Cross-departmental decision-making visualization method and related device for water resource management
By constructing encrypted causal subgraphs and performing vertical federated learning fusion, a global causal structure graph is generated and visualized, solving the problem of difficulty in characterizing causal dependencies in cross-departmental water resource governance and improving the reliability and efficiency of decision-making.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WUHAN UNIV OF TECH
- Filing Date
- 2026-01-26
- Publication Date
- 2026-05-29
AI Technical Summary
Existing water resource governance decision-making methods are mostly based on statistical correlation or single-sector perspectives, which makes it difficult to depict the causal dependencies between cross-sectoral data. This results in insufficient reliability and interpretability of decision conclusions when faced with changes in scenarios or policy interventions, and causal conflicts are difficult to resolve. Existing decision support systems lack the visualization of causal relationships, leading to low decision-making efficiency.
By acquiring the business requirements of the target water area, a local causal subgraph is constructed and encrypted. A vertical federated learning mechanism is used for secure alignment and fusion to generate a global causal structure graph. The decision results are then displayed in a visual manner, meeting privacy protection and compliance requirements.
It improves the reliability, interpretability, and generation efficiency of cross-departmental decision-making, enhances the transparency of the decision-making process, significantly shortens decision-making time, and improves the efficiency of consensus formation.
Smart Images
Figure CN122114356A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of water resource management and intelligent decision support technology, and in particular to a cross-departmental decision visualization method and related equipment for water resource management. Background Technology
[0002] Among related technologies, current water resource management methods mainly suffer from the following problems:
[0003] Existing water resource governance decision-making methods are mostly based on statistical correlation or single-sector perspectives, which makes it difficult to characterize the potential causal dependencies between cross-sectoral data, resulting in insufficient reliability and interpretability of decision conclusions when faced with changes in scenarios or policy interventions.
[0004] In water resource governance scenarios involving multiple departments, raw data cannot be centrally shared due to reasons such as privacy protection and institutional constraints. This results in insufficient credibility and low adoptability of the output results of various departments in the water resource governance process during cross-departmental consultation and decision-making.
[0005] Existing federated learning methods are mainly geared towards prediction tasks and lack explicit mechanisms for handling causal consistency and causal conflicts. When different departments arrive at different causal conclusions based on local data, it is difficult to form an interpretable and negotiable unified understanding, which may lead to repeated arguments in cross-departmental meetings and prolong the time to reach a consensus.
[0006] Existing water resource decision support systems mostly present results as static indicators or charts, lacking visualization of causal relationships and their evolution. This makes it difficult to support communication, consultation, and consensus-building among multiple stakeholders, resulting in low efficiency in water resource governance decision-making and increased decision generation time.
[0007] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention
[0008] The main objective of this application is to propose a cross-departmental decision visualization method and related equipment for water resource governance. This method can provide quantifiable reconciliation evidence for conflicting conclusions while meeting privacy protection and compliance requirements, thereby improving the reliability, interpretability, and generation efficiency of decisions and enhancing the transparency of the decision-making process.
[0009] To achieve the above objectives, one aspect of this application proposes a cross-departmental decision visualization method for water resource governance, the method comprising the following steps: Obtain the water resource management business needs of the target water area; Identify the participating departments associated with the aforementioned water resource management business needs; Obtain local water resource management data from the participating departments, including cross-departmental water resource management related data and water resource management entity information; Under the condition that the local water resource management data of each participating department does not leave the domain, a local causal subgraph corresponding to each participating department is constructed based on the local water resource management data of each participating department, and the local causal subgraph is encrypted. The encrypted local causal subgraph carries the variable data, local adjacency matrix and conditional independence statistics of the local water resource management data of the corresponding department. Secure alignment of all encrypted local causal subgraphs is performed based on a vertical federated learning mechanism. Constraint-based or model-based fusion operations are performed on the aligned local causal subgraphs to obtain a global causal structure graph for cross-departmental water resource governance, which carries weights and confidence levels. Based on the global causal structure diagram, causal reasoning and decision analysis are performed on the water resource governance process to generate corresponding water resource governance reasoning results. The results of the water resource governance reasoning are presented in a visual manner.
[0010] In some embodiments, the cross-departmental water resources governance-related data includes water quality monitoring data, hydrological and water quantity data, pollution source information, meteorological and environmental data, and water use behavior data; the water resources governance subject information includes the department identification information and historical instruction and decision record information of the participating departments.
[0011] In some embodiments, constructing a local causal subgraph corresponding to each participating department based on the local water resource management data of each participating department includes: The local water resource management data of each participating department are registered and mapped to obtain data variables for each participating department; Based on a preset constraint-based algorithm or a scoring-based algorithm, the causal relationships between the data variables of each participating department are learned to obtain a local causal subgraph corresponding to each participating department.
[0012] In some embodiments, the data variables include name information of local water resource management data, unit information of the participating departments, or monitoring frequency data.
[0013] In some embodiments, the secure alignment of all encrypted local causal subgraphs based on a vertical federated learning mechanism includes: Cross-departmental entity alignment is performed on the variable data of the encrypted local causal subgraph so that the summary information in the encrypted local causal subgraph points to the same observation object; A consistency comparison is performed on the conditional independence statistics in the encrypted local causal subgraph to identify common constraints among the participating departments in the causal inference process. Conflict detection is performed on the local adjacency matrix in the encrypted local causal subgraph to identify contradictory causal edges.
[0014] In some embodiments, the constraint-based or model-based fusion operation based on the aligned local causal subgraph includes: In the constraint-based fusion process, the conditional independence statistics of each participating department in the local causal subgraph are aggregated in the dense state, and the global causal direction is determined based on the preset orientation rules. In the model-based fusion process, a global causal adjacency matrix is obtained by jointly training the local adjacency matrices of the local causal subgraphs corresponding to each participating department through a preset secure aggregation algorithm.
[0015] In some embodiments, the causal reasoning and decision analysis of the water resource management process based on the global causal structure diagram includes: Based on the aforementioned global causal structure diagram, a causal path analysis of the water resource management process is conducted. Based on the aforementioned global causal structure diagram, counterfactual reasoning is performed on the water resource governance process; Based on the aforementioned global causal structure diagram, intervention scenario simulations are performed for the water resource management process.
[0016] In some embodiments, the visualization of the water resource management reasoning results includes: The global cause-effect structure diagram corresponding to the water resource governance decision-making scheme is displayed as a guide diagram or a hierarchical diagram through a visualization engine; Use Sankey diagrams, waterfall diagrams, or geographic information overlay views to display the causal contribution and spatial distribution in the global causal structure diagram; The results of the scenario simulation are presented through time-series graphs and dynamic comparison charts before and after the intervention.
[0017] To achieve the above objectives, another aspect of this application proposes a cross-departmental decision visualization device for water resource management, the device comprising: The first module is used to obtain the water resource management business requirements of the target water area; The second module is used to identify the participating departments associated with the water resource management business needs; The third module is used to obtain local water resource management data of the participating departments, including cross-departmental water resource management related data and water resource management entity information; The fourth module is used to construct a local causal subgraph for each participating department based on the local water resource management data of each participating department, under the condition that the local water resource management data of each participating department does not leave the domain, and to encrypt the local causal subgraph. The encrypted local causal subgraph carries the variable data, local adjacency matrix and conditional independence statistics of the local water resource management data of the corresponding department. The fifth module is used to securely align all the encrypted local causal subgraphs based on a vertical federated learning mechanism. The sixth module is used to perform constraint-based or model-based fusion operations based on the aligned local causal subgraphs to obtain a global causal structure graph for cross-departmental water resource governance. The global causal structure graph carries weights and confidence levels. The seventh module is used to perform causal reasoning and decision analysis on the water resource governance process based on the global causal structure diagram, and generate corresponding water resource governance reasoning results. The eighth module is used to visualize the reasoning results of water resource governance.
[0018] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.
[0019] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.
[0020] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.
[0021] The embodiments of this application include at least the following beneficial effects: This application provides a cross-departmental decision visualization method and related equipment for water resource governance. This solution, after obtaining the water resource governance business needs of the target water area, determines the participating departments associated with these needs. Then, it obtains the local water resource governance data of each participating department. Under the condition that the local water resource governance data of each participating department does not leave the domain, it constructs a local causal subgraph corresponding to each participating department based on the local water resource governance data of each participating department. The local causal subgraph is then encrypted, so that the encrypted local causal subgraph only carries the variable data, local adjacency matrix, and conditional independence statistics of the corresponding department's local water resource governance data, without carrying the original data and business needs. The system records data and then securely aligns all encrypted local causal subgraphs using a vertical federated learning mechanism. Based on these aligned subgraphs, constraint-based or model-based fusion operations are performed to obtain a global causal structure graph for cross-departmental water resource governance. This allows for the depiction of causal dependencies between cross-departmental data while meeting privacy and compliance requirements, providing quantifiable reconciliation evidence for conflicting conclusions. Furthermore, based on this global causal structure graph, causal reasoning and decision analysis are performed on the water resource governance process to generate corresponding water resource governance reasoning results. These results are then visualized, effectively improving the reliability, interpretability, and generation efficiency of decisions, and enhancing the transparency of the decision-making process. Attached Figure Description
[0022] Figure 1 This is a flowchart of a cross-departmental decision visualization method for water resource management provided in an embodiment of this application; Figure 2 This is a schematic diagram of the application architecture of the cross-departmental decision visualization method for water resource governance provided in the embodiments of this application. Detailed Implementation
[0023] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.
[0024] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”
[0025] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.
[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0027] Among related technologies, current water resource management methods mainly suffer from the following problems: Existing water resource management decision-making methods are mostly based on statistical correlation or single-sector perspectives, making it difficult to characterize the potential causal dependencies between cross-sectoral data. This results in insufficient reliability and interpretability of decision conclusions when faced with changing scenarios or policy interventions. For example, when analyzing the causes of sudden changes in water quality at a certain cross-section, conclusions drawn by a single sector based on its own data (such as pollution discharge monitoring by the environmental protection department or flow data by the water resources department) often contradict phenomena observed by other departments, leading to a decrease in the credibility of the overall decision.
[0028] In water resource governance scenarios involving multiple departments, raw data cannot be centrally shared due to privacy protection and institutional constraints. This results in insufficient credibility and low adoptability of the outputs from various departments during water resource governance processes in cross-departmental consultations and decision-making. For example, detailed pollution discharge records from environmental protection departments, water use records from water resources departments, and refined rainfall data from meteorological departments are all restricted from being provided to external parties by regulations, rendering the usability of existing centralized causal analysis systems almost zero in cross-departmental scenarios.
[0029] Existing federated learning methods primarily focus on prediction tasks and lack explicit mechanisms for handling causal consistency and conflict. When different departments arrive at differing causal conclusions based on local data, it becomes difficult to reach an interpretable and negotiable consensus. This can lead to interdepartmental meetings becoming bogged down in repeated debates, prolonging the time required to reach a consensus. For example, the environmental protection department might conclude that "corporate pollution is the main cause of water quality deterioration," while the water resources department might argue that "insufficient flow leads to pollutant accumulation." Conflicting conclusions from both sides cause interdepartmental meetings to become bogged down in repeated debates, prolonging the time required to reach a consensus.
[0030] Existing water resource decision support systems primarily present results as static indicators or charts, lacking visualization of causal relationships and their evolution. This makes it difficult to support communication, consultation, and consensus-building among multiple stakeholders, resulting in low efficiency in water resource governance decision-making discussions and increased decision generation time. For example, in water transfer scheme demonstration meetings, decision-makers often face a large number of numerical reports and trend curves, making it difficult to quickly understand the transmission chain between "upstream water intake—midstream ecological flow—downstream water quality," leading to low efficiency in scheme discussions and increased time for identifying key decision points.
[0031] In view of this, this application provides a cross-departmental decision visualization method and related equipment for water resource governance, which can provide quantifiable reconciliation evidence for conflict conclusions while meeting privacy protection and compliance requirements, thereby improving the reliability, interpretability and generation efficiency of decisions and enhancing the transparency of the decision-making process.
[0032] The cross-departmental decision visualization method for water resource governance provided in this application relates to the field of water resource governance and intelligent decision support technology. This cross-departmental decision visualization method for water resource governance can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the cross-departmental decision visualization method for water resource governance, but is not limited to the above forms.
[0033] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0034] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.
[0035] The embodiments of this application will be described in detail below with reference to the accompanying drawings: Figure 1 This is an optional flowchart of a cross-departmental decision visualization method for water resource management provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S110 to S180: Step S110: Obtain the water resource management business requirements of the target water area; Step S120: Identify the participating departments related to the business needs of water resource management; Step S130: Obtain local water resource management data from participating departments, including cross-departmental water resource management related data and water resource management entity information; Step S140: Under the condition that the local water resource management data of each participating department does not leave the domain, construct a local causal subgraph corresponding to each participating department based on the local water resource management data of each participating department, and encrypt the local causal subgraph. The encrypted local causal subgraph carries the variable data, local adjacency matrix and conditional independence statistics of the local water resource management data of the corresponding department. Step S150: Securely align all the encrypted local causal subgraphs based on the vertical federated learning mechanism; Step S160: Perform constraint-based or model-based fusion operations based on the aligned local causal subgraphs to obtain a global causal structure graph for cross-departmental water resource governance, wherein the global causal structure graph carries weights and confidence levels. Step S170: Based on the global causal structure diagram, perform causal reasoning and decision analysis on the water resource governance process, and generate corresponding water resource governance reasoning results; Step S180: Display the reasoning results of water resource governance in a visual manner.
[0036] Understandably, when identifying the target water area requiring water resource management, the next step is to determine the corresponding current water resource management needs, and then analyze and determine the participating departments involved in meeting those needs. Specifically, for example... Figure 2 As shown, current water resource management needs may involve participating departments such as environmental protection, water resources, meteorology, agriculture, and industry. The data obtained by each participating department may be completely identical, partially identical, or completely different. Specifically, local water resource management data includes cross-departmental water resource management related data and information on water resource management entities. Cross-departmental water resource management related data includes water quality monitoring data, hydrological and water quantity data, pollution source information, meteorological environmental data, or water use behavior data; water resource management entity information includes the departmental identification information and historical instruction and decision-making records of the participating departments. Among these, water quality monitoring data includes key indicator data such as pH value, dissolved oxygen, ammonia nitrogen, and chemical oxygen demand. Hydrological and water quantity data includes data such as flow rate, water level, water intake, and water storage. Pollution source information includes information such as discharge volume, pollutant concentration, and discharge time. Meteorological environmental data includes meteorological element data such as rainfall, temperature, humidity, and wind speed.
[0037] It is understandable that each participating department, after collecting local water resource management data corresponding to the target water area, can store this data in its local database. In this embodiment, after determining the current water resource management business needs, this embodiment can, without the local water resource management data of each participating department remaining within its local network domain, register and map the local water resource management data of each participating department to obtain data variables for each participating department. Then, based on a preset constraint-based algorithm or a scoring-based algorithm, it learns the causal relationships between the data variables of each participating department to obtain a local causal subgraph corresponding to each participating department. Specifically, this embodiment can complete the standardized mapping of variables based on semantic rules and a business dictionary, thereby obtaining a mapped subset of variables corresponding to each participating department, such as unifying the variable "ammonia nitrogen concentration". Then, a causal discovery algorithm is executed locally in each participating department to generate a local causal subgraph corresponding to each participating department. For example, this embodiment can employ constraint-based algorithms such as PC (Peter–Clark Algorithm, a constraint-based causal discovery algorithm) and FCI (Fast Causal Inference), or scoring-based algorithms such as NOTEARS (a continuous optimization-based causal structure learning algorithm), to learn the causal relationships between variables from the internal data of participating departments, and then generate a local causal subgraph with confidence scores as the local causal subgraph. The data variables include, but are not limited to, the name information of local water resource management data, the unit information of participating departments, or monitoring frequency data.
[0038] It is understandable that, in this embodiment, when performing causal structure learning locally within the participating departments, the significance level of the conditional independence test can be set to [value missing] when using a constraint-based algorithm (such as the PC algorithm). When using a scoring algorithm (such as NOTEARS), the regularization strength parameter in the model selection criteria... The adjustable range can be [0.01, 0.1] to balance graph sparsity and goodness of fit. In the calculation of causal edge weights in the local causal subgraph, this embodiment can be based on the standardized effect size of the local structural equation model or regression coefficients, and the threshold can be set to... Edges with strong effects are retained, while weak-effect edges below this threshold are filtered out during the local summary generation stage.
[0039] It is understood that, after obtaining the local causal subgraph carrying variable data, local adjacency matrix, and conditional independence statistics of the corresponding department's local water resource management data, this embodiment, in order to protect the data privacy of each participating department, performs secure multi-party computation, homomorphic encryption, or differential privacy processing on the local causal subgraph in the local network domain of each participating department, thereby forming a securely transmittable "causal structure summary". For example, the environmental protection department can generate an encrypted representation of the causal edge "enterprise discharge volume → cross-sectional ammonia nitrogen concentration" and its test p-value as a local causal structure summary (i.e., the encrypted local causal subgraph) and upload it.
[0040] Specifically, when using differential privacy-preserving causal structure summarization, the Laplacian noise scaling parameter is set to... ,satisfy Differential privacy requirements, among which When using Paillier homomorphic encryption, the key length can be set to 2048 bits to ensure the security of transmission and computation.
[0041] It is understood that after encrypting the local causal subgraphs, this embodiment performs secure alignment on all encrypted local causal subgraphs based on a vertical federated learning mechanism. Specifically, this example can perform cross-departmental entity alignment on the variable data of the encrypted local causal subgraphs to ensure that the summary information in the encrypted local causal subgraphs points to the same observation object; perform consistency comparison on the conditional independence statistics in the encrypted local causal subgraphs to identify common constraints among participating departments in the causal inference process; and perform conflict detection on the local adjacency matrix in the encrypted local causal subgraphs to preliminarily identify contradictory causal edges. All operations in this embodiment are performed in a secure or privacy-preserving environment and do not involve the decryption process of the original data.
[0042] This embodiment employs a vertical federated learning framework, combined with secure multi-party computation, homomorphic encryption, or differential privacy techniques. This allows each department to complete data cleaning and feature extraction locally, generating structured causal information which is then encrypted before transmission. This embodiment uses a secure sample ID alignment mechanism to determine the common observations corresponding to variables and performs causal fusion computation within an encrypted domain. This ensures that the original data never leaves the local databases of the participating departments, meeting privacy protection and compliance requirements.
[0043] It is understood that, after obtaining the aligned local causal subgraphs, this embodiment can perform constraint-based or model-based fusion operations on the aligned local causal subgraphs to obtain a global causal structure graph for cross-departmental water resource governance. Specifically, this embodiment can use a constraint-based fusion operation to fuse the encrypted local causal subgraphs sent by each participating department, or it can use a model-based fusion operation to fuse the encrypted local causal subgraphs sent by each participating department. Alternatively, it can simultaneously perform fusion operations using both constraint-based and model-based methods, and then fuse the resulting fusion graphs again to obtain the final global causal structure graph.
[0044] Specifically, in the constraint-based fusion process, the conditional independence statistics of the local causal subgraphs corresponding to each participating department are aggregated in a dense state, and the global causal direction is determined based on preset orientation rules; wherein, the preset orientation rules include, but are not limited to, collision rules, propagation rules, etc. This embodiment uses these orientation rules to gradually determine the global causal direction to improve accuracy. In the model-based fusion process, the local adjacency matrices of the local causal subgraphs corresponding to each participating department are jointly trained through a preset safe aggregation algorithm to learn a "graph fusion model" that can reflect cross-departmental consensus, and a global causal adjacency matrix. The preset safe aggregation algorithm includes, but is not limited to, federated averaging algorithms, safe multi-party computation algorithms, etc. The processing of this embodiment can introduce a confidence weighting mechanism to assign higher weights to participating departments with high confidence, and automatically arbitrate or trigger multi-departmental collaborative review based on statistical significance for detected causal conflicts, finally outputting a global causal structure graph with weights and confidence labels. It is understood that in this embodiment, during the model-based fusion process, the number of iteration rounds for federated averaging or secure aggregation is set to T=50-100 rounds, with no fewer than K=3 participating departments in each round, to ensure fusion stability.
[0045] It is understandable that, in the process of merging cross-departmental causal subgraphs in this embodiment, the confidence level of each participating department's local causal subgraph is considered. It can be determined by its sample size Determined together with the model fit index, the calculation formula is as follows: ,in This represents the goodness of fit of the local causal model. When different participating departments have inconsistent judgments about the direction of the same causal edge, the difference in their confidence levels is considered. If the result is of high confidence, then adopt the result with high confidence; otherwise, trigger manual review or process based on majority voting mechanism.
[0046] As described above, this embodiment employs two fusion strategies: constraint-based and model-based, and introduces confidence assessment and consistency fusion mechanisms. In constraint-based fusion, the conditional independence test results from various departments are aggregated, and unified directional rules are executed in the encrypted state. In model-based fusion, the system learns a "graph fusion model" through secure aggregation technology to find common structures among the local causal graphs. Simultaneously, this embodiment can output a list of node importance and a causal contribution analysis report, providing quantifiable reconciliation evidence for cross-departmental causal conflict conclusions.
[0047] Specifically, the method of this embodiment can be extended to cross-departmental data processing. The number of participating departments and the total number of global variables can be set to no more than [number]. This ensures both the interpretability and computational efficiency of the global causal structure graph.
[0048] It is understood that this embodiment, after obtaining a cross-departmental global causal structure diagram, performs causal reasoning and decision analysis on the water resource governance process. Specifically, this embodiment can perform causal path analysis of the water resource governance process based on the global causal structure diagram; perform counterfactual reasoning of the water resource governance process based on the global causal structure diagram; and simulate intervention scenarios for the water resource governance process based on the global causal structure diagram. For example, this embodiment, by embedding a structural causal model and causal effect calculation engine into the corresponding software system, allows users to set various governance intervention scenarios (such as "upstream water intake reduced by 20%", "discharge standard upgraded by one level", "rainfall reduced by 30%)", and can automatically deduce their causal effects on key target variables (such as "downstream water quality category", "ecological flow satisfaction rate", "water supply guarantee rate"). Simultaneously, this embodiment can identify and quantify the contribution of key causal paths, support parallel simulation and comparative analysis of multiple scenarios, and output causal effect estimates, path decomposition reports, and risk level assessments in a structured form, providing mechanistic explanations and quantitative basis for decision-making. In this embodiment, when performing counterfactual reasoning or policy scenario simulation, the number of samples in the Monte Carlo simulation can be set to M=1000, thereby stabilizing the effect estimation results. In the causal path contribution analysis, the truncation error in the cumulative calculation of path effects is controlled within... Within.
[0049] It is understood that after obtaining the water resource governance inference results, this embodiment can display them visually. Specifically, this embodiment can use a visualization engine to display the global causal structure diagram corresponding to the water resource governance decision scheme as a guide diagram or hierarchical diagram; display the causal contribution and spatial distribution in the global causal structure diagram through Sankey diagrams, waterfall diagrams, or geographic information overlay views; and display the scenario simulation results through time-series curves and dynamic comparison diagrams before and after intervention. All visualization views in this embodiment can be subject to access control and can also be filtered by departmental perspective. The visualization display process in this embodiment can be updated at a preset update frequency, which can be set to no less than f=10 frames / second, and supports real-time path highlighting and node focusing operations to ensure a smooth user experience.
[0050] It is understood that after visualization, this embodiment can also output interactive causal decision graphs, structured causal analysis reports (PDF), and lists of key indicator contributions. It can also connect to the standardized data services of OA systems or command and dispatch platforms through RESTful APIs, providing full-link decision support for cross-departmental consultations, scheme demonstrations, and policy evaluations. This can reduce communication costs and enhance the transparency and explainability of water resource governance decisions.
[0051] As can be seen from the above, the method of this application embodiment has the following beneficial effects: First, this embodiment can use local data from multiple departments as input, and achieve collaborative construction and fusion of cross-departmental causal structures through a federated mechanism without sharing the original data. By introducing a semantically aligned global variable mapping mechanism and encrypted sample ID alignment technology, a unified variable space and association with observed entities are established, ensuring the feasibility and consistency of cross-departmental causal modeling. Compared with traditional centralized causal modeling, this embodiment can significantly reduce the dependence on the output of original data, providing a feasible path for data-driven decision-making in multi-departmental collaborative scenarios such as water resources, environmental protection, and meteorology.
[0052] Secondly, this embodiment addresses the potential conflict between causal conclusions from the same department by introducing a confidence-weighted fusion strategy and a statistical significance arbitration mechanism in the federated fusion stage. By combining constraint-based causal orientation rules and model-based graph aggregation technology, it can efficiently integrate and mitigate conflicts in multi-source causal structures while ensuring privacy, outputting a stable and interpretable global causal graph. Experiments show that in cross-departmental water quality causal analysis scenarios, the method in this embodiment can reduce the causal edge conflict rate by more than 40%, significantly improving the credibility and acceptability of causal inference results in cross-departmental decision-making.
[0053] Third, this embodiment presents the fused global causal structure diagram intuitively through a multi-layered, multi-view visualization interface, supporting interactive functions such as node filtering, path highlighting, contribution decomposition, and scenario simulation. By transforming complex causal relationships into understandable graphical representations, it significantly reduces the cognitive threshold in cross-departmental communication, helping decision-makers quickly identify key influencing factors and assess the effectiveness of policy interventions. In watershed governance consultation scenarios, the method in this embodiment can reduce the average cross-departmental solution discussion time by 35% and improve the efficiency of consensus-building by approximately 50%.
[0054] Fourth, this embodiment comprehensively utilizes privacy enhancement technologies such as homomorphic encryption, differential privacy, and secure multi-party computation throughout the entire process of data access, causal learning, structural fusion, and result output to ensure that the original data and sensitive information of each department never leave the local system. Simultaneously, through a structured and auditable causal output format, it not only meets data security and compliance requirements but also provides transparent and traceable decision-making basis for relevant decisions. Furthermore, the method of this embodiment has been deployed in a cross-provincial river basin management pilot project, supporting six departments, including environmental protection, water resources, and agriculture, to collaboratively conduct water quality causal analysis and water diversion scheme evaluation. Without disclosing any original data, it successfully constructed a global causal graph covering 42 key variables, providing a reliable mechanism explanation and decision support for comprehensive river basin management.
[0055] Fifth, this embodiment also supports modular deployment and parameterized configuration, allowing adjustments to the causal discovery algorithm, fusion strategy, privacy protection strength, and visualization format based on actual governance scenarios. Through standard API interfaces, the method in this embodiment can flexibly connect to existing monitoring platforms and decision support systems, demonstrating good engineering applicability and promotional value.
[0056] By comprehensively applying the methods of this application, the method has achieved significant technical effects in cross-departmental water resource management scenarios: the global causal graph construction time has been shortened from 5-7 days in traditional centralized methods to 2-3 days; the consistency rate of causal edges has increased from approximately 65% to over 89%; the average number of cross-departmental decision-making coordination meetings has been reduced by 40%; and it can support collaborative causal modeling of up to 15 departments and more than 200 variables. These effects fully demonstrate the effectiveness and advancement of this real-time method in improving the efficiency, scientific rigor, and collaboration of water resource management decision-making.
[0057] This application embodiment also provides a cross-departmental decision visualization device for water resource management, the device comprising: The first module is used to obtain the water resource management business requirements of the target water area; The second module is used to identify the participating departments associated with the water resource management business needs; The third module is used to obtain local water resource management data of the participating departments, including cross-departmental water resource management related data and water resource management entity information; The fourth module is used to construct a local causal subgraph corresponding to each participating department based on the local water resource management data of each participating department, under the condition that the local water resource management data of each participating department does not leave the domain, and to encrypt the local causal subgraph. The encrypted local causal subgraph carries the variable data, local adjacency matrix and conditional independence statistics of the local water resource management data of the corresponding department. The fifth module is used to securely align all the encrypted local causal subgraphs based on a vertical federated learning mechanism. The sixth module is used to perform constraint-based or model-based fusion operations based on the aligned local causal subgraphs to obtain a global causal structure graph for cross-departmental water resource governance. The global causal structure graph carries weights and confidence levels. The seventh module is used to perform causal reasoning and decision analysis on the water resource governance process based on the global causal structure diagram, and generate corresponding water resource governance reasoning results. The eighth module is used to visualize the reasoning results of water resource governance.
[0058] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0059] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.
[0060] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0061] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.
[0062] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.
[0063] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.
[0064] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0065] This application provides a cross-departmental decision visualization method and related equipment for water resource governance. The solution involves obtaining the water resource governance business needs of a target water area, identifying the participating departments associated with those needs, and then acquiring the local water resource governance data of each participating department. Under the condition that the local water resource governance data for each participating department does not leave its domain, a local causal subgraph is constructed based on the local water resource governance data of each participating department. This local causal subgraph is then encrypted, so that the encrypted local causal subgraph only carries the variable data, local adjacency matrix, and conditional independence statistics of the corresponding department's local water resource governance data, without carrying the original data or business record data. Next, based on the vertical federated learning mechanism, all encrypted local causal subgraphs are securely aligned. Then, constraint-based or model-based fusion operations are performed on the aligned local causal subgraphs to obtain a global causal structure graph for cross-departmental water resource governance. This allows for the characterization of causal dependencies between cross-departmental data while meeting privacy and compliance requirements, providing quantifiable reconciliation evidence for conflicting conclusions. Based on the global causal structure graph, causal reasoning and decision analysis are performed on the water resource governance process to generate corresponding water resource governance reasoning results. These results are then visualized, effectively improving the reliability, interpretability, and generation efficiency of decisions, and enhancing the transparency of the decision-making process.
[0066] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.
[0067] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.
[0068] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0069] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0070] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0071] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0072] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0073] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0074] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0075] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.
Claims
1. A cross-departmental decision visualization method for water resource management, characterized in that, The method includes the following steps: Obtain the water resource management business needs of the target water area; Identify the participating departments associated with the aforementioned water resource management business needs; Obtain local water resource management data from the participating departments, including cross-departmental water resource management related data and water resource management entity information; Under the condition that the local water resource management data of each participating department does not leave the domain, a local causal subgraph corresponding to each participating department is constructed based on the local water resource management data of each participating department, and the local causal subgraph is encrypted. The encrypted local causal subgraph carries the variable data, local adjacency matrix and conditional independence statistics of the local water resource management data of the corresponding department. Secure alignment of all encrypted local causal subgraphs is performed based on a vertical federated learning mechanism. Constraint-based or model-based fusion operations are performed on the aligned local causal subgraphs to obtain a global causal structure graph for cross-departmental water resource governance, which carries weights and confidence levels. Based on the global causal structure diagram, causal reasoning and decision analysis are performed on the water resource governance process to generate corresponding water resource governance reasoning results. The results of the water resource governance reasoning are presented in a visual manner.
2. The method according to claim 1, characterized in that, The cross-departmental water resources governance-related data includes water quality monitoring data, hydrological and water quantity data, pollution source information, meteorological and environmental data, and water use behavior data; the water resources governance entity information includes the departmental identification information and historical instruction and decision-making records of the participating departments.
3. The method according to claim 1, characterized in that, The construction of a local causal subgraph corresponding to each participating department based on the local water resource management data of each participating department includes: The local water resource management data of each participating department are registered and mapped to obtain data variables for each participating department; Based on a preset constraint-based algorithm or a scoring-based algorithm, the causal relationships between the data variables of each participating department are learned to obtain a local causal subgraph corresponding to each participating department.
4. The method according to claim 3, characterized in that, The data variables include the name information of local water resource management data, the unit information of the participating departments, or the monitoring frequency data.
5. The method according to claim 1, characterized in that, The secure alignment of all encrypted local causal subgraphs based on the vertical federated learning mechanism includes: Cross-departmental entity alignment is performed on the variable data of the encrypted local causal subgraph so that the summary information in the encrypted local causal subgraph points to the same observation object; A consistency comparison is performed on the conditional independence statistics in the encrypted local causal subgraph to identify common constraints among the participating departments in the causal inference process. Conflict detection is performed on the local adjacency matrix in the encrypted local causal subgraph to identify contradictory causal edges.
6. The method according to claim 1, characterized in that, The constraint-based or model-based fusion operation based on the aligned local causal subgraph includes: In the constraint-based fusion process, the conditional independence statistics of each participating department in the local causal subgraph are aggregated in the dense state, and the global causal direction is determined based on the preset orientation rules. In the model-based fusion process, a global causal adjacency matrix is obtained by jointly training the local adjacency matrices of the local causal subgraphs corresponding to each participating department through a preset secure aggregation algorithm.
7. The method according to claim 1, characterized in that, The causal reasoning and decision analysis of the water resource management process based on the global causal structure diagram includes: Based on the aforementioned global causal structure diagram, a causal path analysis of the water resource management process is performed. Based on the aforementioned global causal structure diagram, counterfactual reasoning is performed on the water resource governance process; Based on the aforementioned global causal structure diagram, intervention scenario simulations are performed for the water resource management process.
8. The method according to claim 7, characterized in that, The visualization of the water resource management reasoning results includes: The global cause-effect structure diagram corresponding to the water resource governance decision-making scheme is displayed as a guide diagram or a hierarchical diagram through a visualization engine; The causal contribution and spatial distribution in the global causal structure diagram can be displayed using Sankey diagrams, waterfall diagrams, or geographic information overlay views. The results of the scenario simulation are presented through time-series graphs and dynamic comparison charts before and after the intervention.
9. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method as described in any one of claims 1 to 8.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.