or sudden changes in network traffic), traditional static rules cannot adjust the tracing strategy in real time, leading to tracing delays or misjudgments.

[0021] Since existing improvement schemes (such as reinforcement learning-based source tracing and multi-agent game theory methods) still suffer from slow convergence speed and weak rule generalization, this invention provides a multi-agent collaborative pollutant source tracing system, method, and apparatus. These solutions address these issues by: 1. Optimizing the decision-making process of multiple agents using a chaotic dynamics model, leveraging the ergodicity, randomness, and sensitivity to initial conditions of chaos to overcome local optima and achieve globally optimal or near-optimal selection of the source tracing path; 2. Implementing a multi-agent distributed collaborative mechanism, enabling multiple agents (physical sensor nodes, digital agents, edge computing units, etc.) to dynamically adapt to complex environments through information sharing and task division, thereby improving source tracing efficiency and robustness; and 3. Adaptively fusing sensory data and prior knowledge, adjusting the source tracing strategy in real time through a chaotic and agent collaborative framework to solve problems such as static rule failure and incompatibility with dynamic environments in traditional source tracing.

[0022] To facilitate understanding of this embodiment, a multi-agent collaborative pollutant source tracing method disclosed in this invention will first be described in detail. This method is applied to a multi-agent collaborative pollutant source tracing system. To facilitate understanding of the multi-agent collaborative pollutant source tracing system, this invention provides a structural schematic diagram of the multi-agent collaborative pollutant source tracing system, as shown below. Figure 1 As shown, the system includes: a multi-source heterogeneous data perception layer, a multi-agent collaborative network layer, a chaotic decision-making module, a collaborative source tracing execution layer, and a feedback and adaptive optimization layer. Data flows in the system from top to bottom, and the various layers cooperate with each other to jointly achieve the source tracing of pollutants in rivers and lakes.

[0023] The multi-source heterogeneous data sensing layer is used to collect real-time information about river and lake water bodies through a set of sensors. This information includes water quality, flow rate, and meteorological data. In one implementation, the multi-source sensing layer is located at the top layer of the architecture and contains various sensors, such as water quality sensors, flow rate sensors, and meteorological sensors. These sensors are distributed at different locations in the river and lake to collect real-time data on water quality, flow rate, and meteorological conditions. The data is transmitted to the upper layer via wired or wireless communication.

[0024] The multi-agent collaborative network layer is used to issue task instructions based on river and lake water information and task requirements, to perform intelligent scheduling of various agents, and to acquire target river and lake water information collected by each agent. In one embodiment, the multi-agent collaborative network layer includes: detection agents, analysis agents, and tracking agents. The detection agents send the collected river and lake water information to the analysis agents; the analysis agents generate task instructions based on task requirements and river and lake water information, and send these instructions to the tracking agents. The tracking agents then trace the source of pollution targets according to the sent instructions. In other words, the multi-agent collaborative network consists of multiple agents, including detection agents (monitoring stations), analysis agents (control centers), and tracking agents (such as unmanned surface vessels and drones). These agents are interconnected through a distributed communication network to achieve data sharing and interaction. The detection agents send the collected data to the analysis agents, and the analysis agents issue instructions to the tracking agents according to task requirements.

[0025] Furthermore, each agent in the multi-agent collaborative network layer is an independent unit (physical entity or virtual agent) with sensing, computing, and communication capabilities, categorized into detection agents (monitoring stations), analysis agents (control centers), and tracking agents (drones, unmanned surface vessels). These agents are interconnected via a network of distributed communication protocols (such as MQTT, ZigBee, or custom short-range wireless protocols) to achieve data sharing and interaction. Detection agents send the collected data to analysis agents, which then issue task instructions to tracking agents based on task requirements (e.g., detection agent 1 is responsible for upstream tracing, and detection agent 2 is responsible for downstream tracking).

[0026] Specifically, the detection-type intelligent agents are distributed in different locations in rivers and lakes in the form of fixed monitoring stations and buoys, and send the collected data to nearby aggregation nodes through wireless communication modules (such as ZigBee, LoRa, etc.).

[0027] The aggregation nodes are responsible for collecting data from probing agents within a certain range, and performing preliminary processing and forwarding. Aggregation nodes are connected to each other via wired or wireless communication, forming a distributed network structure.

[0028] Tracking agents, such as unmanned boats or drones, are equipped with wireless communication devices and sensors, enabling them to communicate with analytical agents and other agents. These tracking agents move autonomously in rivers and lakes, interact with detection agents based on mission requirements, and transmit their monitoring data to a central control agent.

[0029] The analytical agent is located at the core of the network, responsible for the coordination and management of the entire system. It connects with other agents through a wireless communication network, receives data from the probing agents and monitoring results from the tracking agents, issues instructions to the tracking agents, and interacts with other systems for data exchange and information sharing.

[0030] The chaotic decision-making module performs chaotic mapping and path optimization on target river and lake water information to obtain decision results. These results include path information for the target source area for pollutant tracing. In one implementation, the chaotic decision-making core is located at the heart of the system, receiving data from a multi-agent collaborative network. The chaotic decision-making core uses the Lorenz chaotic model system to process and analyze the data, generating dynamic weight parameters. These parameters are used to comprehensively assess the possible source direction of pollutants and the selection of tracing paths.

[0031] The collaborative source tracing execution layer is used to execute source tracing tasks based on the decision results of the chaotic decision module. It performs intelligent tracking and monitoring of pollutants in the target source area to obtain the source tracing results. In one implementation, the collaborative source tracing execution layer, based on the output of the chaotic decision core, involves multiple agents collaboratively executing the source tracing task. Tracking agents trace and monitor suspected pollution source areas along designated paths, detection agents continuously collect data and provide support, and analysis agents continue data fusion and path analysis. Information is dynamically shared among the agents, task allocation and action plans are adjusted in a timely manner, and distributed Bayesian inference or federated learning models are used to fuse the local source tracing results of multiple agents (such as the coordinates and types of abnormal points reported by each agent), gradually narrowing the source tracing scope and improving the credibility of the conclusions.

[0032] The feedback and adaptive optimization layer is used to verify the effectiveness of the current source tracing path based on the river and lake water information acquired in real time by the multi-source heterogeneous data perception layer. Based on the verification results, it adjusts the control parameters of the chaotic decision-making module to enhance its adaptability to environmental changes. In one implementation, the feedback optimization layer is located at the bottom layer of the architecture and is responsible for monitoring and evaluating the entire source tracing process. By analyzing the source tracing results and real-time data, the feedback optimization layer transmits perception feedback information to the chaotic decision-making core. The chaotic decision-making core adjusts the parameters of the chaotic model based on the feedback information, continuously optimizing the decision-making process. Simultaneously, the feedback optimization layer updates the knowledge base, storing successful source tracing experiences and cases to provide reference for subsequent source tracing tasks.

[0033] based on Figure 1 The diagram shown illustrates the structure of a multi-agent collaborative pollutant tracing system. This invention provides a detailed description of the multi-agent collaborative pollutant tracing method. (See also...) Figure 2 The diagram shows a flowchart of a multi-agent collaborative pollutant source tracing method, which mainly includes the following steps S202 to S206: Step S202: Obtain river and lake water information and initial weight values ​​for river and lake areas. The river and lake water information includes water quality information, flow information, and meteorological information.

[0034] In one implementation, after obtaining information about river and lake water bodies, the river and lake area is divided into a set of nodes, and the initial weight values ​​corresponding to the edges between each node in the set are obtained. Each node is a fixed monitoring point, the starting point of the detection agent, and the ending point.

[0035] Specifically, the first step is to define nodes: the river and lake area is divided into multiple nodes, which can be fixed monitoring points, starting points for intelligent agents (such as unmanned boats and drones), and ending points (suspected pollution source areas), etc. Next, edges are defined and their weights initialized: the edges between nodes are defined, and the initial weights of the edges can be determined based on traditional distance metrics (such as Euclidean distance, Manhattan distance, etc.) or other factors related to river and lake geography and monitoring.

[0036] Step S204: Perform chaotic mapping processing on the target river and lake water information and initial weight values ​​to obtain the target dynamic edge weights, and use a preset shortest path algorithm to perform path optimization processing on the target dynamic edge weights to obtain the decision results. The decision results include: path information of the target source tracing area to be traced for pollutants.

[0037] In one implementation, chaotic mapping processing can be performed on the target river and lake water information to obtain a sequence of state decision parameters changing over time. Then, the sequence of state decision parameters changing over time is mapped to the dynamic weights of each side of the river channel to obtain the adjusted target dynamic edge weights. The state decision parameters include: the activity of pollutant diffusion, the complexity of source tracing, and the degree of environmental impact.

[0038] Specifically, the chaotic decision-making module receives data from a multi-agent collaborative network, including water quality data, flow data, and meteorological data collected by sensors. This data serves as input to the chaotic model. Furthermore, dynamic weights can be generated through chaotic mapping. The input data is processed using a chaotic model to generate a set of dynamic weight parameters. Chaotic mapping can produce chaotic sequences with both randomness and regularity. Through nonlinear transformation of the input data, weight values ​​reflecting the importance of each factor are obtained. During path selection, the optimal source tracing path can be chosen based on the results of the chaotic model. The path selection algorithm can consider various factors, such as the direction of pollutant diffusion, water flow velocity, and topography, to ensure rapid and accurate tracking of the pollution source.

[0039] Step S206: Based on the decision results, perform intelligent tracking and monitoring of pollutants in the target source area to obtain the source tracing results of the target pollutants.

[0040] In one implementation, a source tracing task can be executed based on the decision results, and intelligent tracking and monitoring of pollutants can be performed on the target source tracing area to obtain the local source tracing results reported by each intelligent agent. Then, the local source tracing results are fused together by a preset distributed Bayesian inference model to obtain the target pollutant source tracing result.

[0041] Specifically, in carrying out collaborative source tracing, the first step is to conduct monitoring by detection-type intelligent agents: fixed monitoring stations and buoys, among other detection-type intelligent agents, continuously collect water quality data from rivers and lakes and send the data to analysis-type intelligent agents. Simultaneously, they monitor the surrounding environment in real time and promptly report any anomalies.

[0042] The tracking agent then performs the tracking: Following instructions from the analytical agent, unmanned surface vessels or drones, or other tracking agents, depart from designated locations and follow predetermined paths to track and monitor suspected pollution source areas. The tracking agents are equipped with high-precision sensors capable of collecting real-time data on water quality, weather conditions, and other factors, and transmitting the data back to the analytical agent.

[0043] Furthermore, dynamic information sharing: During the source tracing process, various agents dynamically share information through wireless communication networks. The detection agent shares real-time collected data with the tracking agent, which in turn feeds back its monitoring results and identified issues to the analysis agent and other agents. Through information sharing and interaction, each agent can better understand the distribution and diffusion trends of pollutants in rivers and lakes.

[0044] Finally, task adjustments are made: Based on real-time data and shared information, the analytical agent dynamically adjusts the tasks of each agent. For example, if the pollution in a certain area is severe, the control center agent can dispatch more tracking agents to that area for focused monitoring; if new pollution source clues are discovered, the tracking direction and path of the tracking agents can be adjusted.

[0045] The multi-agent collaborative pollutant tracing method provided in this invention can significantly improve the accuracy and dynamic adaptability of pollutant tracing.

[0046] In addition, this embodiment of the invention also provides a feedback and adaptive optimization implementation method. Through continuous perception feedback, chaotic parameter adjustment and knowledge base update, the system can achieve adaptive optimization and continuously improve the efficiency and accuracy of tracing the source of pollutants in rivers and lakes. See (1) to (3) below for details: (1) Perception Feedback: Through real-time monitoring by detection-type and tracking-type intelligent agents, the system acquires the latest data and information on pollutants in rivers and lakes. This data is used as perception feedback input to the chaotic decision-making core.

[0047] (2) Chaotic parameter adjustment: The chaotic decision-making core adjusts the parameters of the chaotic model based on the perception feedback information. By changing the initial conditions and control parameters of the chaotic model, the dynamic weight parameters generated by the chaotic mapping are optimized to improve the accuracy and adaptability of the decision.

[0048] (3) Knowledge base update: Successful tracing experiences and cases are stored in the knowledge base, and the rules and models in the knowledge base are updated and improved. The update of the knowledge base can provide reference and guidance for subsequent tracing tasks and improve the overall performance of the system.

[0049] This invention also provides an implementation method for parameter adjustment using chaotic decision-making, as detailed in (A) to (B) below: (A) A Lorenz chaotic system is used to generate dynamic decision parameters for the source tracing path. The state decision parameters x, y, and z of the Lorenz system are extracted from the state variables to provide additional information for path planning. These parameters affect the "effective distance" between nodes. For example, if x represents a high level of pollution diffusion activity, the weights between nodes are adjusted. x represents the activity of pollution diffusion. The standardized value of x is used as the weight parameter for the direction of pollution diffusion. If the x value is large and trending upwards, it indicates that pollution is spreading rapidly, and path planning should prioritize moving in the opposite direction of diffusion (the source tracing direction). y is used to measure the complexity of source tracing. The value of y is mapped to the tortuosity weight of the path. For example, when the y value is large, it indicates high source tracing difficulty. For the weights of all adjacent edges, a value related to y can be uniformly added to reflect the impact of source tracing difficulty on path selection. z is used to measure the degree of environmental impact. The environment is divided into different levels based on the z value. If the z value is in an environmental level unfavorable to source tracing (such as heavy rainfall, strong winds, etc.), the weights of adjacent edges are increased to a larger value. value( It is a constant set according to the severity of the environment; if it is a favorable environment, it can be reduced to a smaller value.

[0050] In one implementation, the differential equations of the Lorenz system are solved using a numerical solution method (fourth-order Runge-Kutta method) to obtain the time-varying sequences of x(t), y(t), and z(t).

[0051]

[0052]

[0053]

[0054] in, It reflects a certain inherent characteristic of pollutant diffusion in river and lake water bodies. It is related to the intensity of the pollution source or the combined effects of environmental factors. It is related to the water body's self-purification capacity or other stabilizing factors.

[0055] (B) For specific details on the source tracing path decision, please refer to (1) to (4) below: (1) Initialization: Set the distance from the starting node s to itself to d(s) = 0, and the distance from other nodes v to the starting node to d(v) = ∞. Create a set of unvisited nodes U, which initially contains all nodes.

[0056] (2) Node selection and distance update: In each iteration, select the node u with the smallest distance from the starting node d(u) from the set of unvisited nodes U. For each neighboring node v of u, calculate the distance d from the starting node to v via u.new (v) = d(u) + w(u,v), where w(u,v) is the adjusted edge weight. If d new If (v) < d(v), then update d(v) = d new (v), and set u as the predecessor node of v. Mark u as visited and remove it from U.

[0057] The calculation method for W(u,v): 1. Adjustment of x-related weights: When the x value is large and the edge... When k is in the same direction as the diffusion, x1 =1+f(x), where f(x) is an increasing function; when the value of x is large and the edge (u,v) is opposite to the diffusion direction, k x2 =1-g(x), where g(x) is an increasing function; when x is small, k x1 = k x2 =1, based on the relationship between (u,v) and the diffusion direction, choose k. x1 or k x2 Then, the initial weights are multiplied by this coefficient to obtain: W x-adjusted (u,v)=K x(u,v) W 0(u,v) 2. Adjustment of y-related weights: Calculate the increment of the weights of the edges opposite to y. , It is an increasing function, and then the weights are updated. .

[0058] 3. z-related weight adjustment: Determine environmental adjustments based on the z-value. .

[0059] If z is in an unfavorable environment level (J is a large positive, for example, j=1).

[0060] If z is in a favorable environment for registration (k is a small positive, for example, k=0.5).

[0061] Finally, the adjusted edge weights are obtained. .

[0062] (3) Termination condition: The algorithm terminates when the set of unvisited nodes U is empty or the endpoint node t is found.

[0063] (4) Path generation: Starting from the endpoint node t, based on the predecessor node information, backtrack step by step to the starting node s to obtain the shortest path from s to t. Use this path as the local tracing direction of the current agent, and transmit the decision result (such as "continue tracking to the point 3km to the northwest") to the cooperative network.

[0064] Regarding the multi-agent collaborative pollutant tracing method provided in the foregoing embodiments, this invention provides a multi-agent collaborative pollutant tracing device. This device is applied to a multi-agent collaborative pollutant tracing system. See [link to documentation]. Figure 3 The diagram shows a multi-agent collaborative pollutant tracing device, which includes the following components: The information acquisition module 302 acquires river and lake water information and the initial weight values ​​of the river and lake area. The river and lake water information includes water quality information, flow information and meteorological information. The chaotic decision module 304 performs chaotic mapping processing on the target river and lake water information and initial weight values ​​to obtain the target dynamic edge weights, and uses a preset shortest path algorithm to perform path optimization processing on the target dynamic edge weights to obtain the decision results. The decision results include: path information of the target source tracing area to be traced for pollutants. The pollutant source tracing module 306 performs intelligent tracking and monitoring of pollutants in the target source tracing area based on the decision results, and obtains the source tracing results of the target pollutants.

[0065] The multi-agent collaborative pollutant tracing device provided in this application embodiment can significantly improve the accuracy and dynamic adaptability of pollutant tracing.

[0066] In one embodiment, when performing the steps of acquiring river and lake water information and the initial weight values ​​of the river and lake area, the information acquisition module 302 is further configured to: after acquiring the river and lake water information, divide the river and lake area into a set of nodes, and acquire the initial weight values ​​corresponding to the edges between each node in the set of nodes, wherein each node is a fixed monitoring point, the starting point and the ending point of the detection agent.

[0067] In one embodiment, when performing chaotic mapping processing on the target river and lake water information and initial weight values ​​to obtain the target dynamic edge weights, the aforementioned chaotic decision module 304 is further configured to: perform chaotic mapping processing on the target river and lake water information to obtain a time-varying sequence of state decision parameters, wherein the state decision parameters include: the activity of pollutant diffusion, the complexity of source tracing, and the degree of environmental impact; map the time-varying sequence of state decision parameters to the dynamic weights of each side of the river channel to obtain the adjusted target dynamic edge weights.

[0068] In one embodiment, when performing the step of intelligently tracking and monitoring pollutants in the target source area based on the decision result to obtain the source tracing result of the target pollutant, the pollutant source tracing module 306 is further configured to: execute the source tracing task based on the decision result, perform intelligent tracking and monitoring of pollutants in the target source area, and obtain the local source tracing results reported by each intelligent agent; and fuse the local source tracing results through a preset distributed Bayesian inference model to obtain the source tracing result of the target pollutant.

[0069] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0070] This invention provides a server, specifically, the server includes a processor and a storage device; the storage device stores a computer program, which, when run by the processor, executes the method described in any of the above embodiments.

[0071] Figure 4 This is a schematic diagram of the structure of a server provided in an embodiment of the present invention. The server 100 includes: a processor 40, a memory 41, a bus 42 and a communication interface 43. The processor 40, the communication interface 43 and the memory 41 are connected through the bus 42. The processor 40 is used to execute executable modules, such as computer programs, stored in the memory 41.

[0072] The memory 41 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 43 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0073] Bus 42 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0074] The memory 41 is used to store programs. After receiving an execution instruction, the processor 40 executes the program. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 40 or implemented by the processor 40.

[0075] Processor 40 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 40 or by instructions in software form. Processor 40 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 41. The processor 40 reads the information in memory 41 and, in conjunction with its hardware, completes the steps of the above method.

[0076] The computer program product of the readable storage medium provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the foregoing method embodiments. For specific implementation, please refer to the foregoing method embodiments, which will not be repeated here.

[0077] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0078] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A multi-agent collaborative pollutant source tracing system, characterized in that, The system comprises: a multi-source heterogeneous data perception layer, a multi-agent collaborative network layer, a chaotic decision-making module, and a collaborative tracing execution layer; wherein... The multi-source heterogeneous data sensing layer is used to collect river and lake water information in real time through a set of sensors, wherein the river and lake water information includes: water quality information, flow information and meteorological information; The multi-agent collaborative network layer is used to issue task instructions based on the river and lake water information and task requirements, so as to perform intelligent scheduling of each agent and obtain the target river and lake water information collected by each agent. The chaotic decision module is used to perform chaotic mapping processing and path optimization processing on the target river and lake water information to obtain decision results, wherein the decision results include: path information of the target source tracing area to be traced for pollutants; The collaborative source tracing execution layer is used to execute source tracing tasks based on the decision results of the chaotic decision module, perform intelligent tracking and monitoring of pollutants in the target source tracing area, and obtain the source tracing results of the target pollutants.

2. The multi-agent collaborative pollutant source tracing system according to claim 1, characterized in that, The multi-agent cooperative network layer includes: a probing agent, an analytical agent, and a tracking agent; wherein... The detection-type intelligent agent is used to send the collected river and lake water information to the analysis-type intelligent agent; The analytical agent is used to generate task instructions based on task requirements and the river and lake water information, and send the task instructions to the tracking agent; The tracking agent is used to trace the source of pollution targets according to the task instructions.

3. The multi-agent collaborative pollutant source tracing system according to claim 1, characterized in that, The system further includes a feedback and adaptive optimization layer; wherein... The feedback and adaptive optimization layer is used to verify the effectiveness of the current source tracing path based on the river and lake water information obtained in real time by the multi-source heterogeneous data perception layer, and to adjust the control parameters of the chaotic decision module according to the verification results, so as to enhance the environmental change adaptability of the chaotic decision module.

4. A multi-agent collaborative pollutant source tracing method, characterized in that, The method is applied to a multi-agent collaborative pollutant source tracing system, and the method includes: Obtain information on river and lake water bodies and initial weight values ​​for river and lake areas, wherein the information on river and lake water bodies includes: water quality information, flow information and meteorological information; Chaotic mapping processing is performed on the river and lake water information and the initial weight values ​​to obtain the target dynamic edge weights. Then, a preset shortest path algorithm is used to optimize the path of the target dynamic edge weights to obtain the decision results. The decision results include the path information of the target source tracing area to be traced for pollutants. Based on the decision results, intelligent tracking and monitoring of pollutants are carried out in the target source area to obtain the source tracing results of the target pollutants.

5. The multi-agent collaborative pollutant source tracing method according to claim 4, characterized in that, The steps for obtaining river and lake water information and initial weight values ​​for river and lake areas include: After obtaining information about river and lake water bodies, the river and lake area is divided into a set of nodes, and the initial weight values ​​of the edges between each node in the set are obtained. Each node is a fixed monitoring point, the starting point of the detection agent, and the ending point.

6. The multi-agent collaborative pollutant source tracing method according to claim 4, characterized in that, The step of performing chaotic mapping processing on the target river / lake water body information and the initial weight values ​​to obtain the target dynamic edge weights includes: Chaotic mapping processing is performed on the target river and lake water information to obtain a time sequence of state decision parameters, wherein the state decision parameters include: the activity of pollutant diffusion, the complexity of source tracing, and the degree of environmental impact. The sequence of changes in the state decision parameters over time is mapped to the dynamic weights of each side of the river channel, resulting in the adjusted target dynamic edge weights.

7. The multi-agent collaborative pollutant source tracing method according to claim 4, characterized in that, The step of performing intelligent pollutant tracking and monitoring on the target source area based on the decision result to obtain the target pollutant source tracing result includes: Based on the decision results, the source tracing task is executed, and intelligent tracking and monitoring of pollutants are carried out in the target source tracing area to obtain the local source tracing results reported by each intelligent agent. By using a pre-defined distributed Bayesian inference model, the local source tracing results are fused together to obtain the source tracing results of the target pollutant.

8. A multi-agent collaborative pollutant tracing device, characterized in that, The device is applied to a multi-agent collaborative pollutant tracing system, and the device includes: The information acquisition module acquires river and lake water information and initial weight values ​​for river and lake areas. The river and lake water information includes water quality information, flow information, and meteorological information. The chaotic decision-making module performs chaotic mapping processing on the river and lake water information and the initial weight values ​​to obtain the target dynamic edge weights, and uses a preset shortest path algorithm to perform path optimization processing on the target dynamic edge weights to obtain the decision results. The decision results include: path information of the target source tracing area to be traced for pollutants. The pollutant tracing module performs intelligent tracking and monitoring of pollutants in the target tracing area based on the decision results, and obtains the target pollutant tracing results.

9. A server, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 4 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method described in any one of claims 4 to 7.

Images